-1CANADA DEPARTMENT OF MINES AND TECHNICAL SURVEYS MINES BRANCH OTTAWA

Fuels Division Technical Memorandum

83/ 58-RBS

A NEW METHOD O F HYDROCARBON STRUCTURAL GROUP ANALYSIS

by D. S. Montgomery and M. L. Boyd

This p a p e r is published with the p e r m i s s i o n of the Director, Mines Branch, Department of Mines and Technical Surveys, Ottawa, Canada.

P a p e r P r e p a r e d for P r e s e n t a t i o n Before the Division of Gas and F u e l Chemistry. American Chemical Society Chicago, Illinois Meeting, September 7 12, 1558.

h
I

(Copy No.

June 1, 1958.

-2

A NEW METHOD O F HYDROCARBON STRUCTURAL GROUP ANALYSIS

D.S.

by Montgomery and M. L. Boyd

F u e l s Division, Mines Branch, Department of Mines and Technical Surveys.

ABSTRACT

A m e t J d o hydrocarbon s t r u c t u r a l group analysis has been developed f o r application t o p u r e compounds in which t h r e e chemical and two physical p r o p e r ties have been e x p r e s s e d in t e r m s of five s t r u c t u r a l groups in a f o r m which m a y b e simultaneously solved by modern high speed computing equipment. The chemical properties i n c l t ~ r l - thn rarho-n and hydrogen content as well a s the numb e r of a r o m a t i c c a r b o n atoms p r e s e n t p e r molecule. The physical p r o p e r t i e s required f o r the analysis are the molar volume and m o l a r refraction. This method has been tested on a selected g r o u p of 114 hydrocarbons whose properties have been determined by A.P. I. P r o j e c t 42. The results of;the application of this s t r u c t u r a l analysis s y s t e m a r e described in detail a n d the accuracies attained have been tabulated.

-3

PRESENTED BEFORE THE DIVISION O F GAS 4ND F U E L CHEMISTRY AMERICAN CHEMICAL SOCIETY 'Chicago, Illinois, Meeting, September 7-12, 1958.

D. S. Montgomery and M. L. Boyd Department of Mines and Technical Surveys, Ottawa, Ontario.

INTRODUCTION

The synthesis and systematic study of the physical p r o p e r t i e s of p u r e hydrocarbons has been in p r o g r e s s f o r a number of , y e a r s to facilitate the determination o f the molecular s t r u c t u r e of hydrocarbons directly o r by analogy, that is by comparing certain p r o p e r t i e s of compounds of unknown s t r u c t u r e with those of compounds whose s t r u c t u r e s are kaown. The value of studying physical p r o p e r t i e s of a series of Compounds, as a means of predicting the p r o p e r t i e s of unknown compounds and affording a means of checking the a c c u r a c y of the physical constants of compounds, has a l s o been repeatedly demonstrated. In general, the study of the physical prope r t i e s of hydrocarbons has been undertaken by a number of independent investigators and usually has been confined to hydrocarbons containing a limited numb e r of types of s t r u c t u r a l groups, o r , alternatively, only o n e physical property of a l a r g e variety of compounds has been examined. This paper d e s c r i b e s a method of simultaneously analyzing c e r t a i n physi c a l and chemical properties of liquid hydrocarbons to s e c u r e s t r u c t u r a l information. Although the system described is confined to specific c l a s s e s of hydrocarbons, i t is m o r e general than any system s o far proposed and involves the simultaneous consideration of t h r e e chemical and two physical p r o p e r t i e s to yield quantitative information concerning five s t r u c t u r a l groups. The investigation was promoted by a d e s i r e to improve existing s t r u c t u r a l analysis systems f o r pure hydrocarbons and to facilitate the study of naturally o c c u r r i n g hydrocarbons. The new possibilities offered by modern high speed computing equipment provided a n n the field of s t r u c additional incentive to re-examine and extend the e a r l i e r work i t u r a l analysis. The most extensively used s t r u c t u r a l analysis s y s t e m applied to hydrocarbon mixtures has been developed by Waterman and his school (13) beginning with the c l a s s i c a l Waterman Ring Analysis of 1 9 3 2 (19, 20) and extending to the n-d-M method of 1947 (14). van Krevelen employed s o m e of the concepts of the Waterman Ring Analysis to develop a system that w a s p a r t i c u l a r l y suited to the study of highly condensed a r o m a t i c s t r u c t u r e s which were assumed to b e the major constituents of coal (7, 8). The chief c r i t i c i s m of the v a r i o u s methods of struc1

!>

ii d

I . ,

b-

t u r d group a n a l y s i s (15) of oil and coal is that when applied to p u r e compounds poor r e s u l t s a r e obtained. These considerations made i t d e s i r a b l e to attempt to f o r m u l a t e a s t r u c t u r a l a n a l y s i s system based on the physical properties of known compounds, but of such a f o r m that i t could b e used w i t h reasonable confidence to ana-lyze t h e s t r u c t u r e of high molecular weight m a t e r i a l .
The method p r e s e n t e d h e r e w a s evolved f r o m a method of carbon type a n a l y s i s published b y van Krevelen (4) in 1952. In this method, van Krevelen divided the carbon a t o m s in a s t r u c t u r e into four main types:-

CH

CH

- CHarornatic; c4 = C, aromatic;
C

f w h e r e C in the denominator of each fraction represented the total number o c a r b o n atoms p e r molecule. He then s e t up the following four quantitative relations hips:
(a) C1 t C2

C3 t C 4 = 1 (carbon balance)

(b) 2 C 1 t C 2 t C3 (c)

= H/C

(hydrogen balance) (ring balance) ( a r o m a t i c carbon balance)

c2

c*

= 2R/C

(d) C3 t C4 = fa

T h e s e equations w e r e not independent, and hence could not be solved simultaneously. van K r e v e l e n solved these equations by giving an equation for C1 as a function of H/ C which represented the statistical probability of the oc: c u r r e n c e of a CHZ group in the molecule. The u s e of this equation w a s open to considerable question and consequently the method was never widely applied. Equations (a), (b) and (d), above, a r e t r u e by definition; however, equation (c), t h e r i n g balance equation, is only valid f o r high molecular weight hydrocarbons w h e r e the factor 2/ C can b e neglected and where two junctions a r e associated with the formation of e v e r y ring. T h e r e a r e s t r u c t u r e s where this relation is not valid, such as in s p i r o compounds, and in three-dimensional s t r u c t u r e s where t h r e e rings p o s s e s s a common side. van Krevelen's system was devised to elucid a t e the s t r u c t u r e of coal and coal-like products where the proportion of saturated c a r b o n atoms was s m a l l o r negligible. While this choice of carbon types was suita b l e f o r the study of coal, i t was undesirable f o r the study of petroleum since it failed to differentiate between chain and cyclic CH2 groups. Hence, a five-type c a r b o n classification w a s chosen which differed from that of van Krevelen by dividing his C1 into two t y p e s . The carbon linkages have therefore been divided into the following five types.

c1 =

number p e r molecule of CH3, CH2, CH, and C groups in linear and branched chains.

- 5C2 =
number p e r molecule 0 . f CHZ groups in saturated rings, including the c a s e where the hydrogen atoms m a y be replaced by branched o r l i n e a r chains.

c3

= - number p e r molecule of CH groups which are junctions

in fused saturated rings, as well a s s i m i l a r l y situated s replaced by l i n e a r o r branched groups where the hydrogen i chains.

c4 c5

n u m e r p e r molecule of CH groups in aromatic rings, including the c a s e where the hydrogen m a y be replaced by branched o r l i n e a r chains. number p e r molecule C groups which a r e junctions in fused aromatic rings, a s well a s junctions between saturated and aromatic rings.

It is possible to r e w r i t e the carbon balance, the hydrogen balance and the a r o m a t i c carbon balance equations in t e r m s of this new classification of s t r u c t u r a l groups. I t should be especially noted that in the van Krevelen system the molecular weight was unknown and hence the s t r u c t u r e was described in t e r m s of fractions of the total number of carbon atoms. However, the p r e s e n t system was designed for the c a s e where the molecular weight ( a s well a s 70 carbon, 70 hydro gen, density, refractive index, and aromatic carbon content) was either 'known o r could be determined. The carbon classification w a s therefore expressed in t e r m s of the actual numbers of the different carbon types C1 C g p r e s e n t in the molecule. The Ring Balance Equation could not be used when analyzing an unknown hydrocarboa, as t h e r e was no a c c u r a t e method of estimating the number of rings in the molecule. The fundamental b a s i s of this s t r u c t u r a l analysis system, therefore, consis te:! in finding two additional physical properties which could be accurately expressed in t e r m s of the above s t r u c t u r a l groups to give five independent equations which could then be solved simultaneously.

Theoretically, any two physical p r o p e r t i e s would b e suitable, provide& i t w e r e possible to express therm a s independent equations in t e r m s of the s t r u c t u r a l types already defined. The two physical properties chosen w e r e the m o l a r volume and the m o l a r refraction (the Lorentz-Lorenz expression). Both of these physical properties of liquids can be approximately described in t e r m s of a linear combination of the atomic contributions, and both quantities have been extensively used f o r the purposes of elucidating s t r u c t u r e ( 7 , 18, 5, 10). It w a s assumed that the s a m e functional form of the equation would apply to both physical properties, and that both properties could be expressed in t e r m s of the s a m e groups of chemlcal types. Owing to the intimate relation between these two quantities, it was felt that th loss in accuracy associated with grouping together so many different chemical linkags might easily yield two expressions which w e r e merely linear combinations of each other. It remained f o r this investigation to demonstrate that the differences between the expression derived for the m o l a r volume and tha't f o r the m o l a r refraction w e r e of such a magnitude that reliable s t r u c t u r a l information could be secured by the simultaneous solution of these expressions.

6-

DEVELOPMENT OF THE FIVE EQUATIONS

It should be emphasized a t the outset that this method a s presently constituted is applicable only to certain specific c l a s s e s of compounds. In the division of the carbon groups into t h e v a r i o u s types, c a r e w a s taken that the hydrogen balance equation should always be satisfied. It w a s recognized that this equation was not s t r i c t l y t r u e a s defined f o r the n o r m a l and branched paraffins. This c a s e w i l l subsequently be discussed. The p r e s e n t s y s t e m w i l l d e a l with monocyclic compounds and fused ring compounds, both saturated aromatic, but w i l l not include polycyclic non-fused structures. Spiro compounds and three-dimensional ring systems are a t o excluded. It can b e shown that the contributions to the m o l a r volume and m o l a r refraction of junction atoms i n pilycyclic non-fused compounds a r e not identical with any of the five types dealt with here, but in fact, r e p r e s e n t a 6th (saturated) and a 7th (aromatic) type of linkage. Another c l a s s of compounds excluded a r e those containing olefinic o r acetylenic bonds. Xx% this work only double bonds existing in arom a t i c rings have been considered. The method depends upon the ability to m e a s u r e the density and refractt one atmosphere p r e s ive index of hydrocarbons in the liquid s t a t e a t 20C. a s u r e , o r on the capacity to c o r r e c t to this standard state m e a s u r e m e n t s nade under other conditions. Except where otherwise specified, the coefficients of the m o l a r volume and m o l a r refraction equations have been determined from the properties of the hydrocarbons prepared by A. P.I. P r o j e c t 42 (16). The Carbon, Hydrogen and Aromatic Carbon Balance Equations The carbon balance, hydrogen balance, and a r o m a t i c carbon balance can b e written a s follows, by definition:

c1 t c2 t c3

c 4 + c5

= C

(1)

The total number of carbon atoms p e r molecule EC was calculated from the-carbon analysis and the molecular weight. The total number of hydrogen a t o m s p e r molecule GH, was s i m i l a r l y calculated from the hydrogen analysis and the molecular weight. The total number of aromatic carbon atoms in the molecule, 6Ca, w a s not quite so readily available, although i t could be determined by spectroscopic means (1). F o r the purposes of the present investigation (testing the method on known compounds), direct m e a s u r e m e n t of this quantitiy was not required.

The Molar Volume Equation T h e following molar volume equation was developed for the purpose of this s t r u c t u r a l analysis system:

M . V . = C1(16. 38 30.61) at 7 20" c. C3(10. 981 20.679) 1 atm. press. Z C + C 5 (5.124 - 5 . 2 3 8 )

C2(13.20

28.48) .GC

+
<c

(4)

+ C4(12. 406 + 14. 042 - 1. 96 5

TC

t 10.1 3C2)

A detailed account of this equation is in preparation. However to clarify the b a s i s of this-method of s t r u c t u r a l analysis, a brief description of the m a n n e r in which this equation w a s developed w i l l now be given.

Initially, on the b a s i s of Kurtz and Sankin's (9) work. the following g e n e r a l form of the m o l a r volume equation w a s assumed:

M.V,

= VICl

2 2

\r3C3 f

v*c4 f

.v5c5 + K

A preliminary investigaticn to evaluate :he coefficients in is expression indicated s e v e r a l difficulties with this functional form. N o s y s t e m of constant coefficients v to v could be found f o r the five s t r u c t u r a l types covered by the 1. 5 p r e s e n t classificatton. This work also revealed that k w a s not constant.

T o overcome t h e s e difficulties, the followkg approach was adopted. F r o m in Equation 1, the following expression was derived by multhe definition of tiplication by k:

a,

When K in equation 5 w a s replaced by the value in equation 6, the following equation was obtained:

u s e was made of the fact that T o determine the coefficients of C to C 1 5' f o r a molecule containing only one species of carbon atom (Ci), Equation 7, a s sumed the following form, as ci was equal toZC:

M.V.

= viCi

Ki

(8)

When the m o l a r volume w a s plotted against C. the resultant s t r a i g h t line had a slope of vi and a n intercept. Df k . . Consequedtly, vi represented the contribution to the m o l a r volume of a Ci gro& in the p r e s e n c e of a n infinite number of Ci groups.

- 0 It w i l l subsequently b e shown that the K's in each b r a c k e t a r e not identical, these have hence been designated kl to k5 respectively. The quantity K in
the above equation e x p r e s s e s the i n c r e a s e in the m o l a r volume that takes place on reducing the m o l e c u l a r weight of a p a r t i c u l a r species. Determination of v1 and kl The m o l a r volume for the 13 n o r m a l paraffins on the l i s t of A. P. I. P r o j e c t 4 2 (16) was plotted against the number of carbon atoms i n the molecule. T h e equation of the resulting s t r a i g h t line was determined by the method of l e a s t s q u a r e s and found to be as follows:

M.V. = 16. 38C1 t 30.61

Determination of v 2 and k 2

(9)

These quantities w e r e calculated in a manner similar to that used for the calculation of v1 and kl, using the physical p r o p e r t i e s given by Ward and Kurtz (21) for cyclopentane, cyclohexane, cycloheptane and cyclooctane. T h e molar volume of t h e s e compounds could b e expressed by the following equation:

M.V. = 13.,20C2 t 28.48

Determination of v4 and k4

(10)

Some difficulties w e r e encountered in obtaining suitable d a t a f o r compounds containing only C4's. T h e following compounds w e r e used: benzene (properties taken from Egloff (3), cyclooctatetraene ( p r o p e r t i e s by Eccleston (2) et al), and cyclopentadiene (Ward and Kurtz (21). A word of explanation about the use of the cyclopentadiene is required. Since it contained one C , group its u s e h e r e is,

found to be:

A slightly different method had to be used to determine v3 and kj, since no compounds existed containing C groups only. Compounds containing C 2 and c 3 groups bad to b e used, and f o r this reason t h e following procedure was

I
4
I

adopted.

9-

was plotted against 3 1&C. The r e s u l t of this plot w a s - a straighfyine whose slope was k3 and intercept.v3. Using the following compounds from the A . P. I. L i s t (16): bicyclooctane 543, decahydronaphthalene 569 and 570, perhydrofluorene 561, perhydropyrene 578 and'perhydrochrysene 575, v3 w a s determined to be 10.981 and k3 to be 20.679. The coefficient of C was undoubtedly a function of the ring size, but the values found.for five- an% six-membered rings w e r e a l m o s t identical.

[(M. V. observed

-C

(13.20 t 2 8 . 4 8 y /

Determination of v5 and k

T h e s e quantities w e r e determined in exactly the s a m e manner a s that used for the determination of v3 and k3. Some difficulties w e r e experienced in obtaining suitable liquid s t a t e m o l a r volume data at 20C f o r fused ring a r o m a t i c compounds that contained only C4 and C5. Liquid m o l a r volume data at 20C. have been determined by Ubbelohde (11, 12) f o r s e v e r a l fused ring a r o m a t i c compounds. This author a l s o determined the m o l a r volume f o r the solid s t a t e a t 20C. M. V. s ) v e r s u s C5/ Cg, it w a s possible to e x p r e s s this shrinkBy plotting (M. V. age i n volume as a function of the d e g r e e of condensation. By means of this r e l a tionship, the solid s t a t e m o l a r volume data of van Krevelen.(l7) for fused r i n g a r o m a t i c compounds was "converted" to the liquid s t a t e a t 20C. Using the data obtained i n this manner for anthracene, chrysene, phenanthrene (Ubbelohde), and for dibenzanthracene, chrysene, pyrene and coronene (van Krevelen's data converted to the liquid state) v5 w a s determined to be 5.124 and k5 to be -5. 238.

The molar volume equation thus obtained w a s used to calculate the m o l a r volume f o r the appropriate c l a s s e s of compounds on the A . P . I. L i s t (16) and to compare the results with the experimental values. The r e s u l t s of this comparison showed that a study of the interactions between various types of s t r u c t u r a l groups had to be made. The m o s t significant interactions w e r e found when C1 and C4 o c c u r r e d in the s a m e molecule and when C2 and c4 o c c u r r e d together. In the f o r m e r c a s e , data f r o m Ward and Kurtz (21) w e r e used to evaluate the ma-mitude of the interaction, and in the l a t t e r c a s e A . P. I. data (16) w e r e employed. In brief, the method of determining the functional form and magnitude of the interaction t e r m s consisted of obtaining the difference between observed and calculated m o l a r volume p e r C4 group and plotting this difference against C / LC in the f i r s t c a s e and C /&C 2 in the second. Two straight lines w e r e obtained, f?om which w e r e derived the two t e r m s in Cl /&C and C 2 / c C which w e r e added to the C t e r m in Equation 4. 9 The accuracy with which this I ' corrected" equation predicted the m o l a r volume of the A. P. I. (16) hydrocarbons is given in Table I.

-10

-. .

Table I - Accuracy of the Authors' Molar Volume Equation Applied to A. P. I. 42-Hydrocarbons Class of:rCompound No. in Class Average - Value of MVcalc - MVobs x 100
MVobs.

~-

Standard Deviation

n-paraffins br. -paraffins Monocyclic saturates Fused ring saturates Monocyclic aromatics Fused ring aromatics

13 38 27 18 21 13

-0.02

'

to. 31

0.07 0.80 0.33 0.80 0.19 0.89

to.

10

to. 66 to. 03

.46

The Molar Refraction Equation The followine m o l a r refraction equation was developed for the purpose of this analysis:

20'C. 1 atm. press.

C3(3. 693 t 0.3395)

Ec,

C5(5. 734'- 14. 333) C C

. . .' . . -. . . . . . . . . . . . .

(12)

In this equation, M. R. r e f e r r e d to the Lorentz-Lorenz expression

f o r the m o l a r r e f r a c t i o n where n w a s the refractive index for the +2 d sodium D line at 20C. Tge coefficients in this equation were determined in exactly the s a m e m a n n e r , and using the s a m e compounds as in the molar volume equation. Attention is drawn to the fact that, in the determination of the C5 term, the m o l a r refraction data for the fused ring a r o m a t i c compounds of van Krevelen (17) r e f e r to m e a s u r e m e n t s made of the compounds in benzene solution and have 1 been r e f e r r e d t o by v a n Krevelen (6) a s ' I hypothetical liquid state data". The accuracy with which this equation predicts the m o l a r refraction of the A. P. I. 42 ' (16) hydrocarbons is indicated in Table 11.

e",-'. 9,

- 11METHODS O F SOLUTION O F THE SET O F F I V E E,2UATIGNS Equations 1, 2, 3, 4 and i?? constitute the three chemical and two physical property equations that, when solved simultaneoasly f o r C1,C2, C 3 , C4 and C5, form the proposed system of analysis. The f i r s t three of these equations a r e linear b u t equations 4 and 12 a r e quadratic in C l and.C This 2 s e t of linear and quadratic equations w a s initially solved ia the followmg manner. T h e solution was obtained by reducing the system to a l i n e a r form by substituting a n initial value C1 = Cz = 0 in the non-linear t e r m s . The resultant s e t of _ . :mear equations was then solved by the standard methods of m a t r i x algebra. Table I1 - Accuracy of the Authors Molar Refraction Equation, Applied to A . P. I. 42 Hydrocarbons

class of Compound

No.& Class

Average Value of MPLc,lc MRObS x 100 MRobs.

Standard Deviation

n-paraffins ST. -paraffins blonocyelic saturates Fused ring saturates Mono cyclic aromatics Fused ring aromatics

9
37

t0.012 +@. 1 9 -0.01 -0.008 $0.16

0,056 0.26 0.20 0.26

26 19 19
8

0. 51
0. 64

to. 1 7

n the n o n T h e new values of C and C from this solution w e r e theii substirated i I linear t e r m s and another sofurion was obtained. This iterative procedure was repeated until two consecutive solutions were equal. The critical aspect of the solution was whether or not the iterative procedure would converge. F r o m the practical point of view the r a t e of convergence was important. The number of iterations varied f r o m three for the paraffins to thirty for s o m e of the fused ring aromatic compounds. Since the- m e l a r volume and the m o l a r refraction equations w e r e quadratic in C1 and Cz, in general t h e r e would be four roots. Since i t w a s c l e a r that the iterative procedure yielded only one root, i t was desirable to obtain a method of solution which would give a l l the roots. F o r this reason the s e t of five equations was solved by a secondmethod. It w a s possible, by simple algebraic re-arrangement of the three linear equations, to express C g , C4 and C in t e r m s of C1 and C These values f o r C C and c5 w e r e then substituted 5 2 3 4

-12

- in the molar volume and m o l a r refraction equations and the following two quadratic equations in C1 and C2 resulted:
1. 96 C;' -

t (0.077 EC
(

1. 9 6 C a

fC

8. 17C2

1.962H

9.3493
)

&z

c1 +

(13)

3.699EC

1.425 <Ca t 7.282%

6.637

sa t 19.280533 -t
EC
5C

1.399 - M . V . )

= 0
and

2 c1

t (2. 516% t 5. 852C2

(

- 0. 396zCa

0.396EH

11. 338)C1 (14)

6C

t 0. 852<Ca

12.973

1.360% -a EC M.R.) = 0

- 1.190EH t

13.312EH

EC

t 4.883c

T h e s e two quadratics w e r e solved graphically by taking a r b i t r a r y values of C2 and s,olving e a c h equation for C1. These values of C 1 w e r e plotted against C2 and the i n t e r s e c t i o n of the molar volume and m o l a r refraction curves represented the roots of the system.

RESULTS
T h e i t e r a t i v e method of solution has been tested by application to 121 known compounds, of which 114 a r e from the l i s t of properties of A. P.I. P r o j e c t 4 2 (16) and s e v e n a r e fused ring aromatic compounds whose properties w e r e determined by van Krevelen (17). All calculations w e r e made by the International Business Machines Company on a Type 650 Magnetic Drum P r o c e s sing Machine. The r e s u l t s a r e shown in Table ILL. In this table, D to D5 r e f e r to the differences between calculated and observed values of Cl to (i.e. [Ci(calc. C i ( o b s e r v e d g ).

d5

Some explanation is required concerning the application of this method to the paraffins. F o r this c l a s s of compounds, Equation 2, the hydrogen balance, is not s t r i c t l y t r u e as given but should be:

2C1 t 2C t C t C x ( H - 2
2.

(15 )

-13-

TABLE III.
..
MCEPl.Yor

orovs I

-rornms

Q8

u
13
14

1 2 . 0 7
13.U 13.99 15.W 15.99 i7.m 11.91
19.59

4.07
4 0 4
4 . O l 4 0 4

0
0 0

4.094

4.094
4.070

a
0

0.425
0.40

4.425
9.030

0
0
0 0 0 0

0.025

9.W
9 . 0 3 0 9.026

4.025 4.030
4.026

Q9
S3l

4.40 -0.016

0 . 4 0 0.026

0
0

S.030 -0.026 4.015 4.016

4.016 4.030 9.024

0 0

0.026

4.426

m
534

15

0 0
0

4.030
0 . m 4

4.45
4.016

4.ol5
4.016

4.015
4.016

4.015 +.E16

0
1

0.015 0.016

16
17 18 20 26

4.01
0 . 0

n5
27
542

d.m&
0.10 0,031
0.U

0 -Q.CO&

0 . m 9 4.m9
4.012

c . m 9 4.ccl9
4.012
4.OU

0 0
0

4.w9 4 . m 9

4 . 0 9
4-01 4-10

4.10 4.031 4.12

0
0

4.012
4.021

0 0 0

0.012 0.ml
0 . 0 2 0

9 . 0 1 )

0 0

4.m1
-0.020

4.m1
4.020

4.021
4.020

4.021

106
Gx-cup II
1

25.90

4,mo

4.m~

brancbd p m m n . 26
34

25.68 33.91 25.40 36.22 22.90

34.19

4.32
4-03

0 0
0

0.22 0.039 0.69 4.13

4.22 Q.039 4.69 4.13

0 0
0

0.099 4.0098
-0.088

4.099 4.W8

0 0
0

0.099

4.099

0 0

4-W9
0.m98

.OX99

107
109

-0.m98 -0.0098 4.19 4.089 -0.092

*.om
9 . 0 6 7
9.092

26 36

-0.60
4 . 2 2

-0.068

0
0

0.089 0.092

u3
163 164

0
0
0

0 0
0

4.092

4-w
-0.40
Q.14

0
0
0

4.w
4.40 0.1L -0.361 0.067 0.10 0 . 0 5 3
0.45

23
34 28

0 . 1 0
4.19
Q.34

a.n
0 . 1 2
4.41 4.0M

9 . n
4.12
4-47 0 . 0 4 4

4.0030 -0.0030
-0.40

4 . 0 0 3 0 4.0030

0
0

o.m>o @.mi0
0 . 0 7 0
Q.TO
4.lb

4.40
.O.ll

la
191
2

2 8 . 3 0
32.10 25.18

0
0 0

0.14 -0.061
0 . 0 6 7
0.10

0
0

0
0 0

4-14 0.161

32
26 26 26
25

rO.13
-0.22

0 0 0
0

-0.061

-0.061 r0.067 4.10

4 . 1 5 )

0.161
-0.061

0.16
0.63 1.33 0.42

4.16 Q.63 11-33 4.42

9.061
9.13

0 0 0 0
0

-0.W
4.13
-0.09

22 23 25 27

2 5 . 2 ~ -0.73

0 0 0
0

0
0 0 0
0

4.10

24.62 24.85 25.62 25.60

25.46

-1.38

0.053

4.053

4-112

-0.15

0.45
0.080

4-45
4.080 4.43
Q.058

9.4s
4.090 4.43
*.a98

4.45
4.180

-0.45
4.080

26 26
26

+.la
4.20
0 . 5 4
-00.28

0-097 4.091
0.U
0 . 4 4

0 0
0

o.*o
a.m
0.098
0.077

3
4

0
0

9.U
e . 4 4

0.073 0.098
0.077

0 0 0 0 0
0 0
0

4.113

4.173

0 0 0

-0.~8
0 . 1 7 7
+.mi6

28 19 26 20 20 14 26 16 13 10 26

27-72 18.68

0 0
0

0 . 2 1 0.31
0 . w

4.21 40.31 4.097 ~ . 6 2 4.21 9 . 0 1 2

0 0
0

4.47 4.0016

4.117

-0.wa Irm
+.mi6

gco
51
5lO

-0.32

0.W14
0 . 0 8 0

0.~016 4.0016 0.080

0 . 2 4 0

2522 4 - 1 8
19.34

4.m
4.M

4.080

0 0 0 0
0

-0.080
4 . 0 1 0
0.02~

4 . 0 8 0

4.66
40.25 4-03 -1.09
-0.32
4.34

0
0

0 . 6 2
0.27
O.OU

0 0
0

0 . 0 4 0

4 . m
4.021

19.75

4.027

4.W
4.018

4.027

4.w *.mi
-0.ai8
4068

N
5)

u.97
24.91 15.68 12.66 9.66 25-17 20.ffi 10.79

0
0

o.o1e
0 . 0 6 8
0.018

0.013

4.015
4 . 0 4

* . a 8
4.068
4.013

1.02 0.30
0.34

.1.m
4.30
40.34

0 0
0
0

9 . 0 6 8
4.018
m 3 0

0 . 0 6 8
0.018

95
9 6 549

0 0 0
0

e.oia
4 . 0 9 4.49 9.011
4.20

4.ola
4.m030
4.09

QmoP 4
0 . 0 5 2
0.079
4.011

QDmP 4 1 w O 3 0
0 0
0

0
0

o o o o p
w.09

4.34

0.29

4.29 4.75

4.09 4.49

O.OQ

55
554
556

4-83
4.14

0.75
0.15
).LO
0.14

0
0

0.079
4.0l.l

0
0
0

-0.119
0.011
0.20

4.49

21
14 18

0 0 0 0 0
0

4.15
r).W

-om1
4.20 4.023

m.ou
4.20

-331

0
0

4 . 2 0

0
0

9.20

557
6

1139 4 - T l
29-70

4.14
4.25

4.m3
0 . 0 4 2

4.023
0 . w

4.023
4 . 0 . 2

0
0

0.023

4.021
4 . 0 . 2 4.096

30
28
26
jz

4-30 9-21

0.25
0.12

0
0

4.w
4.096

0
0

4.w
4&96

63
6 )

27.19 25.94 p.14 30.S

0.12

0.096
0.Y

0.096
0 . u

4.096 4.12

0
0
0 0

6-06
9.26

4.055 0.18
0.09

4.055

0 0
0

4 . 1 2

0
0 0

41y
4.077
4.052

4.12 9.41
9.012

0 0

4.18

0.41
0 . 0 5 1

*on
4 . 0 s

0.m
0.09

. o n
*.on

9.00

.Lo32

-1b-

c r t .

539

8 8

.an us
1.68
9.93

6 6

5.n
6.36

4.19
W.36

0
0 0 0

-0.W -3.039

4&?9
-0.039

4.oE)

4 . -

0 0 0 0 0 0 0 0 0 0

0.029

%e?
90-039

S M

0 . 3 2
-0.q

4 0 3 9 -0&9 0.~039 4 . 0 ~ 3 9

0039 4.m39
0.09

rm
9R
543

ID
. l d

l 0 . S

. s i

6
5

5 w . s.n
5.46

uxw
-0.~6
4.b6

0 . m ~4.0~39
-0.032
0 m .

o
0
0 0

4.0039
9.09

4.031

a 0 3 2 -0.032
0 . 0 n

ULO -0.41

0 0
0 0 0 0 0

4.ou
4.025

4.oll
-0.025

* O n
o.m5
0.038

4 . m

s a
99

l3
17

l 2 . 5 5 -0.45 l6.m 4.20.

i9.o Z -U T
19-U

6 6

64 4 . a
62b

d.B5

-0.m5
4.038
0.017

4 . m 5
30.038

a 2 b

0 . 0 3 8
0.016
O . w
0.063

4.038

-0.038
-.OX?

l u
e l
I l l

Lo
19

5
6

5-26
,5.%
5-05
,

-26

o.01~ 4.017
4.016
*.Me

o
0 0

-ao.oii
4.016

*.mi
4.016
4 . 0 4 4

4.16

0.016
4.04

4.016

n
P
20
20 20

z1.m
z 0 . a

0.0

a05
4.16

-0.w
4.063

0.018

br

Z U 8 4.82

5
6 6 6

5.76
5-49

4.063
-0.097

0 0

0.063 4.W
-0.055
0.0352

0
0

-0.0.06i 4 . 0 6 3
0.47
0.055

im
1oD

4 . 6 0
-0.60

0 . 2
4.29

0
0 0

-0.c9l
-0.055

-0.097 4.055

30.097

1 % 1 1 4.23

6 . 2 9
6.59

-0.055

0 0 0
0 0

4.055

7s 101 76

19.40 19.56 19.9

4 . 5 9

o.m=

*.me
4.025
4.41 4.028

o
0 0

4 . W 9 .

4.ws2
0.w

s.mp
9.044

20

-0.u
4.42

6
6
6 6' 6

6.40

4 . a
4.w
4.3b 4.37 * . L O

0
0 0

4*w -0.W
o.m5
0.031
O.rn8

-0.w 0 . 0 4 4
0.025
4.015

20

b r a
h3L

4.025

0 . 0 2 5

n
18 162 14 0

Lo Lo
20

19-63 -0.37 19.60

-0.40

o
0
0 0 0 0 0 0

o m 1 4.031

6.31
6.10

0
0

0.ms
4.035
-0.mO8

4.028 -0.035

D
0 0 0 0

19-91 4.06
21.1)

-0.035

-0.035

22

-021

5
6 6 5 6 6
6

5-27 6.35 5-70

ClO

437
4.35 -0.30 -00.90 4.23

0
0 0 0 0 0 0

4.W08-0.WO8

-0.om8
4-41
0 . 4 4
4.099

P
P
24

2 Q e
22.38
24.Q

4.38
ad8

0.03T

4.031

0.031
-0.070

91

61-41 -0.44 0.099

is0
69 136

4-81
9-2b

4.w
a.016 -0.099 -0.10

%ow
0.016

0
0 0 0

25

ZC76

6-21
6.01

0.016 -0.w9 4.10

*.016

4
33

2 9 . m
33.8

4 . m
4 . 0 1

a . 0 1
4.01

0
0

-01.099 0 . -

l38
98

6.w

4-10

-0.10

s.m
9.96

o
-0.01
4.14

0 . n
0.22

a.ll
4.22

+ . I 2
-0.27 -0.M
0.20

-0.12

5.88 5.73 5.96

6.20

4 . 0 . 1 2
-00.21

0
0 0 0

0.12

r0.K
'0.21

8
Lb

8 x 6 4.06

0
0

a
0 0 0 0 0
0

-0.27
-0.W

6
6

0.21

sn
5 a

0.081

4.081
-0.061

4 . 0 4
4.20 -00.28

0.M
4.20 0.28 0.59

4.M
4-21

ld
19

l . 8 6
19.64
20.28

0
6 6

4.061
0.64

w.20 -0.28

6 6 6 6, 6 6 6 6 6 6

4.36

a . 6 4
-0.69 ul.045
4.37

-0.28

5.72
hbl

0 0
0

4 . 2 8

io
93
1Ol

19
20 10

*la8
-0.08

4.69
0.w5
0.31

4.59
0.037
-0.016

-0.59
4.031

-0.59
4.01

9 . 5 9
4.037

19.92 19.n

o
0

6.M

4.031

4 . 2 9

-0.076

5.92 5.99
5.98

4 . 0 8
4.01

0
0

0.016
0.Wl

4.46
4JG93 4.021
40.061

19

3
20

1 9 . 6 6
19.R 19.65 19.18 19.m 20.31
2au

4.34

0-35,
0.30
0.u

4.35

0 0 0 0 0

-am93 -o.mg 6.025

-0.obl 0 . 3 2
0.6

1 4

@a28
4.3s
4w1

0 0
0

4 . 3 0
4.L1

-0.025
-0.061

-0.m
-0.06
4.32

0
0 0 0 0

0.025
0.061
-0.32

do
Q
BI
19

20
20
20

5.94
6.32

-0.w 4.w
4-25
465
4 . 9
4.30

4-32

4-12
0 . 6

4A
U 3 1

0 6

4.25
-0.65
-0.9
4.30

4.46
4.35 4.39

6 . 6 4 . 4 6
6-34

4.~6 4.35 4.39

4.53
0.59

Lo
20

0 0 0
0 0

0.35 0.39
0.51

6
6

4.3b 4.39
W.5l

-0.35
-0.39
4%

161
9

ml3
ap

o
0 0 0 0

he39

0 0
0

n
P

2o.n

4.R
4.99
-0.73

6
6

6 n

a m cs

na
2z.n

+
4.21

l4 eLb5
494

4-99
4.73 0.31

boo
5-27
631

-1.m 4.730 a31

9 9 9

n
25

U9b

b
6

0
0

0.73

9 . 7 3

-am

am a095

4-37

431

*3?

. 1 ,

...
c

rroup V

k e d

nmpahsna
4

5 2 1

2.93 U.19 17.29 ll.43

-1.4 4.19
40.71

lo
10 6
lo

lLII

.Lll
-00103

3.96 3.04 1.96 3.88 1.91 1 . a

4.04

4.031

4.031

0 . 0 1 1
0.16

4.031

u 5
118 141
544

12

9.97 6-75 1 0 . 6 9
835

Qib
-0.06

0 . l 6
4.935
4.12

-0.16 -0.035

4.16

. : ;

-.
Zj.

18

40.15 4.69
4.25
4.09 4.08 Q.91

2
4

0
0
0

0
0

0 . 0 3 5
0.U
0.081
0.12

4.45

P
11

4 . 5 7
4 . 0 1

-0.U

4 . u
4.u
-0.u
4.0s

. o . u

-. . . -.
.I

14.53 4.17 a 5 . 0 3

8 8
U
8 8

2
2

-0.09
4 . 1 2

-0.w -0.m
4.12
O . u

4.m
4.u
4.U
4.053,

.-z

175 166

u
8
21 11 15 16
8
0 0
0 0

8.09
u.08

0
0
0

. . ,

, .

0.m
20.08 15.21 ll.63 15.89 1.74 0.41
-0.80
4.44

4 . m
0 . 9 2

6 2 4
4

5 . 8 9
1.95 3.92

-0.U

0
0

0.U

62
111

8.91 9.e

-00.05
-0.06

4.on
-0.bl8

0 . 0 5 3
0.378

4 :
.f

-1.79

r 1 . e
Y.@

0 0 0 0
0
0

-0.018

4.48

193

1 . 1 1
4 . 1 1

8
1 ll

9 . 1 8
1.15 10.45

1 . 8 9
1.95 6.81

-0.u
-0.05
Q.81
-0.01

-0.u
4.W

-0.u
-0.046

0
0

0 . u

4.u
4.016 4-19 4.011 4.moll
-0.oz.5
-0.034
Q . W

>s
19b 95

4-U
-0.55
0 . 6 0

2 6

0.016

-0.26
4-41
-0.80 4-44 -0.40

4.19

-0.19

o
0

0.19
O.OU

U
10 10

U . 6 0
10.80

6
6 6
4

5 . 9 9
6-03 6.m
4.03 2.01

-0.ou -0.oll
aoowT4.m41
0.95

978
577

40.80

0 0
0
0

4.mn
4.025 4.031
*.ma

10.41 921

Q . 4 1
-47

40.02

0
0 0

*.a75 40.034
4.mbr

56l

0 . u )

9
b

4.01
4.01

0.034
0 . W 6 4

9 3
cP.aup V I

* . e d

0 . l U 0 . 3 4

6 . 1 4

3 . 3 4

&io=

561
n 4-

I
0

0.28

4.72

0
10
0

-0.085

-0.085
0 . 4 4

0 . m

1 0 . 8 1

8
2

6.m
l.VO

4.81
4.10
+la41

2
4

1.19
6-10

0.55 4 . 5 5
l0.W

559 16

I I
16 15
15
8

0 . 9 3
-0.4

9.8
0.47

2
0 0
0

1 . 9 3
1.41

4.10
+le41

+m 4.x
-LU
4-23

4-47

8
4
6

9-11
4-21

0 . 5 9

U.91

1
4

2-04

-+I6
-0.52

0 . 2 1
-0.073
2.24 2.12

0.21

4.23

2 2
2
b

1.11
2.4
4.24

168
174 165 119

1 5 . 6 0
14.03
1.67
U,U 2 0 . B

4 . 6 0
-0.91 -0.33 Y.U
4.W

3 . 4 8
1.28

4.on
9-21

3 . 9 3
10.2.
2.12

-0.43
U.2b

6 4 3
4.24

-1.28
4.21 4-18

10
4

10.21 le22 b.11

2
0

Q . U

2
8
6

4.n
4.31
4.24

3.u

0 . 2 2

4-33
-0.24

0 . 0 1
-0.24

Y.67 3.16

6
2

6 . 3 1
2-14

4.33
4.24

in

21

4.n

0 0 0

4.70

-2.10

la

9.30
9.w

4.10

8.10

r.m
4.10

4.095
-1.86 -2.16
4.6b
4.20

-0.095 -1.86

10
1 1

4.10

h 10
U.86

10.14 5 A

1 . 8 6
-2.16
-2.6.

U
b

.1e4
4 . 1 6

-2.16
4.64
-0.20

8
la

h16
14&
h20

0
0
0

9 . 3 6

K
6

a.6e

10
0

%ea
w 2

4 . 2 0

-%za
3.Y

L32

-1.32

* L 3 2

468

- 16'At the outset, the effect of t'nis comparatively s m a l l change was not known and the original hydrogen balance equation was used. The results on the n-paraffins a r e shown in Table IV, An examination of these results indicated that, although in e r r o r , they followed a v e r y recognizable pattern. Since r e s u l t s for all the n o r m a l and branched paraffins on t h s A. P. I. 42 (16) list gave this s a m e pattern, i t w a s foLqd possible to have the p r o g r a m for the Type 650 computer written in such a way that when this pattern occurred the revised hydrogen balance equation 15, would b e substituted in place of the original. Table IV Analysis of n-paraffins to Illustrate the Use of Unco r r e c ted Hydrogen Baianc e Equation

A. P. I. 42 Compd. #
528 529 531 532 534 535 537 540

C1 obs. calc. 12 13 14 15 10.14 11.13 12.09 13.15 11. 12 15.15 16.05 18.17 24.12

obs.
0 0 0 0 0 0

c2

calc. 3.75 3.75 3.78 3.73 3.77 3.72 3-82 3.71 3.74

c3 obs. calc.
0 0 0 0 0 0 0 0 0

obs.
0 0 0 0 0 0 0 0 0

c4

calc. 0.11 0.12 0.12 0.li 0.il 0.13 0.iZ 0.12 0.i3

obs., c5 calc.

