Fuel 84 (2005) 487494 www.fuelrst.

com

A correlation for calculating HHV from proximate analysis of solid fuels

Jigisha Parikha,*, S.A. Channiwalab, G.K. Ghosalc
b a Chemical Engineering Department, Sarvajanik College of Engineering and Technology, Surat, Gujarat 395001, India Mechanical Engineering Department, Sardar Vallabhbhai National Institute of Technology, Surat, Gujarat 395007, India c Chemical Engineering Department, Laxminarayan Institute of Technology, Nagpur, India

Received 20 June 2004; received in revised form 16 October 2004; accepted 17 October 2004 Available online 19 November 2004

Abstract Higher heating value (HHV) and composition of biomass, coal and other solid fuels, are important properties which dene the energy content and determine the clean and efcient use of these fuels. There exists a variety of correlations for predicting HHV from ultimate analysis of fuels. However, the ultimate analysis requires very expensive equipments and highly trained analysts. The proximate analysis on the other hand only requires standard laboratory equipments and can be run by any competent scientist or engineer. A few number of correlations of HHV with proximate analysis have appeared in the solid fuel literature in the past but were focused on one fuel or dependent on the country of origin. This work introduces a general correlation, based on proximate analysis of solid fuels, to calculate HHV, using 450 data points and validated further for additional 100 data points. The entire spectrum of solid carbonaceous materials like coals, lignite, all types of biomass material, and char to residue-derived fuels have been considered in derivation of present correlation which is given as below: HHVZ0.3536FCC0.1559VMK0.0078ASH (MJ/kg) (where FC 1.091.5% xed carbon, VM 0.9290.6% volatile matter and Ash 0.1277.7% ash content in wt% on a dry basis). The average absolute error of this correlation is 3.74% and bias error is 0.12% with respect to the measured value of HHV, which is much less than that of previous correlations of the similar kind. The major advantage of this correlation is its capability to compute HHV of any fuel simply from its proximate analysis and thereby provides a useful tool for modeling of combustion, gasication and pyrolysis processes. It can also be used in examining old/new data for probable errors when results lie much outside the predicted results. q 2004 Elsevier Ltd. All rights reserved.
Keywords: Solid fuels; Higher heating value (HHV); Proximate analysis

1. Introduction The eld of thermo chemistry is one of the foundation stones of modern energy dependant society. The proximate and ultimate analysis of biomass and coal are necessary for their efcient and clean utilization while the HHV of these materials determine the quantitative energy content of these fuels. There exists a variety of correlations for predicting HHV from ultimate analysis of fuel [1]. One such correlation, presented in the earlier work by Channiwala et al. [1], encompasses the complete spectrum of fuel
Abbreviations: ABE, average bias error; AAE, average absolute error; AAD, average absolute deviation; ABD, average bias deviation; HHVc, calculated HHV; MSW, municipal solid waste. * Corresponding author. Tel.: C91 261 3037289; fax: C91 261 2660099. E-mail address: kamal_parikh@yahoo.com (J. Parikh). 0016-2361/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.fuel.2004.10.010

ranging from gaseous, liquid, solid and refuse derived fuels. It is of the form, HHVZ0.3491CC1.1783HC0.1005SK 0.1034OK0.0151NK0.0211A (MJ/kg) with the ranges as, 0%%C%92.25%, 0.43%%H%25.15%, 0%%O%50.00%, 0% % N % 5.60%, 0% % S % 94.08%, 0% % A % 71.4%, 4.745 MJ/kg%HHV%55.345 MJ/kg. C, H, O, N, S and A represents carbon, hydrogen, oxygen, nitrogen, sulphur and ash contents of material respectively, expressed in mass percentages on dry basis. This correlation offers predictions with average absolute error of 1.45% and average bias error of 0%. Although, this generalized correlation which being extremely useful in performance modeling of thermal systems, suffers from a drawback that it needs an elemental analysis as an input data, which needs expensive equipments and highly skilled analysts [24]. There also exists a few correlations (Table 1) based on proximate analysis/chemical analysis of solid fuels