,
0
0 0 0 0

-1..89 -1.88 -1.87

-1.89
-1.89 -1.87 -1.88 -1.88 -1.86

16
17 18 20

0
0 0 0

o
0 0

106

26

-0.11 -0.12 -0.12 -0.11 -0.11 -0.13 -0.i2 -0.12 . -0.13

'

F o r simplicity and also to illustrate that the u s e of the revised hydrogen balance equation ied to accurate results, tl-e revised hydrogen balance (equation 15) w a s used in analyzing groups I and 1 1 (Paraffins) and the original hydrogen balance (equation 2) for the remainder. A summayy of the accuracy of this analysis is given in Table V.
The graphical method of solution which has been described was applied to 17 representative compounds taken i r o n the five m a j o r classes. The relatively s m a l l number of compounds examined by this method was due to the fact that this method w a s much m o r e time-consuming than the iterative procedure. The nature of the roots was essentially the s a m e in all c a s e s , they all possessed only one r e a l root l e s s thanzC. T h e r e was one exception to this rule, A. P. I. compound #133 possessed two real roots but one of these lay in the second quadrant, which has no physical meaning. One representative example of the graphical method of solution showing the intersection between the molar volume and molar refraction curves which yields the single root l e s s thanxG i s given in Figure 1.

-17-

0 M.V. QUADRATIC

M.R. QUADRATIC

I
c

I
20

I 40

I 60

I 80

Figure I - Graphical Solution for A. P. I . Compound #16 to Illustrate the Existence of Only One R e a l Root.

-18

DISCUSSION
Some additional information about the s t r u c t u r e can b e obtained by a detailed examination of the "pattern" of the r e s u l t s in Table 111, within e a c h of the seven groups of compounds into which this table is divided. This information can beLuseda s a valuable a i d when applying the method to unknown compounds. B e f o r e discussing the significance of the e r r o r s , i t should b e pointed out that for a given composition in t e r m s of the f i & c h a r a c t e r i s t i c groups, t h e r e is a fixed relationship between the e r r o r s . If the e r r o r in C1 i s taken a s x and the e r r o r in C 3 as y, then the e r r o r in C 2 will be -(x t y ) , that in C4 w i l l be y and t h a t in C5 w i l l be -y. Considering, f i r s t of all, the branched paraffins, Group II, i t will be noted that whereas C should be zero, i t is in fact positive and sometimes as high a s t3. T h e r e a h e a r e d to be a d i r e c t relationship between the number of branches and the magnitude of the e r r o r . On the average, the e r r o r in C2 p e r branch was t-0.42 Cz group. If, in the analysis of an unknown, c 2 , C3' c4 and C5 were l e s s than 0.1, the compound could be predicted to be a normal paraffin with a high d e g r e e of certainty. However, i f C w e r e between to. 1 and 4-0.4, then the compound would probably have one branch? On the other hand, if C 2 w e r e g r e a t e r - t h a n to. 4, the compound would probably have m o r e than one branch. T h e a c c u r a c y of the analysis of Group I I I , monocyclic naphthenes. was unusually high, due, no doubt, to the uniformity of the s t r u c t u r e s within the group. Little difficulty should be experienced injdentifying this group in the analysis of-unknown s t r u c t u r e s , since C h e r e m u s t b e g r e a t e r than 3 (minimum 2 ring size). A few s t r u c t u r a l effects w e r e noted in the e r r o r s in C1 and Cz. The s i z e of the ring and the number of branches on the ring would appear to be the e f f e c t s having the m o s t influence, whereas branching on a single s i d e chain did not appreciably influence t h e results. T h e a c c u r a c y of the r e s u l t s for Group IV. the monocyclic aromatics, was practically the s a m e as for Group 111, and for essentially the s a m e reasons. T h e a c c u r a c y for compounds with a n o r m a l s i d e chain was v e r y high, a s would be expected, since this was the type of compound upon which the interaction eff e c t of C1 and:C4 w a s based. Where one branched s i d e chain was attached to the ring, the e r r o r m C was about - 0 . 3 and in C about to. 3. T h e l a r g e s t e r r o r for 1 2 the e n t i r e group o c c u r r e d f o r c a s e s where t h e r e w e r e m o r e than one side chain attached to t h e ring. As with Group 111, t h e r e should b e no difficulty in recognizing this c l a s s of compound when analyzing unknown compounds, s i n c e C4 should always be s i x in this class. At this level the e r r o r s w e r e sufficiently s m a l l that no doubt should e x i s t a s to the s t r u c t u r e . Group V, the fused ring naphthenes, presented a much m o r e difficult c a s e to interpret. Since t h e r e appeared to be many m o r e s t r u c t u r a l factors which could influence the analysis, and the examples available w e r e v e r y limited, no r e g u l a r pattern between s t r u c t u r e and the e r r o r s in the analysis could be determined. However, s e v e r a l important a s p e c t s of the e r r o r s associated with this

- 19 -

x
d
3

*
d
m
0

9 0
0

m *
0
3

m *
I

OI
4

x
0
4

*
m

N
Q)

0 0

m *
d

0
I

0
1

Q
I

Cr

OI
3

* 0
0

co

co m
0

I n

m
0

m
N
N

m
4

m
4

+
-

*
Q

. 2
I

rN

(s
!
I

.I

0 *
0 0 0

IC

0 N
d

OI *
. N

m
W
.A

d
rd

rd h

et
& I

20I

group should be mentioned. T h e mean e r r o r and standard deviation w e r e very s m a l l f o r C3, s o that t h e r e should b e no difficulty in detecting the presence of .I a relatively s m a l l number of C3 groups in a saturated compound. An examinaI , tion of the analysis of compounds 543, 561, 577 and 578 indicated that in cases where C 1 was actually z e r o the analysis yielded s m a l l negative values for C 1 ; when this o c c u r r e d , i f the value of C1 was added t o the value obtained for C 2 , the true value of C2 was obtained. This suggested that in the analyses of unknown compounds, i f a negative value of C occurred, It should be replaced by zero and , 1. the negative value of C1 should be added to C2 to give the c o r r e c t amount of C 2 p r e s e n t (#575 was an exception to this rule). Side chains and carbon atoms corn- ; mon to t h r e e r i n g s appeared to have some influence on the accuracy of the analysis.
, t

Group VI consisted of the A. P. I. fused ring a r o m a t i c s . The e r r o r h e r e was, of c o u r s e , higher than f o r the preceding groups, but still s m a l l enough to enable the s t r u c t u r a l analysis s y s t e m to be used. In this c l a s s , where the l a r g e s t number of s t r u c t u r a l groups was p r e s e n t in a single compound, the l a r g e s t number of interaction effects could b e expected. The total number of compounds available in this group was s o s m a l l that it was not possibile to make any general- 1 izations about the relationship between the e r r o r s in the predicted analysis and . the s t r u c t u r e . Attention is drawn to the analysis of compound #179, which exhibited the highest e r r o r in C1 and C2. This r e s u l t was not unexpected, as during the evaluation of the i n t e r a c t i o n effect of C on C i t was observed that #179 did 2 4 not fit the s t r a i g h t l i n e relation on which this interaction factor was based. Group VI1 consisted of fused ring aromatic compounds whose properties had been determined by van Krevelen (17). The e r r o r s in the predicted values of the s t r u c t u r a l groups in this c l a s s a r e considerably g r e a t e r than in the other classe It should be noted that t h e r e s u l t s f o r coronene and dodecahydrotriphenylene w e r e not unexpected. When the original coefficients for C5 w e r e calculated, the data for coronene w e r e not consistent with the o t h e r compounds and w e r e t h e r e f o r e omitted. Consequently, it was expected that the accuracy of the analysis of this compound would not be v e r y high. Dodecahydrotriphenylene represented a type of s t r u c t u r e which the p r e s e n t s y s t e m was not designed to treat. The unusual c h a r a c t e r i s t i c of this s t r u c t u r e was the presence of C and C5 in the absence of C4. It 2 w i l l be recalled that fused ring a r o m a t i c s contaming C2 groups w e r e not successfully dealt with m y determining the interaction effect of C2 on C4. Clearly, t h i s method of s t r u c t u r a l analysis was not applicable to the c a s e where C = 0 and C2 3 and C5 = 0. As in Group V, u s e could be made of the fact that negative values o f c 1 usually indicated a t r u e value for C1 of zero, and that if this negative value o f C1 were added to the value obtained f o r C the resulting revised c2would b e considerably c l o s e r to the t r u e value. The &her high e r r o r for this group as a whole was considered to be d u e to the general unreliability of the physical property data, which has already been discussed. The accuracy of the molar volume and the mola r e f r a c t i o n equations f o r this group was not a s g r e a t as for all other classes. How- e v e r , it was considered n e c e s s a r y t o include this group, because of the s c a r c i t y of data for fused ring a r o m a t i c compounds.

21 -

The average accuracy for all A. P. I. hydrocarbons analyzed was considerably l e s s than one carbon group for each type. This was considered adequate for determining the a v e r a g e number of the various s t r u c t u r a l groups. It was realized, initially, that the m o l a r volume and the m o l a r refraction were v e r y intimately related properties and that the equations developed in t e r m s of the five s t r u c t u r a l groups might not be independent. However, this investigation revealed that these two equations were in fact independent, and that the differences between them w e r e of such a magnitude that the ?elutions obtained w e r e in substantial agreement with the known values of the groups present. Consequently, it was established that i t was possible to deduce a considerable amount of s t r u c t u r a l information f r o m these two closely related properties. The r e s u l t s OF the application of this method indicated the m e r i t s and s o m e of the deficiencies of this type of s t r u c t u r a l analysis. As physical prope r t i e s become available on I f new" types of s t r u c t u r e s , the scope of this method m a y be extended. This approach could be applied to types of s t r u c t u r e s not a l ready covered, by developing analogous equations for other physical p r o p e r t i e s and also by considering additional interaction effects. One of the m e r i t s of this s y s t e m l a y in the manner in which the coefficients in the m o l a r volume and molar refraction equations were determined. This gave confidence in extrapolating beyord the molecular weight range of t h e known compounds that w e r e used to es tablish the system.

22-

,
Burdett, R. A . , Taylor, L. W . , Jones, L. C. Jr., "Molecular Spectroscopy". Report o f a Conference Organized by the Spectroscopic Panel of the Hydrocarbon R e s e a r c h Group of the Institute of Petroleum, London, Oct. 28-29, 1954, p.36Eccleston, B.H., Coleman, H.J., 72. 3866-70 (1950).

Adams, N.G.,

J. Am. Chem. SOC.

Egloff, C. , P h y s i c a l Constants of Hydrocarbons, Vol. 1 1 1 , 25, A, C.S. Monograph, Reinhold, New York, 1946. Krevelen, D. W. van, Brennstoff-Chemie, 33, 260-8 (1952). Ibid.,

34,

167-82 (1953).

Krevelen, D. W. van, Blom, L., and Cherrnin, H. A. G. 1075-6 (1953). Krevelen, D. W. van, and Chermin, H.A.G.

, Nature 1 7 1 .
1

, Fuel, 33.

79-87 (1954).

Krevelen, D. W. van, and Schuyer, J. , I' Coal Science, Aspects of Coal Constitution'' , C h a p t e r s VI and VIIS Elsevier, Amsterdam, 1957. 46, 2186-91 (1954). Kurtz, S.S. Jr., andSankin, A., Ind. Eng. Chem. Lorentz, H.A.,
Ann Physik. 9. 641 (1880).
I

Mahdi, A.A.K. A l ., and Ubbelohde, A . R . , P r o c . Roy. SOC. (London), 220A. 143-56 (1953). Mahdi, A. A. K. Al., and Ubbelohde, A. R. Changements d e P h a s e s , Comptes-rendus de l a deuxieme Reunion Annuelle, SOC. Chim. Phys., P a r i s , J u n e 1952, pp. 360-65. Nes, K. van, and Westen, H. A. van, "Aspects of the Constitution of Mineral Oils'f , pp. 299-314, Elsevier, Amsterdam, 1951. Ibid., pp. 318-49 and pp. 445-53. Ibid.

, Chapter IV.

Schiessler, R.W. , and Whitmore, F. C., Ind. Eng. Chem. (1956); Am. Doc. Inst., Doc. 4597.

47,

1660-5

Schuyer, J., Blom, L., a n d Krevelen, D.W. van. Trans. Far, SOC. 49s 1391-1401 (1953).

(18) (19)

Schuyer, J . , and Krevelen, D. W. van, Fuel,

33, 176-83

(1954).

Vlugter, J. C., Waterman, H. I., and. Westen. H.A. Petroleum Technol. 18. 735-50 (1932).

van, J. Inst.

(20)

Vlugter, J . C . , Waterman, H.I., and Westen, H.A. van, J. Inst. Petroleum Technol. 21. 661-76 (1935).

(21)

Ward, A . L . , andKurt8, S.S. J r . , Ind. Eng. Chern. Anal. Ed. 10. 559-76 (1938).

-25-

PRODUCTION O F PIPELINE GAS BY BATCH HYDROGE3IOLYSIS O F OIL SEIAIE

E. B. Shultz, Jr. and H. R. LLnden Institute of G a s Technology

bY

Chicago, Illinois

P,
J

,
I !

The conversion of oil shale t o high heating value gases by d i r e c t hydrogenation was investigated t o determine if the production of pipeline gas by this method was f e a s i b l e and i f i t offered p o t e n t i a l advantages over a l t e r n a t e methods f o r u t i l i z a t i o n of the l a r g e reserves of this f o s s i l f u e l . Data on the batch hydrogenolysis of a 22.9 g a l . per ton Fischer assay Coloracio oil shale were obtained a t a maximum reactor temperature of 1300F., maximum pressures of 1200 t o 5700 p . s . i . g . , hydrogen-shale r a t i o s equivalent t o 50 t o 200 per cent of stoichiometric requirements f o r complete conversion of the organic carbon plus hydrogen content t o methane, and f o r three p a r t i c l e s i z e ranges. Nearly complete conversion of.orgaaic carbon and hydrogen t o a f u e l gas with a heating value of over 800 8 . t . u . per standard cubic f o o t w a s obtained i n rel a t i v e l g s h o r t residence times a t temperatures of U 0 O o t o 1300 F., with o n l y l i t t l e formation of carbon oxides from mineral carbonate decomposition. In view of the r e l a t i v e l y low material costs, these r e s u l t s i n d i c a t e t h a t serious consideration can be given t o supplementing the f u t u r e supply of natural gas with synthetic high heating value gas from o i l shale, p a r t i c u l a r l y i n areas served by long-distance transmission l i n e s passing i n the v i c i n i t y of the Colorado deposits.

-26-

NOT FOR PUBLICATION Presented Before the Division of G a s and Fuel Chemistry American Chemical Society Chicago, Illinois,Meeting, September 7-12, 1958
PRODUCTION OF PIPELINE GAS BY EATCE HYDRDGEMISSIS OF OIL SHALE

E. B. Shultz, Jr. and H. R. Linden I n s t i t u t e of Gas Technology Chicago , I l l i n o i s

A n exploratory i n v e s t i g a t i o n has been made o f the dry, high-press u r e hydrogenolysis (hydrogasification) of o i l shale as part of a cont i n u i n g program concerned w i t h the production of natural gas supplements and s u b s t i t u t e s from l i q u i d and s o l i d f o s s i l f u e l s . Previous publications in t h i s s e r i e s have d e a l t w i t h the highpressure hydrogasification of petroleum o i l , bituminous coal and T h e l i g n i t e s , and p u r e compounds r e l a t e d t o petroleum purpose of the present work w a s t o determine if the recovery of the organic constituents of o i l shale i n the,form o f high heating value gas would provide an a t t r a c t i v e a l t e r n a t i v e t o the conventional approach of maximizing l i q u i d products recovery. Results indicated that rapid, and nearly complete conversion of the organic carbon plus hydrogen content of o i l shale t o a high methane and ethane content gas o f over 800 B.t.u./SCF (standard cubic f o o t a t 600F. , 30 inches of mercury pressure, saturated w i t h water vapor) heating value can be obtained a t r e l a t i v e l y moderate temperatures and pressures. A Colorado o i l shale of 22.9 gal. per ton Fischer assay was used throughout the study, s i n c e this material appeared representative of the Green River formation deposit of Northwestern Colorado estimated t o contain about 1260 b i l l i o n b a r r e l s of o i l . e A n i n d i c a t i o n o f the need f o r development of economical methods f o r the production o f pipeline gas from the large reserves of s o l i d f o s s i l f u e l s can be obtained from a recent study of f a c t o r s influencing United S t a t e s crude o i l and n a t u r a l gas p r o d ~ c t i o n . ~ The r e s u l t s o f t h i s study show t h a t on the basis of an estimate of average d r i l l i n g r e t u r n ( r a t i o of e s t a b l i s h e d reserves t o footage d r i l l e d ) , domestic crude o i l prices would have t o reach $6 per b a r r e l t o achieve an u l t i mate recovery o f 160 b i l l i o n b a r r e l s , and that a t a maximum price o f $4 p e r b a r r e l o n l y 140 b i l l i o n b a r r e l s would be u l t h a t e l y recovered. These crude o i l prices were computed after allowing f o r n a t u r a l gas revenues ranging from 50 cents per barrel a t present, t o over $1per b a r r e l a t the time u l t i m a t e crude o i l recovery reaches 1 6 0 b i l l i o n b a r r e l s . Since 88 b i l l i o n b a r r e l s of domestic crude o i l had already been discovered a t the end of 1957, this would correspond t o additional discoveries of only 52 b i l l i o n b a r r e l s a t a maximum r i c e of $4 per barrel , o r 72 b i l l i o n barrels a t a maximum p r i c e of $6 per barrel, assuming average d r i l l i n g r e t u r n . A t an expected average f u t u r e recove r y of 6000 cubic feet of n a t u r a l gas per b a r r e l of crude the t o t a l additional gas supply, including present reserves of 247 t r i l l i o n cubic f e e t , would then be about 560 and 680 t r i l l i o n cubic f e e t , respectively. This i s s u b s t a n t i a l l y less than a recent estimate of 1 2 0 0 t r i l l i o n cubic f e e t (corresponding t o an ultimate crude o i l recovery of 250 b i l l i o n b a r r e l s ) , based on geological f a c t o r s without consideration of economic l i m i t a t i o n s on exploration and Thus, i f increased importation of r e l a t i v e l y low-cost foreign crude i s permitted, lack of economic incentives may r e t a r d development of a major portion o f p o t e n t i a l domestic crude o i l reserves. O i l shale would correspondingly gain i n importance as an a l t e r n a t e source

-27-

of pipeline gas in view of the large and well-known proved reserves, approaching in mapitude the thermal value-ofeconomically recoverable coal reserves.ll 2

APPARATUS AND PROCEDKRE

The one-liter Autoclave Engineers high-temperature, high-pressure reactor used in previous batch hydrogenolysis studies was used i n this The procedure was essentially the same as described i n the pure compound study.14 The reactor was charged at room temperatures and placed in the rocking furnace, which was also at room temperature. Heating at full input (4.5k w . ) was maintained throughout the rising temperature portion of the run, with temperature rising at about gF. per minute. Simultaneous temperature and pressure measurements were taken, and gas samples were obtained at intervals throughout t n e course of each run which consisted of a 128 to 152 minute period ?equired to reach the nominal temperature of 1300F., and an additional 30 & U t e period at 1300OF. In all cases reaction was initiated well below 1300OF.; the initiation of rapid gasification appeared to correlate with the appearance of a well-developed temperature dip at 1025OF. In accordance with previous practice, this was designated the inLtial gasification temperature for oil shale and was arbitrarily used as z zero time base for the space-tirne yield calculations. Feed and residue shale samples were subjected to sieve analysis, and to ultimate analysis f o r total carbon and hydrogen.. Mineral carbon ties determined gravinetrically from the carbon dioxide evoLved with ecid, in a technique employed by the Bureau of Mines Experiment Station, L a d e , Wyoming; organic carbon was obtained by difference.17 Carbon dioxide liberation values determhed from residue shale analyses were found to be uniformlg greater than values obtained from gas analysis dati, because of continued evolution of carbon dioxide after runs were terminated; conclusions concerning carbon oxides formation were drawn frco product gas yields and compositions. Product gas samples were analyzed with a Consolidated EngLneerjng Co. Model 21-103 mass spectrometer; heating values and specific gravities were calculated from the analyses. Product gas volumes and heating values were calculated at 60F., 30 inches of mercury absoluts pressure, and saturation uith water vapor,assuming the ideal gas law. Specific gravities were calculated on a dry basis from the average molecular weight of the gas referred to air of molecular weight 28.972. Initial hydrogen volumes were obtained by direct measurement; the reactor was charged with shale and hydrogen to the desired pressure, and the hydrogen was slowly vented through a wet test meter. Use of compressibility data at room temperature permitted the calculation of reactor free space when charged with shale. Product gas volumes during the course of the run at temperatures of 950F. and above were calculated from observed temperatures and pressures and the initial reactor free space, assuming ideal gas behavior. Previous work with coal has shown that gas volumes calculated by this method agree with values measured by wet test meter, with a deviation of about 3 per cent.13 Further, the reasonably close agreement of reported organic carbon and hydrogen conversions based on computed product gas volumes, and organic carbon conversions based on residue ultimate analyses, supports the use of pressure-temperature-reactorvolume measurements with assumption of ideal gas behavior. The feed shale analysis is given in Table 1, and the effects of process variables are shown in Tables 2, 3 , and 4, and Figures 1 and 2.

-28-

Table l.-ASSAY AND ANAL;ysIS O F U . S . BTJREAU OF MINES O I L SHALE SBR58-40X Fischer Assay a Carbon-Eydrogen Analysis b Carbon, w t . % O i l , wt. $ 8.8 water, w t . $ 1.2 Mineral 4 .a8 Spent shale, w t . $ 88.0 Organic 10.52 2.0 G a s + loss, w t . d Total Total lT53 Hydrogen, w t . $ 1.59 Ash, w t . .% 0b1, gal./ton 22.9 68.98 Water, gal./ton YEneral GO*, w t . $ 17.88 3.0 r a v i t y of o i l O 6 0 O ~ . 0 . 917-0. 918

rn
f

" 6

',

Sieve Analysis'

Runs 3, 4, 5, 1 0 , 1 1
U.S.S.
A0

40-100 Mesh Sample

Sieve

Wt. % 0.8

sie;Je h a l y s i s d 5-20 Mesh Sample ,Run 8

U.S.S.
8-10 1 0- 1 2 12 1 4

Sieve

Wt.

40-50 50-60 60-70 70-80

80-100

25.3 20.0

+5 5-8

19.1 16 - 7

-100

Total

1583

13.6 4 -5

1 4- 16 16- 18 18-20
-20

Total Sieve Analysise 140-325 . . Mesh Sample Run 9 U.S.S. Sieve Wt. B +14 0 3.2 140-170 24.5 0.4 170- 200 200-230 32.8 230-270 11.2 270 - 32 5 0.2 -325 27.7 o O . o Total 1

100.0

4.6 41.3 4.7 8.4 %*1 13.4 0.6 13.2 5.7

a) Average of U. S. Bureau of Mines R u n s 53456 and 53457. b, Average of U. S. Bureau of Mines Runs lo291 and 10292. l7 I . G . T . Lab. No. 3910. d, I . G . T . Lab. No. 4012. e ) I.G.T. Lab. No. 4013

-29-

-30-

d 0

d Of N rl

m 0n N In0

! k

OOR

mo m dm

-31-

!
i

i
\

1.

-34EFFECTS OF PRESSURE

At all three pressure levels studled, 340, 735 and 1710 p.3.i.g. ,initial pressure (1215, 2430 and 5540 p.s.i.g., respectively, upon attainment of 13OOOF. ) , the results showed that high heating value product gases were obtained at high organic carbon-plus-hydrogen Conversions as the nominal run temperature of 1300OF. was approached; product gas dilution with carbon oxides was not excessive and practically no liquid products were formed (Table 2 ) . The spent shale was free flowing and had virtually the same sieve analysis as the charge. Rapid attainment of high conversions of the organic matter to gas , , (prlmarily methane, ethane and propane) at temperatures of only 1200O to i3OO0F,was primarily responsible for the low evolution of mineral carbon oxidese; this dFffers significantly from the results obtained in high-tem erature retorting at low pressures and i n the absence of hydrogen."' I: Although higher pressure would be expected -to suppress carbon dioxide evolution, the apparent yield and mole per cent of carbon dioxide was not affected signif'icantly by pressure level (Figure 1 ) . However, total glelds of carbon oxides were decreased at the higher pressures, reflecting the decrease in carbon dioxide conversion to carbon monoxide by the reaction COa + H2 + CO + G O , as the hydrogen content of the product gas decreased. Even at the lowest pressure level, , total carbon oxides content was o n l y 22.4 mole $. (The reported- / . N2 + CO contents were primarily carbon monoxide.) The hydrocarbon hydrogenolysis reactions, an& the sequence of appearance of the stable intermediates Fn methane production from higher molecular weight carbon-containing materials (such as oil shale kerogen), corresponded closely to those observed in hydrogenolyzis of petroleum oilsla and pure compounds related to petroleum o i l s . Propane and higher paraffin hydrocarbons formed in earlier portions of each run were soon hydrogenolyzed with increasing appearance of ethane and methane. Ethane yields and concentrations in turn passed through maxima with increases In time and temperature, as methane, the stable final product, continued to increase. Maximum ethane yields were observed at attainment of l.200F. at all pressures (Figure 1 ) . The effect of pressure increase was to increase the rate of ethane production below 1200F. and the rate of ethane disappearance to methane above 120O0F. Methane yields were considerably increased above 1200OF. with increases in pressure, accompanied by increases i n conversions and space-time yields. Increased hydrogen consumptions at higher pressures, together with lower carbon oxide yields and higher methane yields resulted in considerable increases in product gas heating values. For example, heating values of 792, 871 and 908 B.t.u./SCF were observed at attainment of 1300OF. as pressure w a s increased from 1215 to 2430 to 5540 p.s.i.g., respectively (Table 2 ) . At the two higher pressures studied, as well as in other runs carried out at 1005 of stoichiometric feed ratio, the gasification of organic carbon-plus-hydrogen decreased slightly at temperatures of 1200O tu 1300F., accompanied by a decline in gas heating value 1 4 excess of that corresponding to increased carbon oxide formation. This indicates that insufficient hydrogen m a y have been present in the later portions of these runs to prevent a small amount of carbon formation from product hydrocarbons. This effect appeared to be greater at higher pressures, due to decreased hydrogen concentrations brought about by the higher hydrogen consumptions characteristic of higher pressure operation.

-35-

FZWECTS O F iM3RCGDT-SiWL;E FIE3) FATI0 F i g w e 2 presents gas y i e l d data a t hydrogen-shale feed r a t i o s of (Run 11) , 1005 (RUT 1 0 ) and 200% (Run 5) of s t o i c h i o n e t r i c reqdire1 ~ccl 5 ments for coriversion t o methane; .complete r e s u l t s f o r 3uns 1 are given i n Table 3 and for the key test, 2x1 .lo,i n Table 2 . i t temperatures below 1200F., the extent of g a s i f i c a t i o n at 505 of' stoichiometric hydrogen-shale feed r a t i o was agproxiaately the sram as a t the higher feed r a t i o s so t h a t , i n the absence of excess hy&ogen dLlution, earlier formation of high heating value Froduct gas occurred. Eawever , a t m e higher temperatures, conversions were reduced s u b s t a n t i a l l y by decreases i n hydrogen supply, but not oroport i o n a l l y t o reductions In feed r a t i o . For instance , a t l 3 O O 0 $ . , g a s i f i c a t i o n of organic carbon and hydrogen vas 77 veight 5 a t 505 of stoichiometric, 90 w e i g h t 5 a t 1 0 0 5 of stoichiometric, and complets a t 200% of stoichionetric feed r a t i o . Considerable vapor-Fnase et?bon formation was indicated during the later portion of tine run a t 335 of stoichiometric, and none a t 200s of stoichiometric. This c c q a r e s w i t h evidence of only limited vapor-phase carbon fo m a t i o n , Indicetea value, l a t e by a gradual decline in conversion and product gas heatin the course of the run a t . l 0 0 $ of stoichiometric feed r e t i o . Pressure l e v e l s f o r Xuns 1 0 and 5 a t loo$ and 2005 of s t o i c h i m e t r i c feed r a t i o , respectively, were q u i t e comparable, permitting a direst evaluation of the e f f e c t . o f hydrogen concentration on carbon oxides formation.. Total carbon oxide yields were about the sarxe f o r these t w o m s , b u t increased hydrogen concentration a t 2005 of s t o i c h i m e t r i c feed r a t i o caused much greater conversion of evolved carbon dioxi& ts carbon monoxide. 0 ' Ethane yields were increased by increases in feed r a t i o f r c a 5 t o 200$ of stoichiometric; however, ethane contents were greetest et 100% of stoichiometric feed r a t i o . Dilution of the proauct g t s : , ! i t : ? excess hydrogen reduced the ethane content a t 20075 of' stoichLomet?ic, and pyrolysis reactions favoring methane over ethane formation E~LX&. the ethane content a t 505 of stoichiometric.
50%
~

EFFECTS O F PARTICLE S I Z W G E

In view of the s u b s t a n t i a l cost of o i l shale s i z e r s d u c t i o n , l o it would be desirable t o u t i l i z e r e l a t i v e l y large p a r t i c l e s i z e s tf a p r a c t i c a l hydrogenolysis process can be developed f o r moving- lor 3 . ~ ~ 2 bed operation. In Table 4 , it can be seen that i n s i p d ' i c a n t e E e c t s on gas yields and composition resulted from a v a r i a t i o n in p a r t i c i e s i z e range from 5-20 m e s h t o 140-325 mesh. COPEERCIAL POSSIBILITIES
Production of pipeline' gas from o i l shale may be preferabie t o l i q u i d fuel production because of higher conversion of o r m i c matter (90-100 weight $ f o r hydrogasification, compared t o about 80 weight 5 conversion t o l i q u i d and gaseous products in conventional r e t o r t i n g 2 ) and elimination of c o s t l y l i q u i d p r c 3 x t r e f i n i n g operations. Xydrogasification, in addition t o -producing a free-flowing residue containing l i t t l e organic matter, also Y Z e l d s only negligible quantities of l i q u i d products. T h i s d i f f e r s from o i l h y d r o g a ~ i f i c a t i o n ~ ~ , ~ ~ where substantial q u a n t i t i e s of and pgrolysis of crude shale l i q u i d by-products &re formed. Absence of a@omeration problems should permit the &.?velopinent of continuous moving- o r fluid-bed 09 shale hydrogasification processes. Fixed-bed operation would 'nave the advantr.ge of redwed f zed preparation costs. Hydrogen requirements c o u l d be met with convantiond c a t a i g t i c steam reforming and carbon

-36oxide removal processes, u t i l i z i n g a portion of the p w i f i e d product ' gas f o r feed and f u e l . O n the basis of the results with 22.9 gal. per ton shale a t 50 t o 100% of stoichiometric feed ratios, about 4900 t o 6600 SCF/ton of' 1000 B.t.u./SCF equivalent gas can be produced w i t h 3800 t o 7600 SCF/ton of hydrogen feed. T o t a l product gas requirements f o r hydrogen production, including a l l f u e l requirements, w i l l be about 1900 t o 3800 SCF/ton, leaving a n e t 1000 B.t.u./SCF equivalent gas y i e l d of 2800 t o 3000 SCF/ton. Heat requirements f o r the shale processing s t e p estimated from gas combustion r e t o r t data3, w i l l be about 500,000 B.t.u./ton, so that i f product gas i s used as a source of heat, a net. gas y i e l d of 2300 t o 2500 SCF/ton would final1 be obtained. A t 50 ceni t o n mining cost and 25 cent/ton crushing costJg0 this would r e s u l t i n a r a w material cost of 30 t o 33 cents/1000 SCF of 1000 B.t.u./SCF gas equivalent f o r 22.9 gal. per ton shale. Large deposits of shale average 30 g a l . per ton o r more,' so that raw material cost could be reduced t o l e s s than 30 cents/MCF with the r i c h e r shale. Existing p i p e l i n e systems and requested extensions could supply t h e major West Coast and Middle West marketing areas w i t h pipeline gas produced Fn Colorado. With adequate storage, already under considerat i o n by Congress, the flow of the Colorado r i v e r i s adequate t o provide water f o r a 2 million barrel per day o i l shale industry, which i s , equivalent t o about 8 b i l l i o n cubic f e e t per day of net pipeline gas production. l o CONCLUSIONS
~

'

Nearly complete conversion of the organic matter of a typical Colorado o i l shale t o high methane and ethane content, high heatvalue f'uel gases has been achieved i n batch hydrogenolysis a t 1200 to1300'F. in the presence of s u f f i c i e n t hydrogen t o convert the organic carbon and hydrogen t o methane. Liberation of mineral carbon dioxide w a s kept a t a low l e v e l by operation a t these r e l a t i v e l y low temperat u r e s . P a r t i c l e s i z e range variations from 5-20 mesh t o 140-325 mesh had no s i g n i f i c a n t e f f e c t on g a s i f i c a t i o n r a t e s and yields. Increases' i n pressure t o 5500 p . s . i . g . r e s u l t e d i n more rapid formation o f high hesting value gas, higher gas yields and lower t o t a l yields of carbon oxides. However, pressure increases above 2000 t o 2500 p.s.i.g. did not appear t o a f f o r d advantages commensurate with the cost increases that would be involved i n a commercial application. Hydrogen feed of twice the stoichiometric requirements f o r methane formation resulted i n complete conversion of organic carbon content, but excess hydrogen d i l u t e d the product gas. Hydrogen feed of one-half of the stoichiometric requirements r e s u l t e d i n lower conversions and some vapor-phase carbon deposition, b u t pyrolysis reactions brought about high yields of high heating value gases a t 120O0F. before appreciable carbon deposition appeared t o begin. Low hydrogen concentrations i n the pro- duct gas slowed conversion of carbon dioxide t o carbon monoxide, an undesirable r e a c t i o n which consumes feed hydrogen. Although f u e l gases of pipeline q u a l i t y were produced i n t h i s study without f'urther treatment, i t would probably be economic t o remove carbon dioxide before high-pressure transmission. O n the basis of these r e s u l t s , high heating value gas production by hydrogasification of o i l shale w i t h hydrogen produced from a portion of the product gas appears both technically and econonically f e a s i b l e . I n view of the large reserves of o i l Shale, v a s t l y exceeding estimated ultimate crude o i l reserves and approaching i n magnitude the thermal value of economically recoverable coal reserves, serious consideration t o t h i s a l t e r n a t e s o w ~ of pipeline gas should be given i n studies of f'uture gas supply.

-37\

I
j,
? )
1

ACKNOWI;EM3MElIT

\
i

, data,. j

This study was conducted as p a r t of the basic research program of the I n s t i t u t e o f G a s Technology w i t h f'unds provided by i t s Members and Contributors. 14. A. E l l i o t t , dteector of the I n s t i t u t e and H. M. H x ~ r y ,president of the N.E.G.E.A. Service Corporation were very- lielpfkl i n formulating the progrm-. H . M. Thorne, C u e f of 1 1 , Bureau of Mines, made the o i l shale Oil-Shale Research, Region 1 Supply and analyses available and contributed valuable infornation required for the study. The batch hydrogenation tests weTe performed by R. F. Johnson. D. M. Mason and J. E. Neuzil provid2d the a n a l y t i c a i

h e r . Gag Assoc. , h e r . Petroleum Inst. and Canadian P e t r o l e m ASSOC., Proved Reserves of Crude Oil, Natural Gas Liquids and Natural Gas in the U n i t e d S t a t e s and Canada, December 31, 1957," No. l 2 ., The S o c i e t i e s , New Y : r k and Calgary, Alberta, 1927. Brantley, F. E. and others, High Temperature Shale O i l , Eng. Chem. 44, 2641-50 (1952). Camron, R. J. and Guthrie, B., " O i l from Shale," Chen. Eng.

Progr. 50, 336-41 (1954). Channabasappa, K. C. and Linden, H. R. , "Fluid-Bed Pretreatment of Bituminous Coals an$ Lignite-Direct Hydro enation of the Chars t o Piffeline Gas, Ind. Eng. Chem. 3, t37-44 (1958). Davis , hJ. , A Study of the p t u r e Productive Capacity and Probable Reserves of t h e U. S., O i l Gas J. 56, 105-19 (February 24, 1958). Donne11, J. R. , "Preliminary Report on Oil-ahale Resources o f Piceance Creek Basin Northwestern Colorado, U. S. Geological Survey B u l l e t i n 1042-H, Govt. P r i n t . Office , Washington, D. C.,

Rill, K.

G . and o t h e r s , "Future Growth o f the World Petroleum Industry, The Chase Manhatzan Bank, New York, 1957. Jukkola, E. E. and otf;lers, Thermal Decomposition Rates of Carbonates i n O i l Shale, Ind. Eng. Chem. 45, 2711-14 (1953). Linden, H. R. and o t i e r s , "High TemperaEre Va or-Phase Crack2467-82 (1955). i n g of Hydrocarbons, Ind. EM. Chem. Priec, C. H. and Savage, J. I f . , "A Sha e - O i l Industry i s on i t s Way, Chem. En-. Pro 52, 16-5-26-5 (1956). Rubel, A. C . , " S h a l e T i l - a s a Future Energy Resource, 'I Mines ma^ a z i n e 72-76 (1955) October; a s J. 56, 110-11 "Shale, k a l Gas t o Supplement Natural, O i l G (April 7 , 1958). ahultz, E. B. , Jr. , Channabasappa, K. C. and Linden, H. R., Hydrogasification of Pepoleum O i l s and Bituminous Coal t o Natural G a s S u b s t i t u t e s , Ind. Erg. Chem 48, 894-905 (1956). Shultz, 3. B. , Jr. and Linden, H. R. , "Bat& HydrogenoAysis o m ounds Related t o Petroleum O i l s , Reactions of Pure C En- Chem. &, 2011-18 (1'357). W. and o t h e r s , 'Entraineg-Solids Retorting of Colorado O i l Shale-Equipment and O p r a t i o n , Ind. Eng. Chem. 3, 461-64

1957.

?,

s,

Terry, L. F. End h'inger, J. F., "Future Growth of t h e Natural Gas Industry, The Chase Manhattan Bank, New York, 1957. Thorne, H. M. , Bureau of Mines Experiment Station, Laramie, Wyoming, P r i v a t e Communic$tion, February, 1958. Tihen, S. S. and others, Entrained-Soli$ Retorting o f Colorado O i l Shale-Product Yields and Properties, Ind. E ~ R Chem. . Q,

(1955)

464-68 (1955) .

-39-

i
4
\

PERFORMANCE O F SUPPORTED NICKEL CATUSTS I N CYCLIC S T E A M FEPORMING O F N.A"RAL GAS

bY J. M. Reid and D. M. Mason

i
)

I n s t i t u t e of Gas Technology Chicago, I l l i n o i s

AESTRACT

4
1

,!

Comparatively l i t t l e data are availzble regarding the performance of c a t a l y s t s i n the cyclic steam-hydrocarbon reforming process 1 used by the u t i l i t y gas industry on the Eastern Seaboard, and more 1 extensively in Europe and A s i a . In this study, several commercial types of supported nickel c a t a l y s t s having e i t i e r alumina or magnesia as the base material were subjected t o cyclic process conditims in a laboratory refomzing apparatus. Catalyst perfommi-ce and c a t a l y s t l i f e were signif'icantly a f f e c t e d by the oxygen which ' was present during the heating portion of the cycle. U n l i k e the ' continuous s team-hydrocarbon reforming process used extensiveiy by the chemical industry f o r production of hydrogen and amnonia syn1 thesis gas, performance of the cyclic process w a s not found t o be 4 singularly dependent on the a c t i v i t y of the c a t a l y s t f o r the steauhydrocarbon reaction, but rather under c e r t a i n conditions t o be cont r o l l e d by the r a t e s of oxidation and reduction of the nickel. In these tests,li.fe of alumina-supported c a t a l y s t w a s r e l a t e d t o the formation of an unreactive compound between nickel oxide and the support. W e of the magnesia-supported c a t a l y s t w a s r e l a t e d t o s o l i d solution formation between nickel oxide and the support. The ! inadequacy of present manufacturing specifications and testing pro? cedures f o r nickel c a t a l y s t for cyclic reforming i s i l l u s t r a t e d by these r e s u l t s .

-LO-.