488

J. Parikh et al. / Fuel 84 (2005) 487494

Table 1 Equations available in literature for HHV determination from proximate analysis Sr. no. 1 2 3 4 Equation HHVZ76.56K1.3 (VMCash)C7.03!10 (VMCash) HHVZ354.3FCC170.8VM considering VMCFCCASHZ100 HHVZ0.196FCC14.119 HHVZK10814.08C313.3T where, TZ(FCCVM)
K3 2

Data points 26 24 16 11 nos. [2] nos. [3] nos. [5] nos. [6]

Error Mean diff. 6.34% max 21.2% K7.74 to 4.42% Mean diff. 2.20% !10%

Sample Lignite BiomassCchar Biomass Biomass

HHV, MJ/kg, dry basis; FC, %wt, dry basis; VM, %wt, dry basis.

[2,3,57] but their use and applicability is limited to one kind of fuel or the region only. The present study is focused on the development of a correlation for predicting HHV of solid fuels from its proximate analysis, which is much easy to perform and needs only standard equipments and moderately trained analysts. The correlation has been derived using 450 data points and additional 100 data points used for its validation. The derivation, validation and comparative assessment with published correlations are presented in the subsequent sections.

method for sparking fuels. For coconut shell, ber, pith and bagasse, ash fusion was observed at 950 8C and hence these samples were analyzed for volatile matter contents only up to 750 8C. Determination of the ash content in the test samples was carried out according to ASTM D 3174-89 [12] method in the electric mufe furnace. The xed carbon content of the test samples was calculated by difference. The HHV of these materials was determined according to ASTM D 2015-85 [13] method in a parr microprocessor controlled oxygen Bomb Calorimeter (Model 1241 EF). 2.2. Step 2: selection of suitable data Through, the process of collection and generation of data, information of about 577 materials was collected. Out of these, 450 data points have been used for the purpose of the derivation of the correlation, while 100 were used for validation of the correlation. In order to develop a generalized correlation to the extent possible, care was taken to include data points of the widest range of proximate analysis and HHV. Table 2 presents the summary of few such data points used for derivation and validation of the correlation. In total, 450 data points comprising of 30 data points for manufactured fuel, 13 data points for coal and cokes, 321 data points for biomass material, 25 data points for industrial waste, 11 municipal solid waste, refuse and sludge and 50 data points for chars have been used for derivation of the correlation and 100 data points comprising of 13 coals, cokes and manufactured fuel, 70 biomass materials, 9 refuse material and 8 chars have been used for validation of the correlation. It is worth mentioning over here that equal weightage is given to mature form of fuels and immature form of fuels like biomass. Further, the composite fuels like industrial and municipal solid waste, refuse derived fuels and sludge as well as articially prepared chars are also included as data points in derivation of this correlation. The spectrum of samples is so selected that it approximately represents the relative proportions of their occurrence in nature and thus permits the derivation of a truly useful correlation. The range of data points considered is such that the volatile matter of the samples varied between 0.92 and 90.6%, the ash content between 0.1 and 77.7%, xed carbon content between 1.0 and 91.5%, and the heat content between 5.63 and 39.388 MJ/kg (on dry basis).

2. Derivation of correlation Steps involved in the development and derivation of correlation are listed below: 2.1. Step 1: collection and generation of data To obtain the correlation, proximate analysis data pertaining to different biomass-forest and agricultural wastes have been used as well as the chars resulting from their thermal treatment in an inert atmosphere at different temperature, so as to cover a wide range of chars from low to high degree of carbonization and different coal and lignite samples have also been considered to cover a widely different values of Fixed Carbon, Volatile Matter and Ash content. 2.1.1. Collection of data Data pertaining to large number of fuels like coals/lignite/manufactured fuel/all kind of biomass/industry waste etc. were collected from the published literature. 2.1.2. Generation of data Since there are no standard sampling procedures specied for biomass materials, the samples were collected with due care to get the most representative samples. Preparation of samples was carried out in accordance with ASTM D 2013-86 [8]. The procedure requires samples to be in powder form of up to 250 mm grain size. The proximate analysis of samples were carried out in accordance with ASTM D 3172-73(84) [9] standard. The moisture content in test samples was determined according to ASTM D 3173-87 [10] method in sartorious infrared moisture meter. The volatile matter contents in the test samples were determined according to ASTM D 3175-89 [11] modied