NOT FOR TUBLIGATION Presented Before the Division of Gas and Fuel Chemistry American Chemical Society Chicago, I l l i n o i s Meeting, September 7-12, 1958 P E F P O W C E OF SUPPORTED NICKEL CATALYSTS Ip? CYCLIC STEAM REFORMING O F NATURAL GUS

J. M. Reid and D. M. Mason I n s t i t u t e of G a s Technology Chicago, I l l i n o i s

INTRODUCTION
Catalytic steam reforming of n a t u r a l gas and other low molecular w e i g h t hydrocarbons i s a w e l l established process.6~29132~39y40y41 It has been applied extensively i n the chemical industry where l a r g e q u a n t i t i e s of hydrogen are required, as i n the case of ammonia synthesis. A f e w I n s t a l l a t i o n s have a l s o appeared i n the u t i l i t y gas industry f o r the production of low heating value f u e l The process i s carried out i n a tube furnace in which preheated steam and hydrocarbon are passed through e x t e r n a l l y heated catalystf i l l e d tubes. The hydrocarbon and steam r e a c t t o produce hydrogen, carbon monoxide and some carbon dioxide. This reaction i s h i g h l endothermic. Reaction temperatures range from about 120O0F.t o d O O F . and pressures from atmospheric t o several atmospheres. The most commonly employed c a t a l y s t i s reduced nickel oxide supported by a high surface area r e f r a c t o r y material; nickel concentrations range from several weight percent t o more than 30. Performance of these c a t a l y s t s has been extensively investigated.53031 In addition t o process conditions, the major f a c t o r s influencing the c a t a l y s t behavior have been shown t o be poisons such as s u l w compounds contained i n the feed streams, and physical propert i e s of the c a t a l y s t such as surface area, porosity and c r y s t a l l i t e s i z e . With proper c o n t r o l of process variables, the c a t a l y s t appears t o have e s s e n t i a l l y unlimited l i f e i n commercial operation. Cyclic Reforming Prior t o World W a r 1 1 , the u t i l i t y gas industry i n the U n i t e d S t a t e s was based almost e n t i r e l y on carburetted w a t e r aas moduced . from coke, steam and o i l in cyclic apparatus. The a v a i r a b i h t y of lowc o s t natural gas t o major population centers through long distance pipe lines constructed in the post-war era made processes based on s o l i d f u e l economically unfavorable. Where conditions warranted the continued d i s t r i b u t i o n of low heating value gas, i t was necessary t o f l n d some means f o r converting natural gas. The c a t a l y t i c s t e m reforming process was i d e a l l y s u i t e d f o r this purpose. However, the process as developed by the chemical Industry i s c a r r i e d out i n continuous tube fbrnaces. To adapt t h i s process t o the gas industry would require the c a p i t a l i z a t i o n of e n t i r e new manufacturing plants t o replace the e x i s t i n g carburetted water gas plants. A more a t t r a c t i v e scheme f o r the u t i l i t i e s w a s made possible by the United Gas Improve._. ment Company,which pioneered the development of the cyclic reformlng The cyclic process I process in the United States.20~26~34~35r42944 could be c a r r i e d out by a r e l a t i v e l y low-cost modification of the e x i s t i n g carburetted water gas equipment. A s a r e s u l t , 17 cyclic process I n s t a l l a t i o n s a r e i n use in the eestern p a r t of the United States. 3925

-4lThe Cyclic Catalytic Reforming ( C C R ) process differs from the conventional continuous process, basically, only i n the manner i n which the heat requirements are supplied. F i r s t , the c a t a l y s t bed located i n one of the refractory-lined s h e l l s from the carburetted Water gas apparatus i s heated t o reaction temperature by the passage of hot products of combustion supplied by e i t h e r o i l or gas burners. This Ls followed by a reforming s t e p i n w'nich steam m-d hydrocarbon a r e passed through the c a t a l y s t bed. The heat stored in-tine refractor y shapes used f o r process steam preheat a d in the c a t a l y s t bed during t'le heating s t e p i s used during t i e reforming step. Steam Purges are normally used t o separate the h e a t k g and refoming s t e p s . The e n t i r e cycle sequence is generally completed i n less thm f i v e minutes. The c a t a l y s t employed i n the cyclic reforzing process i s similar t o continuous reforming c a t a l y s t i n that it i s metallic nickel supported by a refractory material. Properties of t'ne r e f r e c t o r y supccrt a r e necessarily more s t r i n g e n t f o r the cyclic process because e i ' the thermal shock associated w i t h cyclic heating and cooling of t'ze c a t a i l y s t bed and because of a tendency f o r the bed t o l i f t o r move . s l i g h t l y with cyclic flow changes. The asteria1 used a l m o s t ur2:iers a l l y i n commercial operation consists of fused spherss 1 / 2 t o 1 inch in diameter, of impure alumina (90%) having medium porosity ( W - ' + O $ ) . The c a t a l y s t i s prepared by impregnating the spherical s i a q o r t w i t h 2. nickel s a l t solution and then decomposing the nickel s a l t t o nickel oxide by heating i n air t o about 6 O O O C . Performance of the c a t a l y s t has been commerciall'y ecceptable for u t i l i t y operation, but the c a t a l y s t has decidedly short l i f e coqxred. t o c a t a l y s t s used in the continuous reforming process.25 Ordiczrily more than 50% of the o r i g i n a l c a t a l y s t a c t i v i t y i s l o s t after 2000 t o 4000 hours of operation w i t h one inch diameter c a t a l y s t . For 1/2inch diameter c a t a l y s t ,where r e l a t i v e light c a t a l y s t loading i s used, somewhat longer l i f e i s obtained. Plant capacity i s obviously a f f e c t ed by l o s s in c a t a l y s t activity,and replacement of a t l e a s t p a r t of the c a t a l y s t i s required annually. Efforts t,o improve c a t a l y s t performance in the cyclic process have recently become of considerable interest t o u t i l i t y compazies using the CCR process.z5 In addition, extension of tine CC?, p=.ocess t o l i q u i d hydrocarbon operation and the development of severel new e the gas manufacturing processes which incorporate i ~ . s o a forn principles of cyclic c a t a l y t i c steam reforming of hydrocarbons 'nave focused a t t e n t i o n on the performance of c a t a l y s t under c y c l i c condit i o n ~ . ~ ' ~ ~ ~ ~ , ~a t ~the' ' I ~n s~ In ~ a recent study t i t u t e of G a s Technology, the cyclic performance of several supported nickel c a t a l y s t s was investigated under closely controlled conditions i n the laboratory. It w a s the object of this study t o determiie the f a c t o r s unique t o the cyclic process which governed c a t a l y s t performance and were responsible f o r r e l a t i v e l y short c a t a l y s t l i f e . Significant res u l t s of this study a r e presented here.
EXPERIMENTAL

il

This study was limited t o commercial c a t a l y s t s containing approximately 5 w e i g h t er cent nickel. Support materials were e i t h e r fused alumina ( a -AL203p or fused periclase ( Q O ) All of the c a t a l y s t p e l l e t s were i n the form of nominal 1-inch diameter spheres except f o r one sample w i t h high-purity alumina support which consisted of irregu l a r l y shaped, 1-inch lumps. Only the lower p u r i t y alumina base c a t a l y s t contained magnesium oxide promoter. Properties of the

-42-

1
~

unused c a t a l y s t s a r e shown below. Catalys t Designation Acid Soluble Nickel, w t . 5 Magnesium Oxide Promoter, w t . $ Support Composition, w t . $ a203 si02 CaO Fe P e l l e t Shape
MgO;

5.14

1.81 86.85
12.90 0.25 Sphere

B 4.48 None

---

99c

-----

Irregular

Sphere

-0 . 3 - I - 95.5 -3.0 1.0 -0 . 2 _ L

C 4.99 None

5-16 None

J
j

Sphere

Catalyst n i c k e l concentration data appearing throughout this paper r e f e r t o t h a t n i c k e l portion of the c a t a l y s t which w a s soluble in n i t r i c acid. The n i c k e l content was determined by boiling a pound , (minus-100 mesh) sample w i t h concentrated n i t r i c a c i d u n t i l the disap pearance of brown rUmes, followed by f i l t r a t i o n and gravimetric determination by a standard dimethylglyoxime method. X-ray d i f f r a c t i o n patterns of the c a t a l y s t s were obtained by the Debye-Scherrer powder camera method. Reforming t e s t s were conducted i n the apparatus shown i n Figure 1. The reactor consisted of a 3.125-inch I.D. x 102-inch long, Type 310 stainless s t e e l tube with a c e n t r a l l y located 0.675-inch O.D. thermowell of the same a l l o y inserted through the bottom. The reactor tube was suspended i n a Smith a l l o y wound e l e c t r i c mrnace with f o u r independently controlled heating zones. The temperature in each zone was regulated by potentiometric temperature indicator-controllers i n comchromel-alumel thermocouples. The control thermocouple bination w i t h was welded t o the outside s k i n of the reactor tube a t the center of each heating zone. Additional chromel-alumel chermocouples located i n t h e i n t e r n a l thermowell were used t o measure t h e c a t a l y s t temperature at three points w i t h i n the bed and the feed gas stream temperature immediately before e n t e r i n g the bed. a s f e d by a Provision was made t o weigh d i s t i l l e d water,which w chemical proportioning pump through an e l e c t r i c a l l y heated steam generator t o the top of the r e a c t o r tube. N a t u r a l gas w a s fed from highpressure cylinders through a pressure regulator and gas meter t o the top of the reactor tube,where i t w a s mixed with the steam feed. Product gas was withdrawn from the bottom of the reactor tube through a water, cooled t u b e and s h e l l condenser,where excess steam w a s removed. The cooled product gas stream w a s measured w i t h a second gas meter before being discharged through a back- pressure regulator t o the sampling system. A l l measured gas volumes were corrected t o standard cubic f e e t (SCF) a t 6 0 O ~ , . 30 inches of mercury pressure, and saturated W i t h water vapor. A side stream of the product gas w a s collected f o r a recording calorimeter and f o r Orsat and mass spectrometer analyses. For each reforming t e s t a 0 . 1 cubic f o o t sample of c a t a l y s t was placed i n the lower two heating zones of the reactor furnace and formed a 24-inch deep bed. I n e r t periclase spheres were used below the c a t a l y s t t o properly space the bed i n the reactor tube. The apparatus as described w a s s u i t a b l e for conducting the continUOUs reforming process. For cyclic operation, additional equipment was required. Provisions were made f o r supplying and metering b o t h nitrogen ( f o r purging) and a i r in a manner similar t o the natural gas feed system. The n a t u r a l gas, nitrogen and a i r f e e d l i n e s were equipped . w i t h e l e c t r i c a l l y operated solenoid v a l v e s which were opened and closed by a repeat cycle sequence timer. The steam feed system, r e a c t o r and product gas handling system were i d e n t i c a l f o r cyclic

-43, . ,

GENERATOR

UAX GAS

OWELL

OD

NAT. GAS CYLl N DE R METER

PRQDUCT G A S M E T E R

CONDENSER

3.125"IKl-REACTOR

Fig. 1.- W L S A T O R Y FCZEWRJEUYG ITNIT

operation as f o r continuous reforming t e s t s . A simple switching system made i t possible t o change from continuous operation t o c y c l i c operat i o n without i n t e r r u p t i o n . Reforming t e s t s were long enough t o insure attainment of equili- i brium. The test period w a s normally about three t o f o u r hours a f t e r , steady state w a s reached. FACTORS INFLUENCING CATALYST A C T M T Y IN CYCLIC REFORMING I n the continuous reforming process the c a t a l y s t i s a t a l l times i n the reduced state and a t a constant temperature. 'ke major react i o n which i s a f f e c t e d by the condition of the c a t a l y s t i s t h a t between the hydrocarbon and steam:

Subsequent r e a c t i o n between p a r t of t h e product carbon monoxide and steam has l i t t l e influence on r e a c t i o n (1) or on the a b i l i t y of the c a t a l y s t t o promote reaction (1) :
CO

+ H20

The overall c a t a l y s t performance i s therefore governed by the a b i l i t y of the c a t a l y s t t o promote the steam-hydrocarbon reaction a t the temperature maintained i n the c a t a l y s t bed. Cyclic reforming i s an exceedingly complex process by comparison. Neither c a t a l y s t temperature nor composition of t h e atmosphere remain constant. I n a s l n g l e cycle the c a t a l y s t i s f i r s t heated,then cooled; oxidized, then reduced. The extent of these changes varies widely i n actual practice. I n some cases, combustion i s controlled during the heating portion of the cycle so that the combustion products contacting t h e c a t a l y s t a r e e s s e n t i a l l y n e u t r a l . The quantity of c a t a l y s t which i s oxidized in this instance i s small. Another method of operation employs a c e r t a i n amount of excess air i n the combustion products t o take advantage o f the heat released i n the c a t a l y s t bed by combustion of n i c k e l metal.7 Other forms of c y c l i c reforming require the passa e of undiluted air through the c a t a l y s t during the heating cycle.2g I n a l l methods o f operation, process heat i s stored i n the massive c a t a l y s t bed during the heating p e r i o d f o r use during t h e make period. The c a t a l y s t , therefore, undergoes a significant chznge i n temperature throughout the cycle. Because of the changes i n atmosphere throughout the cycle, other reactions besides r e a c t i o n (1)occur i n the c y c l i c process and are dependent upon the c a t a l y s t . Although the c y c l i c process i s not f u l l y understood, a t l e a s t two other reactions of importance are those betweer m e t a l l i c n i c k e l and oxygen and between nickel oxide and a reducing agent.
2Ni

COz

+ Ha

(2)

'

(4) Reaction ( 4 ) represents o n l y the most probable stoichiometric relationship,and i s not intended t o indicate the a c t u a l mechanism of reduction in the process. Other c o n s t i t u e n t s of the r e a c t i n g gases, such as sulfur compounds i n the hydrocarbon feed, and n i t r i c oxide in the combustion products used for heating,affect the process.

(2n

5) N i O . + CnHm -(2n

O2 - 2 N i 0

(3)

$ ! )

Ni

nCOz

+m 2

HzO

'

:
I

-wIt i s apparent that the o v e r a l l c a t a l y s t perfo,mace in t h e cyclic process i s governed by the behavior of the c a t a l y s t i n a11 o f the s i w i c a n t reactions which occur during the cycle, together w i t h the change i n c a t a l y s t temperature over the course of a cycle. The r a t e s of one o r more of these reactions may be controlling f o r a sel e c t e d cycle so that the process i s not s i n g u l a r l y dependent oi-i reaction (1)as i n the case of' continuous reforming. To determine the extent t o which the overali c a t a l g s t performame cyclic operation W f e r s f r o m continuous reforminga s e r i e s of cyclic tests were made in the laboratory r e f o m + @apparatus ; at react o r tube w a l l temperatures Prom 1400OF. t o lgOCoP. (Ta.ble. 1). The c a t a l y s t samples used were f i r s t determined to have approximately equal a c t i v i t y for contbuous reformtcg a t a s e t of standard a c t i v i t y t e s t conditions in the same apparatus ( s e e Table 2 ) . Tne cycle siml a t e d conditions f o r a process presently under development a t tbe Institute.28 After each c y c l i c test, the a c t i v i t y of the c a t a l y s t was a g a h checked a t continuous conditions t o a s c e r t a i n that n o Fermanent change i n the c a t a l y s t had occwTed during the course of t'ne tests. Overall- c a t a l y s t performance i s i l l u s t r a t e d by Figure 2, LL which the percentage of hydrocarbon conversion is p l o t t e d against reactor tube wall temperature. It c m be seen that al+&ough the 2ctivi t y of a l l of the c a t a l y s t samples was considered equivalent f o r tEe continuous process, under c y c l i c conditions their performance differed widely. mot ordy were the l e v e l s of hydrocarbon conversFon d i f f e r e n t f o r each catalyst,but +&e e f f e c t of temperatme on conversion v y i e d s i g n i r i c a n t l y f o r the a f f e r e n t c a t a l y s t s . Several general c h a r a c t e r i s t i c s of' c a t a l y s t behavior under cyclic conditions can be observed from the data i n Table 1 and Figure 2 . The magnesia-supported c a t a l y s t s used in t'nis study wepe l e s s S f f P C t i v e . f o r cyclic reforming th+n were the. aluIzlina-supported c a t a l y s t s . N q t . only d i d the magnesia-supported cata-lysts (curves C ( b ) and D,. Figure 2 ) give lower conversion a t equivalent tenperature and natural gas feed space velocity than did the alumina supported c a t a l y s t s (curves A and B), but the decrease i n conversion w i t h reduction in temperatme was also much more severe. There was a l s o considerable evidence that the portion of nickel entering i n t o oxidation and reduction reactions during the cycle w a s s e a t e r ? o r the a l m i n a supported c a t a l y s t s . This was indicated by the c a t a l y s t temperature, carbon dioxide-carbon monoxide r a t i o in the product gas, and agreement between the q u a n t i t i e s of water clecornposition calculated from nydrogen and oxygen balances, In several instances w i r h the alumina-supported ca-lalyst, especially a t high reactor tempera.-%mes (see Runs 19, 2 0 , 28 and 29, Table lj, the c a t a l y s t bed ternperatwe approached or exceeded the reactor tube w a l l temperatme indicating a sizable release of exothermic r e a c t i o n heat from oxidat i o n of metallic nickel. I n these same instances, the r a t i o of carbon dioxide t o carbon monoxide i n the product gas vas considerably 'higher than the carbon oxides r a t i o normally r e s u l t i n g from water-gzs shift reaction ( 2 ) . This i s due t o carbon dioxide formation a t the beginning of the cycle by reaction ( 4 ) i n which nickel oxide i s recluced t o metallic nickel. The quantity of water deconposition f o r t h e process calculated from both hydrogen and oxygen balances should agree i f only reactions (1) mri ( 2 ) were s i g n i f i c a n t . I n the runs where oxygen was transferred by the c a t a l y s t , the calculated quantity of steam decom2osition based on a hydrogen balance appears lower than actual. Cofiversely, the oxygen content of carbon dioxide formed b s reaction (4! w i l l make the balculated quantity of stem decomposition based on & ' I oxygen balance appear higher than a c t u a l . These observations i n
~

I!

-47-

REACTOR

TUBE

WALL

TEMP,

OF.

Fig. 2.-COMPARISON OF REFORMIFIG CATALYST PETIFORTWICE IN CYCLIC LABORATORY OPERATIOB

5.5

5.0

2
W X

4.5 4.0

z
J

v 3.3
3.0
I

m.
0

3 2.5

2.0

0 a

1.5
I

.o
0

0.5

20

40

60

80

1 0 0

EXPOSURE T I M E . HR.

Fig. 3 . D E C R ? U . S Z I N CONCE1\TTRATION OF ACID S0LWL;E NICKEL IN ALUMINA SUPPORTED CATALYSTS ON EXPOSlJKE AT 1 8 0 0 F IlI AIR

-48the case of the alumina-supported c a t a l y s t seem t o i n d i c a t e higher rates of oxidation and reduction of nickel thm occurred w i t h the magnesia-supported c a t a l y s t . It can a l s o be noted that the general l e v e l of hydrocarbon conversion and the apparent q u a n t i t i e s of nickel c y c l i c a l l y undergoin oxidation and reduction were higher f o r the high p u r i t y , unpromote% alumina-base c a t a l y s t (B) than f o r the lower p u r i t y alumina-supported, magnesia,promoted c a t a l y s t ( A ) . It is not possible t o d r a w conclusions regarding the effect of base impurities and p r m o t e r from these data, however. Overall c a t a l y s t performance as shown i n Figure 2 does not indic a t e the s p e c i f i c effect of individual c a t a l y s t properties, but rather r e f l e c t s the cumulative e f f e c t of the complex f a c t o r s Fnfluencing c y c l i c reforming operation which are not readily apparent from charact e r i s t i c s of the c a t a l y s t under continuous reforming conditions. O n e might conclude from t h e s e data t h a t high catalyst performance under c y c l i c conditions i s c l o s e l y related t o the s u s c e p t i b i l i t y of the c a t a l y s t t o oxidation and reduction of i t s nickel content. This con-. clusion would seem logical, since i t i s evident that the steam-hydrocarbon reaction (1)i s primarily dependent upon the presence of some form of metallic n i c k e l as the c a t a l y t i c agent. Recent work in t h e f i e l d of s o l i d state physics dealing w i t h the r e l a t i o n s h i p s between the electronic properties of semiconductor metal oxides and t h e i r c a t a l y t i c behavior has shed additional light on the f a c t o r s which influence the oxidation and reduction reactions of nicke1.4~10~12~3~~37y These 45~* fa ~ c~ to 4r7 s can be summarized as follows : Source of n i c k e l oxide. Nickel oxide prepared from different nickel salts can differ s i g n i f i c a n t l y i n chemicalpropert i e s . 2 J 27 Treatment of n i c k e l oxide. Temperature, pressure and atmospheric exposure have been demonstrated experimentally t o chamge the susce t i b i l i t y of nickel oxide t o reduction by hydrogen.29J18,g7 It has been found that the combination, sequence, duration and rate of change of treatment conditions are a l l important in determining i t s behavior during subsequent reduction.18 Thus the treatment of nickel oxide during preparation of the catalgs%,and the treatment received i n operation i n the c y c l i c process,are both important in determhlng its behavior. Promoters o r impurities i n the nickel oxide. Studies have shown t h a t the inclusion of foreign m e t a l ions i n the c r y s t a l l a t t i c e of nickel oxide has a pronounced e f f e c t on i t s chemical reactivity, including reduction t o m e t a l l i c nickel. 37 Support material. The s u s c e p t i b i l i t y of nickel oxide t o reduction by hydrogen is influenced by the e l e c t r o n i c nature of an adjacent materia1,even thou@ there is no chemical r e a c t i o n o r s o l i d s o l u t i o n between t h e two materials. 143*339

I t should be noted that the foregoing f a c t o r s are a l l r e l a t e d t o changes %or v a r i a t i o n s of, the c r y s t a l s t r u c t u r e of nickel oxide. Present-day considerations i n the manufacture of supported nickel reforming c a t a l y s t s include nickel concentration and d i s t r i b u t i o n , support porosity, surface area and the mechanical properties of the support. The s e l e c t i o n o f promoters i s based l a r g e l y on i n i t i a l a c t i v i t y considerations. Little a t t e n t i o n , however, i s given t o

-49-

conditions c o n t r o l l i n g tne c r y s t a l structure of the nickel and nickel oxide. In view of the significance of the oxidation-reduction re-

In the continuous reforming process the c a t a l y s t , as long as i t i s not poisoned by sulfur o r similar impurities in the feed stream, apparently has unlimited l i f e . Since this has not been the case in commercial operation of the cyclic reforming process i t was suspected t h a t l o s s of c a t a l y s t a c t i v i t y i n cyclic use was r e l a t e d t o the opera t i n g differences i n the two processes already c i t e d . These a r e c y c l i c v a r i a t i o n in atmosphere and c a t a l y s t temperature which are not encountered i n continuous reforming. All types of c a t a l y s t employed i n the cyclic process have been subject t o permanent l o s s i n a c t i v i t y , although not n e c e s s a r i l y a t the same rate. For example, 1-inch d i a m e t e r aiunina-base promoted c a t a l y s t , and 1-inch diameter magnesia. bese unpromoted c a t a l y s t have been observed t o l o s e a major p a r t of t h e i r a c t i v i t y a f t e r 2000 t o 4000 hours of operation. O n the other hand, some 1/2-inch alumina base unpromoted c a t a l y s t has been reported t o have given e s s e n t i a l l y constant performance f o r about 20,000 hours. 25

- E
Since the atnosphere i n the continuous reforming process i s allrays reducing, an investigation w a s made of the e f f e c t of an oxidizing atmosphere on the c a t a l y s t . A s e r i e s of tests w a s made i n the laboratory reforming apparatus w i t h both magnesia-supported and elumina-supported c a t a l y s t . Each c a t a l y s t was treated for 4 t o 6.5 hours i n a stream of a i r a t 1700 and 1900OF. reactor tube wall temperature. Before and a f t e r each exposure, an a c t i v i t y test was made a t comparable continuous reforming conditions. Results of these tests a r e shown i n Table 2 . With the alumina-supported c a t a l y s t (A, Tests 4 , 5 and 6), l i t t l e change i n reforming a c t i v i t y w a s observed even a f t e r exposure a t lgOOF. Chemical analysis of the c a t a l y s t following f i n a l exposure indicated a n i t r i c a c i d soluble nickel concentration o f 4.08 weight per cent, compared t o 5.14 weight per cent before exposure. An x-ray d i f f r a c t i o n p a t t e r n of the c a t a l y s t a f t e r exposure contained l i n e s corresponding t o n i c k e l aluminate ( s p i n e l ) which were not present in the pattern of the new c a t a l y s t . It appeared that some of the nickel content o f the c a t a l y s t had formed an insoluble spinel compound with the support material during the period of exposure,but that the reduction in concentration of available nickel was not s u f f i c i e n t t o s i g n i f i c a n t l y a f f e c t the reforming a c t i v i t y . , With the magnesia-supported c a t a l y s t ( C , Tests 1, 2 and 3 ) , no reduction in reforming a c t i v i t y was observed a f t e r exposure a t 17OO0F., b u t following exposure a t 1900OF. the l?eform?ing a c t i v i t y was approxi n a t e l y 2 / 3 that of the o r i g i n a l c a t a l y s t . Chemical analysis indicated a n i t r i c a c i d soluble nickel concentration of 4.63 weight per cent following f i n a l exposure compared t o 4.99 weight per cent i n the new c a t a l y s t . The x-ray d i f f r a c t i o n p a t t e r n o f the exposed catalyst

-51-

contained only l i n e s corresponding t o periclase (Tilgo). With the magnesia-supported c a t a l y s t , exposure i n a i r a t elevated temperature had a more serious e f f e c t on reforming a c t i v i t y than in tne case of the alumina-supported c a t a l y s t , althou& the reduction i n n i t r i c a c i d soluble nickel content was less, and no compound formation w a s detected f o r the magnesia c a t a l y s t . AE exainination of the phase d i a g r m f o r the nickel oxide-magnesium oxide s g ~ t e t n does ~ ~ ' not ~ ~ indicate tne e x i s t ence of a compound; however, Eolgersson and Karlsson confirmed the existence of a completely miscible system of s o l i d solutions between these two materials.15 Since both nickel oxide end magnesium oxide have face-centered cubic l a t t i c e s w i t h nearly the same unit c e l l length, low concentrations of s o l i d solution, if present, would probably not be detectable in the x-ray d i f f r a c t i o n pattern. Although the s p e c i f i c r e l a t i o n between c a t a l y s t changes and reforming a c t i v i t y w a s not defined i n these tests, the indications f i t the circumstances. In the case of the alumina-supported c a t a l y s t , combination of the nickel content w i t h the alumin2 support t,o form spinel would be expected t o reduce reforming a c t i v i t y , s i n c e s p i r e 1 compounds a r e very unreactive,uld reduction t o metallic nickel wi-ch hydrogen or hydrocarbons would be unlikely. I n the case of the nrgi ? t'ne unnesia-supported c a t a l y s t , solution of the nickel content i reactive magnesium oxide l a t t i c e could e a s i l y m a k e the nickel unavailable f o r reduction by hydrogen or hydrocarbons. Both cases vould be unique t o the cyclic reforming procesg since e i t h e r would proceed oLnly w i t h nickel present as nickel oxide. Spinel Additional tests were conducted t o determine if r;he rete of spinel formation w a s rapid enough t o contribute materially t o loss of catel y s t a c t i v i t y a t conditions of the cyclic reforming process. m l e s of b o t h the high p u r i t y alumina-base unpromoted c e t a l y s t (B), and The lower p u r i t y alumina-base magnesia-promoted c a t a l y s t (A), were crushed t o pass 100 mesh. Each c a t a l y s t was heated t o 1800~.in ul e l e c t r i c furnace for an extended period in contact w i t ; ? a i r . A t i n t e r v a l s , samples were withdraxm f o r determination of n i t r i c a c i d soluble nickel concentration. The change in n i t r i c a c i d soluble nickel content wizh time of exposure i s shown i n Figure 3 . ATter 96 hours a t the t e s t conditions both c a t a l y s t s showed a marked decrease i n the concentration of a c i d soluble nickel. It i s i n t e r e s t i n g t o note the difference in r a t e of decrease f o r these two c a t a l y s t s . In curve B f o r the high-purity alumina support the rate of decrease was almost a l i n e a r flmction of time. Since only nickel oxide and alumina were present i n t h i s c a t a l y s t , nickel aluminate would be the only suspected compound which would l o w e r the concentrat i o n of a c i d soluble nickel. This w a s c o n f i n e d by the x-ray d i f f r a c t i o n pattern of the c a t a l y s t after exposure f o r 96 hours. O r d g l i n e s Corresponding t o a - a l u a i n a , nickel oxide and spinel were detectsd. No spinel l i n e s were observed in the p a t t e r n f o r the new c a t a l y s t . Curve A f o r the lower purity alumina base c a t a l y s t shows a d i f f e r e n t t h e nickel concentration relationship. The decrease in a c i d soluble nickel content was very rapid i n i t i a l l y , but approached &\e r a t e in curve B after about 30 hours. The major materials present in t h i s c a t a l y s t in addition t o alumina and nickel oxide were 12.90 weight per cent s i l i c a i n the support'and 1.8 weight per cent magnesia added as promoter by coimpregnation as n i t r a t e w i t h the nickel. An x-ray dlffraction pattern of this c a t a l y s t b d i c a t e d t h a t m e silica w a s in the form of aluminm s i l i c a t e (jAl2O3.2SiO.02). In addition t o spinel lines, the pattern of the c a t a l y s t after 96 hours of exposure contained unidentified l i n e s corresponding t o 5.4, 4.7 and 1.84A. !These do

-52-

not correspond t o any nickel compound f o r which x-ray data a r e readily available. The difference between curves A and B could be a t t r i b u t e d e i t h e r t o the formation of insoluble nickel compounds i n addition t o nickel aluminate,or t o an increased rate of formation of nickel aluminate due t o the presence of s i l i c a o r magnesia.

>

Solid Solutions 1 To determine the relationship between c a t a l y s t a c t i v i t y and the possible formation of s o l i d solutions between nickel oxide and the magnesia c a t a l y s t support as indicated by the atmosphere t e s t s described above, a number of samples of the same magnesia-supported catalyst , C having various operating h i s t o r i e s and showing d i f f e r e n t levels of a c t i v i t y under comparable continuous reforming conditions were subjected to microscopic analysis. Descriptive data f o r the samples selected are itemized below:
IGT Sample No. Acid Soluble Nickel, w t $ Approximate R e l at i v e Continuou s Reforming Activity, $ History

3 %
N e w
100

3664
4.47
1 20

1 6 1 5
4.63
70

I 3665'
4 . 0 1

29 hours Oxidized P i l o t Plant in Tube Operation Furnace a t 1400f o r 9.5 1660OF. hours a t 17001900 F.

15 29 hours Pilot Plant Operatic a t 1830-

1890~~.

For each sample t h e catalyst p e l l e t was impregnated w i t h Canada balsam,after which a t h i n section through the center w a s prepared. Photographs of the sections made w i t h transmitted l i g h t are shown in Figures 4 , 5, 6 and 7. The original pore s t r u c t u r e of the support material w a s preserved and shows as white areas i n these photographs. I n Figure 4 , the new c a t a l y s t , two d i s t i n c t phases a r e apparent; a colorless i s o t r o p i c phase having an index of r e f r a c t i o n s l i g h t l y below that of a 1.74 immersion medium ( r e f r a c t i v e index of periclase: 1.734-1.737) and a black opaque phase ( n i c k e l oxide). The nickel oxide i s concentrated near the surface of the p e l l e t and especially around the perimeter of the l a r g e r pores. P a r t i c l e s i z e of the peric l a s e ranged from a maximum of approximately 50 microns down t o a very f i n e dust. The n i c k e l oxide p a r t i c l e s i z e wastoo s m a l l t o be determined Figure 5 shows a section of the c a t a l y s t a f t e r brief cyclic operat i o n i n a p i l o t p l a n t a t r e l a t i v e l y l o w temperature. T h e activity under continuous reforming conditions i n the laboratory apparatus was found t o be noticeably higher than that of the new c a t a l y s t . Two phases a r e apparent, as i n the new c a t a l y s t . However, the nickel oxide seems t o be more uniformly d i s t r i b u t e d throughout the pellet,and instead of being concentrated around the pore perimeters i t appears t o have d i f h s e d i n t o a t h i n l a y e r i n the grain boundary surrounding each periclase c r y s t a l l i t e . Some grains have a r e f r a c t i v e index s l i g h t l y above 1.74, whereas others have a r e f r a c t i v e index lower than 1.74. Since nickel oxide has a r e f r a c t i v e index of 2.37, the grains w i t h a r e f r a c t i v e index higher than 1.74 undoubtedly contain nickel oxide i n s o l i d solution. S o l l d solution formation i n t h i s sample i s evidently in i n i t i a l stages and far from complete. Figure 6 shows a section of the c a t a l y s t following a i r treatment i n the laboratory reforming apparatus a t 1 9 0 0 ' F . reactor tube w a l l

-53-

a.

4.-NEW CATAIYST C, TIVE ACTIVITY = 10%

Fig. 5.-USED CATATXST C,

FELATIVE ACTIVITY
=

124
... . , .

Fig.

6.-Tf(EATED CATALYST C , ACTMTY = 7 4

Fig. 7.-USED CATALYST C, FIEUTIVE ACTIVITY = 15%

-5Ltemperature (Test 3 , Table 2 ) . This sample w a s found t o have approximately 2 / 3 of the a c t i v i t y of the new Catalyst. Two phases can be detected i n t h i s sample,although the amount of black opaque phase has g r e a t l y diminished. Only a few grains were observed w i t h an index of i r e f r a c t i o n less than 1.74, indicating a substantial amount of s o l i d solution formation. Figure 7 shows a sample o f c a t a l y s t which was subjected t o high temperatures i n c y c l i c p i l o t plant operation. The a c t i v i t y of t h i s material under continuous reforming conditions was very low, estimated a t about 15% of the new c a t a l y s t . Only one phase can be detected in t h i s sample, the nickel oxide phase having completely disappeared. Ref r a c t i v e indices of a l l grains were w e l l above 1.74, indicating that . s o l i d solution of the nickel oxide i n magnesia w a s e s s e n t i a l l y complete. No grains with an index of r e f r a c t i o n greater than 1.78 were observed, indicating that the solutions formed were always d i l u t e i n nickel oxide. The l a r g e s t p a r t i c l e s i z e i n the sample was approximately four times that in the new c a t a l y s t , indicating that some recrystallization , may have occurred. This series of analyses seems t o indicate the path of the change which occurs i n the magnesia-supported c a t a l y s t w i t h use. It i s int e r e s t i n g t o note that the e a r l y changes i n s t r u c t u r e corresponding t o nickel oxide migration throughout the pellet,and incomplete s o l i d solution formation,are accompanied by an increase i n reforming a c t i v i t y r a t h e r than a decrease. This r e s u l t agrees w i t h the findings of HUttig w i t h mixtures of metal oxide cata1yst.l Htlttig describes a s e r i e s of states t h a t occur during the formation of a s o l i d solution from a mixture of metal oxides. The i n i t i a l states constitute diffus i o n of the more mobile l a t t i c e over and around the less mobile l a t t i c e . IPuttig relates the highly disordered condition which i s created t o increased c a t a l y t i c a c t i v i t y . The f i n a l state,constituting the complete s o l u t i o n and f i l l i n g i n of l a t t i c e defects, he r e l a t e s t o W i s h i n g catalytic activity.
i

CONCLUSIONS
It appears from the r e s u l t s of this study that the f a c t o r s which influence o v e r a l l performance of the c a t a l y s t i n the cyclic reforming process,as well as c a t a l y s t life, a r e q u i t e d i f f e r e n t from significant f a c t o r s f o r the continuous reforming process. For this reason, catal y s t properties which have generally been considered desirable in continuous reforming c a t a l y s t s may not be s a t i s f a c t o r y for the cyclic process. Likewise, the methods commonly used t o evaluate catalyst performance for continuous reforming do n o t appear adequate t o predict the behavior of c a t a l y s t s under cyclic conditions. Although the c y c l i c process i s extremely complex, the data from t h i s study lead t o several general conclusions regarding performance of supported nickel c a t a l y s t s of the type employed i n commercial practice. F i r s t , the s u s c e p t i b i l i t y of the nickel content of the c a t a l y s t t o rapid oxidation and reduction appears t o be closely rel a t e d t o good o v e r a l l refortning efficiency i n cyclic operation. Second, c a t a l y s t a c t i v i t y i s decreased by any changes which r e s u l t i n decreasing the amount of n i c k e l available f o r oxidation and reduction. For alumina-supported c a t a l y s t s a major f a c t o r contributing t o l o s s in a c t i v i t y appears t o be the formation of spinel by reaction between nickel oxide and the support. For magnesia- supported c a t a l y s t s a similar e f f e c t appears t o r e s u l t from solution of n3.ckel oxide i n the support. In both instances, the undesirable changes must occur during the heating portion of the cycle,when all or p a r t c/f the nickel i s present as the oxide.

i i>
I

-55-

Many other f a c t o r s may contribute t o s h o r t c a t a l y s t l v k ,i n t'ile One f o r vhich very l i t t l e specif'$ ckta \ a r e available i s the e f f e c t of coatings b u i l t up on the surface of 4 the c a t a l y s t w i t h use,from dust constituents in the combustion air, o r from ash content of o i l feed o r f u e l . Aaditional studies w i l l be ' necessary t o define these eff'ects,as well as t o determine methods fc,r improving c a t a l y s t performance and l U e .

1 cyclic reforming process.

AKXO'rnrnhT

' Plan. by the i Belyea ' edged.

J !

Permission f o r publication of this paper and the p i d u l c e given Project Supervising Comi-ttee under the c h a i m m s h i g of' A. 2. and with the assistance of N. K. Chaney are g a t e r ' u l l g 2ckric:.?lThe authors wish t o express t h e i r a g p e c i a t i o n t o the meny staff members of the I n s t i t u t e of cas TPchnology who contributes to this work. The counsel of H. R. .Linden, Research Director, ?,iEs especially helpful. J. S. G r i f f i t h of the &mom Research Fowkztion o f the I l l i n o i s I n s t i t u t e of Technology conducted the petrogagiiic analysis of c a t a l y s t s ( t h e r e s u l t s of which are presented i n t i n i s paper). Catalyst samples f o r this study were supplied by Catalysts , and Chemicals, Inc., The Harshaw Chemical Company and Westvaco itGiisr2l ( Products Division of the Food Nachinery and Cheaical Coqoration.

T h i s study w a s conducted under the sponsors'hip of Gas ' Operations Research Cotmittee of the American Gas Association wi-ih 5 funds provided by the Association' Promotion-Advertising-iiesearch
the s

1
'\

LITERATURE CITED

Barnes, J. M. and Bergex, E. S. , "Eastern U t i l i t y I n s t a l l s Cat- , a l y t i c Reforming U n i t s , Gas 3 29-42 (1957) December. The Reduction of Nickelous Benton, A. F. and F m m e t t , T Oxides by Eydrogen," J . h e r . Chem. SOC. 46, 2728-37 and (1924). (3) Calderwood, G. L., "The 'ONIA-GEGI' Process f o r Cyclic Catalytid Cracking of Liquid Hydrocarbons," Paper CXP-38-11, 0Fratb.g Seck t i o n , American G a s Association, 1958. ( 4 ) de Boer, J. H. and Vervey, E. J. W . , "Sernicondfctors with P a r t i a l l y and Completely F i l l e d 3d-Lattice Bands, Proc. Phys. SOC. ' (London) 49J 59-73 (1937). Dirksen, H. A. , Linden, H. R . and Pettyjohn, 5. S., "CFacking Catalyst A c t i v i t y i n t h e Presence of Hydrogen Sulfide, I n s t i t u t e of Gas Technology Research B u l l e t i n No. 4 (1953) February. ' Eiclaneyer, A. G. and Marshall, W. H., Jr. , "Aihonia 2ynthesis Gas Generation b Pressure Reforming of Natural Gas, Chern. En?. Prom-. 51, $18-21 (1955): Er?c!csEny H . T . and Hartzel, F. W. ( t o United Gas pprovenent G o . ) , Msthod of Carrying out C a t a l y t i c Reactions, U. S. Patent 2,759,505 (Augus; 21, 1956). Fensha;, P. J . , Solid Solutions of Lithium Oxide i n Nickel Oxida, J . Amer. Chem. SOC.2, 969-71 (1954). Foex, M., "PropeFty Anomalies Accompanying A-Type Transformations Bull. SOC. Chim. France, 377-79 ($952). of Nickel Oxide . E. 'Adsorption and Catalysis on Oxides, Discussions Garner, W Farada SOC 6 211-15 (1950). &.s)"Ney Developments i n the U . G . I . Cyclic Catalytic Reforming Process, A.G.A. Proceedings-1955, 829-33. P i v i t y and Catalysis Gray, T. J. and Darby, i n t h e Kickel Ozide System,';' 'J. Phys Chem. 60, 209-17 ( 1956) gperating Experience with t h e C.C.R. Set a t Griggs, E. R . , Paterson, N. J . , A.G.A. Proceedings-1954, 740-48. H i l l , F. PJ. and Selwood, P. W., "Structure and A c t i v i t y of Sup orted Nickel Catalysts," J. Amer. Chem. SOC.71, 2522-29 (19i9). Holgersson, S. and Karlsson, A., "X-Ray ExaFination of a System of Mixed C r y s t a l s with Monoxide Components, Z . Anorg. Allgem. Chem. 182, 2 5 - 7 1 ,(1929). z f i m , S. W., R e f o F n g Natural Gas by the Continuous Cata l y t i c Crac!dngllProcess, A. G.A. Proceedings-1951 , 591-601. C a t a l y t i c A c t i v i t y and Composition o f Oxide Huttig, g. F., Systems, Discussions Faraday SOC. 8, 215-22 (1950). Igaki ,lfK., "The E l e c t r i c and MagnetTc Properties of Nickel Oxide, B u l l . Naniwa Univ. g, 113-23 (1955). Innes, W. B . , "Catalystl,Carriers , Promoters' Accelerattrs, Poisons and I n h i b i t o r s , i n Ehtnett, P. H. , Catalysis, V o l . 1, pp. 245-314, Reinhold, New York, 1954. Janeway, P. W. , Jr. ( t o The Unittd Gas Improvement Co.), "Method f o r Making Gas Rich i n Hydrogen, U. S. Patent 2 , 7 4 3 , l n (April 24, 1956). King, G. R., " I n s t a l l Modern Catalytic Reformers," A.G.A. Monthly 37, 33-36 (1955) Septeplber. o r Regenerative Catalytic ReRnoblock, K. D. , "Duluth Plant : forming of Liquid Hydrocarbons, Paper CEP-58-5, Operating Section, American Gas Association, 1958. Landholt-Bornstein, "Physikalish-Chemische Tabellen," 3rd Erg., V O l . 1, p. 453, Springer Verlag, Berlin, 1935.

k:

-57Levin, 3. M., ;flr,pt]urdie, 5. F. zpd 3211, F . p . , "':?ase -. ~ ~ ~ ~ - p ~ ; : : s f o r Ceramists, 'I p . 54, T'ne -td3ericzn Zer.-nic ~ c c i e t y ,~ ~ o i c . " -, u~ Ohio, 1955. Milbourne, C . G . , "CCR Catalyst Ikvelo~ments," Pecer p ~ s - . n i a d a t A.G.A. Operating Section Com%renca, Mzg 0, 1 3 : , 3 , ?!eT:; Yo-&. Milbourne, C . G. and G l o ~ e r ,C . B . , "Cyciic Cet3lyi;ic 3efo:mizg Process," Ind. %ng. Chen. 49, 337-p (1.957). Prasad, M. and TendulkarA 31. G . , Tne P r e w r e t i o n 2nd E)?o-r;art i e s of Nickelous Oxide, J. Chern. Soc. 1403-07 (1931). Reid, J. E . , Bair, W. G. and Llndzn, 3 . R., "Prcductic-?_of 3iz5B t u O i l Gas i n a Cyclic-Regenerative Tilot Unit with XySrogen 2s Carrier Gas, I! Paper CEP-57-14, O?erating Section, Amriczn Gas Association, 1927. Reidel, J . C . , pmrnonia Nmufacture by G a s Reromiiig, O f 1 52s T. 54, 106-07, 110, 113 (1956)March 26. R i e s z , C. H., Dirksen, R. A. a2d Er'Lcpatrick, X. J . , "Sulfur Poisoning of Nickel C a t a l m t s , I n s t i t u t k of Gas Technology 11951) Septen'aer. Research B u l l e t i n No. R i e s z , C. H., Dirksen, d. A. and P l e t i c k a , Y. J., " I q r c v e n e n t o f Nickel Cracking Catal sts, I m t i t u t e or" Gas Tschnoiozi 2sseerch B u l l e t i n No. 20 (1952y October. R i e s z , C. H,.,Komarewsky, V. I . , Kane, L. J . , E s t e s , F. zrC! L u r i e , P., Catalytic Gasification of Eigher R~-drocarbors, I' A.G.A. Monthly 28, 159-64, 190 (1946). E i e s z , C. H., Lurie, P. C . , Tsaros, C . L. and Pettgjohn, 3. S . , P i l o t Plant Catalytic Gasificatlcn of ETdracarisons , I 1 I n s t i t u t a of Gas TechnoloallResearch S u l l e t i n W . - (1923)July. Schlegel, C. A., Thz UGI Cyclic C a t a l y t i c i ? e r o i ~ - ~ n ??roczss g 2sforming Natural Gas, Paper presented a t E e w %nzlend Ges Asss.-,lst i o n , Operators ;ection ?{eeting, January 18, 1359, Bostoc, Schlegel, C . A . , . Try Cyclic C a t a l y t i c Ref'Gm-ing P r o c e s s f o r ieforming Natural Gas, 26, 26-29 (1.350)FebruarS. Schwab, G. M . , l , " " ~ e r i m e n t ~ C o n n e c t i nSmiconductcr g ?ro;srtiss and Catalysis, i n Conference on t;e Physics o f Sazicozductor I' Surfaces, R . H. Kingston, Editor, Semicoaductcr Surfsce P'.;;sks, pp. 283-96, Univ. or" Pannsy1v;txh Press, Fhilsdelphis, 1957. SchTiaS, G. PI. and Block, J., Oxidation of Carbon ??CDGXT& zzd Decomposition of Nitrous Oxide on Defined SerLconductins Cxides:, '' Z . PhTrsik. Chem. 1,42-62 (1954). Selwobd, P. W., H i l i , F. N. and Boardman, H-., "A B e w 1:iagnettic Effect Applied t o the Structure of C a t a l y t i c a l l y Active Soli",s," J . h e r . Chem. SOC. 68, 2055-59 (1946). Selwood P . irl. a i d L y z , L., "Structure of Oxide Catalyst Sgs terns; 'I Discussions Faraday SOC. 8, 222-30 (1950) . Shearonh I { ,. H., Jr. and Thom?son, 9 .L., " - h n o n l u a t lCC0 Atr;los-pheres, Ind. Eng. Chem. 44, 254-64 (19522. Sherwood, P. W., "Water Gas from Methane, Petroleum P~ocessing,
"

10

Smoker, E. H., The UGI .Cyclic C a t a l y t i c Refoming Process.-Recent Developments," A.G.A. Proceedin,cs-1952, 543-44. Smoker, E. H., '%xperiences i n ::he YrcducLion of High 3tu Gas from FTaptha $nd Kerosene Using t h e UGI Cyclic Catalytic. Refom-i n g Process, A.G.A. Froceedings-1953, 852-53. Taussig, J . El., Jr. ( t o The Urii.tedllGas Improvement Go,..),. "Method of Reforming Gaseous Hydrocarbons, U ' . S. Patent 2, 665,979 (January 12, 1924). Taylor, H. S . , Catalysis : Retrospect aEd Prospect-," D?-scusslcris Faraday Soc. 8, 9-18 (1959).

5, 1308-11 ( 1 9 5 : ) .

(46)

Vol!!censhtein, T. F., "T'ne 14echanism o f Catalytic Action by Semiconductors, I z v e s t . Almd. Piauk. SSSR, Otdel, LRi~. Naul;. (1957) 924-28,No. 8. (47) Wright, R. IT. and Andrews, J . P . , "Temprature Variation of the Z l e c t r i c a l Pro?erties of Nickel Oxide," P r o c . P h p . Soc. (London) 624, 440-55 (19'i-9).

I4

i'

-59

ABSTRACT P.PPLYIT?G FUPIDAiVZNTfi CGNCFPTS

TO THE ETTGDTEZ3ING DESIGN OF PLPLIUTCE 5WJEFL3

by Walter B. Kirk and James C . G r i f f i t h s American Gas Association Laboratories Cleveland, Ohio

Fundmental concepts r e l a t i n g burner desi,g f a c t o r s azld ;as ccmpositicn t o t h e conventional flame c h a r a c t e r i s t i c s of l i f t i n g , yellow t i p p i q , and f l e s h back are discussed. Design r e l a t i o n s h i p s a r e developed from t h e basic c r l t i c a l bcu??ar)r v e l o c i t y gradient theory with respect t o l i f t i n g f l a n e a . Such burner desizn f a c t o r s a s p o r t size, depth, and spacing; p o r t loading an& p r i n a n j zeracion; and gas conposition are taken i n t o account.

Not f o r Publication Presented Before t h e Division of -Gas and Fuel Chemistry American Chemical Society Chicago, I l l i n o i s , Meeting, September 7-12, 1958 APPLYING FUND.U~ALCONCEPTS TO TEE EECINEXFIING DESIGN OF APPLIANCE BUIWEBS by Walter B. Kirk and James C . G r i f f i t h s American Gas Association Laboratories Cleveland, Ohio There are many a r e a s i n t h e g a s industry where t h e r e i s a d e f i n i t e need t o bridge t h e gap between fundamental concepts and engineering application. A s a n example, t h e design o f atmospheric g a s burners f o r domestic appliances requires a judicious balance o f o_oposing c h a r a c t e r i s t i c s . A design d i r e c t e d totrard elimi n a t i p z any propensity for l i f t i n g flames might introduce a tendency f o r yellow t i p p i n g flames, or a design f o r high primary a i r i n j e c t i o n could be l i a b l e t o flashback. A Paoxledge o f t h e c u a n t i t a t i v e e f f e c t s of a l l p e r t i n e n t variables is, therefore, necessary t o echieve a d e s i r e d balance i n design. The following i s an aztempz t o i n t e r p r e t fundamental concepts i n such a manner that they n i g h t b e e a s i l y a2plied i n t h e engineering design of burner p o r t areas.

. me n a t u r e of t h e port area f o r any given a p p l i c a t i o n i s d i c t a t e d by l i m i t i n g condizions which produce c r i t i c a l flame c h a r a c t e r i s t i c s such a s 1iL?ing, yellow tipping, and flashback. k t y p i c a l flame c h a r a c t e r i s t i c diagram is shown i n Fi5gxe 1, which d e s c r i b e s l i m i t i n g conditions i n terms of primary zeration, e x g e s s e d as a per c e n t of t h e a i r required i n a stoichiometric mixture and port l o a d i w i n 9 t u per hour per square inch of p o r t area. Any change i n variables such a s port size, depth, or spacing w i l l d i s p l a c e these l i o l i t curves so as t o e i t h e r increase or decrease t h e a r e a of t h e s t a b l e , b l u e flame zone. Ideally, of course, t h i s zone should be a s l a r g e a s possible.
The flame c h a r a c t e r i s t i c diagram f o r a given burner operating on a given gas i s fixed. If at any time t h e operating point of t h e burner, a s defined by p o r t loading and pri-mary aeration, l i e s above t h e l i f t i n g l i m i t c w e , l i f t i n g flames xi11 occur. !.ken t h e operating point l i e s below t h e yellow t i p limit, t h e flames v i l l e x h i b i t yellow t i p s , and flashback w i l l t a k e place i f t h e operating point l i e s t o t h e leIft of t h e flashback l i m i t curve.

!Be general effects of burner design v a r i a b l e s and gas composition on t h e s e l i m i t i n g flane c h a r a c t e r i s t i c s have been known f o r many years. Lifzing tendencies are reduced by t h e use of larger, deeper, more closely spaced ports. Y e l l o v t i p p i n g of flames, on t h e other hand, i s reduced by using smaller, more widely spaced p o r t s . The use of several rows of p o r t s generally raises t h e yellow t i p l i m i t , but does not affect l i f t i n g tendencies, provided even d i s t r i b u t i o n of air-gas mixture t o all t h e p o r t s is provided. P o r t depth does not appear t o have a n effect on yellow t i p g i n g tendencies. Flashback tendencies appear t o be reduced by t h e use of smaller, eeeper p o r t s . The use of c l o s e r p o r t spacings apparently results i n a h o t t e r operating burner head temperature, which i n t u r n promotes flashback. A t t h e same a i r - g a s mixture temperature, hoirever, port spacing does not appear t o a f f e c t flashback.
Burning v e l o c i t i e s of f u e l gases a f f e c t l i f t i n g and flashback tendenc i e s . Slower burning gases are more c r i t i c a l i n regard t o l i f t i n g flames, while flashback occurs more r e a d i l y with f a s t e r burning gases. Esch f u e l gas has i t s own yellov t i p p i n g c h a r a c t e r i s t i c s .

-61L i f t i n g and flashback a r e flame s t a b i l i t y phenonem. As such, they a r e dependent on t h e r e l a t i v e v e l o c i t i e s o f a i r - g a s ci+-ure floiring out of t h e port, and t h e counter propagation of flame i n t o t h e air-gas nixcure i n t h e p o r t s an5 burner head. !,lith normal flames, equilibrium betveen These o??osing f a c t o r s gene r a l l y t a k e s Glace a short d i s t a t x e .above t h e burner port. T h i s d i s t a n c e betveen t h e burner port and t h e base o f t h e flame i s generally r e f e r r e d t o 8 s t h e "Ceaa space". Yellow tiGping, on t h e other hand, i s a completely d i f f e r e n t phencmenon. Each f u e l gas r e q u i r e s a c e r t a i n amount of a i r t o conpletebj e l j a i n 2 t e yellosi tios from appearing i n i t s flames. Tnis a i r can be o b t a i n e i e i t h e r a s prLnar?j e i r o r a s secondary a i r d i f f u s i n g t o t h e point ?here yeLLo,r tips appear i n t h e flanes. Port size, spacirg, and t h e number o f rows of ~ o r t s a f f e c t t h e degree t o x h i c i secondary a i r can be u t i l i z e d t o eliminate yellow t i p o i n z . This, i n turn, determines t h e primary air necessary t o make up a deficiency of seccncarj a i r , and hence e s t a b l i s h e s t h e so-called yellow t i p l i m i t of t h e burner. m e n s i v e s t u d i e s of f l v n e c h a r a c t e r i s t i c s and burner desi,m have been made a t t h e A.G.A. Laboratories, tkie U . S.. Bureau of Mines, and o t h e r instizu2ior.s. I n February, 1958, A.G.A. Laboratories Research 3 u l l e t i n ' 77, "IEfluerce cf ?art Design and Gas Composition on Flame C h a r a c t e r i s t i c s of Atmospheric 3uiTLers",xis published. 'This b u l l e t i n not o n l y describes recent burner design stu&ies, but c o r r e l a t e s t h e r e s u l t s of those s t u d i e s with t h e results of previous inveszizat i o n s . For t h e purposes of t h e present discussion, only l i f t i n g c h a r a c t e r i s t i c s of flames w i l l be consiidered. The c r i t i c a l boundaFj velocity gradient theory was u s 4 as a b a s i s o f studies of l i f t i n g flames. This theory states t h a t b l o y o f f ( l i f t i 7 4 ) ;rill occur when t h e flow velocity gradient a t the.wa3-l cf a port exceeds t h e burning velocit:: a t a l l p o i n t s around t h e port. The expression f o r t h e boundary velocity gradieEt, g, i s derived by equating the pressure drop per u n i t le=th of channel t o t h e retardim viscous f o r c e a t the w a l l per unit length of channel:

where:

/ i = viscosity, pound second per sauare f o o t , g = boundary v e l o c i t y gradient, l/sec, R = p o r t radius, f e e t , (Ap/A) = pressure drop p e r u n i t length of port, pounds 'per square f o o t p e r foot, = mass density, slugs per cubic foot, f = f r i c t i o n factor, dimensionless, and V = volumetric flow rate, cubic feet p e r second.

Introducing Reynolds' number as R e = 2 V e P T ( R , equation (1)becomes:

'

f V Re

166R3

-62Equation (2) is, therefore, a generalized equation f o r t h e boundary v e l o c i t y g r a d i e n t . S u b s t i t u t i n g t h e Hagen-Poiseuille r e l a t i o n s h i p f o r parabolic (laminar) flow, f = 6 4 / ~ e , i n equation (2) obtains:

T h i s i s t h e expression f o r t h e boundary v e l o c i t y gradient f o r deep c i r c u l a r ports. If t h e value of t h e volumetric f l o w a t a condition of l i f t i n g flames, VL, i s s u b s t i t u t e d i n equation (3), t h e equation then d e f i n e s t h e c r i t i r c a l boundary v e l o c i t y g r a d i e n t f o r l i f t i n g , gL, f o r deep c i r c u l a r p o r t s .

It should be emphasized t h a t t h i s c r i t i c a l boundary v e l o c i t y gradient f o r l i f t i n g i s a f u n d m e n t a l c h a r a c t e r i s t i c of t h e given gas, and a s such i s in&* pendent of burner design. I n other words, f o r any given p r i m a r y aeration, there i s a d e f i n i t e c r i t i c a l boundary v e l o c i t y gradient at which l i f t i n g will OCCUT, regardless of t h e design of the p o r t . Advantage can be taken of t h i s f a c t t o obtain expressions r e l a t i n g port geometry t o l i f t i n g flames f o r shallow ana now c i r c u l a r p o r t forms.
Equation industry as: g = srhere : g I a A/G R H
= boundary v e l o c i t y gradient, l / s e c , = port loading, Btu p e r hour per square inch of p o r t area,
= port area, square inches,

( 3 ) may be expressed i n terms more generally used i n t h e g a s

1.92 I a (1 + A/G) flR3 H

= air-gas ratio, = p o r t radius, f e e t , and = heating value of f u e l gas, Btu per cubic foot.

For an i n d i v i d u a l port, a =

rlR2

and equation (4) reduces t o :

A t t h e l i m i t i n g conditions f o r l i f t i n g flames f o r a given f u e l gas and a given primary a e r a t i o n , t h e f r a c t i o n (1+ A/G)/H i s a constant, K, so t h a t equat i o n (5) can be f u r t h e r reduced t o :

K gL =
where :

3.84 IL f o r deep c i r c u l a r p o r t s D

K = a f u n c t i o n of t h e primary a e r a t i o n and heating value, H/(1 + A b ) , = c r i t i c a l boundary v e l o c i t y gradient f o r l i f t i n g , l / s e c , = p o r t loading a t t h e l i f t i n g l i m i t , Btu p e r hour per square inch of p o r t area, and D = port diameter, inches.

.. ..

, . .

-63

The c r i t i c a l boundary v e l o c i t y gradient f o r a given gas -&th a given primary a e r a t i o n i s a constant. Equation (6) then s t a t e s t h a t t h e l i m i t i n g port loading a t t h e l i f t i n g l i m i t i s d i r e c t l y p r o p o r t i o r s l t o t h e p o r t d i a n e t e r . The equation a l s o i n d i c a t e s t h a t l i f t i n 4 l i m i t curves ( p l o t t i r z per cent primary a e r a t i o n versus port loading) irould be p a r a l l e l cur-res. F i e w e 2 demonstrates t h a t t h i s i s t r u e f o r port s i z e s which might be used i n ccntemporarj. d r i l l e d p o r t burners.

It w i l l b e noted t h a t t h e l i f t i n g l i m i t curve i n Figure 2 f o r t h e 1/L inch diameter p o r t i s not p a r a l l e l t o t h e other curves, but f a l l s o f f nore abru3tly a t higher p o r t loadings. Uneven d i s t r i b u t i o n of a i r - g a s n i x t u r e flo? v e l o c i t i e s i n l a r g e r p o r t s lovers t h e l i f t i n g l i m i t of t h a t p o r t . L i f t i n g i l cccur l as ar:r point around t h e p o r t rim where t h e c r i t i c a l boundarj v e l o c i t y gradient f o r l i r ' t ing i s reached, even though flames nay be s t a b l e a t a l l o t h e r p o i n t s arow-6 t h e port. In c a l c u l a t i n g p o r t loading, even d i s t r i b u t i c n must be assmed 2i-L the port loading obtained simply by d i v i d i M t h e heat input by t h e porz area. Uneirer. d i s t r i b u t i o n r e s u l t e d i n "incipient l i f t i n g " , i.e., l i f t i n g a t one poins on she port with t h e flame hanging on t o t h e remainder of t h e o o r t r b . %is cor.CirLcn was a l s o noted with l a r g e rectangular _oorts. :Jithin t h e Fort s i z e s generelby used i n contemporary d r i l l e d p o r t burners (No. 50 t o "Io. 20 CXS), hc-*-e.rer, 52r21l e 1 l i f t i n g l i m i t curves f o r d i f f e r e n t p o r t diameters vere observed.
It i s possible, therefore, t o represent t h e lifsiI?? l M $ s o f varicus individual p o r t s of various s i z e s by a s i n g l e l i m i t curve b y p l o s t i z ,d IL -rersu5 primary aeration a t l i f t i n g , where IL i s t h e limitiw p o r t l c a d i x and ,d i s B multiplying f a c t o r dependent on t h e p o r t geometry. ?or deep c i r c u l a r _torrs, fi would equal 1/D, where D is p o r t diameter.
It was a l s o observed that p a r a l l e l l i f t i n g limit c w r e s Tere obraired f o r various port depths, up t o a l i m i t i n g depth. The 1iIXi- l i m i t of p o r t s deeper than t h i s l i m i t i n g depth iiere found t o be t h e s a % . P i g i r e 3 i l l u s r r a t e a a t y p i c a l example of t h i s trend. Similarly, p a r a l l e l l i f t i n g l i m i t c m - e s x r s ' observed f o r rectangular p o r t s of various -widths, lengths, and depths, m e n t-*-o of t h e s e geometric f a c t o r s were held constant and t h e t h i r d varie?. IC a??esreci, then, t h a t geometric multiplyirg f a c t o r s could be obtaiced f o r t h e s e varicus ~ c r r designs so that a s i n g l e l i f t i n g l i m i t c w e could be used f o r a l l such 2 c r ~ designs.
A s a f i r s t step i n these determinasions, t h e c r i t i c a l boundary -relocis;gradient f o r Cleveland n a t u r a l gas was calculazed (by use of e q k i t i o n (39 fro2 observed l i f t i n g l i m i t s with very deep s o r t s . A l l o f these liftix s t u d i e s were made with a r e l a t i v e l y cold burner, since t h i s i s t h e nost c r i t i c a l comiition f o r l i f t i n g flames. Figure 4 shows t h e s e c r i t i c a l boundary v e l o c i t y gradient's f o r Cleveland n a t u r a l gas.

Equation (2), when applied t o a condition of l i f t i n g flames, can be modified t o include a term f o r p o r t loading t o become:
f Re ILa

gL =

1 + A/G

16(rlR3

(7)

A t any given primary a e r a t i o n f o r Cleveland natural gas, equation (7) would reduce to:

where:

D = p o r t diameter, inches, and X = a function of t h e given primary aeration.

The values of f a t l i f t i n g conditions f o r various p o r t geometries were t h e n evaluated by observing t h e l i f t i n g limits f o r a number of p o r t s . For each corresponding value of primary a e r a t i o n a t t h e l i f t i n g condition, values o f t h e c r i t i c a l velocity gradient, gL, were picked from t h e curve of Figure 4 . Values o f f could then be solved f o r i n equation ( 8 ) , since all of t h e o t h e r factors i n t h e equation a r e lmovn from measurement. F'igure 5 shows t h e observed empirical r e l a t i o n s h i p between t h e expression f Re and p o r t geometry f o r various c i r c u l a r p o r t s . The h o r i z o n t a l segment of t h e curve r e p r e s e n t s data obtained with r e l a t i v e l y deep p o r t s i n which laminar flow takes place. The sloped p o r t i o n of t h e c w e , which can be represented by t h e expression f = 305 D/Re a p p l i e s t o r e l a t i v e l y shallow p o r t s . Substit u t i o n of the l a t t e r r e l a t i o n s h i p i n equation (8) obtains:

4 a ,

As was previously pointed out, t h e c r i t i c a l boundary v e l o c i t y gradient for a given gas and primary a e r a t i o n i s a function of t h e burning c h a r a c t e r i s t i c s
of t h e gas, and as such is independent of t h e p o r t geometry. and ( 9 ) can be equated so that:
%us,

equations (6)

IL for. deep D
circular ports
=

IL f o r shallow 7 4 circular ports

(8)
,

= 1/D,

If the multiplying f a c t o r f o r deep c i r c u l a r p o r t s is taken4as then this f a c t o r f o r shallow c i r c u l a r p o r t s would be @ = 4.7714.

It i s r e a l i z e d t h a t t h e t r a n s i t i o n from "deep" t o "shallow" c i r c u l a r ports takes place i n a gradual process, r a t h e r than a t a sharp breaking point. This t r a n s i t i o n zone i s i l l u s t r a t e d i n Figure 5 by t h e curved l i n e joining t h e s t r a i g h t horizontal and sloped l i n e s . It was found,. however, t h a t a n arbitrary l i m i t i n g point could be determined by t h e intersectLon of t h e extensions of the s t r a i g h t horizontal and sloped l i n e s of Figure 5 . The "deep" p o r t multiplying f a c t o r can be used with good accuracy for p o r t s t o t h e l e f t of t h i s a r b i t r a r y point, and t h e " s h a l l o w " port expression can be used f o r p o r t s t o t h e r i g h t of this point.
A similar treatment was m a d e u l t h rectangular ports, and empirical geometrical multiplying f a c t o r s were determined. Eqressions' obtained f o r t h e f a c t o r @, f o r circular and rectangular p o r t s are summarized i n t h e following:

Deep c i r c u l a r ports (D/ *<0.21),

= 1

(U)

Deep rectangular p o r t s ( De <l.O),

= 1.44 ( n

+ 11
c

n
%&ow where: rectangular p o r t s (%
=

> 1-01,6 G L & d

D = p o r t diameter of c i r c u l a r ports, inches, d = port depth, inches De = equivalent port diameter of rectangular ports, inches = 2 n W/(n + l ) , W = rectangular p o r t width, inches, and n = r a t i o of rectangular p o r t length t o width.These expressions f o r were applied t o a l l t h e served f o r Cleveland n a t u r a l gas with individual p o r t s t o generalized l i f t i n g l i m i t curve i l l u s t r a t e d i n Figure 6. used i n contemporary burners, t h e observed l i f t i n g limits 3 percentage u n i t s of primary a e r a t i o n of t h e curve. l i f t i n g l i m i t data ooobtain t h e averzge For 3 o r t s i z e s generall;. generally f e l l - . t h i n

The use' of multiple d r i l l e d p o r t s a f f e c t s t h e l i f t i n g l i m i t of a burzer. 0 1 ' course, t h e r e i s a maximum p o r t spacing above which t h e p o r t s can be considered t o a c t a s irxhvicluel p o r t s . This e f f e c t i s apparently due t o a reduction i n t h e quenching of t h e f l v e r e a c t i o n by the surrounding a i r above t h e p o r t .

In general, c l o s e r p o r t spacing r a i s e s t h e l i f t i n g l i m i t .

A n e l e c t r i c a l l y heated c o i l w a s placed around a p o r t .

Laboratories U s t r a t e i this e f f e c t . The c o i l -"zs so spaced that it would not i g n i t e t h e air-gas mixture.issuing from t h e port. The 1 i L 3 i n g l i m i t of t h e p o r t was r a i s e d considerably when t h e c o i l was heated e l e c t r i c a l l y .

A simple experiment a t t h e A.G.A.

With a multiple d r i l l e d p o r t burner, adjacent f l a u e s apparently c r e a r e t h e same e f f e c t a s t h e heated c o i l . The end p o r t s of bar burners, which a r e influenced by adjacent p o r t s on o n l y one side, generally lift f i r s t . Some such l i f t i n g can be t o l e r a t e d i f combustion o r carryover i s not impaired when t h e burner i s f i r s t lighted. Because of t h i s , t h e l i f t i n g l i m i s - o f rmilti_ole p o r t burners was a r b i t r a r i l y chosen a s t h e point a t which t h r e e o r four of t h e end ports lifted.

Figure 7 i l l u s t r a t e s that l i f t i n g l i m i t curves f o r multiple p o r t burners were found t o be e s s e n t i a l l y p a r a l l e l curves. A multiplying f a c t o r can then be determined f o r various p o r t spacings i n order tb o b t a i n a single l i f t i n g l i m i t curve.
The values of t h e s e f a c t o r s , i n r e l a t i o n t o t h e i n d i v i d u a l port curve of Figure 6, were determined t o be: 0.88 f o r 1/4 inch o r g r e a t e r ; 1.0 f o r 3/16 inch; 1 . 1 2 f o r 1/8 inch; and 1.82 f o r 1/16 inch spacing of p o r t s edge-to-edge. The u s e of multiple rows does not appear t o a f f e c t l i f t i n g characteri s t i c s , provided the p o r t spacing multiplying f a c t o r f o r t h e smallest spacing between p o r t s (whether i n t h e rows o r between rows) i s used i n e n t e r i n g Figure 6.

-66It has been observed chat within t h e span o f p o r t loadings generally used i n contemporary d r i l l e d p o r t burners (10,000 t o 60,000 Btu per hour per square inch of p o r t area), l i f t i n g l i m i t curves f o r various gases were essenc i a l l y p a r a l l e l curves. This trend was noted i n burner p o r t design s t u d i e s conducted a t t h e U. s. Bureau of Mines, and i n i n v e s t i g a t i o n s conducted a t t h e A.G.A. Laboratories. Figure 8 i l l u s t r a t e s t h e l i f t i n g limits obtained for various gases a t t h e U. S. Bureau of Mines.
Thus, t h e l i f t i n g c h a r a c t e r i s t i c s of a f u e l gas may be expressed i n

terms of a n equivalent rate f a c t o r r e l a t i v e t o the l i f t i n g c h a r a c t e r i s t i c s of a

reference gas. The e f f e c t s of gas composition on l i f t i n g c h a r a c t e r i s t i c s were calculated f o r a limited number of gases. The p r i n c i p l e involved, however, can b e applied t o o t h e r gases which might be used i n t h e f i e l d . Information which can be used t o determine t h e equivalent r a t e f a c t o r f o r a number of gases can be found i n s e v e r a l p u b l i c a t i o n s . Two such publications a r e t h e U. S. Bureau o f Mines' Report of I n v e s t i g a t i o n 5225, "Fundamental Flashback, Blowoff, and Yellow Tip L i m i t s of Fuel Gas-Air MLxtures", and A.G.A. Laboratories Research Bulletin 36, "Interchangeability o f Other Fuel Gases with Natural Gas". The method o f u s i w t h e information contained i n t h e s e r e p o r t s i s described more f u l l y i n A.G.A. Laboratories Research B u l l e t i n 77. Tables have been prepared which r e l a t e t h e l i f t i n g c h a r a c t e r i s t i c s of designs generally used i n contemporary burners t o those of an a r b i t r a r i l y chosen reference burner. Several f a c t o r s a f f e c t i n g t h e l i f t i n g c h a r a c t e r i s t i c s of burners a r e s t i l l r e l a t i v e l y unexplored. Uneven d i s t r i b u t i o n of a i r - g a s m i x t u r e t o t h e p o r t s w i l l always r e s u l t i n a lower l i f t i n g 1Mt than values calculated f r o m t h e equat i o n s previously discussed. Port loading must b e considered a s simply t h e t o t a l heat input t o t h e burner divided by t h e t o t a l p o r t a r e a i n design c a l c u l a t i o n s . Overrated p o r t s w i l l , therefore, lift a t a lower primary a e r a t i o n than The calcul a t e d value. The design methods described i n A.G.A. Laboratories Research Bulletin Combustion chamber environment has some e f f e c t on l i f t i n g c h a r a c t e r i s t i c s . A current water heater research i n v e s t i g a t i o n a t the A.G.A. Laboratories i s considering t h i s aspect of t h e problem.

77 a r e f o r open room conditions.

Burner design research i s a continuing process. By applying fundamental concepts, t h e engineering design of appliance burners is, with time, becoming l e s s of a n a r t and more of a science.

JCG:MEIP

5/27/58

,
'I

'.

-67-

i I

//

YELLOW

TIPPED FLAMES ZONE

PORT UIADINO, BTU PER HOUR PER 9Q. IN. W PORT AREA

Figure 1

Typical Flame Characteristics D i a g r a a ~ , Shoving L i f t i n g , Yellow Tipping, and Flashback Limit Curves.

D
PORT KIADINQ.
103am PER HOUR PER SOUARE INOH OF PORT AREA

Figure 2
I

Lifting Limits of Various Individual Circular Ports wlth Cleveland Natural Gas.

-68-

Figure 3

L i f t i n g L M t s of a No. 30 DMS D r i l l e d Port with Various Port Depthe with Cleveland Natural Cea

OAS CONCENTRATION, F, FRACTION OF STOICHlOHETRIC COP(~ENfRATIOPI

Figure 4

Critical ~ i f t i n g ~oundaryv e l o c i t y Gradient $or Cleve18pd Natural Gas.

-69-

w
0

Figure 5

F r i c t i o n Factors f o r Various Deep and Shallov Circular P o r t s with Cleveland Natural Gas a t t h e Lifting L i m i t .

. 0 IL x

I C

Figure 6

L i f t i n g L i m i t s of Slotted and Circular P o r t s w i t h Cleveland Natural Gas.

-70-

>
POAT LOADINO,

103 B N

PER HOUR PER SCUARE INCH OF PORT AREA

Figure

L i f t i n g Limit Curves with Cleveland Natural Gas f o r Experimental, Multiple Port Burners.

mm
Figure

L~AOINO. 101OTU PZR FOUR FSR

WINRE

wcn of

PORI AREA

L i f t i n g Limits Obtained a t t h e U. S. Bureau of Mines f o r Various Fuel Gases.

-71-

Measurement of t h e Tnemal P r o F e r t i e s of Carbonaceous Materials

Bjr

J. D. Batchelor, P. M. Ya-rorsky ar,d 3. Gorin CO8SOLIDATIOB COAL i r G K P P X Y Research and Development Division L i b r a r j , Tennsylvani a

Three parameters o f i n t e r e s t f o r t h e m a l processing of 2 rbo-aceous m a t e r i a l s have been stuCied. The mean s p e c i f i c n e a t c a p a c i t y from r o c s ~ e ~ e r z t u r e t o temperatures of 500" t o 9 0 0 ' F have bee3 shown t o range from 0.34 t o 0.41 f o r c o a l from the P i t t s b u r g h Seam and 0.28 t o 0 . 3 1 f o r i t s LTC cher. These E -Lemal measurements were made i n a simple high t e - v e r a t u r e c a l o r h e t e r . % c o n d u c t i v i t y of coal-char-picch b r i q u e t s w a s found t o follow a c a t z e r n s i r i l z r i l v a l u e I'rom r o m zeT:era<ze to coal. being coked, e x h i b i t i n g a 20-fold i n c r e a s e i t o h o t coke temperature. The c o n d u c t i v i t i e s were measured b 7 . a very d i r e c x An e s t i m a t i o n of t h e s u r f a c e f i h method using a compensated conductometer. sFhere . a s h e a t t r a n s f e r c o e f f i c i e n t f o r f l u i d i z e d s o l i d h e a t i n g of an %ersed c a r r i e d out experimentally u s i n g t h e unsteady s t a t e h e a t i n g curve. Values fzr 5 o f t h e o r d e r of 30 Btu/hr f t 2 F a were obtained whinh e r e c o n s i s t e n x x i r n t h e range g e n e r a l l y found i n t h e l i t e r a t u r e f o r w a l l and tiioe c o e r ' f i c i e n s s .

:i
-72Not f o r P u b l i c a t i o n P r e s e n t e d Before t h e D i v i s i o n of Gas and F u e l Chemistry American Chemical S o c i e t y Chicago, I l l i n o i s , Meeting, September 7-12, 199 Measurement of t h e ?"nemal P r o p e r t i e s of Carbonaceous Materials

J. D. Batchelor, P. M. Yavorsky and E. Gorin

CONSOLIDATIOB COAL COMPANY Research and Development M v i s i o n Library, Pennsylvania , INTROWCTTON

Many o f the processes f o r t h e treatment of c o a l , e i t h e r i n p r e s e n t day use o r under development, i n v o l v e thermal conversion. One such process, the fonncoking o f coal-char b r i q u e t s , i s d i s c u s s e d i n a companion paper. It i s obviously d e s i r a b l e , i n the development of such p r o c e s s e s , t o have a v a i l a b l e themal d a t a o n t h e materials b e i n g handled. The work d e s c r i b e d h e r e was designed t o supply such thermal data f o r carbonaceous ,naterials o f i n t e r e s t . Heat c a p a c i t y measurements have been made o n a bituminous c o a l from t h e P i t t s b u r g h Seam and i t s low t e n p e r a t u r e char. Thermal c o n d u c t i v i t y measurements have been c a r r i e d out on a coal-char-pitch briquet. F i n a l l y , an estimate h a s been made of t h e hear. t r a n s f e r film c o e f f i c i e n t from a f l u i d i z e d h e a t i n g medium t o a s p h e r i c a l b r i q u e t . %is l a t t e r work k'as c a r r i e d o u t so t h a t t h e thermal parameters could be applied t o t h e c a l c u l a t i o n o f h e a t i n g rates f o r carbonaceous b r i q u e t s , an a p p l i c a t i o n which is d e s c r i b e d i n aaother paper.
L i t e r a t u r e Survey The heat c a p a c i t y d a t a f o r c o a l recorded i n t h e l i t e r a t u r e have been reviewed by Qendenin, e t a l . 3 I n t h i s review a c o r r e l a t i o n equation f o r r,he s p e c i f i c heat c a p a c i t y of m o i s t u r e - f r e e c o a l i s presented, based on d a t a f o r 23 American bituminous c o a l s . The e q u a t i o n r e l a t e s h e a t c a p a c i t y t o temperature and v o l a t i l e natter c o n t e n t as follows:

= 0.200

0.0015 V.M.

0.00088 t

where V.M. t

= percentage v o l a t i l e matter on a moisture-free b a s i s = temperature, "c

This equation cannot be expected t o h o l d i n t o t h e temperature range i n which c o d becomes p l a s t i c . I n f a c t the very l i m i t e d d a t a included i n t h e c o r r e l a t i o n for temperatures above lGQC &e i t s u s e above about 1 5 0 C a n e x t r a p o l a t i o n t h a t may not b e j u s t i f i e d .

me heat c a p a c i t y o f a bituminous c o a l w i t h i n i t s p l a s t i c range may be a somewhat v a r i a b l e q u a n t i t y depending upon t h e d e f i n i t i o n o f t h e q u a n t i t y p d t h e method of measurement. %e l i t e r a t u r e contains numerous papers devoted t o t h e measurement o f two heat q u a n t i t i e s . One, t h e t o t a l "heat of carbonization", i s d e f i n e d t o i n c l u d e all the s e n s i b l e and r e a c t i o n h e a t s involved i n carbonization; t h e o t h e r , o f t e n c u e d t h e "heat of decomposition", refers t o t h e r e a c t i o n heats o n l y after t h e s e n s i b l e heats have been removed by c a l c u l a t i o n o r by t h e use of a d i f f e r e n t i d c a l o r i m e t r i c method. The most n o t a b l e of t h e d a t a i n t h e l i t e r a t u r e

-734-5 are t h o s e of Terres and co-workers 11 ' , X e p a n L B , m d Burke and 1 5 , Davis, e t a l . Par+. Work has a l s o been r e p o r t e d upon t h e heat c a p a c i t y of c h a r s produced i'ron c o a l b y thermal treatmentI3. i n g e n e r a l , the data found i n the l i t e r a t u r e fall s h o r t of p r o v i d i n g a p p l i c a b l e and c o n s i s t e n t d a t a f o r materials t y p i c a l of P i t t s b u r g h Seam bituminous coal.

Published information on the thermal c o n d u c t i v i t y - a p p l i c a b l e t o n a t e r i a l s w i t h which w e d e a l is a l s o verf meager. B;r t h e tine our work had p r o g r e s s e d i o c o n d u c t i v i t y measurements, w e 'were prFroarilji i n t e r e s t e d i n compactiocs of c o d , c h a r and p i t c h . D a t a on c o a l s and cokes are s u m a r i z e d by Clendenin, e t ~ t . b~ 3t , again t h e s e d a t a are p r i n c i p a l l y low temperature measureaents below lCOC. Sore d a t a by Terres'l cover t h e whole coking range, b u t s i n c e t h e values are reportec? as "mean" c o n d u c t i v i t i e s t h e y probably r e p r e s e n t average v a l u e s o v e r a b i d e :eEp e r a t u r e rarge. Millard' has published c o n d u c t i v i t y d a t a f o r beds of ga-.lLa%ed Coal (and coke, subsequently produced) over a 100-gOO"C t e q e r a t u r e m g e . \;nile t h e experimental u n i t used by Millard w a s adequate f o r zeasuremenzs on pzc-rea ->e<-= , it was not a p p l i c a b l e t o s o l i d c o m p a e t i ~ n f . Work by B i r c h a.nd 2iark1 3 n :ne coxd u c t i v i t y of rock samples up t o 9 0 0 ' F provided the b a s i c design for she corcl~c:smeter used i n our work.

N o work has been found i n t h e l i t e r a t u r e which p r e d i c t s d i r e c c l y zhe s u r f a c e f i l m h e a t t r a n s f e r c o e f f i c i e n t between f l u i d i z e d s o l i d s and s m l l spheres immersed i n the bed. Numerous i n v e s t i g a t i o n s and c o r r e l a t i o n s ere a-.-ail&le f a r h e a t t r a n s f e r c o e f f i c i e n t s between f l u i d i z e d s o l i d s and the w a l l s of the c o n t a i c i c g v e s s e l and some work i s r e p o r t e d f o r i n t e r n a l tubes. No e f f o r t h s beec macle zo c r i t i c a l l y e v a l u a t e a l l t h i s i n f o m a t i o n , b u t t h e range o f v a l u e s f o r h 'AouGYt 20 be a p p l i c a b l e i s from 20 t o 50 Btu/hr f t 2 Fa.
Heat Capacity Measurements Method
A v a r i a t i o n o f t h e s t a n d a r d c a l o r i n e t r i c method o f n i x t u r e s v a s uszc i n t h i s work. The bituminous c o a l samples under study were known to .undergo irr e v e r s i b l e changes upon thermal treatment i n +%e tercperature r m g e or' i n t e r e s t ; for this reason, it was e s s e n t i a l t h a t a high temperature c a l o r i c e t e r be used i r which t h e sample i s heated and t h e c a l o r i n e t e r cooled. This procecure i s e x a c t l y t h e opposite of the one g e n e r a l l y used i n which a h o t sample i s cooled i n a c o i d calorimeter. The method adapted measured the t o t a l heaz r e q u i r e d t o raise t h e sample from room temperature t o t h e temperature i n v e s t i g a t e d ; s e n s i b l e and reactio?. heats are thus lumped t o g e t h e r . This i s t h e important q u a n t i t y , however, f o r c o s t p r a c t i c a l applications.

The use of a h i & temperature c a l o r i m e t e r nade i t h p r a c t i c a l t o reduce the thermal l o s s e s to a n e g l i g i b l e l e v e l . I n s t e a d , i t was necessary t o c o n t r o l and s t a n d a r d i z e t h e l o s s e s a t a'reasonably high l e v e l . In a n e f f o r t t o keep t h e u n i t as simple as p o s s i b l e , an unsteady s t a t e method vas used, i n which t h e efr'ect of a n added sample on the normal c o o l i n g curve of a h e a t e d metal block w a s observed. This method proved adequate, and allowed t h e use of simple equipment; however, cons i d e r a b l e c a r e and e x t e n s i v e c d l c u l a t i o n were needed t o produce these s a t i s f a c t o r y results.

The h e a r t of t h e c a l o r i m e t e r was a n aluminum block 1-1/2" 0.d. by 2-718" l o n g w i t h an a x i a l l y bored h o l e 314" i n diameter and 2-112'' deep. Six h o l e s were d r i l l e d on a 1-1/8"c i r c l e i n t h e t o p of t h e block; two of these h o l e s were tapped

f o r support rods, and t h e o t h e r f o u r were used as thermocouple w e l l s 1/2", l", 1-1/2" and 2" deep, r e s p e c t i v e l y . This aluminum block was wound with a r e s i s t a q c e k-ire h e a t e r and suspended by s m a l l i r o n rods from a transite d i s c . The block was hung down i n t o a glass Dewar f l a s k , wnich i n t u r n was p l a c e d i n a tube furnace. m e furnace t e m p e r a t u r e was c o n t r o l l e d a t a l e v e l near t h e o p e r a t i n g temperature of t h e block t o r e d u c e t h e heat l o s s . A "seal-ring'' o f transite, s l i g h t l y s m d l e r ' +han t h e i n s i d e diameter o f t h e Dewar f l a s k , was placed between t h e block and i t s supporting disc. m s " s e a l - r i n g " , which contained an e l e c t r i c h e a t e r powered fro? a c o n s t a n t v o l t a g e s o u r c e , w a s provided t o reduce t h e end l o s s e s upward from t i e block and t o improve t h e temperature symmetry of t h e system. A l o n g glass tube ( was used t o p r o v i d e a n i n l e t p a t h f o r t h e t e s t samples. A n i t r o g e n purge was , metered i n n e a r the bottom o f t h e Dewar f l a s k t o i n s u r e a n i n e r t atmosphere. m e 1 c a l o r i m e t e r is shown assembled i n F i g u r e 1 .
't
~

The thermocouple c i r c u i t s used t o g e t a s e n s i t i v e measure o f the temperatures of t h e system c o n s i s t e d of two b e l e m e n t thermo-piles made of iron-constant& The f o u r h o t j u n c t i o n s o f one thermo-pile were p l a c e d i n the h o l e s d r i l l e d i n t h e aluminum block, and because of t h e i r spacing i n the block y i e l d e d a good average temperature d i r e c t l y . These j u n c t i o n s had t o b e i n s u l a t e d e l e c t r i c a l l y from t h e block, and this w a s done w i t h a bead o f alundum cement t o which a l i t t l e waterglass had been added. The h o t j u n c t i o n s o f t h e second thermo-pile were h e l d about 1/16" a w v from the i n n e r w a l l of the Dewar f l a s k , spaced a t 90" around t h e block at a l e v e l n e a r t i e middle of t h e block. The s e n s i t i v i t y o f each o f t h e s e thermop i l e s , between 500" and 900F, i s o f t h e o r d e r of 0.12 m i l l i v o l t p e r degree Fahrenheit, so that b y measuring e a c h o u t p u t t o t h e n e a r e s t 0.01 m i l l i v o l t w i t h a Leeds and Northrup p o r t a b l e p o t e n t i o m e t e r t h e temperatures w i t h i n t h e block and e x t e r n a l t o it could b e determined t o t h e n e a r e s t 0.1'F.
~ ~

I n t h e normzl experimental procedure, the o u t e r furnace and t h e e a l r i n g h e a t e r were used t o e s t a b l i s h a n e l e v a t e d "steady s t a t e " temperature i n t h e c a l o r i meter block. From t h i s l e v e l t h e block h e a t e r w a s used t o raise t h e block temp e r a t u r e to a l e v e l a b o u t 20F higher. A f t e r c o n d i t i o n s were steady a t this upper temperature l e v e l , the b l o c k h e a t e r was switched o f f , and the block allowed t o cool towards i t s lower s t e a d y s t a t e temperature. After about t h r e e minutes had elapsed a powdered s o l i d sample, h e l d i n a s m a l l aluminum f o i l c y l i n d e r about 5/8" diameter and 1-1/2" l o n g , was dropped i n t o the block. The block and e x t e r n a l temperatures were read a l t e r n a t e l y , each on a 30-second schedule, u n t i l the c o o l i n g rate became r e l a t i v e l y small so that l e s s f r e q u e n t readings were s u f f i c i e n t . The c a l c u l a t i o n o f t h e e f f e c t o f t h e added sample r e q u i r e d that a "normal cooling curve b e determined f o r each temperature l e v e l w i t h o u t any sample a d d i t i o n or o t h e r e x t e r n a l s t i m u l u s . By g r a p h i c a l d i f f e r e n t i a t i o n o f t h e block temperature curve d u r i n g "normdl" cooling, t h e block c o o l i n g r a t e w a s p l o t t e d as a function of t h e i n s t a n t a n e o u s temperature d i f f e r e n c e between t h e block and i t s surroundings. A t each l e v e l a good straight l i n e r e l a t i o n e x i s t e d , as p r e d i c t e d by t h e simple Newtonian c o o l i n g law; a t y p i c a l p l o t i s shown i n Figure 2. The slope of this l i n e depends on t h e p r o p e r t i e s of the system, b u t t h e i n t e r c e p t depends, p r i n c i p a l l y , on t h e i n p u t t o the seal r i n g heater. When used as a c o r r e c t i o n of t h e observed cooling after a sample has been added t o the c a l o r i m e t e r block, t h i s Newtonian p l o t must b e e x t r a p o l a t e d i n t o t h e h e a t i n g domain where d t block is p o s i t i v e and de ( tblock-texternal) i s negative. To prove t h a t t h i s e x t r a p o l a t i o n w a s reasonably accurate, a d r y i c e p e l l e t w a s added a t one temperature l e v e l t o f o r c e the block temperature below the externdl temperature; t h e observed h e a t i n g rate f o r the block w a s j u s t that p r e d i c c e d b y e x t r a p o l a t i o n .