J. Parikh et al. / Fuel 84 (2005) 487494 Table 2 Summary of few salient data used for derivation and validation of correlation
Sr. no. Raw material Proximate analysis [% by wt. dry basis] FC 1 Coals/coke 1 Northumber land no. 8-Anth. coal 2 German-Anna 3 Coal sample 4 German Braunkohole lignite 5 L01 6 L14 7 Charcoal 8 Coke 2 Manufactured fuel/wood 9 Cotton shells briquettes 10 Lantana briquettes 11 Press mud briquettes 12 Block wood 13 Ply wood 3 Biomasspit/shells/seeds/cobs 14 Peach Pit 15 Macadamia shell 16 Pistachio shell 17 Hazelnut shell 18 Coconut shell powder 19 Cotton shells 20 Spire-mint 21 Corncob 22 Corncob 23 Cornstover 24 Corn cob 4 Biomasswood/energy crops 25 Wood Chips 26 Canyon live Oak 27 Red wood 28 Softwood 29 Spruce wood 30 Es 31 Pine wood 32 Subabul wood 33 Eucalyptus 34 Eucalyptus-Grandis 35 Sudan Grass 36 Subabul 5 Other biomassbarks/prunings 37 Douglas Fir 38 Loblolly Pine 39 Eucatlyptus bark 40 Almond 41 Cabernet Sauvignon 42 Walnut 43 Olive twigs 44 Wood chips 45 Coffee chaff 46 Tapero root skin scale scrapping 5 Other biomassstraws 47 Wheat straw 48 Paddy straw 49 Rice straw (ground) 50 Wheat straw 84.59 79.60 30.00 46.03 27.72 39.37 89.10 91.47 17.10 11.90 8.60 14.59 21.80 19.80 23.68 16.84 28.30 20.58 16.90 11.80 18.54 12.50 17.60 16.80 23.50 11.30 19.92 28.10 28.30 17.90 15.70 18.52 21.30 16.93 18.60 13.80 25.80 33.90 15.30 21.54 19.20 20.80 10.73 15.40 19.60 11.40 23.50 11.80 16.20 24.00 VM 7.09 12.00 30.00 49.47 38.32 52.05 9.88 0.92 77.80 20.80 54.70 83.32 74.20 79.10 75.92 82.03 69.30 79.07 68.50 70.10 80.10 86.50 78.70 82.10 76.40 88.20 79.72 70.00 70.20 82.00 73.60 81.02 75.35 82.55 72.75 85.20 73.00 65.70 65.70 76.83 78.63 78.50 88.25 83.40 75.80 35.10 63.00 72.70 68.30 69.60 ASH 8.32 8.40 40.00 4.50 33.96 8.58 1.02 7.61 5.10 67.30 36.70 2.09 4.00 1.10 0.40 1.13 1.40 0.35 14.60 18.10 1.36 1.00 3.70 1.10 0.10 0.50 0.36 1.70 1.50 0.10 11.30 1.20 3.35 0.52 8.65 1.00 1.20 0.40 19.00 1.63 2.17 0.70 1.02 1.20 4.60 39.20 13.50 15.50 15.50 6.40 Ultimate analysis [% by wt. dry basis] C 83.67 82.62 63.89 92.04 89.13 46.9 49.14 54.41 48.79 52.9 37.23 46.58 49 48.1 47.84 50.64 52.1 51.9 47.3 48.15 46.04 48.33 44.58 56.2 56.3 51.3 46.59 48.2 45.5 35.97 H 3.56 3.02 4.97 2.45 0.43 6.07 6.34 4.99 5.91 5.6 5.34 5.87 5.4 5.99 5.8 5.98 6.1 6.1 6 5.87 5.82 5.89 5.35 5.9 5.6 5.29 5.85 6.25 5.1 5.28 O 2.84 3.66 24.54 2.96 0.98 43.99 43.52 39.69 43.41 42.7 33.38 45.46 44.6 45.74 45.76 42.88 41 40.9 46.5 44.75 44.49 45.13 39.18 36.7 37.7 40.9 43.9 43.24 34.1 43.08 N 0.55 0.92 0.57 0.53 0.85 0.95 0.48 0.36 0.56 1.4 5.95 0.47 0.4 0.08 0.07 0.05 0.2 0.3 0.1 0.03 0.3 0.15 1.21 0 0 0.66 0.83 1.61 1.8 0.17 S 1.05 0.73 0.48 1 1 0 0.02 0.01 0.01 0.01 0 0.01 0.03 0 0 0.01 0.01 0 0 0.01 0.04 Measured HHV Dry basis MJ/kg 32.856 33.000 14.772 25.100 15.630 24.080 34.388 31.124 19.055 7.687 11.972 18.261 19.720 19.423 21.010 19.260 19.300 19.675 16.376 15.530 18.770 17.000 17.800 18.795 19.916 18.981 20.720 20.000 20.100 20.080 16.644 19.777 18.640 19.350 17.390 16.660 22.098 21.772 15.195 20.010 19.030 19.967 18.699 20.031 17.686 9.228 17.000 14.522 15.614 18.905 Calculated HHV MJ/kg 30.951 29.952 14.973 23.953 15.511 21.969 33.038 32.428 18.136 6.926 11.282 18.132 19.245 19.324 20.206 18.734 20.800 19.601 16.541 14.960 19.033 17.898 18.464 18.731 20.220 17.742 19.469 20.836 20.939 19.112 16.938 19.170 19.253 18.852 17.851 18.155 20.494 22.227 15.505 19.582 19.031 19.588 17.544 18.438 18.712 9.197 18.026 15.386 16.255 19.287 (continued on [14] [15] [16] [15] [4] [4] [17] [18] [16] [16] [16] [PS] [16] [19] [20] [20] [5] [16] [16] [16] [20] [5] [5] [16] [21] [19] [20] [5] [5] [3] [16] [PS] [PS] [20] [20] [16] [22] [23] [16] [20] [20] [16] [6] [16] [16] [16]