-75A t y p i c a l p a i r o f temperature curves f o r a m i i n which a so1i.d sample w a s added i s shown i n F i y r e 3 . For s e l e c t e d v a l u e s of time 'he temperature d r i v i n g f o r c e , ( t b l o c k - t e x t e r n d ) , WBS noted and a e c o d c o o l i n g rate dete-nixed 'from t h e Newtonian p l o t . The i n t e g r a l expression, ( d t block) d6, vas
dQ )r,onzl. determined g r a p h i c a l l y f o r each run, and s e r v e d as a c o r r e c t i o n l o r t h e o b s e r r e c temperature drop of t h e block. P i e c o r r e c t e d temperature e f f e c t of t h e s a a p l e was t h e a l g e b r a i c d i f f e r e n c e of t h e observed temperature drop a d t h i s i n t e g r a l n o z a l temperature drop; %e c o r r e c t e d temperature e f f e c t should approach a s t e 2 . i ~f i i z l value after a reasonable p e r i o d of t h e , as shown i n a.t,-ypical cu-rve i c P i G r e 5 . I n some c a s e s i n which no f i n a l s t e a d y v a l u e w a s approached, i t vas n e c e s s a r j t o s h i f t t h e Newtonian c o o l i n g p l o t , h o l d i n g i t s sloFe c o n s t a n t , u n t i l t h e t o t a i This f o r c e d b a l a x e was e q u i v a l e n t t o assuming e f f e c t was forced t o l e v e l off t h a t the s e a l r i n g heater i n p u t v a r i e d while t h e i n s u l a t i n g p r o p e r t i e s of t h e system d i d n o t . This assumption appeared reasonable s i n c e the c o r r e c t i o n s were random and no aging t r e n d w a s n o t i c e d d u r i n g t h e p e r i o d s p e n t on measurenents a t a given temperature l e v e l .

f(

A f t e r t h e t o t a l c o r r e c t e d e f f e c t o f a sample on t h e c a l o r i m e t e r block iias found, t h e r e m q n i n g c a l c u l a t i o n s were f a m i l i a r . ,Powdered a z o 3 , a substance o f ---knownh e a t capacity, w a s run t o e s t a b l i s h a c a l o r i m e t e r constant. ,The h e a t c a p a c i t i e s o f o t h e r materials were determined b y u s i n g t h i s c a l i b r a t i o n as a reference.

Results
The h e a t c a p a c i t y o f a P i t t s b u r g h seam c o a l (38.6$ V.M. and 7.05 ash, . tempera',ure from t h e Arkwright mine i n West V i r g i n i a ) w a s determined t o a uxxdrum of 8 0 0 " ~ . Swelling and f r o t h i n g of t h e c o d made i t impossible t o use higher temperatures. A c h a r produced f r o m this c o a l by f l u i d i z e d c a r b o n i z a t i o n a t 9,5O"F w a s used f o r measurements t o 890F. This Arkwright c h a r c o n t a i r e d l3.2$ V.>I. Magnesium oxide was used as a secondary s t a n d a r d -co check t h e accuracy of tae Zetnod, by comparison between observed and l i t e r a t u r e v a l u e s f o r h e a t c a p a c i t y from 6 j O " TO 8 0 0 " ~ . The r e s u l t s f o r these t h r e e materials i s shown i n Figure 5; t h e nean s p e c i f i c h e a t c a p a c i t i e s a r e , o f course, averaged from room temperature t o ;he f i n a l t e c q e r a t u r e plotted. Most p o i n t s OK t h e curves shown i n Figure 5 are the average o f two o r more determinations. The v a r i a b i l i t y of i n d i v i d u a l values from t h e uean w a s generall;; less than 2 d p f o r t h e h e a t s t a b l e materials, b u t rose as high as 5$ i n one i n s t z n c e f o r the c o a l sample i n the middle of t h e p l a s t i c temperature range. The agreenext be-cween t h e observed and l i t e r a t u r e v a l u e s f o r MgO adds confidence i n t h e Ee+$od. For comparison w i t h d a t a found i n t h e l i t e r a t u r e , two o t h e r curves are shown i n Figure 5. One r e p r e s e n t s t h e v a l u e s t h a t would be prediczed by applj5ng t h e equation derived by Clendenin, e t ale3; The o t h e r curve i s from d a t a by P o r t e r and Taylors, t h e s e d a t a being t h e h i g h e s t temperature values included i n t h e c o r r e l a t i o n o f Clendenin. No comparable d a t a a r e a v a i l a b l e f o r char. The d a t a obtained h e r e b j i use o f a high temperature c a l o r i m e t e r app13- t o t h e p a r t i c u l a r case i n which t h e sample i s heated very r a p i a y and t h e $roduc-.s of carbonization a r e c a r r i e d o u t of t h e system. This corresconds q u i t e well t o t h e p a t t e r n i n f l u i d i z e d low temperature c a r b o n i z a t i o n of c o d , the a p p l i c e z i o n f o r which t h e s e d a t a were o r i g i n a l l y determined, b u t may d i f f e r s u b s i z c t i a l l j f r c ~ t h e h e a t i n g regime i n some o t h e r processes. Reaction e f f e c t s , p a r t i c d a r l y f r o m e x t e n s i v e secondary r e a c t i o n o f t h e t a r products, may cause some variance i n t o t a l s p e c i f i c h e a t c a p a c i t y under o t h e r t r e a t m e n t conditions.

-76T a e r n a l Conductivity Measurements

,
Method
\

Absolute t h e d c o n d u c t i v i t i e s of a m a t e r i a l a r e b e s t determined by d i r e c t measurement o f t h e te.mperature g r e d i e n t which exists f o r a known r a t e of h e a t t r a n s f e a c r o s s a u n i t a r e a o f material under s t e a d y state conditions. From t h e F o u r i e r h e a t , t r a n s f e r equation

t h e thermal c o n d u c t i v i t y c a n be c a l c u l a t e d d i r e c t l y . The thermal conductometer u s e d ; i n t b i s work was modeled a f t e r one d e s c r i b e d by Birch and Clark1, and allowed us t o measure t h e s e needed f a c t o r s f o r compactions made of char, c o a l and p i t c h .

' A c r o s s - s e c t i o n a l view of t h e conductometer i s shown i n F i g u r e 6. m e l o c a t i o n of t h e l e a d - i n wires h a s been omitted. The apparatus was s y m e t r i c a l l y 1 square about t h e v e r t i c a l axis. The specimen w a s a f l a t s l a b approximately 2" , s q u a r e and 112" t h i c k , c u t from a s i n g l e l a r g e b r i q u e t , o r formed of c l o s e f i t t i n g r e c t a n g d a r s e c t i o n s from small b r i q u e t s . It was placed between the h e a t e r block , arnd c o o l i n g g l a t e , b o t h of which are made of copper t o a s s u r e an even temperature , d i s t r i b u t i o c . A 30-gauge nichrome wire h e a t i n g element i n t h e h e a t e r block h e l d i t s 1 temperature a3out 10C above t h a t of the c o o l i n g p l a t e when a s t e a d y s t a t e w a s reached. Heasure-3ent of this temperature d i f f e r e n c e a c r o s s t h e specimen was made w i t n c h e m o c o l q l e s p l a c e d a t p o s i t i o n s i n d i c a t e d by t h e h e a v j dots. Provided that t h e r e i s no h e a t t r a n s f e r from the h e a t e r block o t h e r than through the specimen, t h e c o n d x t i v i t y is simply c z l c u l a t e d . The c o n d i t i o n of no extraneous h e a t loss f r o m the b h c s w a s approached as c l o s e l y as p o s s i b l e by u s e of a thermal guard dome of cooper which covered t h e heater block as sham. The @ard dome contained a h e a t i n g element by which the dome's temperature w a s a d j u s t e d t o e q u a l t h e temperature of t h e h e a t e r block. Heat losses t o the surroundings t h e n came from t h e guard dome and n o t from t h e h e a t e r block. Lateral h e a t losses from t h e specimen were minimized by t h e guard r i n g of i n s u l a t i n g material ( t r a n s i t e ) t h a t surrounds t h e specimen. Small. ceramic s p a c e r s prevented a c c i d e n t a l metal-to-metal c o n t a c t between the dome and h e a t e r block.

'

'

The e n t i r e conductometer was enclosed i n a s t e e l box which was purged w i t h n i t r o g e n t o p r e v e n t o x i d a t i o n o f the copper o r t h e specimen a t high temperat u r e s . The e n t i r e u n i t vas i m e r s e d i n a f l u i d i z e d sand b a t h t o provide t h e cont r o l l e d e l e v a t e d ambient temperatures. S t a b l e power f o r t h e h e a t e r block was provided by two s i x - v o l t s t o r a g e b a t t e r i e s i n s e r i e s . The c u r r e n t was manually c o n t r o l l e d w i t h r h e o s t a t s t o g i v e t h e d e s i r e d temperature d i f f e r e n t i a l a c r o s s t h e specimen. With c u r r e n t s of 100200 milliampere i n t h e b l o c k h e a t e r , t h e r e was no d i s c e r n i b l e d r i f t i n t h e c u r r e n t d u r i n g t h e course o f a measurement. The less c r i t i c a l power f o r the w a r d dome w a s provided by a Variac w i t h a r h e o s t a t f o r f i r e c o n t r o l . A t y p e K Leeds and Northrup p o t e n t i o m e t e r w a s used f o r a c c u r a t e measurement of t h e thermocouple p o t e c t i a l s and the v o l t a g e and c u r r e n t of t h e h e a t e r block. Voltage measurements a c r o s s a s t a n d a r d one-ohm r e s i s t o r i n the h e a t e r c i r c u i t gave direct measurements of t h e h e a t e r current.
Results

Thermal c o n d u c t i v i t i e s were determined f o r s e c t i o n s o f b r i q u e t s made from c o a l , char and p i t c h . The composition was 25dp c o d ( P i t t s b u r g h Seam from Montour __

-77

mine i n Western Pennsylvania), 63.5% l o w temperature c h a r produced from this c o a l by f l u i d i z e d carbonization a t 950F and 11.5% p i t c h . mese b r i q u e t s were pre-coked a t s e v e r a l temperature l e v e l s , t h e i r conductivity measured a t that temperature, and then a l s o a t lower temperatures. This procedure s e p a r a t e l y demonstrates b o t h t h e e f f e c t of coking on c o n d u c t i v i t y a t a p a r t i c u l a r t e n p e r a t u r e and t h e temperature dependence of conductivity of s p e c i f i c a l l y coked b r i q u e t s . These data are p l o t t e d i n Figure 7. Trustworthy r e s u l t s could n o t be obtained at temperatures above 1 6 0 0 ~ ~ on t h e f u l l y coked b r i q u e t s because adequate f l u i d i z e d sand furnace temperature c o n t r o l was not r e a l i z e d . However, t h e values obtained at o t h e r temperatures on this f u l l y coked m a t e r i a l a l l o w a reasonable e x t r a p o l a t i o n of the curve t o 1800F. Figure 8 shows a comparative p l o t of the p r e s e n t d a t a with those presented by Millard' f o r coal.. Millard m a d e no attempt t o e x p l a i n the d i f f e r e n c e i n the c o n d u c t i v i t i e s of t h e two c o a l s , which cane from d i f f e r e n t mines. It i s r e a d i l y seen from t h e p l o t that there is no r a d i c a l d i f f e r e n c e i n the b r i q u e t conductivity as compared t o t h e coals. Both m a t e r i a l s show a v e r y l a r g e i n c r e a s e i n c o n d u c t i v i t y with temperature and a marked i n c r e a s e w i t h degree of coking. Estimation o f Surface Film C o e f f i c i e n t f o r Heating W i t h F l u i d i z e d Solids

The a p p l i c a t i o n of t h e t h e d parameters r e p o r t e d above very o f t e n req u i r e s t h e estimation of a s u r f a c e f i l m h e a t t r a n s f e r c o e f f i c i e n t p e r t i n e n t t o the thermal system t o be used. One of t h e thermal arrangements of i n t e r e s t f o r t h e carbonization of b r i q u e t s i s t h e h e a t i n g with c i r c u l a t i n g , h o t f l u i d i z e d s o l i d s . Considerable work has been done by t h e authors and t h e i r a s s o c i a t e s on t h e treatment of carbonaceous b r i q u e t s i n this type of system. I n an e f f o r t t o b r i n g o r d e r and p r e d i c t a b i l i t y t o some of the results noted, and f u r t h e r t o check t h e v a l i d i t y of assumed values of t h e h e a t t r a n s f e r f i l m c o e f f i c i e n t used i n machine c a l c u l a t i o n s made on this type system ( r e p o r t e d i n a companion paper), an experimental me-hod f o r t h e estimation of t h e f i l m c o e f f i c i e n t was t e s t e d . This experimental method is simple, e s p e c i a l l y so under t h e s p e c i a l conditions i n which it has been applied here. The r e s u l t s obtained r e p r e s e n t o n l y a r a t h e r cursory examination o f the t r a n s f e r c o e f f i c i e n t , b u t i t is f e l t that t h e method deserves d e s c r i p t i o n .

me method used w a s to observe t h e temperature rise of s p h e r i c a l o b j e c t s after Fmmersion i n a constant temperature, f l u i d i z e d sand bath. Although, g e n e r a l l y speaking, t h e use of an unsteady s t a t e system l e a d s t o a r e l a t i v e l y low degree of p r e c i s i o n , i t is a l s o t r u e t h a t these measurements are r a p i d and may b e made q u i t e well i n a very simple system, much simpler than would be needed for s t e a d y state measurements.
Examination of mathematical. expressions f o r this system shoved that i n order t o r e l a t e these results t o the f i l m c o e f f i c i e n t it w a s necessary t o r e s t r i c t our a t t e n t i o n t o m a t e r i a l s and temperature ranges f o r which t h e themal d i f f u s i v l t y may be t r e a t e d as a constant. I n t h e s e cases, t h e F o u r i e r equation can be solved a n a l y t i c a l l y . A s o l u t i o n which has been published may then be a p p l i e d t o this p r 0 b l e m ~ , 1 ~ . This s o l u t i o n reduced f u r t h e r to a very simple form f o r the case of 1). o b j e c t s with a high thermal conductivity such as aluminum or copper (Nu

<<

Under these s p e c i f i c conditions t h e temperatures of t h e c e n t e r and surface, respectively, of a sphere of r a d i u s r o a t i n i t i a l temperature TI which is suddenly surrounded by a medium a t a constant temperature To are given by t h e equations

-? 8-

The N u s s e l t number, Nu,

is d e f i n e d as

and

is a dimensionless t i m e =
ro2

or

p c/

k t

c2

The N u s s e l t number can be v e r y simply determined i n t h i s c a s e b y p l o t t i n g experimental. %is v a l u e s of I n Y v e r s u s 8 . A straight l i n e should b e o b t a i n e d whose s l o p e is03 Nu. i s t r u e f o r b o t h the c e n t e r and s u r f a c e temperatures.
Experimentally, all measurements were made b y plunglng an aluminum sphere i n t o an 8'' d i a m e t e r f l u i d i z e d sand bath. An aluminum sphere w a s chosen so that t h e s i m p l i f i e d form o f t h e F o u r i e r s o l u t i o n equation c o u l d be applied; i.e., the thermal c o n d u c t i v i t y o f aluminum i s h i g h and almost e x a c t l y independent of temperature, and t h e thermal d i f f u s i v i t y i s r e l a t i v e l y c o n s t a n t . It w a s assumed that t h e nature of t h e s p h e r e ' s s u r f a c e w a s unimportant t o t h e result, so that v a l u e s of h c a l c u l a t e d h e r e would b e a p p l i c a b l e t o carbonaceous b r i q u e t s .

TJO thermocouples were used t o s e n s e t h e temperatures of a n t e r e s t i n t h e

sphere, one a t t a c h e d t o the s u r f a c e and one p l a c e d i n a small h o l e bored t o t h e specimen c e n t e r . small d i a m e t e r thermocouple w i r e (28 gauge) was used t o minimize error caused by heat conduction down the wire. A . t h i r d thermocouple i n the f l u i d i z e d sand b a t h w a s used t o s e n s e and c o n t r o l i t s temperature. All t h r e e temperatures were recorded on a m u l t i - p o i n t s t r i p c h a r t recorder. A l l the tests used a nominal f l u i d i z e d s v l d temperature of 1000F. F i v e tests were run u s i n g a 2" sphere. Four of these t e s t s were made u s i n g a f l u i d i z a t i o n v e l o c i t y ( s u p e r f i c i a l . ) o f 0.6-0.7 f e e t p e r second; one t e s t used a reduced l i n e a r v e l o c i t y o f 0.34 f p s . S i n g l e tests were made with a 1"and a 5'' sphere. A t y p i c a l p l o t of t h e temperature h i s t o r y recorded by the themocouple p l a c e d a t the c e n t e r of t h e sphere i s shown i n F i s r e 9.
A p l o t of t h e temperature approach d a t a used f o r t h e e s t i m a t i o n of some o f t h e f i l m c o e f f i c i e n t s , h, is sham i n F i g u r e 10. S t r a i g h t l i n e s are obtained as p r e d i c t e d by the s p e c i a l s o l u t i o n f o r t h e basic F o u r i e r equation. Similar results may b e o b t a i n e d when t h e s u r f a c e temperature i s treated i n a similar manner, b u t t h e s u r f a c e temperature i s more d i f f i c u l t t o measure w i t h p r e c i s i o n .

The l i n e s i n F i g u r e 10 do n o t show t h e u n i t value of y - i n t e r c e p t as pred i c t e d b y theory. P a r t of t h e cause o f t h i s discrepancy i s thought t o be a thermocouple error caused b y conduction o f h e a t down t h e thermocouple wire, thus producing high temperature readings. It can e a s i l y be shown that t h e t o t a l h e a t conducted down t h e w i r e i s a n e g l i g i b l y s m a l l f r a c t i o n of that conducted through t h e aluminum. F u r t h e r , i t can b e shown t h a t i f the thermocouple e r r o r i s proport i o n a l t o t h e temperature d i f f e r e n c e between t h e surroundings and t h e c e n t e r of t h e sphere (a logic^ assumption), then only t h e i n t e r c e p t of t h e F o u r i e r s o l u t i o n equation i s s h i f t e d and i t s s l o p e i s unaffected.

-79

The c a l c u l a t e d v a l u e s of t h e f i l m c o e f f i c i e n t , n, are l i s t e d i n Table I. The p r e c i s i o n of t h e r e p l i c a t e t e s t s on t h e 2 " i n c h sphere i s n o t t h e b e s t , b u t t h e r e s u l t s a r e c o n s i s t e n t and reasonable. The t e s t made a t a rediuced f l u i d i z i n g v e l o c i t y shows a somewhat lower v a l u e f o r h, which i s also reasonable. Table I Experimental Values of Heat T r a n s f e r C o e f f i c i e n t Between F l u i d i z e d Sand and Aluminum Spheres

Thermal Parameters f o r C a l c u l a t i o n s :
Sphere D i m et e r Fluidization Velocity 0.72 f t / s e c 0.72 0.72

k = 117 Btu/hr. f t . a = 5.33 ft.2/hr.

"F

Run No.

F l u i d i z e d Sand ,Temperature

Calculated Heat T r a n s f e r CoeffLcient, h

Al-1 AI.-2

3"
2" 1" 2" 2" 2" 2"

950'F
1000'F 10500~

38 Btu/hr f t ' 31
27

F"

Al-3 Al-4 d-5

Al-7

0.61
0.61 0.34 0 . 6 1

AI.-6

95'F 99O'F 90F 950'F

36
34 23

30

The v a l u e s measured h e r e might be expected t o l i e i n t h e upper range o f v a l u e s r e p o r t e d because of the high d e n s i t y o f t h e f l u i d i z e d sand; on the o t h e r hand, t h e f l u i d i z a t i o n v e l o c i t i e s used were r a t h e r low, e s p e c i a l l y f o r a dense s o l i d , azzd t h i s f a c t o r would lower t h e v a l u e of h. The average v a l u e of h neasured h e r e i s about 30 Btu/hr. f t . 2 F", and .thus w e l l w i t h i n t h e range of 20-50 Btu/ilr. f t ? Fa. chosen f o r t h e c a l c u l a t i o n s of h e a t i n g rates o f Carbonaceous b r i q u e t s i n f l u i d i z e d s o l i d s systems.
. I

--.

SUt4MRY

The v a r i o u s thermal d a t a r e p o r t e d h e r e are not e n t i r e l y comparable i n t e n s of materials and temperature ranges. However, t h e y all b e a r d i r e c t l y on t h e p r o b l a of h e a t t r a n s f e r t o carbonaceous b r i q u e t s c o n t a i n i n g c o a l , c h a r m d pktch. The h e a t c a p a c i t y d a t a measured f o r c o a l and c h a r may be coinbined.with v a l u e s i n ?;he l i t e r a t u r e f o r p i t c h t o provide composite values a p p l i c a b l e t o b r i c p e t s . The ;hem& :ond u c t i v i t y d a t a were o b t a i n e d d i r e c t l y on b r i q u e t samples. The h e a t t r a n s f e r f i L q c o e f f i c i e n t , h, w a s e s t i m a t e d f o r a metal of h i g h thermal c o n d u c t i v i t y ( a l : ~ ? i r m ) f o r s i m p l i c i t y , b u t no reason is a p p a r e n t why t h i s e x t e r n a l f a c t o r carmot be a p p l i e d d i r e c t l y t o t h e h e a t i n g o f carbonaceous shapes of much lower c o n d u c t i v i t y . ,These data have been used f o r machine computations of b r i q u e t h e a t i n g r a t e s as r e p o r t e d i n a companion paper.

-80BISLIOGWEE
1

Birch, F. and Clark, H., Burke, S. P.,

s. J. of

Sci.,

2
3

u l d P a r r y , V. F . ,

E .

=.e., 9,

a, 529 (1940).
l5-& =O (1927).

Clendenin, J . D., B a r c l a y , K. M., Dondd, H. J., Gillmore, D. W., and Wight, C. C., Trans. of t h e Seventh Annual A n t h r a c i t e Conf. of Lehigh U n i v e r s i t y , May 5 - c l m t h l e h e m , Pennsylvania.

4 5

. D., Davis, J
Davis, J . D.
Davis, J. D.,

s.m. m.,16, 726-30 (1924). and P l a c e , P. B . , Ind. En4. m., 6 . l , 589 (1924). Place, P. B. and Edeburn, P., w, 5, 286-99 (1925).
2 . Inst.
of Fuel, 28, 345 (1955).

I n g e r s o l l , L. R . , Zobel, 0. J. and I n g e r s o l l , A. C., "Heat Conduction". University of Wisconsin P r e s s , Madison, Wisconsin (1954). Millard, D. J . ,

8 3

P o r t e r , H. C. and Taylor, G. B.,

E.3 . m.,5, 289-93 (1913).

Addison-Wesley P r e s s ,

10

Schneider, P. J., "Conduction Heat Transfer". Cambridge, Massachusetts (1955). Terres, E., Terres, E.,

11
12

Proc. Second I n t e r n . Conf. Bituminous

m, 2,

657-84 (1928).

m,

et d l . ,

Beiheft

2 .E . Deut.

Chem.,

No. 10, Angew &em.,

17-21 (19353Gas und Wasserfach,

13

W r r e s , E. and Biederbeck, H.,

a, 265-8,

297-303,

320-5, 338-45 (1928)14

Terres, E. end Meier, M.,

G a s und Wasserfach,

E,457-61,

490-5,

99-23 (1928)
15

Terres, E. and V o i t u r e t , K . ,

Gas und Wasserfach, 74, 98-101, 122-8,

1 4 8 9 , 178-82 (1931).
16

W e ,

G . ,

Journal SOC. Chem.

s., Zp,

168-9 (1920).

-81-

I'

-82-

'J
FIGURE 3 OBSERVED DATA FOR A CALIBRATION TEST

BLOCK TEMPERATURE EXTERNAL TEMPERATURE

4
COWNO TIYE. MINUTES

20

23

5 a

0.0:

10

CWLlNe TIME. MINUTES I3

20

25

30

-12.0

. . .

,I

-83-

FIGURE 3 MEAN SPECIFIC HEAT CAPACITIES OF COAL AND CHAR

FIGURE 6 THERMAL CONDUCTOMETER

-85-

FIGURE 9 CENTER TEMPERATURE OF ALUMINUM HEATED BY FLUIDIZED SAND BED

1 . 0

2.0

3 . 0

4.0

5.0

b*

TIME. YlNUTES

-87

Heat Transfer and Associated E f f e c t s i n The Carbonization of B r i q u e t s

BY
P. M. Yavorsky, R. J. F r i e d r i c h and E. G o r i n

CONSOLIDATION COAL COMPANY Research and Development Division Library, Pennsylvania

ABSTRACT
_i

Temperature-time p a t t e r n s f o r t h e carbonization af coal-char b r i q u e t s were computed with an e l e c t r o n i c digital. computer f o r various b r i q u e t s i z e s , f i l m neat t r a n s f e r c o e f f i c i e n t s and two shock h e a t i n g methods h3t f l u e gas and hot fluid2zed s o l i d s . This represents s o l u t i o n of an unusual h e a t transfer problem wherein conEeating r a t e cord u c t i v i t y and s p e c i f i c h e a t are s t r o n g functions of temperature. r e l a t i o n s evolved from the computed results permit extension o f t h e d a t a to conditions not included d i r e c t l y i n the computations.

Comblnation of the thermal p a t t e r n s w i t h experimental b r i q u e t expansivity data y i e l d e d information on t h e r e l a t i v e magnitude of t h e d stresses i n briqueT;s undergoing carbonization. Excessive stresses l e a d t o d e l e t e r i o u s b r i q u e t f r a c t u r i n g . The assembled d a t a supply the necessary background f o r estimation o f o p e r a b i l i t y limits f o r b r i q u e t carbonization by shock h e a t i n g procedures and can be used t o d e f i n e physical conditions and dimensions i n the design o f carbonization units i n formcoking processes.

-88Not f o r P u b l i c a t i o n Presented Before t h e Division of Gas and Fuel Chemistry American Chemical Society Chicago I l l i n o i s , Meeting, September 7-12, 1 9 9

Heat T r a n s f e r and Associated E f f e c t s i n The Carbonization of Briquets

P. M. Yavorsky, R. J. F r i e d r i c h and E. Gorin

CONSOLIDATION COAL COMPANY Research and Development Division Library, Pennsylvania

Th work r e p o r t e d h e r e i s p a r t of a program aimed a t t h e development of a continuous process for the production of formcoke s u i t a b l e f o r b l a s t furnace use. %e p o t e n t i a l advantages of such a process are lower investment c o s t s / t o n of coke, the a b i l i t y t o u t i l i z e t h e v a s t r e s e r v e s of non-metallurgical grade c o a l , and t h e production of a product coke having uniform s i z e and q u a l i t y . Experimental work c a r r i e d o u t i n these l a b o r a t o r i e s has shown t h a t such a process can b e developed based on b r i q u e t t i n g followed by continuous coking of The b r i q u e t s a r e formulated from l o w temperature char, c o a l and a t h e briquets. The b r i q u e t t i n g a s p e c t s of t h i s program will b e discussed i n subp i t c h binder. sequent papers. The p r e s e n t paper is concerned with t h e carbonization of t h e b r i q u e t s and p a r t i c u l a r l y w i t h h e a t t r a n s f e r problems a s s o c i a t e d with t h i s process. The b r i q u e t s r e q u i r e very c r i t i c a l c o n t r o l i n the carbonization process t o y i e l d acceptable formcoke. On t h e one hand, shock h e a t i n g i s necessary t o prevent p l a s t i c deformation, and on t h e o t h e r , too severe shock h e a t i n g causes f r a c t u r i n g of t h e b r i q u e t s . Deformation and binding of b r i q u e t s would cause ino p e r a b i l i t y i n any continuous process and f r a c t u r i n g of t h e formcoke i n t o s m a l l p i e c e s would make it unacceptable f o r b l a s t furnaces.

There are two imaediate o b j e c t i v e s of t h i s work. One i s t h e determination of t h e h e a t i n g rates i n b r i q u e t s subjected t o various methods o f shock heating. These r a t e s are required for t h e r a t i o n a l design of l a r g e s c a l e carbonizing equipment, The o t h e r o b j e c t i v e i s t o determine, by way of c a l c u l a t e d thermal p a t t e r n s wizhin b r i q u e t s undergoing carbonization, t h e r e l a t i v e magnitude of thermal stresses. Excessive thermal s t r e s s e s are b e l i e v e d responsible f o r d e l e t e r i o u s f r a c t u r i n g . Knowledge of t h e s e s t r e s s e s produced under various conditions and methods of shock h e a t i n g would b e u s e f u l i n t h e s e l e c t i o n o f t h e most favorable process f o r producing i n t a c t formcoke. Some empirical data, gathered from s m a l l s c a l e laboratory experiments, o u t l i n e t h e g e n e r a l thermal regime required f o r successful carbonization of briquets. The c a l c u l a t e d temperature d i s t r i b u t i o n within these b r i q u e t s during carbonization can t h u s s e r v e as a guide f o r s e l e c t i o n of processes and equipment f o r l a r g e s c a l e equipment. The g e n e r a l p r i n c i p l e t h a t wil.1 b e adapted, as will be d i s cussed l a t e r , i s that t h e temperature gradient w i t h i n a b r i q u e t undergoing carboniz a t i o n s h a l l b e no g r e a t e r than t h a t corresponding t o t h e o p e r a b i l i t y limits pres c r i b e d by s m a l l s c a l e work.
Attainment of both of t h e above o b j e c t i v e s hinges upon s o l u t i o n of t h e pcoblem of h e a t t r a n s f e r t o and through a b r i q u e t . This problem d e f i e s s o l u t i o n i n r % rigorous a n a l y t i c a l form because both t h e thermal conductivity and t h e s p e c i f i c a r e stron@Y temperature dependent, as is e s t a b l i s h e d i n a companion paper.

-89Therefore, recourse w a s had t o an approximation method o r n m e r i c a l s o l u t i o n of the d i f f e r e n t i a l equations which involved use of an IBM 650 D i g i t a l Conputer. The machine computations were c a r r i e d out a t the University of ? i t t s b u r g h Computing Center and were progrmmed t o d e f i n e t h e t n e r n a l p a t t e r n i n a b r i q u e t a t various times as a f u n c t i o n of b r i q u e t s i z e , f i l m heat t r a n s f e r c o e f f i c i e n t , and t;io shock h e a t i n g methods - with n o t f l u e gas znd with hot f l u i d i z e d s o l i d s . Zstimation of the f i l m c o e f f i c i e n t f o r a h e a t i n g medium then allows s e l e c t i o n o f f h e appropriase h e a t i n g p a t t e r n f o r tine b r i q u e t s . An empirical c o r r e l a t i o n vas deri-red f r o n t k e computed r e s u l t s which r e l a t e s t h e rate of heasing oi Sriquecs t o Zil the i m p o r t a t variables. With t h e l a w of squares f o r heated bodies, the c o r r e l a t i o n permiss r a p i d e x t r a p o l a t i o n to. many h e a t i n g cases not d i r e c t l y covered i n t h e computer program. The accuracy of t h e c o r r e l a t i o n i s w i t i i n 7%. Dilatometer measurements irere made t o deternine t h e masnitude or r h e m a l expansion and c o n t r a c t i o n t h a t occurs during carbonizarion of 5 r i q u e t s . Ey coupling these d a t a with t h e computed t h e m a l p a t f e r n s , t h e r e l z t i v e .zagcitude of <?e-~ai. s t r e s s e s can be e s t i n a t e d f r o n t h e Tixosherko theory of e l a s t i c i t j . f o r heated i s o t r o p i c and e l a s t i c bodies. Direct experimental d a t a f o r t h e te,mperature rise of zhe c e n t e r o f 2 heated b r i q u e t a r e conpared with the computed r e s u l t s i r ? a few cases -Acre a conparison can be made. Reasonable agreement with t h e comyrablc c2ses i e z d s some degree of confidence t o t h e o t h e r conputed r e s . d t s . Mary of :he bric_ilet neazinz cases i n v e s t i g a t e d by machine conputation cannot b e e a s i l y e m n e d experinezfai.1;on a sffiall s c a l e . THE PROBLEM MID GEXER4L PROPERTIES OF TIE SOLbiONS
The problem of h e a t conduction i n carbonaceous n a t e r i d s 5 2 s keen attac..eC previously. Burke e t al. discussed soze years ago roathematiczl r e l a t i o n s h i g s f o r ?owever, i z the r a t e of heat conduction through c o a l undergoing c a r c o r i z a t i o n . o r d e r t o a r r i v e a t a n a n a l y t i c a l s o l u t i o n , these authors t r e i i e a %e case zhzf c o r responded t o t h e i l l e g i t i m a t e asslumption t h a t t h e t h e n a l d L i l f u s i v i ~ y , ( # ) , sizs independent of temperature. A constant o ( . f o r car3ozaceous m a t e r i a l s c a m o t be assumed. The s t r o n g temperature dependence of t h e r n a l conductivity ( k ) nzd s p e c i f i c OC i s given Sy a = k / p c vhere p h e a t (c) has been shown i n a companion paper. i s t h e density. The non-constant conductivity of cozi w a s a l s o reported by Millard2 who attempted s o l u t i o n of t h e Fourier heat-flow equation by ilse o f an e1ec:ricd e a u l d o b t a i n agreement between c a l c u l a t e d d a t a and experianalog technique. H mental t h e r m d p a t t e r n s i n a coke oven only by i n j e c t i n g s i z a b l e heats-of-carbonization into the calculations.

The problem t h a t concerns u s i s t o d e f i n e t h e temperature p r o f i l e as a function of t i m e within a s p h e r i c a l briquet undergoing carbocization, taking full account of t h e tem?erature dependence of the t h e m a l paracceters the conducrivizy and s p e c i f i c heat. Two d i f f e r e n t merhods of d i r e c t shock heafing are considered, . namely, with hot gas &d hot f l u i d i z e d s o l i d s .

A t l e a s t i n p r i n c i p l e , the problem c a n b e handled by s o l u t i o n of he Fourier heat conduction equation. The equatioc, i n p o l a r coordinates, f o r t h e general case of a sphere wherein t h e thermal d i f f u s i v i t y i s temperature dependent, is

-90-

This equation must b e s o l v e d with t h e s e boundary conditions:

T = TI a t t =

f o r all Values of r.

d x

To, t h e ambient temperature o f t h e h e a t i n g medium surrounding t h e b r i q u e t , v a r i e s u n l e s s the h e a t content o f t h e nedium g r e a t l y exceeds t h a t of the b r i q u e t . The v a r i a t i o n of To K i t h time can b e derived from h e a t balance, y i e l d i n g

"he above r e l a t i o n holds f o r continuous carbonization where concurrent flow of h e a t i n g medium and b r i q u e t s p r e v a i l . It a l s o holds f o r batch carbonization where no temperature g r a d i e n t exists within t h e h e a t i n g medium.

It thus becomes evidenz, t h a t s o l u t i o n of t h e heat flow equation (l),wi'& t h e com?lex boundary c o n d i t i o n s , as w e l l as t h e non-constant parameters k and c, cannot be obtained by the usual. a n a l y t i c a l methods. A numerical method of s o l u t i o n i s consequently necessary.
Although an a n a l y t i c a l s o l u t i o n of t h e above equations cannot be obtained, c e r t a i n i n t e r e s t i n g p r o p e r t i e s of t h e s o l u t i o n s follow f r o m t i e form of equation (1) and t h e boundary conditions.
L ? t new v a r i a b l e s b e introduced, namely X = r/ro, t ' = t/ro2, and h' = hro. I t can sow b e r e a d i l y v e r i f i e d t h a t t h e temperatures, as expressed i n equations (I) and ( 2 ) , becoae f u n c t i o n s only of X, t ' , k , h ' , and M. It follows t h a t t h e solution is a function only of X and t ' i f k , M, and h r o are maintained constant. ConseqttentPy, a ? eqilivalent temperzture p r o f i l e i s e s t a b l i s h e d within a b r i q u e t a t corresponding values of t h e r e l a t i v e r a d i u s X a t a t i m e i n v e r s e l y proportional t o t h e square o f t h e r a d i u s of t h e b r i q u e t . Tnis i s a more generalized expression of t h e l a w of squares Finich has been previously discussed2 and w i l l become u s e f u l l a t e r .

Before proceeding with t h e numerical analyses, i t i s necessary t o e s t a b l i s h t h e values o f the parameters needed. These parameters a r e f o r b r i q u e t s of PitLsburgh Seam c o a l and product char. The Lhermal c o n d u c t i v i t y measured as a f u n c t i o n of temperature f o r b r i q u e t s is reported i n t h e comparion paper. Tne s p e c i f i c h e a t f o r b r i q u e t s w a s obtsined f r c m t h e a d d i t i v e equatior. f o r the componentsthus,

where t h e nmbers correspond t o t h e weight f r a c t i o n of t h e b r i q u e t conponents. The s p e c i f i c heats of c o a l and char were taken from measurements given i n the companion paper. The s p e c i f i c h e a t o f p i t c h w a s taken from Hynan and Kay3. These d a t a were adequate t o cover t h e range from room temperature t o 500C. Above .%O"C, it was assumed t h a t t h e b r i q u e t was carbonized and t h e d a t a of Terres4 f o r char and coke w a s used.

Graphical r e p r e s e n t a t i o c s . o f t h e s p e c i f i c n e a t and t h e m d conductivity as functions of temperature were reduced t o a l g e b r a i c expressions priroarily f o r purposes of machine computation. The @-Taphs were f i t t e d t o polynomial equazions by a standard l e a s t - s q u a r e s sub-routine used i n t h e IBM 550 D i g i t a l Computer. The a n a l y t i c a l expressions were found t o be

f o r t h e s p e c i f i c h e a t of b r i q u e t s and

for t h e thermal conductivity of b r i q u e t s .

T i s expressed i z

OC.

Tne c o e f f i c i e n t s and exponents were s t o r e d i n the computer merrorjr section. %e appropriate values of C ( T ) and k(T) could be detemdned b y The =chine whenever needed i n t h e o v e r a l l computations.
The measured d e n s i t y o f r a w b r i q u e t s i s 0 . 8 gn/cm3. It vas assim?d r?ot t o change during carbonization. This assumption i s acceptable s i n c e shrinkage conpe2s a t e s approximEtely f o r t h e loss i n weight ( v o l a t i l i z a t i o n ) during c a r b m i z a t i o n . Evaluation of t h e f i l m c o e f f i c i e n t s i s a l s o explained i n t h e c o n p a i o n paper. Values of 20 and 50 Btu/hr f t 2 Fo were assigned f o r the conputations on s o l i d s heating, t o b r a c k e t expected values f o r h e a t t r a n s f e r from f l i l i t i z e d . s o l i C s t o briquets. For the gas film h e a t t r a n s f e r c o e f f i c i e n t , 9.5 Stu/nr f t 2 F" was employed i n t h e c a l c u l a t i o n of t h e h e a t i n g rate of two-inch b r i q u e t s by not f h e gas. This value derives from c o r r e l a t i o n s by Gamson' e t al. and by Wil!!eg e t d. and corresponds t o a flow rate of 375 l b s . g a s / f t 2 hr. Tnis i s c o n s i s t e n t w i t 5 a p h y s i c a l s i t u a t i o n i n which a l 5 - f t . high shock h e a t i n g zone, containirig b r i q u e t s , is i n j e c t e d with 2 2 0 0 ' F f l u e g a s and i n which t h e b r i q u e t residence tine is 80 minutes.

The problem of 'the rate of h e a t i n g of b r i q u e t s can be solved by a p p l i c a t i o n of numerical methods of a n d l y s i s similar t o the s t e p methods descri'aed by Izge:scl17. I n a generalized problem as complex as t h e one encountered here, hand c a l c ~ u l a t i o n s would become p r o h i b i t i v e l y long and tedious. Only t h e speed of modern computing machines a l l o w s one t o consider attempting such numerical s o l u t i o n s . I4achiLir.e computing time f o r a s i n g l e s e t of conditions, a s i n g l e h e a t i n g case, w a s an hour and a half.
I n i t i a l attempts with t h e D i g i t a l Computer t o solve t h e general numerical problem by applying "relaxation" operations t o t h e s t e p method7 proved impractical. programming t h e machine t o make judicious p r e d i c t i o n s of temperature changes f o r small t i m e i n t e r v a l s w a s d i f f i c u l t and i t required t o o much t i m e i n working through many erroneous guesses before s t r i k i n g upon ones s u f f i c i e n t l y c o r r e c t . Therefore, a "guess-free" s t e p method w a s developed t o t r a n s l a t e the numericd. procedure i n t o a form more s u i t a b l e for machine computations.

To f a c i l i t a t e numerical s o l u t i o n of t h i s problem, t h e physical. process of h e a t transfer i s a r t i f i c i a l l y resolved i n t o two d i s t i n c t , s e q u e n t i a l processes f i r s t , isothermal flow of h e a t from one s e c t i o n of matter t o another f o r a s h o r t

-92-

time, '$ad second, a r e s u l t a n t change i n tempeqature of t h e s e c t i o n based upon i t s h e a t balance during t h e iso=kermal period. This a r t i f i c e is most h e l p f u l i n re,] ducing t h e h e a t transfer problem i n t o simple f i r s t o r d e r d i f f e r e n c e equations t h a t 1 a r e e a s i l y t r a n s l a t e d i n t o t h e b a s i c langoage of digitd. computation. In r e a l i t y , j l t h e flow of heat and change of temperzture are not s e q u e n t i a l , as pictured here, b u t occur simultaneously, so t h a t o u r s o l u t i o n of t h e problem i s an approximate : one, tinough of very c l o s e approximation. The numerical s o l u t i o n approaches t h e rigorous one as t h e increments of time and space that e n t e r t h e computations a r e '/ made smaller f o r i f t h e i n c r e x e n t s become i n f i n i t e s i m a l s , the s o l u t i o n would be a t r u e c a l c u l u s i n t e g r a t i o n of the d i f f e r e n t i d equations. The use of an e l e c t r o n i c \ d i g i t a l computer, w i t h i t s extremely r a p i d computation permits computation of a , s e t o f equations f o r a g r e a t number of very small time and space increments; thus, o u r s o l u t i o n i s v e r y n e a r l y rigorous.
)t

'
t
1
I

D e f i n i t i o n s o f symbolic terms used i n the following c a l c u l a t i o n s and discussions appear i n tkne Appendix.

I n t h i s approximation method, a b r i q u e t i s considered as being made up of lTlo. t e n concentric s p h e r i c a l l a y e r s having i n i t i a l temperatures of lT1, 1T2 The b r i q u e t i s heated by h o t gas o f i n i t i a l temperature lTo. For t h e duration of ' a s h o r t time i n t e r v a l , A t , t h e s e temperatures a r e assumed constant w h i l e h e a t con- ' duction proceeds, t n e d r i v i n g f o r c e being d i f f e r e n c e s between lT1 II Tlo. , The h e a t t r a n s f e r r e d i s given by

",

651

--A N (6);
Z

f o r t h e t r a n s f e r fron gas t o the f i r s t l a y e r , and by a; ( 7 2 Ti+:1 of far SC to*/ P A

-&4&

- XC,~;.) ~ C L ,

"= 1 . '

/o
(7)-

f o r t h e t r a n s f e r from l a y e r t o l a y e r . t h e end o f a tine i n t e r v a l i s

The balance of h e a t l e f t i n i ' t h l a y e r a t

A&* =

''
/

- $5 c>/

(8)

which upon s u b s t i t u t i o n of equations ( 6 ) and (7) l e a d s t o

(9) - 8% f o r t h e gas s i n c e can l o s e heat. F o r t h e f i r s t l a y e r , t h e h e a t balance is A$, = so,/ - el,+ = ( x - : t j A t - &$) a,cr- :& 1A f

A$,

it

only

A&/=

R(J)&:
A%,

h.
)

AR

(101

= 6, (z

where by d e f i n i t i o n

- a, (,T,-,Ta

For the remaining l a y e r s , t h e r e s i d u a l h e a t i n each l a y e r i s

(12 1

i'

-93Equations (9) through (12) now give t h e balance of h e a t l o s t by t h e gas and gained Fy each l a y e r during t h e a r b i t r a r y s h o r t time i n t e r v a l , At. Now, the temperatures ?f t h e gas and l a y e r s are allowed t o change by v i r t u e of t h e heat l o s t or gained

so
\ \

that

Ago

7n. =e

(8%

-aL*

iand
I

4%;

= mic ( & G -#c)

A&
A from equations (9) and (13) and defining .
= MoCo y i e l d s

-Elimination of 4

,
'i

4 (;r, -85 1 -Bk.+

Be

( 1 5

-,TI
4 = mic

(15)
(lTi) y i e l d s

k i m i n a t i o n of 4&.from

equations (12) and (14) and d e f i n i n g

At ('.c l T 1=

(tc+-tT ) - 4. ( 8 T - ;rL.+i )
+.4* -4),7;:
+

(16)

,Since t h e 1T values are known from chosen i n i t i a l conditions the 2T's are t h e only hknowns involved i n equations (15) and (16), both of which can b e re-arranged t o $he form,

itI

A'#zq = B & ,q,

. ' 3 1

,7;*+,
.

(17)

4'
:which i s completely general i f it i s remembered t h a t 8,.-/ does n o t e x i s t when i = o F d t h a t Bo has a d i f f e r e n t form than all t h e o t h e r B i values ( s e e equation (11)). ;mu., t h e temperature of t h e gas, and of each l a y e r i n t h e b r i q u e t , a t t h e end of t h e f i r s t t i m e i n t e r v a l , are computed d i r e c t l y from equation (17). These end temperatures, (2Ti), then become t h e i n i t i a l temperatures, (lTi), f o r t h e next t i m e :interval, and t h e computations are repeated t o determine t h e temperatures a t t h e lend of t h e second i n t e r v a l . I t e r a t i o n of this procedure, f o r many t i m e i n t e r v a l s , !in t h e computer y i e l d s the temperature p r o f i l e of t h e gas and t h e l a y e r s of t h e b r i q u e t as a function of t i m e . Ten-second t i m e i n t e r v a l s were used i n computations 'for 2 and 3-inch b r i q u e t s and one-second i n t e r v a l s f o r t h e 1-inch b r i q u e t .
1

Eight b r i q u e t h e a t i n g cases were solved d i r e c t l y on t h e computer, covering $ a r i a t i o m of t h e following parameters , i n i t i a l temperature of t h e h e a t i n g medium, ,mass r a t i o of h e a t i n g medium t o b r i q u e t s , f i l m c o e f f i c i e n t , and the b r i q u e t radius. ,The l a w of squares, mentioned e a r l i e r , can be used t o apply t h e r e s u l t s t o d i f f e r e n t 'size spheres by a d j u s t i n g t h e value of h t o m a i n t a i n ' h r o = constant. The cases studied are o u t l i n e d i n Table I.

: The f i r s t f o u r cases, involving high i n i t i a l temperatures and a low value ,of h describe shock h e a t i n g with h o t gas. The s p e c i f i c h e a t and molecular w e i g h t of t h e h e a t i n g m e d i u m i n t h e s e cases was chosen t o correspond t o that of f l u e gas.
i
)

The o t h e r cases i n Table I describe carbonization of b r i q u e t s with h o t f l u i d i z e d s o l i d s , such as char.

I The temperature d i s t r i b u t i o n p a t t e r n s solved f o r t h e eight programmed ,cases are shown i n Figures 1 through 8. Figures 1 through 4 show t h e d i s t r i b u t i o n bbtained f o r shock h e a t i n g with hot gas. Figures 5 through 8 show t h e d i s t r i b u t i o n f o r t h e cases where a f l u i d i z e d s o l i d h e a t c a r r i e r i s used f o r shock heating. These rigures i l l u s t r a t e immediately t h e h e a t i n g times involved i n carbonization of diff e r e n t s i z e b r i q u e t s under various conditions.

-9LThe time s c a l e i n each case i s given f o r e i t h e r a 1, 2 o r 3-inch b r i q u e t ' as noted i n t h e p a r t i c u l a r f i g u r e . The conversion f a c t o r t o convert t h e time s c a l e t o o t h e r b r i q u e t sizes by t h e l a w of squares i s a l s o noted on each f i g u r e . I The expected q u a l i t a t i v e trends a r e c l e a r l y amination of these f i g u r e s . A f a s t e r rate of heating being equal, when 1)t h e shock heating temperature is c o e f f i c i e n t becomes l a r g e r and 3 ) t h e b r i q u e t s become i n evidence upon close exi s observed, other things increased, 2) t h e film smaller.

\
I
(

Some of t h e s e t r e n d s can be demonstrated more q u a n t i t a t i v e l y by combining t h e s a l i e n t f e a t u r e s f o r s e v e r a l cases on individual p l o t s . f i g u r e 9, for example,: shows t h e e f f e c t o f b r i q u e t s i z e on t h e r a t e of r i s e of t h e b r i q u e t center temperat u r e with various v a l u e s of h. The l a w of squares s t a t e s that the rate of temperat u r e r i s e i s i n v e r s e l y p r o p o r t i o n a l t o t h e square of b r i q u e t radius when hro i s constant. If h is h e l d constant, t h e r a t e here does n o t decrease q u i t e as rapidly.: The r a t e , i n t h i s case, decreases roughly as t h e 1.75 power of t h e b r i q u e t radius. It i s c l e a r , that as h becomes very l a r g e , t h a t the l a w of squares w i l l again become valid. A very l a r g e value of h corresponds t o t h e case i n which t h e surface temperature i s h e l d equal t o t h a t of t h e heating medium. Conversely, as h becomes very small t h e temperature within t h e b r i q u e t w i l l tend t o become uniform a t all p o i n t s and t h e rate of temperature rise becomes i n v e r s e l y proportional to t h e f i r s t power of t h e b r i q u e t radius. The e f f e c t of t h e value of t h e h e a t t r a n s f e r c o e f f i c i e n t on t h e r a t e of temperature rise o f t h e s u r f a c e and c e n t e r temperatures f o r t h e case of a two-inch . As t h e value of h i n c r e a s e s beyond 50 Btu/hr ftb r i q u e t may be seen i n Figure 9 t h e rate of temperature r i s e tends t o become independent of h.

The e f f e c t of the f i l m c o e f f i c i e n t on t h e time required f o r t h e b r i q u e t c e n t e r t o reach a given t e m p e r a t u r e i s i l l u s t r a t e d i n Figure 10. The time required t o reach temperature a g a i n becomes independent of h f o r large values.
Some experiments w e r e conducted t o observe t h e temperature r i s e a t t h e b r i q u e t c e n t e r f o r comparison with t h e t h e o r e t i c a l behavior. The experiments were arranged so t h a t the results could be compared w i t h Cases I V through VI11 where char w a s used as t h e h e a t i n g medium. These computed cases w e r e set up w i t h a decreasing temperature of t h e char f r o m an i n i t i a l value of l35O0F t o a f i n a l equilibrium temperature of lllO'F. The above temperature p a t t e r n (set up as repres e n t a t i v e of a continuous process) could not be conveniently reproduced i n t h e laboratory.

The experiments w e r e , therefore, conducted a t a uniform 1200F i n t h e 8" f l u i d i z e d sand bath. A thermocouple w a s i n s e r t e d i n t o t h e c e n t e r of t h e b r i q u e t which was plunged i n t o t h e sand b a t h and t h e temperature h i s t o r y w a s continuously recorded.
Measurements of t h i s kind were made w i t h l", 2" and 3" diameter s p h e r i c a l b r i q u e t s . The l i n e a r f l u i d i z i n g v e l o c i t y of t h e sand b a t h w a s maintained constant at 0.26 f p s f o r 1 and 2-inch briquets. I n order to check t h e e f f e c t of l i n e a r v e l o c i t y a second and h i g h e r v e l o c i t y of 0.45 f p s w a s a l s o employed f o r the 2-inch b r i q u e t s . Since no e f f e c t of f l u i d i z i n g v e l o c i t y was found i n t h i s range, threeinch b r i q u e t s were measured only a t 0.45 f p s . The experimental measurements a r e compared with the c a l c u l a t e d r a t e of temperature r i s e i n Figure 9 f o r all t h r e e b r i q u e t s i z e s . The curves shown a r e t h e c a l c u l a t e d curves f o r d i f f e r e n t values of h as parameter.

-95It should be remembered here t h a t t h e work reported i n t h e previous paper f o r a l u m i n u m spheres i n d i c a t e s t h a t t h e c o r r e c t value o f h i s i n t h e neigbborhood of 30 Btu/hr f t 2 .
It i s seen +chat i n all cases t h e rate of temperature rise i n i t i a l l y i s greater than t h e c a l c u l a t e d f i g u r e s . This phenomenon i s undoubtedly n a i n l y due t o a conduction thermocouple e r r o r . This w a s shown by t h e following experiaer.t. The exposed extrusions of a thermocouple i n j e c t e d i n t o t h e c e n t e r o f - a 2'' b r i q u e t -das heated e l e c t r i c a l l y t o LLOO'F. .Liotiier unheated couple was i n s e r t e d a l s o t o tce b r i q u e t center at a 90" angle t o the former orie. It was found t h a t t h e heated thermocouple would read as much as 100F above t h e urheated codpole. This m c h thennocouple e r r o r i s s u f f i c i e n t t o b r i n g about agreement between t h e lower teaperat u r e experimental p o i n t s and a c a l c u l a t e d curve expected f o r an h of JO for 2" b r i q u e t s , as can be seen from Figure 9.

As t h e temperature of t h e b r i q u e t rises the 'hernocouple e r r o r n a z a r a l l y would diminish q u i t e r a p i d l y both because of t h e smaller t a p e r a t i i r e d i f f e r e n t i a l and t h e increase i n t h e d conductivity of the b r i q u e t material.
The lowest value of h a v a i l a b l e from &&e c a l c u l a t i o n s f o r t h e 1" sphere is 40. Extrapolation by t h e a i d o f the curve shown in Figure 10 t o a n h value of 30 shows that q u i t e good agreement e x i s t s between theory and e x p e r b e n t a f t e r allowing f o r the i n i t i a l thermocouple e r r o r . It i s a l s o noted i n t h e case of t h e two-inch b r i q u e t , that t h e e f f e c z o f f l u i d i z i n g v e l o c i t y on t h e experimental r a t e of temperature r i s e is n e g l i a o l e . It is seen likewise, t h a t good agreement between theory and e q e r i a e c t i s oozzized. after t h e temperature r i s e s above j 0 0 " C by assigning a lower than g r e d i c t e d vzlde of h of the o r d e r of 20.
The agreement a t higher temperatures becomes r a t h e r poor i n the Zase of t h e t h r e e - i n c h . b r i q u e t s i n c e one must a s s i g n a value below lJ t o h t o obtair. agreement i n t h i s case.
The explanation o f t h i s higher temperature discrepancy i s thou&: a ; lie i n the r e t a r d i n g e f f e c t of v o l a t i l e matter r e l e a s e on t h e p e n e t r a t i o n o f zezt :hrou&i t h e surface of t h e briquet; i.e., e f f e c t i v e l y on t h e value of h. This e f f e c t vas neglected i n t h e c a l c u l a t i o n s because of the added conplexity i t would ilave int-roduced. It should be noted, however, t h a t t h e rate o f v o l a t i l e matter r e l e a s e p e r u n i t of b r i q u e t surface increases p r o p o r t i o n a t e l y t o the b r i q u e t radius, nakizg i t more serious f o r the l a r g e r b r i q u e t s .
EMpIRlCAL COilRFLATIONS FOR E W T I N G RATES

It i s desirable, i f p o s s i b l e , t o have a v a i l a b l e a s i m p l i f i e d c o r r e l a t i o n encompassing t h e c a l c u l a t e d r e s u l t s . If such a c o r r e l a t i o n can be derived it would simplify extrapolation t o cases t h a t were not direc'dy considered and the a p p l i c a t i o n of t h e c a l c u l a t e d rate of h e a t i n g t o many design problems.
A c o r r e l a t i o n w a s developed t o f i t t h e f o u r cases, V through VIII, cons i d e r e d f o r s o l i d s heated briquezs. !These cases correspond, f o r a two-inch s p h e r i c a l b r i q u e t , t o a range of values f o r h o f 20 t o 7 5 Btu/hr f t 2 Fo.

The c o r r e l a t i o n i s based on t h e use of t h e empirical equation

-96-

T i s approximately the memL b r i q u e t t e q p e r a t u r e and Ts i s the temperature of the f l u i d i z e d s o l i d s m e d i u m . T i s e x a c t l y defined b?- t h e equation
-

(
//

To i s t h e i n i t i a l b r i q u e t

r=

9 / r

(19 1
I,

temperature, Q i s t h e amount of h e a t absorbed by t h e b r i q u e t and C is t h e mean s p e c i f i c h e a t of t h e b r i q u e t over t h e whole carboniz a t i o n range. T would be e x a c t l y e q u a l t o t h e mean b r i q u e t temperature i f the s p e c i f i c h e a t of t h e b r i q u e t were independent of temperature, which it i s not.

It i s c l e a r t h a t e q u a t i o n (18) i n view of equation (19) can a l s o b e w r i t t e n as follows

d Q t

= E-KCT,-T)

,
(20)

This equation i s obviously f a l l a c i o u s s i n c e t h e e x a c t equation i s

dt
Where TI i s t h e temperature of s u r f a c e of t h e b r i q u e t .

(21)

I f equation (20) i s v a l i d , i t can only mean t h a t t h e i n c r e a s e i n t h e thermal conductivity of t h e b r i q u e t with temperature, i s such t h a t T follows, f o r tuituouslj-, ?;he r e l a t i o n s h i p

The t e s t f o r . e q u a t i o n (18) i s shown g r a p h i c a l l y i n Figure ll where t h e l o g (Ts T) i s p l o t t e d a s a i n s t t i m e . The p o i n t s shown a r e derived from t h e computed r e s u l t s of Cases V through VIII. I n the p a r t i c u l a r cases s t u d i e d here, h e a t balance considera t i o n s give r i s e t o t h e r e l a t i o n s h i p

--

T, - r =

t g.73 - 1 . 2 3 7

The slooe o f t h e s t r a i g h t l i n e s shown in Figure 1 1 are equal t o K of equation (18) m u l t i p l i e d by 1.23.

To i s t h e i n i t i a l temperature of t h e h e a t i n g m e d i u m i n degrees centigrade.

It i s c l e a r t h a t t h e c o r r e l a t i o n holds with adequate accuracy as noted by t h e l i n e a r i t y of t h e p l o t s . Tne c a l c u l a t e d slopes K'(= 1.23 K ) a r e given f o r t h e d i f f e r e n t h values on t h e graph.
It now remains, t o complete the c o r r e l a t i o n , t o account f o r t h e v a r i a t i o n o f K w i t h h and with briquer; r a d i u s r. The v a r i a t i o n w i t h h, at constant r of one inch, i s adequately expressed by t h e empirical equation

where a = .00834, b = .0369,

t h e time i s i n minutes and h has u n i t s of Btu/hr f t 2 Fa.

The t r a n s p o s i t i o n of equation (24) t o o t h e r b r i q u e t s i z e s i s c a r r i e d out r e a d i l y by the l a w of squares. The f i n a l c o r r e l a t i o n i s given below where r is i n f e e t and t i n minutes.

This e q u a t i o n f i t s t h e computer c a l c u l a t i o n s shown i n Figure 1 1 with an accuracy

of 5%.

-97THEWIL STRESSES
Thermal s t r e s s e s can a r i s e during t h e coking operation as a result of the temperature d i s t r i b u t i o n produced within t h e briqilet combined with either a cont r a c t i o n o r expansion of t h e b r i q u e t material. To prsvide a b e t t e r urrderstanding of t h e nature of the t h e d s t r e s s e s e x i s t i n g within t h e b r i q u e t , a b r i e f i i i a tometer study w a s made of the thermal. expansion and c o n t r a c t i o n c h a r a c t e r i s t i c s ol' t h e b r i q u e t material. The dilatometer used is simply an e l e c t r i c a l l y heated v e s s e l about 1 c i ~ I.D. containing a 2 c m high sample. B e sample supports a rod which i s counterbalanced by a s m a l l weight a t t a c h e d t o a s t r i n g which p r e s s e s over a pulley. P-e p u l l e y i s f i t t e d with an i n d i c a t i n g needle whose displacement can be calibrated. i n terms of l i n e a r expansion o r c o n t r a c t i o n o f t h e sample. Using a s l o w r a t e of heating, about 3 t o 5 F0 p e r minute, t h e sample i s assumed t o be a t t h e same temperature as t h e container, and a t e s s e n t i a l l y uniform temperature throughout.

The dilatometer s t u d i e s were made on sanples of b r i q u e t s o f m a t e r i a l from Pittsburgh Seam coal, p a r t i c u l a r l y from two of our West Virginia mines, t h e Arkwrighz and t h e Moundsville mines.
The r e s u l t s of two slow h e a t i n g runs on m a t e r i a l derived f r o m A r k w r i & t c o a l appear i n Figure 12. I n one case the s-le w a s c u t from a r a w b r i q u e t and i n t h e o t h e r case, l o o s e mix (not b r i q u e t t e d ) w a s used. The b r i q u e t sample conzracted ~ an o v e r a l l l i n e a r shrinkage of 90 upon s h a r p l y a t nominally 400 and 8 0 0 ' ~with c a l c i n a t i o n t o 1500F and subsequent cooling. The loose mix d i d n o t snow as nu& shrinkage. This is probably a r e f l e c t i o n of the less i n t i m a t e c o n t a c t of t h e p a r t i c u l a r ingredients i n t h e mix. The c o a l and p i t c h m y becone coked as s e p a r a t e p a r t i c l e s i n t h e mix whereas they envelop o r p e n e t r a t e t h e char i n the b r i q u e t making it one s o l i d body upon coking. Figure 13 shows the negative thermal expansion of a b r i q u e t sample of material. f r o m Moundsville coal. It had an o v e r a l l shrinkage of 13.54 and e x h i b i t e d an abrupt c o n t r a c t i o n a t about 700 t o 8 0 0 ' ~as i n the above Case..
.#

Figure 14 shows t h e r e s u l t s o f a run which i s an attempt t o simulate shock heating. The sample and c o n t a i n e r were set i n t o t h e furnace at l p F which immediately dropped t o ca. 1000"F, which i s t h e shock h e a t i n g t e n p e r a t u r e wanted. The furnace w a s maintained a t this temperature f o r 20 minutes a t which time the c e n t e r of t h e sample w a s n e a r l y a t t h e same temperature as the wall (lOOOF). The temp e r a t u r e was then raised t o 1500F. I n this case the sharp c o n t r a c t i o n a t about 8 0 0 ~ seen ~ i n Figure 1 w a s not evident, probably because the whole sample vas not a t that temperature a t any one time.

It w a s hoped t h a t use could be made of the above d a t a v i t h exact methods that are a v a i l a b l e f o r c a l c u l a t i o n of t h e thermal stresses developed w i t h i n an i s o t r o p i c and elastic body upon h e a t i n g o r cooling. The t h e d stresses are a f u n c t i o n of t h e temperature d i s t r i b u t i o n w i t h i n t h e body, t h e shape o f t h e body, the c o e f f i c i e n t of t h e 4 expansion , t h e modulus o f e l a s t i c i t y E and Poissons number t/

The t a n g e n t i a l s t r e s s dt at' a r a d i a l p o s i t i o n r of a sphere o f radius ro i s given by t h e following expression due t o TLmosheL=koe,

-90The expression f o r t h e c a s e of a hollow sphere w i t h an i n n e r r a d i u s a and a n outer r a d i u s b is given b y 'pimoshenkoe as follows

\ m e use of t h e s e expressions t o c a l c u l a t e t h e exact magnitEde of the stresses e x i s t i n g i n b r i q u e t s during coking i s n o t possible. %e rnain d i f f i c u l t y i s that we are d e a l i n g w i t h a aaterial t h a t i s not homogeneous and that i s changing i n chemIca3. and p h y s i c a l s t r u c t u r e w i t h time and temperature. Likewi-se, t h e physical , parameters ac E, and V are not known exactly. I n f a c t , t h e above d a t a show t h e r e i s no constant d

m e b r i q u e t s actually undergo shrinkage r a t h e r than expansion as the temg e r a t u r e r i s e s due t o t h e above mentioned p h y s i c a l changes. If we pennit ourselves a rough approximation of t h e dilatometer r e s u l t s shown i n Figure 24, t h e shrinkage may then be t r e a t e d as l i n e a r with t h e temperature. Under such condLtions the Timoshenko equation can be employed t o c a l c u l a t e r e l a t i v e thermal stresses by t r e a t i n g the f a c t 3 r ( &E/~-Y) as an ~nlmownparameter f o r d i f f e r e n t coiting cond i t i o n s , b u t which i s assumed to be constant. m e b r i q u e t mix remains in a p l a s t i c condition due t o s o f t e n i n g of t h e ?$+ah and coal u n t i l a rigid coke bond. is formed. Therefore, thermal s t r e s s e s can only be s e t up e t h i n t h e r i g i d p o r t i o n of t h e b r i q u e t , while t h e imer p l a s t i c region ilndergoes r e l a x a t i o n of any imposed s t r e s s e s by flow i n t o t h e r i g i d s h e l l . !the problem c a t h e r e f o r e be handled by a p p l i c a t i o n of equation (27) f o r t h e case of a hollow sDhere. The a p p l i c a t i o n of t h i s method r e q u i r e s t h a t a loore o r l e s s a r b i t r a r y d e c i s i o n be made r e l a t i v e to t h e temperature at which p l a s t i c i t y of r;he mix disappears. It has been assumed in w h a t follows t h a t t h e mix becomes r i g i d a t 8 0 0 " ~ . This may be i n some error, b u t d e f i n i t i o n of t h e s o l i d i f i c a t i o n temperature is not too important, however, s i n c e we are merely concerned with r e l a t i v e s t r e s s e s . The method adopted t h e r e f o r e was t o compute t h e r e l a t i v e thermal s t r e s s over t h e coked p o r t i o n of t h e b r i q L a , i.e., over t h e s h e l l where t h e temperature w a s 8 0 0 and ~ ~ h i g h e r as a f u n c t i o n of t i m e , coking conditions and b r i q u e t size. For s i m p l i c i t y , t h e c a l c u l a t i o n s were r e s t r i c t e d t o computing t h e tangentiall s t r e s s a t t h e s u r f a c e of the b r i q u e t only. Under t h e s e conditions, it can be s h a m that equation (27) reduces t o

where c = a/b and y = r/b. The camputed temperature d i s t r i b u t i o n p a t t e r n s given i n Figures L.through 8 w e r e employed Kith this equation t o o b t a i n t h e r e l a t i v e stress results discussed below.

The a p p l i c a t i o n of t h e !rfmoshenko equqtion t o t h e c a l c u l a t i o n of thermal stresses can be readily c r i t i c i z e d s i n c e t h e equations apply t o an e l a s t i c body of c o n s t a n t chemical structure vhich i s n o t t h e case here. It i s c l e a r , however, t h a t

-99

i n any case t h e thermal. s t r e s s e s are g r e a t e r t h e s h a r p e r t h e temperature gradients within t h e body. The Timoshenko equations merely provides a convenient framework upon which t o make a s e e - q u a n t i t a t i v e evaluation of r e l a t i v e thermal g r a d i e n t s f o r Lifferent heating patterns.

TO e s t a b l i s h a background of comparison f o r determining which r e l a t i v e stresses can be expected t o exceed the f r a c t u r i n g limits, a s u c c e s s f u l regiqe t h a t has been worked o u t experimentally f o r shock carbonization of b r i q u e t s without f r a c t u r i n g follows. I n z a c t and non-deformed two-inch foniicoke has been produced by shock heating i n a f l u i d i z e d sand bath, provided t h a t t h e sand temperamre was w i t h i n t h e range of 900 t o l150F. The proper regime f o r h o t gas f o m c o K i n g is d i f f i c u l t , if not n e a r l y impossible, t o i n v e s t i g a t e on a IaboratorJ scale.
The c a l c u l a t e d r e l a t i v e thermal stress i n 2-inch b r i q u e t s heated *atin gas i s shown as a f u n c t i o n of t h e i n i t i a l gas t e q e r a t u r e i n Figure 15. I n all cases t h e equilibrium temperature w a s maintained n e a 1000F by s e l e c t i o n of t h e quzntrLji o f gas. The maximum thermal stress i n c r e a s e s as expected with i n i t i a l gas t e i q e r a t u r e , b u t a t a r e l a t i v e l y low r a t e . The thexmal stress f o r char heated b r i q u e t s as a f u n c t i o n of b r i q u e t s i z e , t i m e and film c o e f f i c i e n t i s shown i n Figure 15. The equilibrium temperature i n d l cases was constant a t 1100F. It i s noted t h a t the thermal stress i s of t h e same order, o r higher, than i n t h e gas cases even where 2600'F gas was used. One may conclude, a t least t e n t a t i v e l y , from t h i s t h a t two-inch b r i q u e t s may b e succ e s s f u l l y coked without f r a c t u r i n g even when 2 6 0 0 " ~gas i s used.

The o t h e r r e l a t i o n s h i p s i n Figure 16 show t h e a n t i c i p a t e d i n c r e a s e i n . The m a j o r thermal stress with i n c r e a s i n g b r i q u e t s i z e using a f i x e d value of h i n c r e a s e i n maxhum stress i s between the one and two-inch s i z e with a r e l a t i v e l y small i n c r e a s e between t w o and three inches. This complies w i t h e x p e r i n e n t a l f i n d i n g s , that 1"b r i q u e t s s u r v i v e t h e successful regime e s t a b l i s h e d f o r 2" briquets.
Experimentally, it is d i f f i c u l t t o produce f r a c t u r e - f r e e 3'' carbonized b r i q u e t s by shock h e a t i n g i n f l u i d i z e d sand. The slight i n c r e a s e i n themal s t r e s s i n going from 2 t o 3" b r i q u e t s , as shown i n Figure 16, must be c r i t i c a l . Certain q u a l i t a t i v e conclusions are s i g n i f i e d by t h e s e results. It i s r e a d i l y seen f r o m equation (28) t h a t i f t h e temperature d i s t r i b u t i o n wit.hizi a b r i q u e t i s i d e n t i c a l with r e s p e c t t o t h e r e l a t i v e r a d i u s y t h a t the zhhe-rmal stress should b e i d e n t i c a l and independent of b r i q u e t s i z e . The l a w of squares stases t h a t the temperature d i s t r i b u t i o n versus y w i l l go through e x a c t l y t h e s m e iequence when p l o t t e d a g a i n s t t h e reduced time scale t/ro2 i f h r o i s constan',. Snce t h e non-dependence of stress upon b r i q u e t s i z e i s n o t observed e i t h e r e x p e r i x x t a l l y o r from the computed results (Figure 16) it can only b e concluded that h does riot decrease i n v e r s e l y with r. Actually, i t is f e l t t h a t h decreases less rapidly, and therefore, the thermal stress does i n c r e a s e with s i z e .

ACKNOWLFEEMENT
!Thanks are extended t o Dr. W i l l i a m Kehl and Nicholas Sabers of t h e University of P i t t s b u r g h Computing Center f o r helpfuL d i s c u s s i o n s and t h e f i n a l d e t a i l e d programming o f t h e problem i n t o computer language:

-100-

L1'IXRATUR.E CITED
1

Burke, S. P. and Others. "'Ole Physics of Coal Carbonizat.ion". Am. G a s Assoc. Proc., 1930, p. 820 f p. 700 .A5p 1930. Millard, D. J . '"A Study of Temperature Conditions i n a Coke Oven". J . I n s t . of Fuel, 28, 345 (1955).
m a n , D. and Kay, W. B.,

3
4

Ind. Eng. Chem.,

41,

1764 (1949).

Terres, E. and Biederbeck, H. "Specific Heat of Char". Gas und Wasserfach, 2, 265-268, 297-302, 303-3G5,

338-345 (1928).
5

Gamson, B. W . a n d Others. "Correlation of G a s Film C o e f f i c i e r t " . Trans. of t h e Am. i n s t . of Chem. Engrg., 9, 1 (1943).
Wilke, C. R. end Hougen, 0. A. "Correlation o f Gas Fiim Coefficient". Trans. An. I n s f . Chen?. Eogrs., 41, 445

(1945).

I n g e r s o l l , L. R. "Heat Conduction". Madison, Wisconsin U n i v e r s i t y of Wisconsin Press, 1954.

Tinoshenko, S.

"Theory of E l a s t i c i t y " . McGraw-Hill, 1934. &A 931 .T5.

New York;

1%

LIt

qi, i

a u a n t i t y o f beat t r a n s f e r r e d during a T h e i c + , e r r d . fron the i 1a:ier t o t h e i f 1 i a y e r o r I r o a f l u i d rnediim r,c t h e f i r s t l a y e r i n t h e 'c..riTaet i f i = o. Radius i n general. Thickness a? a b r i q u e t Layer. Radius of t h e s o l i d s p h e r i c a l b r i q u e t . Surface area or' b r i q u e t . E f f e c t i v e heat t r a n s f e r area f o r tramfer f r o m t h e i-i t o t h e i layer. It i s given by ai = k r r i - , ri.

Density.of b r i q u e t mterid, assumed constan; l o r t h e conputer problem.

Mass o f briq'det l a y e r , taiceE as ( i i p h ) . Qiemal conductivity of b r i q u e t material, a f u c t i o n of terperature
I

S p e c i f i c heax of' briqilet m a t e r i e l , a f l u x t i o n o f t e q e r a x r e . F i l a n e a t t r a n s f e r c o e f f i c i e n t f o r trazsfer ?ma gas or fluidized solids t o the briquet.
Number o f Layers i n the briquet,

k/,pc,
MZSS

the t h e m a l d i f f u s i v i t y ,

f i s c used

for e x p z r s i v i t y .

r a t i o of h e a t i n g rneciimj3riquet,

Russelt Nuiber =

/
/'

, .
r

, .
Q

-102'

OC t/ro2.
Roots o f equation, Nu = 1
Mean Briquet T e a p e r a h r e .

- % c o t Mno

Initial Briquet Temperature.

A l l o t h e r symbols i d e n t i f i e d as used.

cu

9 9 9 M 0 I n
M
P-

9 0

rl

9 0

9 0
r4

PI

11
H

-104-

-105-

-106-

-107-

FIGURE 7 TEMPERATURE PATTERNS IN FORMCOKINB CASE VI1

TIME.

SECOMS

-108-

I'

-107

FIGURE I I

CORRELATtON OF RATE OF HEATINO OF BRIOVETS

nyE,

mums

IO

12

II

FIGURE 1 3 THERMAL CONTRACTION OF YOUHOSVILLE BRIQUET. SLOW WEATINB

FlSURI I4

t n I R N A l COWTRACTION O f ARKWRIBWT BRIOUET.

SHOK)c(( HEATIN0

FlGURE 1 9
RELATIVE THERMAL STRESS IN GAS HEATED z-iwn SRIQUET AS A FUNCTION OF TIME AND INITIAL GAS TEMPERATURE

FIOURE 1 6

RELATNE THERMAL STRESS IN CHAR HEATD BRIOUETS AS A FUNCTION OF T I E . BRlOUEP SIZE AND b

-113

Chemical, Plastic, and Coking Properties of "Natural. Bitumen"

by J. D. Clendenin and J. G. Price Applied Research Laboratory, United S t a t e s S t e e l Corporation MonroeviYe, Pennsylvania

Abstract

In the course of evaluatin@;p o t e n t i a l l y useful blending materials f o r improving t h e coking strength of coals, t h e Applied Research Laboratory of t h e United S t a t e s S t e e l Corporation investigated t h e p o t e n t i a t i e s of a coal-like material i d e n t i f i e d o n l y as "natural bitmen." From this study, the material. has been tentat i v e l y i d e n t i f i e d as impsonite, probably resulting from the weathering of grahamite, both of which are a s p h a t i t e s .
" N a t u r a l bitumen," found t o be of very low fluidity, possessed cokeimprovement properties similar t o those of Lower Kittanning low-volatile coking coal and somewhat b e t t e r than those of Pocahontas l a r - v o l a t i l e coking coal. Coked alone, "natural bitumen" produced a carbonization pressure of over 4 0 p s i and a volume expansion of about 32 per cent.

4.4Not for hblication Presented Before the Division 01' Gas and Fuel Chemistry American Chemical Society Chicago, Illinois, Meeting, Sept. 7 to 12, 1958 Chemical, Plastic, and Coking Properties of "Natural Bitumen"

JI
J
I
I

by J . D . Clendenin and J . G . Price Applied Research Laboratory, United States Steel Corporation Monroeville, Pennsylvania
Introduction

In the course of evsluating potentially useful materials for use in blending ! in zoke manufacture and for improving the coking strength of coals, the Applied Research, Laboratory of the United States Steel Corporation investigated the potentirnties of impsonite, a coal-like material initially identified only-as %aturk bitumen." A preliminary examination was completed because of the unique properties possessed by this mkterial.
The efficacy of pitches and asphaltic materials as coke-strengtheningagents for the poorer coking coals has been known for maqy years. Coy-tar pitches have been used with Utah coals in commercial operationsl,2) and in tests3 with poorly coking Frexb and Italian coals. A highly fluid asphaltite fr Argentina has been blended in zarbonization tests with weakly coking Chilean C O ~ L 5 .e ~ effectiveness ~ of these bleixking materials is usually attributed to their high Fuidity, as compared with that of the poor or weakly coking coals with which they are blended.

In this article the chemical and physical properties of "natural bitumen" are described, its tentative identification is presented, and its coking properties whez coked alone and in blends with good and poor coking coals are demonstrated.
Experimental Three different samples of "natural bitumen" were received for study. 5 e first sample, consisting of only a few grams of about 200-mesh by 0 material, was used in determining Gieseler Fluidity, Free Swelling Index, and Proximate Analysis. The other two samples, on which most of the evaluation was made, were received in 3-inch by 0 size. In this size, the "natural bitumen" could be easily mistaken for a bright coal, Figure 1 . Chemical and petrographic analyses were performed on representative samples of "natural bitumen." Proximate and ultimate analyses were determined, as were solubilities i n selected organic solvents. The rrbitumen't was carbonized i n blends w i t h several coals and coal blends in the 30-pound Pressure-Test and Sole-Heated Ovens for pressure and volume-change determinations, respectively; the resulting coke from the 30-pound Pressure-Test Oven was tested to provide comparative-strengthdata. The Laboratory's 30-pound Pressure-Test Oven has been described in several published . S. Bureau paprs;5,6) the Sole-Heated Oven used is similar to that developed by the U of lbes7) and the first reported in 1 9 3 8 .
Coke strengths were determined by the Laboratory's Modified Tumbler Test, in w M c h 1 0 pounds of 1-1/2by 1-inch coke is tumbled for 500 revolutions in the standard ASTM Tumbler Drum. ple tumbled coke is screened and the er cent retained on a 3/4-inch screen is designated as the Modified Tumbler

'

1 1 . 5 -

Results and Discussion Examination of the "natural bitumen" a s received showed it t o be codl-like and lustrous i n appearance, extremely f r i a b l e , greasy t o t h e touch, and d i r t y t o w e . Macroscopically, banding was not evident, but the presence o f cleavage and the checkered appearance suggested some kind of structure. Microscopically, a conchoiddl-like frdcture was c l e a r l y identified, and t h e surface resembled that found on hard pitches a d other similar amorphous organic materials. Smooth, irregular, jet-black, highly r e f l e c t i n g surfaces were proninent, Figure 1 .
The physical prope-dies of t h e "natural bitumen" were similar t o those of Lower Kittanning low-volatile coal, Table I. I n the 3-inch by 0 size, this material had a bulk density of 49 l b per cu ft, w h i c h i s only a l i t t l e lower than that of Lower Kittanning coal, and it floated a t a specific gravity a s low a s 1.23. This "natural bitumen" was s l i g h t l y soluble i n carbon disulfide, but the insoluble residue disintegrated i n t o a f i n e powder. The behavior in xylene was s i m i l a r : a l a p of "bitumen" disintegrated i n t o an insoluble, f i n e powder, while the solvent became f a i n t l y strawcolored,

I n t h e flame t e s t , a lurp of "natural bitumen" decrepitated and burned, and Incipient fusion w a s detectable. Burning ceased when t h e lump was removed from $he flane; this indicated a high-ignition emperahre. Wen t h e "bitumen" was subjected u s t be pulverized t o t o the ASTM F'ree-Swelling Index Test,8j in which t h e sample m 60 mesh, t h e r e s u l t i n g "button" w a s strong, coherent, dull black, and non-swollen, and had an index of 1-1/2, see Figure 2. However, when coarse, minus 1/4-inch p a r t i c l e s ( r a t h e r than t h e standaxd 60-mesh p a s t i c l e s ) were used in the ASTM Free-SweUlng Index Test, a strong, dense, coherent, shiny black button was formed that had an apparent Index of 3 t o 4. %e outlines of the s m a l l lumps were s t i l l distinguishable on the outside surfaces of the Pree-Swelliag Index button, Figure 3. The proldmate analyses i n Table 11 show that t h e ash content W+S low, but mjght be somewhat variable, whereas the sulfur content ranged from 1-1/2 t o 2 per cent. Although this materid could be designated i n "he ASTM d a s s i f i c a t i o n as a mediumv o l a t i l e coal on t h e basis of (1)i t s proximate andlysis and (2) t h e f a c t t h a t the material w a s agglomerating, petrographic examination, wbich revealed 110 plant s t r u c t u r e v i s i b l e by e i t h e r transmitted o r r e f l e c t e d l i g h t , indicated that the material i s not of plant origin. None of t h e structures ordinarily found in coal of t h e same v o l a t i l e matter and fixed carbon contents were present. The material. appeared homogeneous, with 1 1 0 banding or recognizable c e l l structure being observed in e i t h e r the polished o r thin section; however, i s o l a t e d masses and c r y s t a l s of pyrite were noted.
Because of i t s poor s o l u b i l i t y in carbon disulfide, high percentage of fixed carbon, and high s p e c i f i c gravity, and because 110 petrographic e n t i t i e s c h a r s c t e r i s t i c of coal could e found, this so-called "natural bitumen" has been t e n t a t i v e l y i d e n t i f i e d as impsonite,g one of the asphaltic pyrobitumens, which i n d u d e e l a t e r i t e , u u r t z i l i t e , and albertite, ail characterized by i n f u s i b i l i t y and insolubility, o r poor t o s l i g h t s o l u b i l i t y in carbon aisulfide. Impsonite represents t h e f i n a l stage i n the metantorphosis of a s p h a l t i t e s and asphaltic pyrobitumens. Outcrops of grahamite,10) which i s an asphaltite, metamorphize readily i n t o impsonite. "Weathered asphaltites closely resemble impsonite i n t h e i r physical and chemical properties and may be c l a s s i f i e d as such,"g) The softed-lng and fusion exhibited by t h e "natural bitumen" upon being heated is the only property not c h a r a c t e r i s t i c of i q s o n i t e . This may indicate t h a t t h e m a t e r i d i s indeed weathered grahamite, since v a r i e t i e s of this asphaltite exhibit softening, possess good s o l u b i l i t g in carbon disulfide, and contain l e s s than 55 p e r cent fixed carbon.