489

Reference

[5] [16] [16] [16] next page)

490 Table 2 (continued )

Sr. no. Raw material

J. Parikh et al. / Fuel 84 (2005) 487494

Proximate analysis [% by wt. dry basis] FC VM 80.60 74.40 65.40 60.90 87.30 75.10 66.58 70.90 86.30 86.60 75.00 67.80 57.20 77.40 69.45 87.20 47.70 76.50 62.54 74.20 83.60 72.40 55.40 79.00 85.50 87.30 70.40 85.00 12.70 45.00 8.10 9.09 67.90 83.41 67.30 74.70 83.17 90.60 75.10 14.30 25.00 35.70 ASH 7.70 2.50 17.30 17.20 0.50 2.10 3.72 22.10 1.80 2.80 20.00 21.50 17.30 7.70 4.72 0.40 48.70 1.70 9.40 16.50 0.20 2.60 32.70 13.50 0.30 10.80 1.50 1.40 39.60 54.00 72.10 77.70 6.90 1.61 17.60 3.30 0.25 0.20 4.80 59.80 70.00 60.30

Ultimate analysis [% by wt. dry basis] C 39.47 44.23 50.29 36.2 39.75 19.12 47.67 48.12 49.37 48.6 42.66 39.59 49.5 49 48.67 H 5.07 6.61 5.05 4.72 5.55 2 6.13 6.55 6.4 5.5 6.05 5.26 5.7 5.98 6.03 O 39.14 46.25 39.63 37.49 46.82 25.96 43.16 35.93 42.01 39.5 49.5 36.38 41.3 44.75 44.99 N 1.2 0.51 0.45 4.29 0.17 4.22 1.33 0 2.02 0.5 0.18 2.09 0.2 0.05 0.06 S 0.02 0.16 0 0 0 0 0.01 0.04

Measured HHV Dry basis MJ/kg 17.355 19.436 15.830 17.489 19.097 18.356 20.050 14.258 18.167 17.192 13.730 14.232 18.125 17.410 21.055 19.297 8.887 19.842 20.600 15.890 18.502 20.930 13.727 15.153 18.000 14.806 21.097 17.100 17.657 7.685 8.385 5.700 21.010 17.483 16.420 19.228 19.950 18.934 18.928 11.712 5.630 7.183