...

Deposits of inpsonite a r e reportedlo) in Oklahoma, Arkansas, and Nevsda, Deposits of grahamite and copper-ben-:ing i q s o n i t e has been reported in Hch&an.

-116a r e reported i n West Virginia, Texas, Oklahoma, and Coloraao. Dietrich, i n a recent publication=) r e f e r s t o impsonite from t h e Carbonife ous Jackforth sandstone as o r i g i n a l l y being called "grahamite. I' Abraham c-ites12T t h e same deposits (LeFlore County, Oklahoma) and l i s t s it as grahamite, but he notes t h a t graharnite can weather i n t o i q s o n i t e . Perhaps Dietrich may have obtained a saruple o f weathered grahamite. "Natural bitumen" exhibits p l a s t i c properties and very low fluidity similar 1 , and possesses t o those of Lower Kittanning seam low-volatile coking coal, Table 1 good coke-improvement properties comparable w i t h those of Lower Kittanning i n blends with high-volatile coking coals, see Table III. This "natural bitumen," l i k e 1 1 , Lower Kittanning coal, e x h i b i t s a prohibitively high coking pressure, Table 1 when coked alone i n t h e 30-pound Pressure-Test Oven. Also, the "bitumen" showed n estimated volume change (expansion i n the soleextremely high expansion; it has a heated oven of approximately 9 p e r cent, whereas t h e Lower Kittanning low-volatile coal used f o r comparison i n this evaluation showed expansion of about 14 per cent, at bulk densities of 55 t o 58 pounds of dry coal per cubic foot. A photograph of a lump of coke from "natural bitumen" t h a t w a s coked alone, i s shown i n Figure 4 . The extremely f i n e c e l l structure of the coke, as well as its fingery character, a r e evident. Chemical analysis and a s u l f u r balance made on "natural bitumen" and the r e s u l t a n t coke from the 30-pound Pressure-Test Oven indicate s u l f u r , elimination from the during carbonization, Tables I1 and 11J is l i s t e d i n Table I1 for the r a w "natural. bitumen.: r e l a t i v e l y low. Forms of Carbonization in t h e 30-pound Pressure-Test Oven of blends containing "natural bitumen" showed t h a t this material was as e f f e c t i v e a s Lower Kittanning coal i n improving the strength o f cokes from the highly f l u i d Pittsburgh Seam high-volatile coal, but somewhat l e s s e f f e c t i v e than Lower Kittanning i n inrproving t h e strength

.-----

of cokes from t h e low-fluidity U t a h high-volatile coal, see Table 1 1 1 . Replacement of n e a r g half the Pocahontas low-volatile component by "natural bitumen" i n a blend % . E O XSnTuclw S p l i n t Coal shows that t h e "bitumen" possesses greater strength-improving power than Pocahontas and t h a t it increases the coking pressure o f t h e blend.
Conclusions Coking t e s t s reported by Plompsonl) showed that a 15 p e r cent addition of c o a l - t a r p i t c h t o Utah coal s i g n i f i c a n t l y increased t h e S t a b i l i t y Factor of this coke over that of coke containing no p i t c h (24 a s opposed t o ll). A s shown i n Table 1 1 1 , similar additions of "natural bitumen," and of Lower K i t t a n n i n g low-volatile coal, both having low f l u i d i t y , upgraded t h e strength of t h e coke from Utah high-volatile coal t o a degree comparable with t h a t found f o r coal-tar p i t c h additions i n the e a r l i e r work cited. Consequently, these results lend no support t o t h e long-held hypothesis t h a t improvement of poorer coking coals i s best accomplished by the addition of "fluid" a y be more dependent components. It appears, therefore, t h a t coke-strengthening power m on some combination of optimum coking pressure and f l u i d i t y , rather than on f l u i d i t y alone. Also, t h e high fixed-carbon content of "natural bitumen" (76$) contributes t o coke improvement: a f t e r the "bitumen" has passed through the f l u i d s t a t e , s u f f i c i e n t carbonaceous residue remains, adding t o t h e strength of the c e l l w a l l s of the resultant coke, and not m c h material i s l o s t by devolatilization (which promotes shrinkage). The folloving t h r e e factors, then, probably explain the success of "natural bitumen" as a coke-improvement additive: (1)coking pressure, t h a t helps consolidate a 6 s during the coke-formation process , ( 2 ) s u f f i c i e n t f l u i d i t y , t h a t agglomerates the m t h e non-coking e n t i t i e s of t h e coal in the c e l l w a l l s of t h e r e s u l t a n t semi-coke and (3) a comparatively l a r g e amount of carbon remaining a f t e r decomposition and s o l i d i f i c a t i o n of the f l u i d m a t e r i a l (compared t o that of pitches and asphaltites), t h a t contributes strength t o t h e c e l l w a l l s of t h e f i n a l , finished coke and thus minimizes t h e weakening e f f e c t of shrinkage.

-117References

Tnonompson, J. H. "Benefication of Blast R r m c e Coke a t the Kaiser S t e e l Company, Fontana, California," B l a s t Furnace and S t e e l P l a n t , Vol. 34, pp. 225 230, 350 354, 4'75 479, 584 586, 624; 1946.

Waggoner, C. L., "Observations on Coke Oven and Blast =.Pace B l a s t Furnace and S t e e l Plant, pp. 325 Geneva Coke Plant"

--

- 328,

Practice a t t h e

35, 1947.

Powell, A. R., "Improving Coking Qualities of Coal by Additior? o f Vario-& Materials," A Study of the Iron and S t e e l Industry i n Latin America, Vol. IT, Proceedings o f the Expert Working Group held i n Bogota, Published by t h e United mations Department e w York, 1954, pp. 91 to 96. of Economic Affairs, N Powell, A. R., "Argentine Asphaltites as Blending Material f o r Poorly Coking Coals," ope C i t . pp. 99 100.

Harris, E. E,, b k g e l , E. W., and Howell, R. Bo, "A S m a l l Pressure-Test Oven for Yiasuring Expansion Presswe Developed During the Carbonization of Coal," Froceedings, Blast Furnace, Coke Ovea, and Raw Materials Conf'erence, AIME, Volume 13, 1954, ppe 127 141.

Price, J . G., Shoenberger, R . W., and P e r l i c , B., 'Use of a QxLrty-Pomd Tesz Oven f o r Rapidly Assaying t h e Coking Strength af CodLs," Proceedings, m a s t r k n a c e , Ccke Oven, and Raw Materials Conference, AINE, 'Tolme 17, 1958.

Anvil, H, S . , and Davis, 5 . D., "Determination of the Swelling Properties of Coal During the Coking Process," Uo S. Bureau of Wines, Report of Investigations 3403,

1938.
&TM Procedure ~ 7 7 0 - 46 f o r determining n e e - s w e m n g Index. Abraham, H., "Asphalts and A l l i e d Substances," De Van Nostnzud Co., Edition, Vol. I, p. 298,
Ibid.,

Inc.,

Fifth

p. 270.

Dietrich,

a.

V.,

Economic Geology, V o l - 51, Xovember,

1 9 5 6 , P- 653.

Abraham, E . ,

op. C i t . p. 274.

B r i t i s h Standard Method ~31016 as Revised Under Publishea date March, 1957. Plso A. R. Powell, Inil. and Eng. C n e n e m , Vol. I 2 (l92O), p. 887. ASTM Procedure Dk

52 for DetermFning Czrbon Disulfide Insolubles,

"Methods of T e s t i c g Coal T a r Products,R Barrett Division, Allied Chemicd and 27. Dye, 1950, Test Method B7 f o r Benzene Insolubles, pp. 24

'/

Errattun:

! -

O n pagel&mmgraph 4 , t h e first 3 words of l i n e read "Coal-like i n f i n e structure."

...

7 should be changed t o

hi
i'
I

,
I(

-118-

0
b

rl

3
w

P 0 c,

8
0

=I 8l 2 2

2.

L c9
f0

Proximate Analysis, ultimate Analyeis, Sulfur Content, and Heating Value of "Natural Bitumen,'' a l l on the dry basis:
Sample No.

1 -

3 22 -0

Lower Kittaming
Low-Volatile Coal - -

9oximate Analysis, $ dry basis: Volatile Matter Fixed Carbon Ash Ultimate Analysis, $ Carbon Hydrogen Nitrogen Wgen

21.6 77.6

0.8

76.3 1.7 85 -7 5.3

1.9

15.1 76.8 8.1

83.4 4.3

Ash Sulfur
?eating Value, BTU per lb

-, . 1.66

1.2

3.6 1.7 1.82

3 .O 8-1
1.01

Gieseler Fluidity ana ,Free-Swelling L?dex of "Natural Bitumen", and Lower Sunnyside Sean (Utah and Lower-Kittaming Seam Coals) : Natural Lower S m y s i d e Bitumen High-Volatile Ccal (Sample N O . 1) P l a s t i c Properties: Sortening Temerature, C S o l i d i f i c a t i o n Ternyeratwe, C P l a s t i c Range+, C Maximum Fluidity, dial div. per min. Temperature of &ximum Fluidity, C Free-Swelling Index Lower Kittanning Low-Voht i l e Coal

435 480

45

377 453 76
3

449

504 55
1.1

0.6
458
1 4 2

L13
2-1/2

h77
2-1/2

S o l i d i f i c a t i o n temperature minus softening temperature

Forms o f Sulfur in "Natural Bitumen" Sulfate Sulf'ur P y r i t i c SLiLfbr "organic" i z j u m l r Total

0.04$

3.0%

0.70$

0.57%
1 31%

43 -55
100 -0%

-120Table I11 Coke Strength and Carbonizatix Pressure fkom 30-pound Presswe-Test Oven Tests with "Natural Bitumen" and Coking Coals 'Modified Tumbler Approximate ASTM S t a b i l i t y Index, cum $ plus 3/4 in. Factor
'/

Blend*

Carbonization Pressure, p s i

'

X& "Natural Bitumen"

-a=
*P

--

40
30

100% Lower Kittanning Seam Low-Volatile

----

45$ Kentucky Splint High-Volatile 55$ Pocahontas Low-volatile 55% Kentucky Splint Kigh-Volatile

55
62

3.5
6.7

45%

"Natural Bitumn"

53% Kentucky Splint %i&-Volatile

30$ Pocahontas Low-Volatile

20% "Natural Bitumer,"

63
59

--

4.5
2.3 1

835 Kentucky
I C @

Splint High-Volatile

? I "Natural $ Bitumen"

Pittsburgh Seam High-Volatile P i t t s b q g h Seam

43 63

I?$'

20% "Natural Bitumen"

&$ Pittsburgh Seam

20s Lower Kittanning Seam
~ O O $mer m y s i d e Seam Zigh-volatile

61
21

1.5
1

9 %Lower

m y s i d e Seam

10% "Natural Bit\unen:'

P
43

1.4
2.2

60% Lower Sunnyside Seam =3$ "Natural Bitumen"

90% Lower sunnyside Seam
LC$ Lower Kittanning Seam

41
50

1 . 1

80% Lower m y s i d e &am

20% Lower Kittanning Seam

35

1.8

*
+ic

Bulk Density as charged 50 t o

53 lb per cu ft

Proximate analysis and sulfur content o f coke per cent (dry basia): Volatile Matter 3.0 Fixed Carbon 92.7

Ash
S u l k

4.3
2.12

1
I

-121-

SURFACE MAGNIFIED 5x N e g a t i v e No. 2-3578E-2

LUMP MAGNIFIED 1X N e g a t i v e No. 2-3578E-1

Figure 1. LUMP OF "NATITRAZ, BITLTMEN", AS RECEIVED

UNITED STATES STEEL

-122-

MAGNIFIED lx N e g a t i v e No. 2-3574B-1

CROSS-SECTION, MAGNIFIED 5~ N e g a t i v e Na. 2-3574B-2

Figure 2 . FREE-SWELLING INDEX BUTMN - INDEX 1-1/2 - FROM ''NATURAL BITUMFN" I N T J S 60 MESH PULVERIZED TO M

UNITED STATES STEEL

-123-

CROSS-SECTION MAGNIFIED 5x Negative No. 2-3573B-2

Figure 3.

FREE-SWELLING INDEX BUTTON - INDEX "NATUFNL BITUMEN" PARTTCLES

4-

F R O M MINUS

1/4 INCH

I'

)I

p
/I I

UNITED STATES STEEL

-12b-

FlTLL LUMP MAGNIFIED lx Negative No. 2-3577B-1

SECTION OF LUMP MAGNIFIED 5X Negative No. 2-3577B-2

Figure

4.

LUMP OF COKE FROM "NATURAL BIl'UMEN" ChRBOMZED IN 30-POUND TEST O m

UNITED STATES STEEL

The Reaction

Gaseous Acetic ----gdride With Coal

1 Paul Fugassi, Ruth T r a m e l l and P h i l i p Hasciantonio

Coal Research Laboratory, Departanent of Chemistry Carnegie I n s t i t u t e of Technology, P i t t s b u r g h (13) Penna.

A number of
acetylated a t

coals of t h e bituminous and sub-bitminous

ranks have been

45"C, using a c e t i c anhydride vapor a t a r e l a t i v e pressure o f unity.

Acetylation by t h i s procedure gives hydroxyl contents leas than t h a t determined

by t h e t r i m e t h y l s i l y l e t h e r method f o r bituminous c o a l s but higher hydroqyl

c o n t e n t s f o r sub-bituminous coalso

It i s suggested t h a t a c e t i c anhydride r e a c t s

with carboxyl groups t o g i v e mixed a c i d anhydrides and t h e higher values obtained f o r sub-bituminous coals is due t o t h e c a r b o w l group content of such coals. Oxygen g a s c a t a l y z e s t h e a c e t y l a t i o n of sub-bituminous coals and u n t i l t h e e f f e c t

of oqygen i s i n v e s t i g a t e d i n d e t a i l , a c e t y l a t i o n o f lower rank c o a l s w i t h a c e t i c

anhydride should be done i n t h e absence of oxygen.

h e s e n t address, Depto of Chemistry, Chatham College, Pittsburgh (13) Pa.

-126Not f o r P u b l i c a t i o n ?resented Before t h e Divisioi? of G ~ and s the1 Chemistry d n e r i c a n Chemical S o c i e t y Chicago, I l l i n o i s , Meeting, Septeniber 7-12, 1958 The Z e a c t i o n o f Gaseous Acetic Anhydride With Coal

Paul Fugassi, iluth T r a m e l i1 and P i i i l i p Kssciantonio

Coai Research Laboratory, Dept. of ChemLstry Carnegie I n s t i t u t e of Technology, P i t t s b u r g h (l3), Pa.

1 . P r e s e n t address, Dept. of Chemistry, Chatham College, P i t t s b u r g h ( i 3 ) , Pa.

Introduction The hydroxyl c o n t e n t s of c o d s have bees xeasured by a number o f d i f ferent, methods and t h e residts af such vork have been swnmarized by Elom, Zd?lhailsen and V a n Krevelen( 1). i n g e n z r a l methods using d i f f e r e n t reageiits Even i n cdsas g i v e d i f f e r e n t h y d r o q l conteiits f o r c o a l s of t h e same r.-nk. khers t h e same re+,,snt has Seen emplcyed by d i f f e r e n t i n v a s t i g a t o r s t h e r e i s n o t ciosa agreemsnt on t h e hydroxyl content.

It i s tP.e o b j e c t o f tiiis comxnictrtion t o r5Sm-t some expeii.menta1 work bitumino.x 2-n.d sub-bitmLious coels vharein t h e c o a l s a r e reacted a t kSoC. witin gaseous a c e t i c mhyilride a t a r e l a - l i v e prassure c l o s e t o unity. 3 conparison t o 0th4r ecet;rlation work on c o a l using a c e t i c a.rdiydride, the temperaturesain$.G;-Y~ h a r e a r e mucn lower and tila r a d c t i o n times much longs?. Tha extent of the x a c t i o r . i s followed gravimetrically.
JZ

EQerLxental R p p r a t u s : Yeasurements ware m d e using LA33in-3akr balances housed i n a air t h e r n o s t & h e l d a t 45C. 3 . 1 " . B e c x i e of t h e corrosive ndture of Acetlc mh.yd=ide--acet,ic a c i d ?riix;tures, s p i n g s m d e from f i n ? tungsten w i r e vere used. These s p r i n g s had s e n s i t i v i t i e s ransing f r o n 3.81 t o L.86 milligrams p e r mllimeter extension. S p r i n g displacements were measured t o 0.1 mm using a c a t h a t m e t e r . iciibh sampld m i g h t s o f t h e o r d e r of 500 milligrams t h e p r e c i s i o n i s about one p a r t i n d 5housand. Cheliiicals: h c e t i c anhydride, benzene, and pyridi,ie were f r e s h l y d i s t i l l e d before use and. s t o r e d o v e r Drierite. 71.2 foliowing c o a l s o f bituminous rank were used: Upper Eittanning, Gpper Freaport, Pittsburgh, Sruceton and aruceton a n t h r q l o n . Tha coals o f sub-bituminous rank were Wyoming (Zlkol), Xyoming (2ock Spriags), I l l i n o i s IJo- 6 (Clinton), and I l l i n o i s Mo. 6 ( J e f f e r s o n ) . The analyses f o r ;JyOfing (Rock Springs), Bruceton, Bruceton a n t h r w l o n , and I l l i n o i s No. 6 (Jefferson) have been p r e v i o u s l y p u b l i s h e d ( 2). The analyses f o r t h e o t h e r c o a l s are given i n Table I.

-127TzEe I Ultimate Analyses (1-ioisture iree a a s i s ) Coal Upper Kittanning Upper Freeport Pittsburgh Iniyort~ng( E l k o l ) I l l i n o i s No. 6 (Jefferson)
C

N 1.1 0.9

s
1.3

Ash

80.6

3.8
4.3

3.7
4.6

78.7 76.0 74.3 65.2

5.1 5.3
1.9

1.5
1 . 2
1.2

7.6
i5.2

1.7 1.5

i r . 6
4.0

9.5 9.3 3 .2 3.L

1 0 . 1

l4.3

The samples of Upper Kittarrlling, U?per Fresport, and P i t t s b u g h codis a r e standard samples of t h e Coal Zesearzh Eoard, Colaionwealth of Fennsylvania, and were used as received. Samples of #iyomining(3ock S?rir.gs), I l l i n c i s XO. 6 ( J z f f e r s o n j and Brucaton c o a l s were furnished by she 3ruceton S t a t i o n , U. 3 . ?DP=Q of I i n e s 2r.d were used as received. Vyozing (3lkolj and I l i i n o i s KO. 5 (CLinton) riere a v a i l a b l e i n bulk. These svnples were ground by passage t h r o u g a h m e r x i i l l axi used unsieved. Irocedure: The e m p l e s of coal, placed i n s m a l l g l a s s buckets 7.i:Ys veizhed t o 0.1 mg. and alloued t o s t a n d o v e r k g h t i n a desiccator. The s a p l s s , a f t e r beiag reweighed, were attached t o t h e s p r i c g s m d d o w e d t o stand f o r one hour, with an atmosphere of air present, t o come t o thermostat temperatwe. The e x t e m i o n o f t h e s p r i n g was then measwed. It i s a s s m e d t h a t t h e weight of t h p s m p 1 ~ had not changed during t h e hour waiting period. The apparatus was then evacuated and t h e extension of t h e sp,-ings measwed. UsudLy s a q l e s c m e t o c o n s t s a t weight in 8 hours, however evacuation was continued f o r a miii>mof 2L, hours.
After t h c system w a s closed off from t h e punping system, z c e t i c anhyfiride was a d a i t t e d a t a r e l a t i v e pressure of unity (about 12 m.) and the system allowed t o starxi. Beadings of t i i a s?ring exLansion iere taken .uti1 t h e weights of t h e sample were p r a c t i c a l l y c o n s t a t . i C , t h i s p i n t , t h e systen was evacuated f o r s a v e r a l days a x i t h e weight of t h e sangle r,,aaswed. ;~eec;us a c e t i c aid?y-lride was t h s n re-adinitted arrd tila s z q l e s allowed t o s t m d anti1 t h e i r weights reached a constant v&l?Ae. Zvacuation o f t h e system f o r s e v e r a l days followed by measurenect of t h e k-eights of t h e svnples then gave t k e i x r e a s e i n x e i g h t of t h e sample due t o a c e t y l a t i o n . I n most czses s m p l e s of coal were subjected t o two consecutive anhrdride treatments before t h e f i n a i reading w a s taken. I n a few cases t h r e e consecutive anhydride treetments were .s@oyed bcit l i t t l e o r no increase i n weight r e s d t e 6 from t h e t h i r d treatmznt with a c e t i c vlhydride.

Discussion Catalysis: The vapor phase a c e t y l a t i o n o f c o a l s a t 45OC. i s a very Slow p-ocess so a number of q u a l i t a t i v e experinents were made with I l l i n o i s KO. 6 (Clinton) coal t o d e t s r n i n e i f t h e r a t e could be accelerated. Coal i s a g e l and it i s known t h a t f o r o t h e r gels, such as c e l l u l o s e , a c e t i c anhydride i s a poor swelling agent and t h a t f o r c e l l d o s e t h e r a t e of a c e t y l a t i o n can be increasod by t h e use of s u i t a b l e swelling agents. The behavior of a c e t i c anhydride with r e s p e c t t o c e l l u l o s e scggests t h e use of swelling agents i n coal a c e t y l a t i o n . i.iethano1 swells coal. so s e v e r a l experiinents :+ere ,made i n which t h e coal beiore a c e t y l a t i o n w a s exposed t o methanol vapor a t 45C. and t h e methanol

-128vapor iesorbed before a c e t y l a t i o n . i n a c d t g l a t i o n r a t e was noted. Kethanol i s r e v e r s i b l y desorbed and no i n c r e a s e

Pyridine swells c o a l but pyridine i s sorbed i r r e v e r s i b l y by coal. Coal t r e a t e d with p y r i d i n e vipor, evacuated, and t h e n a c e t y l a t e d showed an i.n.creased rate of a c e t y l a t i o n . Unfortunately t h e i r r e v e r s l b l e s o r p t i o n o f pyridine i s l a r g e and although t h e aiount of t h i s i r r e v e r s i b l e s o r p t i c n on c o a l can be measured accurately, it i s not p o s s i b l e t o a s s m a t h a t t h e amount o f pyridine sorbed i r r e - r e r s i b l y by t h e c o a l does n o t char,se as t h e c o i l i s a c e t y l r t e d . F o r tkis rzsson t h e o v e r d l weight chaqge rnight not be a measure of t h e e x t e n t of acetylation. Benzeni vapor a t 45C. i s i r r e v e r s i b l y sorbed by c o a l i n amounts ranging frm 0.0 t o 23.4 milligrams p e r gram of coal. A c o a l s s u p l e exposed t o benzene vapor, evacuated, and t h e n a c e t y l a t e d appezrs t o a c e t y l a t e f a s t e r . However, t h e i n c r e a s e i i i rate o f a c e t y l a t i o n i s not s t r i k i n g . Consequently tfii experirnents m2de here used a c e t i c anhydride d o n e . As a consequence of t h e a c e t y l a t i o n r e a c t i o n a c e t i c acis. i s present and serves t:, some e x t e n t as a swelling agent. On I l l i n o i s No. 6 (Clinton) c o a l it has been shown t h a t a c e t i c a c i d vapor i s revei-sibly desorbed on c o a l previously a c e t y i a t e d with a c e t i c h y d r i d e .

-E f f e c t sf O ~ g e n : I n c a r r y i n g o u t t h e a c e t y h t i o n s i t was noted t h a t w'nile

,

thec o z l s o f bituminous r i n k gave reproducible i n c r e a s e s i n weight w i t h i n 2 53, the c o a l s ~f sub-tjit7minous rank would sonetimes g i v s h i g h values and i n o t h e r exp e r i r ~ ~ e n low t s values. Aithough a l l sam>les wsre evacuated f o r a t l e a s t 24 hours using a two sta.ge u e r c u r y vaqor p ~ ~ p t h ,e syzten could not be baked out and small b1.1.t v a r i a b l z vlorrnts of o x y g m might be present i n i n d i v i d u a l s e t s of experiments. Shenever a small arount o f oxygen, 5 m., w a s added t o t h e system t h s r e s u l t s becane more r e p r o d u c i b l e and i t appears, f o r subbituminous coals, sqrFen has a c a t a l y t i c e f f e c t . However t h e presence of oxygen i n t r o d u c e s the p o s s i b i l i t y c f Ln oxidation of t h e coal. To check t h i s point f i v e s 3 a l samples a f t e r t h e u s c a l evecuation were exposed t o 5 m. of o q g e n , with no b j d r i d a present, f o r a period of 2 1 days. The experimental r e s u l t s a r e t a b u l a t e d i n Table I1 and a r e t h e i n c r e a s e s i n weight a f t e r evacuation. Tabla Ii Oxidatior, of Coals; 5 inn. 02, 21 days a t 45C. Coal W t . .Increase mg/gram Upger IlittarJling Upper Freeport Pittsburgh 2.0 Illinois (Clinton) 'dyoming .(Elkol)

4.1

1.8

1 1 . 2

13.5

It w i l l be noted t h a t t h e f i r s t t h r e e coals 03 bituminous rank show l i t t l e 6r no

change i n veight. The t w o sub-bitwinous coals, I l l i n o i s (Clinton) 2nd. :4r>-oming ( E l k o l ) show r e l a t i v e l y g r s a h r changes in weight. Pressure changes during these e q e r h e n t s were n a g l i g i b l e and it was concluded t h a t o-ddation under t h e exp e r k e n t a l conditions used was a very slow p r s c e s s r d a t i v e t o t h e slow acetylation reaction. Acetylatior.: Kost of t h e a c e t y l a t i o n e q e r i n e n t s were aade with 5 mn of o q g e n p r e s e n t but s u f f i c i e n t d a t a w i t h no added o-xygen a r e a v a i l a b l e t o i n d i c a t e t h a t t h e presence of oxygen has l i t t l e o r no e f f e c t on t h e weight i n c r e a s e r " o ~

-129bituminous coals. The a c e t y l a t i o n d a t a are tabulated i n Table 111. These data a r e a l l f o r experiments a t 45"C., kiith a r e l a t i v e pressure of a c e t i c anhydride of u n i t y and incll.de, i n t h e average weizht increase, e-xperinents with and htthout added oqgen. All sanpl2s were given a t l e a s t two consecutivs treatments with a c e t i c anhydride and i n some cases t h r e e consecutive treatments. The spread of the individual values from the average i s 7 d t h i n . t 5%. Reaction times ranged from 23 t o 37 days. Table I11 Acetylation Data (45C.)
coal

Average W t . Increase (aiiliigrams) gram

Upper Kittanning Upper Freeport Pittsburgh Bruceton P n t h r q l o n Bruceton I l l i n o i s No. 6 (Clinton) I l l i n o i s No. 6 (Jefferson) Ttryoming (Rock Springs) 'rlyoming ( U o l )

32

ur

3c
13

78 4 0 158

143 175

s ' it i s sssume-d Gnat t h e increasa i n weight Calculation of Hydroml Content: I of t h e coal sample i s caused mainly by replacement of a hydrogen ator? from an hydroxyl group by t h e a c e t y l r a d i c a l then the weight increase divided by 42 -.iiU. give m i l h o l e s of hydroxyl groups per gram of coal (m.f. basis). The implicatioils of such an assumption w i l l be discussed. Processes leading t o -3 weight increase other than replacement of hydrogen b j t h e a c e t y l group might be 1. I r r e v e r s i b l e . I r r e v e r s i b l e s o z t i o n of a c e t i c a c i d resultTLng sorption of a c e t i c anhydride, 2 from t h e acetylation process and 3. Fteaction of a c e t i c acid k<th salts of weak acids, such as p y r i t e s , present i n the coal. There i s no incependent evidence xhich indicates t h e magnitude s f i r r e v e r s i b i l i t y of a c e t i c anhydrida sorption on coal. However the-hydroxyl content as c a l c d a t e d assuming no trapped a c e t i c anhydride i s already l o w r e l a t i v e t o the values obtained 5 , - o t h e r methods f o r coals of t h e b i t m i n o u s rank. The e f f e c t of trapped a c e t i c anhydride would be t o make these vaiues lowar still. WLth regard t o a c e t i c a c i d it has been shown t h a t a c e t i c s c i d sorption i s r e v z r s i b l e on acetylats6 I l l i n o i s coal. T h i s f a c t suggests t h a t the amount o f a c e t i c acid trapped must be sudd. Tce r e a c t i o n o f a c e t i c acid with p y r i t e s can be assumed t o giva sulfur, i r o n a c e t a t e , and hydrogen s u l f i d e . Assuming a 500 milligram sample containing 1% o f i r o n s u l f i d e , t h e weight Licrease would be a r o m d 2.9 niilligrams per gram or' coal i f hydrogen sulf i d e i s l o s t cn evacuation. Frorc other experiments it i s known t h a t the zvaporation of elemental s u l f u r a t 45C. i s very slow.
?or four of t h e coals used data on t h e hydroxyl content a r e a v a i l a b l e
as determined by t h e t r h e t h y l s i l j - 1 e t h e r method(2).

with the data obtained by acetylation i s given

iri

A corngarison of these dat2 Table I T .

-130Table Coal Bruceton Anthraxylon Bruceton Wyoming (Rock Springs) Illinois No. 6 ( J e f f z r s o n )

I V
Trimethylsilyl Ether

H y d r o x y l Content of Coals ( m i l k h o l e s per gram)

Acetylation (Vapor)

0-3
1 . 9

3 . 4
3.8

2 . 3 2 . 3 2 . 8

3 . 4

For t h e bituminous c o a l the h y d r o w l content ds detemuned by acetylat i o n is lower than t h a t obtained by t h e t r i m e t h y l s i l y l e t h e r method. One presumes t h a t c e r t a i n s t e r i c a l l y hindered hydroxyl groups r e a c t very s l o w l y with a c e t i c anhydride. Or. t h e o t h e r hand thr: h y d r o q l content 2 s determined by a c e t y l a t i o n is higher f o r tiro sub-bitminous coals. Although it i s knoin;(3) t h a t c e r t a i n polycyclic s t r u c t u r e s photo-ofidize r e a d i l y i n ttie presence o f a c e t i c anhydride . Acetylation experiments and oxygen such a r e a c t i o n i s excluded here because 1 w i t h l i g h t excluded gave about t h s same weight increases as w i t h l i g h t present and 2. No decrease i n p r e s s u r e caused by o-wggen conswption was observed. i-IoweveP it i s b c w n t h a t sub-tituminous coals contain carboxylic aciZ groups and t h e p o s s i b i l i t y exists t h a t :Vi.th sub-bituminous coals a c e t i c anhrdride r e a c t s with a c i d i c groups t o f o m mixed anhydrides i n addition t o r e a c t i n g with hydroxyl group3 t o form e s t e r s . The mixed h y d r i d e would be expected t o r e a c t with wdter more rapidly at a given ternperature than t h e e s t e r . Accordingly a s e r i e s of f i v e coals were acetyl37;ed and then exposed t o water vapor a t 20 mm pressure a t 45C. The r e s u l t s of t h e s e hydrolysis experiments a r e shown i n Table V. Table V H f l r o l y s i s of Acetylated Coals coal Upper Kittanning Exposure Time

(45C.)
Relative %eight

H20 (20

UUU.)

0 hours 1
22

Upper Freeport

70
0

22

1 3
0

Pittsburgh

70

I l l i n o i s No. 6 (Clinton)

22 70
0

1 3

3
Wyoming (Elkol)

22 70 0

1 3
22 70

1 . 0 0 1 . 0 0 1 . 0 6 1 . 0 0 0 . 9 0 1 . 0 0 1 . 1 0 1 . 1 5 1-15 0- 89 1 . 0 0 1 . 0 8 1 . 0 8 1 . 0 8 1 . 0 0 1 . 0 0 0 . 9 5 0-95 0 . 9 1 0 . 8 6 1 . 0 0 0 . 9 4 0.92 0 . 9 0 0 . 8 1

-131Conparison of these d a t a i n d i c a t e t h a t t h e t h r e e Situmirous coals Shoh small changes i n height on trectment with water even f o r periods a s long .is 70 hours. This iildicates t h a t the e s t e r grodp present i s rela+,ively s t b b l e t o t h e a c t i o n of n t h e other hand t h e two sub-bitminous coals show g r e a t e r decreases water vapor. O i n weight under t h e came conditions. It appears p r s t a b l e t h a t a c e t y l a t e d subbituminous coals contain groups which d i f f e r i n t h a i r behavior 1 5 t h respect t o water s t a b i l i t y and a s has been merationed t h e mixed anhydride group from i t s known properties wo1d.d f u l f i l l this requirement. Acetylates sub-bituminous coals w x l d contain a c e t a t e e s t e r s an6 mixed anhydrides.
Su-nrary
A number of coals of t h e bituminous aiiri scb-bitminous ranks have been acetylated a t 45C. usk.g a c e t i c anhydride vapor a t d r e l a t i v e pressure of unity. Acetylation b y t h i s procedure gives hydroqrl contents less ihvl t h a t detemined by t h e t r i m e t h y l s i l y l e t h e r method f o r bituminous coals but higher hydrox;;l cont e n t s f o r sub-bituminous coals. It i s suggested t h a t a c e t i c a r i j d r i d e r e a c t s with carboxyl groups t o give r i x e d acid snhydrides and tine higher values obtained f o r sub-bituminous c o a l i s due t o t h e carboxyl group content of such coals. ilxygeri gas catalyzes the acetylatio'i of sub-bitmkous coals .ad u n t i l t h e e f f e c t of oxygen i s investigated i n d e t a i l , a c e t y l a t i o n of lower rank -coals with a c e t i c anhydride should be done i ?t h e absence of oxygen.

4-cknowledpcni The authors wish t o thank BCil, Inc. and t h e Coal Research Bcli-d, Commonwealth of Pennsylvania, for furnishing coal samples and analyses. They a l s o wish t o thank 3r. 3. Friedman and D r . Irving Mender, Bruceton S t a t i o n , U. S. Bureau of Mines, f o r a r n i s h i n g c o a l samples. References

1 . Blom, L . , Edelhausen, L., and V a n Krevelen, D . i i . , Fuel (London) 135 (1957). 2 . FYiedmn, S . , Steiner, W- A-., Raymond, R., &IC. X z ~ d e ~I., , ? r e p r i c t , Gas and

s,

. , Traminell, Ii., 3. Fugassi, P., Kasciantonio, F

Fdel Divisioa, !Jew Pork Xeeting, September, 1 9 . 5 7 . P r e p r i n t Li

f 2 3

volume.

-123_ .

Not f o r ? u b l i c a t i o n Presented Before t h e Division of G a s and h e r i c a n Ciierical S o c i t t y Chicago, I l l i r i o i s , Keeting, September The Photo-Oxidation of Polycyclic Hydrocarbons Paul Fugassi, P h i l i p , ~ ~ . ~ a s c i a n t o n i and o, 2uth Tramell Coal Research Laboratory, Dept. of Cheaistr;. Carnegie I n s t i t c t e of Technology, P i t t s b u r g h ( i 3 ) , Pa.
SUEE2-g

Solutions 3f anthracene and t e t r a c e n e i n a c e t i c a n k ~ d r i d e2r.d o t h e r s o l v e n t s i n t h e presence of a i r and u l t r a - v i o l e t l i g h t a t room temper2tux-e photo-oxidize t o t h e 7,lO and j,12 quinones respectivelg. Gther p o l y c y c l i c hydrocarbons such as chrysene, pyrene and ti-iphenylene do not photo-ofiddize under t h e s e conditions. !With naphthalene a d phenanthrene oxidation t c k e s p l a c e but t h e r e a c t i o n appears t o c o n t i m e p a s t t h e quinone stage. The photo-oxidation r e a c t i o n i s zztalyzed by bass5 anb i s not i n h i b i t e d by sodium 5.gdroger. s l i l f i t e i n d i c a t i n g t h a t t h e nechar,ism i s probably not a f r e e r a d i c a l c k i n reaction. Oiscussion I n an i n v e s t i g a t i o n now under%-iy i n t h e C o a l Zesearch Laboratory involving t h e vapor phase acetyiat,ion of various c o a l s wlth a c e t i c anh7i!?-ide it was n0te.i t h a t w i t h c e r t a i n c o a l s a vaay slow r e a c t i o n (w2eks) w a s s u p r k p s e d on t h e main aeaction (days) so t h a t a f t e r a r e l a t i v e l y r a p i d u e f g t t change, ths weight of t h e c o a l sample increased e r a d u a l l y over d period OF :asny ~ i z k s . Ss t h e x i n r e a c t i o n ,carA be a s s u e d t o be t h e a c e t z l a t i o n of t h a hj-2ro;cg.l 2nd t r a c e s of amino-groups present i n t h e c o a l i t became nacessary to c..r_aider t h e p o s s i b i l i t j r t h a t gaseous a c e t i c anhydride vas r e a c t i n g l i r e c t l y i r i t h ?,ydroc~-bor, s t r u c t u r e s present i n tile coal. The a c e t y l a t i o n o f c o a l i s being c a r r i e d out usin; an adzorption t s i apparatus. The colirse of tile acetjriafioa 1s followed d i r e c t l y b 7 use o f I:c&;nEikr s p r i n g balances. ill a-qeriments a r e being nade a: L5"C. ~ 5 t h tha reiitive pressure of a c e t i c anhydride being fiked a t unitg. The vary s i o x r e a c t i s n i s p a r t i c u l a r l y noticaabie -"hen s n a l i amounts ol oq~;5er.a r e ; j ~ e s e ; i ts o t h a t xost expe;irrimts have h e n irtade with 5 i i ~ .o f oxygei; gas present in ti,e s y s t a x . U n d m t h e s e conditioris t h i c o d sample i s pi-esezt is larg: sxcess r s l e t i v e ,to tiie a k e t i c anhydride. i?cetic a c i d i-asulting fi-GX the z c e t y h t i o r ? process is &so : ' hj5iwcarcsn s t r u c t u r e s present. I n i n atternct t o d c t e m i n e ti?o type o r +,;yes 0 which would r w c t -hth n i x t u r e s of s?catic arkydride and a c e t i c a c i d a se-les o f e q e r i a e n t s using szrious polycyclic h;rdrocarbcnj L v - s been X&E i i ? ;Xck ti!e e:qeaixental csnditions a p p r o x i ~ ~ ~ those a t e u t i l i z e d in the -,rork 0% coal.

If a s a t u r a t e d s o l u t i o n of .ziit,hracene i;i a c a t i c aiiny2ride i s allo-JTed t o stand i n a m e s f l a s k , loosely sts?pa:eci so thLt z i r I?SS access to t h e f l a s k , and e.ypsed 'io l i g h t t h e r e appears i . ligl-.: t a n , c;-:rstzliina p r a a i 3 i t a t s TLiick an i s o l a t i o n proved t o be anthraccr.a-9,lC-qui.no;le. !,lo o t h e r product. i n appreciablz :':.s anounts could be i s o l a t a l froin tip? reacttoil riiYc!F? except s t a r t i n g ziatarial. t h e reactioil invokriag 3nthracer.e progresses t h c solutioii becomes l i g h t y s l i ~ ~ y.. ~,rth o t h e r hydrocarbzns the s o l u t i o n becoz.es d.rzply colorad. ;he i n coior. . coloring niaterial i s present i n such small amourits t h a t it can not be i s o l a t e d

-134and might w e l l be due t o traces of i m p u r i t i e s p r e s e n t i n t h e s t a r t i n g material. I . s y s t e m a t i c i n v e s t i g a t i o n of t h i s r e a c t i o n w a s undertaken t o d e t e r n i n e t h e cond i t i o n s under which it t a k e s place. Z f f e c t of o q g e n : S a t u r a t e d s o l u t i o n s of anthracene i n a c e t i c anhydride a r e -placed i n smalliyrex tubes, evacuated, and sealed. The tubes are then exposed t o sunligli: and t o t h e r a d i a t i o n from a 36DL f l u o r a s c e n t lamp. After s e v e r a l months exposure a t rooirl temperature t o t h e .altra-violet r a d i a t i o n t h e s o l u t i o n s remained colorless. 4 w h i t e c r y s t a l l i n e product which p r e c i p i t a t e d from t h e solut i o n ?roved t o b e t h e photo-dirrer o f anthracene. These e x p e r h e n t s i n d i c a s e t h a t o w g e n from t h e air is t h e o x i d i z i n g agent far anthraquinone formation. Solutions of -E f f e c t of u : anthracene i n aceticaanhydri.de s a t u r a t e d with a i r and kept i n t h e d a r k a t room temperature f o r s e v e r a l mont!is r m a i n c o l o r l e s s and i s o l u t i o n of a?ttrracene i n a c e t i c no quinone (negative vat t e s t ) i s present. : anhydride was refluxed t h r e e weeks under conditions where l i t t l e l i g h t reached t h e system. The s o l u t i o n becomes r e d in c o l o r blit no quinone i s present. After cooling t h e mixture w a s exposed t o the r a d i a t i o n from a 3603L lamp. After s e v e r a l weeks anthracene-9,lO-quinone p r e c i p i t a t e d and t h e red c o l o r g r a d u a l l y disappeared. Thess experiments i n d i c a t e t h a t t h e a a r k r e a c t i o n i s v e r y slow. The cause of t h e c o l o r i s not known. Effect -of temperature and l l p h t i n t e n s i t x : From q u a l i t a t i v e experiments samples kept a t room temperature a2pearsd t o r e a c t slower t h a n sanples exposed t o sunlight and kept outdoors during t h e w i n t e r zonths. The temperature c o e f f i c i e n t of t h e Since s u n l i g h t r e p r e s e n t s ?hoto-oxidation, as one would expect, must be s m a l l . much higher i n t e n s i t i e s t h a n t h e r a d i a t i o n from l a m p it seems l i k e l y t h a t t h e rate o f t h e r e a c t i o n i n c r e a s e s with i n c r e a s i n g i n t e n s i t y of i n c i d e n t l i g h t . Catalysts: It has been found t h a t t h e rate o f anthracene-9,lC-quinone formation i s i n c r e a s e d by t h e a d d i t i o n o f bases t o t h e a c e t i c anhydride solution. S u i t a b l e c a t a l y s t s are the sodium o r potassium salts o f o r g a n i c a c i d s such as formic, a c e t i c , p h t h a l i c , and benzoic a c i d s and pyridine bases such as pyridine, quinoline, and a c r i d i n e .

I n h i b i t o r s : Kost of t h e common organic i n h i b i t o r s would be expected t o r e a c t w i t h a c e t i c anhydride. S m a l l amounts of sodium hydrogen s u l f i t e edded t o t h e a c e t i c anhydride s o l u t i o n s appear t o have l i t t l e o r no e f f e c t upon t h e r a t e of quinone fornation. It i s t h e r e f o r e u n l i k e l y t h a t t h e photo-oddation r e a c t i o n i s a chain reaction. Types o f hydrocarbons: The following polycyclic hydrocarbons have been employed: naphthalene, anthracene, phenanthrene, tetracene, pyrene, chryseria, and triphenylene. !.Jith anthracene and t e t r a c e n e t h e o n l y products are anthracene-9,lCquinone and tetracene-5,12-quinone. It has been previously reported (1) t h a t

1 . C. Dufraisse and 3. liorclois, aull. SOC. C h i m . France (5) 3 , 18G3 (1936)

. 1 2 quinone. Pyrene, chryt e t r a c e n e i n xylene s o l u t i o n photo-oxidizes t o t h e 5 sene and triphenylene gave no appreciable anount o f any r e a c t i o n product although t h e s o l u t i o n s became colored. Naphthalene and phenanthrene gave a mixture of r e a c t i o n products which have not been completely characterized. Solutions i n a c e t i c anhydride of phenanthrena-9,lO-quinone are n o t s t a b l e i n t h e presence of oxygen and u l t r a - v i o l e t l i g h t b u t apparently undergo f u r t h e r oxidation. On t h e o t h e r hand s a t u r a t e d s o l u t i q n s of anthracene-9,lO-quinone i n a c e t i c anhydride have been exposed t o t h e r a d i a t i o n from t h e 360X lamp f o r several. months and no r e a c t i o n has been noticed. The u l t r a v i o l e t a b s o q t i o n s p e c t r i of naphthalene and phenanthrene begins r a t h e r far i n t h e u l t r a - v i o l e t region (2) where t h e

2 . 1 . Clar, Aromatische Kohle~iwasserstof f e , Springer, S e r l i n 1952, pagss 131

absorption by Pyrex i s appreciable and where t h e i n t e n s i t y o f such r a d i a t i o n from t h e 360EL lamp i s low. The photo-oxidation r e a c t i o n s f o r t h e two hjarocarbons w i l l need t o be i n v e s t i g a t e d i n q u a r t z f l a s k s with lower wave l e n g t h r a d i a t i o n . Solvents: The photo-oxidation of anthracene proceeds in t h e follo-drrg solvents: a c e t i c anhydrids, a c e t i c a c i d , mixbui-ss o f a c e t i c a c i d a n t anhydritie, cerhon d i s u l f i d e , methanol, pyridine, quinoline, and Jimethjlformamide. The f o r n a t i o n o f anthracene-9,lO-quinone was provsn by i t s i s o l a t i o n in a l l c i s e s except divethylformamide. For this solvent a p o s i t i v e v a t t e s t f o r quinone was obtained using sodium h y d r o s u l f i t e as t h e reducing agenl. The f o r n a t i o n of antk.racene-9,lGquinone i s very slow i n benzene and cyclohexane a s i n d i s a t e d by t h e v a t t e s t . It has been shown (3) t h a t t h e photo-oxidation of dnthracene i n cerbon d i s u l f i d e

9.

C. Dufraisse and E. Gerard.

B u l l . 3oc. C h b . France L,,

2052 (1937)

s o l u t i o n l e a d s t o t h e formation o f a photo-oxide. Carbon d i s u l f i d e i s one of t h e few s o l v e n t s which i n h i b i t s t h e formation of t h e photo-dimer of anthrscene. Xccording t o H. E. Ocamp t h e photo-oxidation of anthracene in nitrobenzene

(e),

__

4.

H. E. Ocanipo, C. A . 40, 6458 (1945)

s o l u t i o n g i v e s anthra~ene-9~10-quinone. Kechanism: The most recent i n v e s t i g a t i o n of t h e photo-oxidLtion o f anth-acene has been made by E. J. Bowen and h i s c o l l a b o r a t s r s ( 5 ) . T h e i r work has done

5. 4. J. 905ren e t al., Trans. Faraday SOC., Discussions Trans. Faraday SOC., 5 1 , 475 (19.55)

2,143,

l46 (1953);

p r i m a r i l y i n benzene s o l u t i o n although solvent mixtures were a l s o used. Lczordiig t o Bower t h e i r r a d i a t i o n o f benzene s o l u t i o n s of anthracene s a t u r a t e d d t h oxygen gives t h e following reactions: 1 . Pormaticn o f t h e photo-dimer, 2. ?o,mation o f t h e photo-oxide, and 3. 3 e a c t i o n with c e r t a i n adckd s o l v e n t s sach as CSlk. C u r work i n d i c a t e s t h a t with long exposure t h e s t h e r e i s a f o u r t h p o s s i b i l i t y , t h e formation of t h e quinone. The e x p e r i n e n t a l conditions Cor t h e two s e t s o f axperi2s ments a - d q u i t e d i f f s r e n t . Eowan used benzene as h i s prbkary solvent ~ h i c l ? has been i n d i c a t e d i n h i b i t s quinone formation, used high light i n t e n s i t i e s , used s h o r t e r times of r a d i a t i o n (hours r e l a t i v e t o days), used higher o-xygen pressures, and used d i l u t e s o l u t i o n s of anthracene. 3owen p o s t u l a t e d t h a t t h e forrnation o f t h e photo-oxide of anthracene i s a non-chain process. .From t h e r e s u l t s with sodium hflrogan s u l f i t e it is believed t h a t quinone forrnation i s a non-chain process. I n t h e work described here t h e photo-dimer w a s not d e t e c t e d i n t h e photo-oxidation reaction. The pnoto-dimer i s known t o show B photo-stationary s t a t e i n d i c a t i n g t h a t t h e reverse reaction, d i s s o c i a t i o n of t h e d h e r i n t o anthracene, has a f i n i t e v e l o c i t y constant, a t room temperature. During t h e long exposure times used i n this work any photo-dimer could be e v e n t u a l l y converted completely i n t o t h a quinone. Experimental

i Pyrex f l a s k s using t h e Apparatus: The expsriments described here were mads L r a d i a t i o n from a 360BL fluorescent lamp. A fesr experiments used s u n l i g h t . The transmission c h a r a c t e r i s t i c s o f ?y-rex glass and t h e d i s t r i b u t i o n o f r a d i a t i c n fron t h e 360% lamp have been pi?hlished by L. R. Koller (6). The r a d i e t i o n of this 6. L . 2. Koller, U l t r a v i o l e t iladiation, J o h X l e y a d Sons, Inc., pages 68 and 147.
Xew Pork,

1952,

-136Pyrex g l a s s cuts o f f lamp i s i n t b e r e g i o n 32GO-l+l+OO and peaks at 36OG around 3009 A. It i s probable t h a t t h e e f f e c t i v e r a d i a t i o n i s i n t h e region 3000-4000 A although w i t h t h e l o n g p e r i o d s of exposure used appreciable aqounts of s h o r t e r wavelength r a d i a t i o n might have passed through t h e s o l u t i o n s . Chemicals: The hydrocarbons used were i n s t o c k a t t h e Coal Research Laboratory and were o r i g i n a l l y from American and f o r e i g n manufacturers. Anthracene (bluewhite fluorescence) - i k s an Esstnan product and was used without f u r t h e r p u r i f i c a t i o n . A l l o t h e r hydrocarbons were r e c r y s t a l l i z e d a t l e a s t once and i n some cases two o r t h r e e times. R e c r y s t a l l i z a t i o n s were continued a n t i 1 t h e a e l t i n g point o f a given hydrocarbon ,vas l e s s t h a n one degree from t h e inelting p o i n t c i t e d by E. Clar ( 2 ) f o r this hydrocarbon. A l l s o l v e n t s were f r e s h l y d i s t i l l e d before use. Procedure: The g e n s r a l procedure was t o m k e a s s t u r a t e d s o l u t i o n of t h e hydrocarbon i n t h e d e s i r e d s o l v e n t and place it i n a Pyrex f l a s k , l o o s e l y stop?ered L t h a n aluminun: f c i l covered cork so t h a t air had access t o t h e s o l u t i o n . The f l a s k was t h c n placed one f o o t from t h e 360% lamp. The contents o f t h e f l a s k were a t room tzniperatme. A t times s m a l l p o r t i o n s o f t h e s o l u t i o n were withdrawn and addad t o an a l k a l i n e aqueous s o l u t i o n of sodium h y d r o s u l f i t e . Tha mixture W ~ S t h e n heated and t h e c o l o r changes observed. After exposure t o t h e r a d i a t i o r i f o r u s u a l l y a period of a week or more, t h e f l a s k was p l l c e d i n a water bath and the s o l v e n t eudpozated under vacuum w i t h a water I s p i r a t o r . iilith s o l v e n t s of lower v o l a t i l i t y t h e water b a t h w a s warmed t o about 50C. The s o l i d r e s i d u e was washed w i t h water t o remove c a t a l y s t , i f t h e salt of an organic a c i d was used, and then d i s s o l v e d i n methanol o r some o t h e r s u i t a b l e s o i v e n t such as benzene. The product was r e c r y s t a X i z e d a t l e a s t twice i n all cases. Yields of product were as high a s 7 4 of t h e o r e t i c a l 2nd depended mainly on t h e time of exposure and t h e intens i t y of t h e l i g h t source. S u n l i g h t always gave t h e higher y i e l d s for equal exposure times. The product from t h e anthracene o x i d a t i o n melted a t 287-289"C. which was i d e n t i c a l wiGh t h e melting p o i n t of m a u t h e n t i c sample of anthracene-9,lOquinone. A mixture of a u t h e n t i c sample and r e a c t i o n product gave t h e same melting p o i z t , 287-289O. Using t h e r e e c t i c n product f r o m t h e photo-oxidation o f anthracene, benzanthrone (1,9-10) w a s prepared by r e a c t i n g t h e product with g l y c e r o l i n 92-935 s u l f u i c a c i d - w a t e r s o l u t i o n using i r o n powder as t h e reducing agent and a r e a c t i o n temperature cf 118 t o 120C. A f t e r one h o w h e a t i a q t h e r e a c t i o n m i x ture was poured i n water, p r e c i p i t a t e d s o l i d recovered and r e c r y s t a l l i z e d from methanol. The a u r i f i e d benzanthrone was b r i g h t yellow, melted a t 167-169O, gave an orange colored s o h t i o n i n s u l f u r i c a c i d which shows a b r i g h t r e d e r a n g e fluorescence. These p r o p e r t i e s agree w i t h t h e p r o p e r t i e s of benzan:hrone c i t e d by J. Xouben (7), NhO lists 170" as being t h e melting point of t h i s compound.

i.

7. J. Houben, 3as Antiracen mil Die Anthrachinone, G. Thrime, Leipzig, 1920,

page 575.

For t e t r a c e n e t h e r e a c t i o n product from a c e t i c anhydride s o l u t i o n vas compared - d t h t h e r e a c t i o n product u s i n g xylene solution. The r e a c t i o n product from t h e photo-oxidation of t e t r a c e n e i n xylene has been i d e n t i f i e d as tetracene1 ) . The r e a c t i o n product f r o n a c e t i c anhyckide had a melting p o i n t 5,U-qmone ( of 293-295" and t h a t from xylene had a melting point 01 296297". The melting p o i n t f o r a mixture o f t h e two products was found t o be 295-296".
The r e a c t i o n product which p r e c i p i t a t e d by r a d i a t i o n of an anthracene s o l u t i o n i n a c e t i c anhydride which had been placed i n an evacuated and sealed t u b e gave a melting p o i n t of 245-250"C. J. Houben (8) l i s t s f o r t h e photo-dimer

-1378. J. Houbent loc. c i i . ,

page 135

of anthracene tile melting p o i n t s , 242-2&4", 2&", 272-274", and 27G-260". The wide v a r i a t i o n i s d t t r i b u t e d t o t h e f a c t t h a t 3n heating, t h e photo-diner diss o c i a t e s slowly i n t o anthracene and t h e melting point obtainad depends on t h e r a t e of heating. Tha sample on being allowed t o cool and reheated gave a melting point of 208". This behavior t s g e t h e r with t h e w h i t e c o l o r and t h e l a c k of fluorescence by t h e s o l i d , can be considered as evidence t h a t :he r e a c t i o n product obtained i n t h e absence of O2 i s t i e photo-diaer of aithracene.

-139Not f o r h b l i c a t i o n d s e n t e d Befor5 t h e Division o f Gas and f i e 1 Chemistry American Chemical Society Chicago, I l l i n o i s , Neetins, S e p t e d x r 7-l2, 1956 The k i n e t i c s of t h e Sorption a d Desorption of Kethanol on Coals of Various M s .George Ostapchenko and P a u l Fugassi Coal Research Laboratory, Dept. of C n k s t r y Carnegie I n s t i t u t e of Technology, Pittsburgh (131, Pa.

The k i n e t i c s of t h e sorption and desorption of methanol were measured on e i g h t American c o a l s varying i n rank from lignite t o bituminous and were found t o obey a second order r a t e equation. A mechanism f o r t h e s o r p t i o n process i s propoaed which explains t h e v a r i a t i o n of t h e experimental v e l o c i t y c o n s t v l t s with pressure and with t h e rank af the c o d . A n eqcilibrium sor;tion isotherm derived f r o m t h e sorption mechanism permits t h e calculation o f t h e noles of sorption s i t e s per grm of coal, A. The value.of A f o r t h e eigh? coals i s found t o be r e l a t e d t o the o q g e n content, 0, i n moXas per gram of coal by t h e r e l a t i m s h i p , . & = Oa667(0). This r e l a t i o n s h i p i s consistent' with tine hypothesis t h a t sorption occurs on hydroxyl groups, e i t h e r .massociatsd o r associate< Wi_tii ca.r*'oonyl groups. r o n t h e avidence gathered fxin this r e l a t i o n s h i p , it i s conchded t h a t s o q t i o n sccurs on s p e c i f i c sites, probably hydroxyl groups, t-hroighout t h e c o a l substance.

There has been much discussion i n t h e literature concerning the measweuient of the surface a r e a of coal and t h e i r k e r a c t i o c of polar molecules with t h e coal substance. A g r e a t d e a l of this discussion has been suimarized receEtly(1) and it was concluded t h a t the sorption of p o l a r molecules is agparently corqlicated by swelling and imbibition, involving weak bonds between t h e sorbate and y o k r groups i n t h e coal.

was f o w d t o obey a second order r a t e equation.

A k i n e t i c study of the sorption cjf methznol on coal w a s m d e ( 2 ) acd it ?'his r a t e s q m t i o n XZJ a i 3 0f o ~ i c ? tx be d e s c r i p t i v e i n cases of water and methanol sorg'iion' o~ sach thin~sas c e l l u l o s e and proteiiis(3j. I n t h e case of methmol sorption oil c a l l u l o s e a n e c h a i s n was proposed uMch l e d t o a nex sorption iso%s-rm equation t i . < contaim as one of t h e parsmeters, t h e moles of sorption s i t e s per gym of s c ~ b l n g i m t e r i a l . T h i s paper is concerned with the k i n e t i c s of t h e soypiion of r w t h n o l on col-1s of various r a n k i n order t o obtiiin q-czqtitati-te data concernirrz 5 3 2 i n t e r a c t i o n of r:izL&iiol with coals as t h e rank: c i ~ m s f sfrom l i g n i t e t o bitminous.

Tile apparatus and t h e proceduxc used i n t h i s k i n e t i c study ?rere the sane a s previous1;T described( 2) t r i t i ~t h e exceptio;i t h a t t h e e d s o r $ i s n tubas comectacl LE s e r i e s ;*rere used instead o f one. Tiia spying constants r e r e 1.92, r.5k ar?d 1.55 r r i l l i g r a a s p i - m i l l i a e t e r e,utansion.

The coal sa!?les used i n t h e sorption experiments a r e l i s t e d along . d i t ' ' t h e i r u l t h a t e analyses i n Table I.

Table I Ultimate Analyses of t h e Coal Samples "Coal Sample Upper Kittanning Lower Kittanning Upper Freeport Wyoming Kincaid L i g n i t e I l l i n o i s !io. 6 Hendrix Koundsville
.-

'

C -

H 3.8 4.7
4.8

N 1 . 1 0.9 0.9

Ultimate Analysis%S 0

3.7 4.1 4.6

1.3
2.3
.

.4sh 10.1

80.6

9.5

77.9 78.7 74.3 67.2 65.2

81.8

5.3 4 . 6
4.9 5.5 5 . 2

1.7
0.6

1 . 2
1.0 1.2

15.2

9.3
3.4
6.2

20.4
10.4

74.5

1.7 1.4
*Koisture Free

6.8 5.7

0.6 4.0
1 . 1 42

u.3
3 . 1
9-0

A l l tine coal sample3 were u3ed i n an unsieved, powdered f o m . The adsorbate, absolute methyl alcohol, a n a l y t i c a l reagent, was obtained from the Fisher S c i e n t i f i c Company and was used k i t h o u t p u r i f i c a t i o n .

i n s e t t i n g up t h e mechanism f o r methanol s o r p t i o n on coal, t h e mechanism has been mads t o conform t o t h r e e requirenents: (1)explanation of t h e second order r a t e equztion, (2) explanation of t h e variance of t h e experimental v e l o c i t y constant 16th pressure and ( 3 ) explanation o f having one molecule on one s i t e a t e q u i l i b r i u n conditions. This t h i r d requireme9t i s necessary since t h e eqQation f o r t h e equilibrium isotherm i s a function o f t h e pressure t o the f i r s t power. The following sorption mechanism i s postulated:

I.

+-

G-*S

Reactions I , 2 and 3 represent t h e adsorption of msthznol vapor, G, on t h e s-wface s i t e s , S, o f t h e coal. A l l reactions involving sirrface adsorption of t h e physical type a r e h o - m t o be r a p i d both on t h e o r e t i c a l and e,xperimental grounds. The doubiie a x o w i s used t o i n d i c a t e t i a t these r e a c t i o n s a r e i n e q u i l i b r i u m React i o n 4 i s t h e slow r a t e 4 e t e r m i n i n g s t e p f o r t h e sorption process. A molecule from a s i t e - o f t h e GneS t y p e migrates i n t o t h e i n t e r i o r and i s held on two i n t e r n a l , D, s i t e s , where n may be 1 , 2, j o r combinations of 1, 2 , The exact. value or" n, o r combination of n! s, depends on the' nature of 3 t h e coal's s u r f a c e LYd i n t e r i o r s i t e s . The f r a c t i o n of surface sites covered w i t h n molecules i s a function of t h e pressure and s i n c e these s i t e s precede i n t e r n a l sorption, t h e i r concentration w i l l e n t s r ilrto t h e experimental r a t e equation. Reaction 5 i s t h e reverse of r e a c t i o n and i s necessary t o account for t h e r e v e r s i b i l i t y of t h e reaction. Xeaction 6 i s a rapid e q u i l i b r i w i % m e r e a c t i o n which accounts for t h e fact t h a t one molecule i s held on one sit? vhen equilibrium i s reached.

....-.

....'..,
! ,

The r a t e ecpation derived from t h e aiiove sequence o f r e a c t i o n s i s

i n t h i s equation k i s the t r u e v e l o c i t y constant, % i 3 th i e , f r a c t i o n of surface s i t e s covered by n4 molecules, We i s t h e moles of adsorbzte per gram of sorbent at equilibrium, t i s t h e time, and f i s t h e f r a c t i o n of t h e r e a c t i o n completed a t time t.

The equilibrium sorption equation derivad i n a previous paper(3) i s

I n this equation, A i s t h e moles of sorption s i t e s per gram of sorbent, K i s t h e equilibrium constant f o r the d i s t r i b u t i o n of nethanol between t h e surface s i t e s and the iritei-ior sites, Kl i s the equilibrium constant for the i n t e r a c t i o n of gas molecules and t h e surface sites containing one molecule, K 2 is the equilibrium constant f o r t h e i n t e r a c t i o n of gas molecules and t h e surfac? sites containing two molecules, po is t h e vapor pressure of t h e absorbate a t t h e t e q e r a t u r e of t h e experiment and c i s t h e r e l a t i v e pressure.
X 3 U L T S .AN3 DISCUSSIOE -The experimental v e l o c i t y constants, a r e derived f r o m t h e experi. f / ( l - f ) i s pl p o t t e d a g a i n s t t and a s t r a i g h t mental data by using equation 1 l i n e 14 obtained whose slope i s k g e . Since We i s known experimentally, t h e value of % i s r e a d i l y obtained. A t y p i c a l p l o t of f / ( l - f ) i s shown i n

Figure 1 for methanol sorption a t 45C. on ;dyoinin& c o a l at a r e l a t i v e pressure of C.485. The k i n e t i c d a t a p l o t t e d in t M s way f o r t h e e i g h t coal samples f e l l reasonably w e l l on s t r a i g h t Lines which covered 9% o f t h e reaction.

k = kkgn. The e x p e r h s n t a l v e l o c i t y constant i s given by t h e e q r e s s i o n , Since t h e k4 is a constant % should vary w i t h t h e rzdktive pressure, c, i n t h e , .I n tha d e r i v a t i o n of t h e equilibrium s o q t i 0 3 equation same manner as 0 i s given by the following relationships:

e,,

where

, and 6 , a r e p l o t t e d against c, using values of 1 , 5 and I n Figure 2, 10 f o r Klpo and assuming K2po = 1 . I n t h e region KIP. = 5 t o Klp = 10, the curve f o r has a m a h m a t abodt c = 0.3, t h e curve f o r 0 2 gas a nz5nm a t about c = 0. j and t h e curve f o r has a , d n m at sbout c * 0.7. The experimental curves o f kx versus c f o r JQaning coal, Kincaid @ , i le t i and Illinois coal show a m ~ h m a t about 0.3, i n d i c a t i n g t h a t 8 , c o n t r i b u t e s most i n t h e k i n e t i c mechanism. The c u y e s f o r Zendrix and Moundsville coals snow a maxirim a t about c = 0.5, indi2ati.q t h a t e % c o n t r i b u t e s nost t o t h e :kinetic :.wchznism. The curves f o r Upper and Loder Kittanning coals shsw a m7-?dmun at about c = 0.7,

e, e.2.

e,

4 2 i n d i c a t i n g t h a t 6 3 c o n t r i b u t e s most t o t h e k i n e t i c luechanism. The c u v e for Upper Freeport c o a l appears t o be a s t r a i g h t l i n e i n d i c a t i n g t h a t n e i t h e r predominate br?t perhaps a l l t h r e e contribute i n such a way as to 8 ,,nor, give a s t r a i g h t !line.

e,

I n Figures 3 t o 6 a r e p l o t t e d t h e vcrlues of %n f o r each of t h e coal samples studied. These curves were drawn using t h e experimental values f o r X 1 derived from t h e equilibrium sorption equation. Also p l o t t e d in t h e s e figures a r e t h e values of g k which shoiild equal e,, at corresponding values of c. The agreement of t h e s e4values is q u i t e good. The vdues of k4 used i n t h e calculations . a r e l i s t e d i n Table 2 Table 2 Values of t h e True Velocity Constants

Coal Samvle -Upper K i t t a m i n g

t , " c .

4 , mole CB30H - hr

gms. coal

45
40 35

Lower Kittanning

45 40 35 4 5 35 30

Kbcaid L i g n i t e

45 35 30

Illinois No. 6

' 4 5 35
30

Hendrix

45 35 30

Moundsville

4 5 35 30

65,000

50,000

45 000

I n comparing t h i s phenomena with t h e rank of the coal, i t i s seen t h a t as t h e rank of t h e coal goes from l i g n i t e t o bituminous t h e controlling factor i n t h e sorption mechanism goes from 0, through to It car, be p o s t d a t e d t h a t this change i s c o n t r o l l e d by t h e r e l a t i v e a t , t r a c t i v e 2ower of the surface sites and t h e i n t e r i o r s i t e s f o r the methanolmolecule. The surface of coal can be imagined as naviing rmiy adsorption s i t e s , each having a d i f f e r e n t a f f i n i t y f o r t h e methanol molecule. The s i t e s could be t h e sane atcm o r group of atoms 'out having d i f f e r e n t a f f i n i t i e s due t o s t e r i c f a c t o r s o r i n t e r a c t i o n s with other s i t e s

e , .

The s i t e s having t h e lowzst a f f i n i t y would be capable of holding o n l y one metharol , and those having a s l i g h t l y g r e a t e r a f f i r i t y would be capable of molecule, 8 , holding two methanol molecules 8 , , e t c . The i n t e r n a l s i t e s w o d d then a t t r a c t i h a weakest held n e t h a o l riiolecule on t h e s u r f a c e i n e r e a t e s t concsntration.. It is postulated t h a t t h e methaxol molecule m o s t vea,c;ly held on t h e s u r I a c e i s t t e t h i r d molecule going en a s i t e holding t h r e e molecules, ?.he ne<kL would be t h e second molecule going cn a s i t e hoiding two I T I O ~ ~ C U ~ EtS h,e ne* would t h e %olecule going on a s i t e capable of holding o n l y one molecule. %er. +,he a t t f a c t i i i e power of t h e i n t e r n a l sites i s high, t h e i n t e r r i a l s i t e -dll a t L r a c t t h e m e t k n o l molecules on s i t e s o f most abundance. A s t h e a t t r a c t i v e power o f t h e i n t e r n a l sites decreases t h e s u r f a c e s i t e s with one methanolmolecllle becoEe s t a b l e and t h e i n t e r i o r s i t e s a t t r a 2 t t h e next n o s t a.F.wd.-n.t and Less w e ; * j held metb=nol molecule o r Thi3 argument can be applied agair, f o r e-controlling t h e r e x t i o n mechanism. A measwe of t'his a t t r a c t i v e power would be t h e d i f f e r e n c e i n free energy change f o r t h e methanol vapor going on t h e s u r f a c e ana t h e f r e e e x r g 7 change f o r t h e methanol vapor going i n t o t h e i n t e r i o r . This information i s a v a i l a b l e from t h e equilibrium s o r p t i o n equation which w i l l be discussed l a t e r and i s shown i n Table 3.

e%.

Table 3 Average Standard Free Energy Differences at 3Oo-45"C. Coal Sample Lower Kittanning Upper Kittanning Upper Freeport Moundsville Wyoming Hendrix Kincaid L i , @ . i t e I l l i n o i s No. 6

Free E n e r a Differences,
+23 - 6 -407

cal./mole

8 u
-918
,

-708

-la7
-1282

It can be seen from this Table and from Figures 3 t o b t h a t 2s t h e f r e e energy d i f f e r e n c e goes from a high t o a low negativs value, t h e type of s i t e conthrough to The sites t r o l l i n g t h e k i n e t i c mechanisn changes from 8 , f o r =ethanol s o r p t i o n on c o a l are probably t h e same type in t h e high rarik c o a l s as i n t h e low rank coals. I f it were assumed that a l l tile p r o p e r t i e s of t h e coals were t h e same, except f o r t h e number of sites, t h e n f r e e energy c t a r y e s would a l l be equal. However, t h e r e i s evidence(4) t h a t a s t h e rank of t h e s u b s t a w e incrsases, the. degree of compactness of t h e c o a l m k e l l e s increases. I C , i s probable then, a s t h e coal a i c e l l e s become more compact t h e free energy change of t h e methanol molecule due t o s o r p t i o n w i l l decrease.

e% e,.

The energies o f a c t i v a t i o n , Ea, f o r t h e s o r p t i o n process were c d c d a t e d n k versus S/T and zeasuring t h e slope. from t h e v d u e s in Table 2 by p l o t t i n g l k The Tralues of Ea f o r t h e coal sau.?les a r z snc-m i n Table 4.

'i

-&Table 4 Bctivatiori Energies f o r t h e Sorption Process

Coal Sample Upper Kittanning Lower Kittanning Upper Freeport

W Y O U

Ea, cal./mole

Ed, cal./mole
7035 7191

Za

Ed, cal./nole

1625 8798 5513

6 6 u 5395
5826 523 5 5501

-5409 1h7

4947 4795
4850 2991 U62

1640 355

Kincaid Lignite

545
2835

Illinois No. 6
Hendrh Moundsville

2 5 U

3073 2987

The case of methanol s o r p t i o n on Uppi- ,Freeport coal i s s i m i l a r t o t h a t reported e a r l i e r ( 2) f o r P i t t s b u r g h Edenborn coal. The v e l o c i t y constar;t f o r sorpt i o n did not vary w i t h pressure, making it CZfficult t o determine t h e degree of coritribution o f 8, , 8, , and e 3 f o r t h e rate determining step. I n this case t h e average v d u e of kxwas used i n determining t h e a c t i v a t i o n energy for sorption. These values a r e reported in Table 5. Table 5 " q e r i m e n t a l Velocity Constants f o r Upper Freeport Coal

Coal Sample Uppsr Freeport

m s . coal

t , " c .
45
40

k, (av. ) mole CH3GH

2717 2265 2010

- hr.

35
DESORPTION

The v e l o c i t y constants f o r t h e desorption of methamol from c o a l were a i l independent of t k e p r e s s u r e i n d i c a t i n g t h a t t h e experiinental v e l o c i t y cons t a n t i s t h e t r u e v e l o c i t y constant. The v e l o c i t y constants f o r t h e desorption of methanol f o r t h e e i g h t c o a l samples a r e l i s t e d i n Table 6.

-u15Table 6 Velocity Constants f o r Desorption Coal Sample Upper K i t t a m i n g

t "C. -

Lower Kittanning

45
40

35
Upper Freeport

45 40 35 45 35
30

Wyoming

Eincaid 'Lignite

45 35 30 45 35 30 45 35 30 45 35 30

.1,640 1,370 1,ljc1

Illinois

Koundsville

Hendrix

The energies of a c t i v a t i o n of desorption calculated from these valces a r e l i s t e d i n Table 4, along with the differences i n t t e a c t i v a t i o n anergies. T h i s difference i s t h e heat of r e a c t i o n f o r t h e r3te-deta-r!nhbg step (equation 4 of t h e sorption mechanism). Considering t h e a s s m p t i o n s nede m d t h e experimental e r r o r s involved,

it i s , d i f f i c u l t t o draw any conclusions concerai-ng the d i f f e r e n c e s in energies f o r t h e d i f f e r e n t coal samples. Howsver, i t can be noted t h a t t n e magnitude of t h e
energies involved excludes compound fom&tion of methanol with c o d a d i n d i c a t e s t h a t the type of bonding i s probably hydrogen bonding.

Equation 2, t h e equilibrium sorption equation contains 3 2arameters A, K and K 1 which can be evaluated by solving a s e r i e s of sinultzneous equations. The values f o r A, K and K 1 f o r t h e e i g h t coal smples a r e l i s t e d i n Table 7 .

l 4 &
Table 7 Constants f o r t h e Equilibrium Sorption Equation Coal Sample Upper Kittanning Lower Kittanning Upper Freeport

t, 'C.

moles of s i t e s x A, p. coal

104
K -

Kl -

45 40 35 45
40

16.4 16.7 16.4 15.7 15.6

15.2

2.56 2.31

2. 2- 43

35 45
40

Wyoming
Kincaid Lignite I l l i n o i s No. 6 Hendrix Pioundsville

35 45 35 30 45 35
30

45 35 30 45 35 30 45 35 30

22.9 23.2 22.9 61.4 66.0 67.7 73.7 82.7 83.2 53.0 57.6 61.0 25.9 25.2 23.9
22.8 21.7

2.09 2-33 1.90 1.62

1.15 1.06 1.04

6.82 9.06 10.33 5.73 8. l l 9.43 8.46 11.69

14.10

1.52

1.04

0.84 0.90 0- 79
0.80

1 4 . 30
25.34 28.29 14-48
21.12

9-19 15.71 20.49

20.4

0.69 1.60 139 1 . 2 1 1.64 1.68 1.35

32-29 14.48
21.12

32.29
12.00

17.16 26.10

It can be saen fror: Table 7 tht t h e values f o r A, t h e moles of s i t e s p e r gram of coal are independent of temperature and t n a t t h e values decrease a s a n ! ! of t h e coal increases. It has been suggested(3) t h a t the atom which tne r forms t h e s i t e f o r a e t h a n o l sorption on coal i s probably oxygen. I n Figure 7 i s p l o t t e d t h e m d e s o f s i t e s per gram o f coal versus t h e moles o f oxygen per gram of coal. The r e l a t i o n s h i p obtained is l i n e a r h i t n a slope o f 2/3 which ind i c a t e s 2/3 of t h e omgen i n coals from high rank t o low rank, form s i t e s f o r methanol sorption. S e v e r a l workers( 5) (6) hava analyzed coals of varying rank i n C=O, and non-reactive 0 groups i n an effort t o follow terms of OH, COOH, OCH t h e changes occurring a&ing t h e c o a l i f i c a t i o n process. Using t h e data of Blom e t a L ( 6 ) and assuming v a r i o u s combinations of oxygen groups, it vas found t h a t t h e suin of OH and COOH groups gave t h e b e s t agreement with the number o f sorption s i t e s i n t h e coal. These d a t a a r e p l o t t e d i n Figure 7 and the agraement i s quite good considering t h e experimental d i f f i c u l t i e s i n detei-mining t h e oxygen groups i n coals. The f a c t t h a t t h e sites responsible f o r s o v t i o n remain a t a constant r a t i o with respect t o t h e t o t a l oxygen during t h e c o a l i f i c a t i o n process indicates t h a t t h e oxygen i s l o s t i n a very s p e c i f i c manner as coal ages. The exact manner i n which the o q g e n i s l o s t is d i f f i c u l t t o determine s i n c e l i t t l e i s known about t h e manner of t h e changes of t h e carbon and hydrogen during c o a l i f i c a t i o n .
Since t h e equilibrium c o n s t a t s X and K 1 a r e known a t t h r e e temperat u r e s , t h e values o f my AFO, and hs c o d d be determined. These values i n cal/mole f o r &I and AF" and cal/degree f o r IS a r e shown i n Table 8.

-147

is -

Table 8 Values of oY, M 0 and IS i o r Z q u i l i b r i m S o r p t i o n Coal Sample Ugper K i t t a r n i n g Lower Kittanning Upper Freeport Wyoming Kincaid L i g n i t e I l l i i i o i s No. 6 Hendrix Moundsville
I s 3 -

nr' -620 2.1 -all

195 1 543 4429 1131 u9l 2331 3843 3445

-1215
-1372 -LjO -U04

-522 -540
-566

-21* 7
-21. j

7.9

2. L
i .4 4.2

-9751 -99E3
-10,2@Lb

-459 -51s
-408

4.4
12.6
13.2

-302

-1303 -1375 -1351 -1531

2 i . 7 -27.2 -27.0
-27.2 -27.1 -27.0 -27.0

-2

5 6 67 37 25 27 1 1

13.5
3.8 3.8 3.8

-u+,Oz -1627
-1820

-1621

-27 * 7 -27.7
-27.7

3.8
3. a 3.9
6.3

-7928

-1ss3
1 9 3 %
-2c12

-19.6
-19.7 -19.5 -29.7 -3c. 9
-27.3

126 224 -797

-203

6 151

-11,229
-9503

-1690
-Lo^$

7.1
7.0 12.0 i3.1 13.1
E .8

-2094

-1622
-2006

-2L. 8 -&L
-24.6

-113
-212

-10,827

-SOL5 -1572

-312 -179

12.2

1 : G

-29.1
-29.5
-29;

11.9

-1965

These values show t h a t t h e e n e r s associated with t b e s o q t i o n of ;!e%?.a??ol on c o e l s i s small compared t.o energies due t o cheinicai reacbions ziti, a s Tolntsd ont i n the discussion of t h e r a t e equation, a r e probably due t o h p i r o g s r ; 'condhg. TheFa seems t o be no corre1dtio;l of t h e s e values x i t h t h e rank sf t h e coai.

The s o r p t i o n o f rtiethanol on coal t a k e s place i n . t w o steps. (1) ?he adsorption of methanol inolecules on t o t h e s l l r f a e of t h e c o a l i n 3 wLtiizger ranner. (2) Migration of t h e moleculss f r o m t h e surface iE5o t h e i n t e r l o r of t h e c o a l substance. The f i r s t s t e p i s r a p i d and t h 5 nunber of s i t e s involved i s smll com2ared t o t h e t o t a l nwnber of s i t e s responsible f o r s o r p t i o n s i n c e the second o r d e r r a t e equation i s obeyed d w i n g ' ; h e e n t i r e s c r p t i o n process. The migration of t h e methanol molecules i n t o t h e i n t e r i o r occurs p r i m a r i l y f r o 3 only one t 3 ~ e of surface site. This s i t e i r a ~ be hoidiiig z i t h e r one, Lwo o r three methanol molecules, depending o n t h e rank o f t h e coei s a q l e . As t h e rank of t h e c o a l increases iron? l i g r A t a t o b i t d n o u s , t h e t;Te o f s i t e determiring changes from a s i x f a c e s i t e holding one m t h a n o l molecule t o a SUTface s i t e holding t h r e e methanol molscules. This change i s r e l a t e d t o t h a j ' r e e energy changes f o r methanol s o r p t i o n caused by t h e c o a l i f i c e t i o n process. The males of s i t e s per g r m o f coal, A, x s p o n s i b l e f o r methanol sorpt i o n cai? be calculated from t h e equilibrium sorption i s o t h e r m The value oi A

decreases a s t h e rad.! of t h e c o a l i n c r e a s e s and i t was found t h a t A c a r i e s d i r e c t l y with t h e oxygen content of t h e coal, 0, i n moles per gram according t o t h e reiat i o n s h i p A = 0.667 (C). Since t h e oxygen content of c o a l cannot be obtained dj-rectly by a n a l y t i c d ~r.=am, And i s obtained by d i f f e r e n c e , %ne s o r p t i o n o f sethano1 oil c o a l might be used es a means of determining oxygen d i r e c t l y .
. The v a r i a t i o n o f t h e value of 4 . with t h e rank of t h e c o a l i s consister,t with t h e hypothesis t h a t t h e s i t e s rsaponzible f o r a o r p t i o n a r e hydro.uyl groups e i t h e r i s o l a t e d o r i n conjunction with a carbonyl group. T h i s i s corroborated by t h e e n e r g i e s derived from t h e s o r p t i o n d a t a w'hich show that; t h e type of bonding which occurs during methanol s o r p t i o n i s probably hydrogen bonding. The loss of o.q-gen i n c o a l during t h e c o a l i f i c a t i o n process occurs i n such a manner t h a t f o r every two h y d r o q l groups l o s t one o t h e r type of oxygen group i s a l s o l o s t . The l o s s of oxygen i n t h i s n v l n e r i s q u i t s s p e c i f i c and f u r t h e r i n w s t i g a t i o n s along t h e s e l i n e s &gilt provide q u a n t i t a t i v e d a t a concerning t h e c o a l i f i c a t i o n . process.

I
I

1
8

This communication has been a b s t r a c t e d frcm t h e t h e s i s o f George Gstapchenko subrfitted t o t h e Graduate School, Carnegie I n s t i t u t e of Technolo*, in p a r t i a l f u l f i l l m e n t of the requirements f o r t h e degree, Doctor o f Philosophy. The authors wish t o acknowledge and tnank BCZ, Inc. and t h e Coal Research Board, Commonwealth of ?ennsylvania, f o r furnishing samples and a n a l y t i c a l d2ta Toi-t h e K i t t a m i n g and l r e e p o r t coals. They also wish t o thaiik Dr. K a r t i n Neuwortn, Pittsburgh-Consolici=tion -Coal Co., forsunFles of Hendrix a d Noundsville coals.

1 . Anderson, R. 9., B a l l . 1 1 . K., Lecky., J. A., end S t e i n , K . C., J. o f Phys. Chem. 60, 1543 (1956) 2. Fugassi, J. P., Ostapchenko, G. and Trammell, R., Fuel, 36 (1958) . a d Ostapchenko, G . , unpublished m a t e r i a l 3. h g a s s i , J. P 4 . Bailgham, D. H., Franklin, R. E., Hirst, U . and ld!aggs, F. PA. ?., fiel, 28, 231 ( 1949 5. Ihnato\licz, L , ?race Glownego Inst., Gornictwa (Katowice) , Komm., (1952) 6. Slou., L . , Bdelhausen, L. and van Krevelen, D. Y., Fuel, 36 (1957)

x,

f I-f

t , hrs.
I

I
I

FIGUiLE 1

- .UTE

PLOTS
..

0.2

0.2

0. E

IO

0.4

0 . 2

0,2

0.4
P Po

0.6

0.8

1.0

FIGURE

e AS A EVNCITON OF

-151-

v
kx ks

I
Kincaid

I
Lignite

0 0.6 - 0
0.4

0-45.e

el
0.2

0
kx k4

0.4

0 1 0.2

0.2 P PO

0.6

0.8

FIGURE 3

AS A FUNC'ITON O F c

-152-

0.8
kx k4

1 Ilinois N0.6

0.6

0.4

0.2

0.2

0.4

P Po

0.6

0.8

I .o

FIGURE 4

el AS A

FUNCPON

OF

-153-

0 - 4 5 "6
kx ks

M oundsvil I e

0.2

Qe

0.1

0
Hendrix
kx ks

0.2

Qo

0.

0.2

0.4 P
Po
FIGU;IE

0.6

0.8

I .o

e AS
2

A FUNCTION

OF c

-154-

0 - 4 5 "c
kx -

k*

0.2

- 4 0 "C

0-35"C
Lower K i t t anning

I 8 s 0.

-8

j
0.4
P Po

0
kx
k4

0.2

8 3

0.I

0.2

0.6

0.8.
.

1.0

FIGURE

e AS A rn1c~101.r OF
3

'I

\ a

0
CD
t

\*

0
4 -

T3

0 W

c
0
c

c
P
L

E
0

0 d-

a
I x
-\

I
0

cu

0 CD

0
u)

cu e loo:, UOJB

sa+!s 40 s a p w