Calculated HHV MJ/kg 16.643 19.748 16.178 17.104 17.920 19.754 20.853 13.356 17.648 17.227 13.305 14.186 17.799 17.275 19.924 17.976 8.330 19.622 19.599 14.728 18.760 20.107 12.590 14.863 18.348 14.198 20.900 18.050 18.538 6.948 7.702 5.479 19.443 18.284 15.694 19.364 18.827 17.376 18.778 10.921 5.120 6.510

Reference

51 Wheat straw 11.70 52 Coconut stem 23.10 5 Other biomassstalks 53 Cotton stalk 17.30 54 Musturd stalk (A) 21.90 55 Eucalyptus stalk 12.20 56 Mulberry stick 22.80 5 Other biomassbrous material/leaves/grass 57 Coconut coir 29.70 58 Bagasse 7.00 59 Bagasse 11.90 60 Bagasse pith 10.60 61 Sweet sorghum bagasse 5.00 62 Moringa-oleifera (leaves) 10.70 63 Sena leaves 25.50 64 Sugar cane leaves 14.90 65 Olive marc 25.83 66 Miscalthus (elephanta grass) 12.40 67 Dal lake weed 3.60 68 Tea bush 21.80 5 Other biomasshull/husk/dust 69 Sal seed husk 28.06 70 Bamboo dust 9.30 71 Eucalyptus saw dust 16.20 72 Saw dust 25.00 73 Saw dustCmusturd 11.90 5 Other biomassothers/misc. 74 Mentha Piperita 7.50 75 Grewia optiva (Bhimal) 14.20 76 Water hyacinth 1.90 77 Ceder cones 28.10 6 Biomass waste material 78 Tea waste 13.60 79 Bamboo stick waste 47.70 80 Tannary waste 1.00 81 Fly ash (Bagasse fuel) 19.80 82 Pine needle (40% clay) 13.20 83 Castor seed cake 25.20 6 Biomass waste materialmillgin industry waste 84 Cotton gin waste 14.97 85 Cottongin Trash 15.10 86 Alabama Oak wood waste 21.90 87 White Fir 16.58 88 Tan Oak 9.20 6 Biomass waste materialrefuse/MSW 89 Industrial waste (stalla) 20.10 90 Poultry pure waste 25.90 91 Municipal Solid Waste-Bareilly 5.00 92 Municipal Solid Waste-Morada4.00 bad 7 Biomass chars 93 Redwood char-790-1020 8F 67.70 94 Oak char-820-1185 8F 59.30 95 Coconut shell char-750 8C 87.17 96 QrC550 82.20 97 PhC 300 31.30 98 EsC700 91.50 99 Rice husk char 41.20 100 Rose apple char 12.20 PS, present study.

[16] [16] [20] [16] [16] [16] [20] [16] [16] [16] [16] [16] [16] [16] [6] [16] [16] [16] [24] [16] [16] [16] [16] [16] [16] [16] [16] [5] [16] [16] [16] [16] [16] [PS] [20] [25] [20] [19] [16] [16] [16] [16]

30.00 25.80 9.93 14.70 68.10 6.60 5.90 22.20

2.30 14.90 2.90 3.10 0.60 1.90 52.90 65.60

75.6 67.7 88.95 87.1 57.8 92.7

3.3 2.4 0.73 2.4 5 1.6

18.4 14.4 6.04 6.9 36.5 3.3

0.2 0.4 1.38 0.5 0.2 0.4

0.2 0.2 0

28.844 24.796 31.124 32.720 22.840 32.204 14.944 7.577

28.598 24.874 32.349 31.333 21.680 33.369 15.076 7.263

[25] [25] [PS] [3] [3] [3] [16] [16]

J. Parikh et al. / Fuel 84 (2005) 487494 Table 3 List of algebraic equations tested for the development of correlation Sr. no. 1 2 3 4 5 6 7 8 9 10 Proposed equation HHVZaFCCbVMCcAsh HHVZaFCCb(FC/VM)CcVMCdAsh HHVZaFCCbFC2Cc(FC*VM)CdVMCeVM2CdAsh HHVZa1FCCb1VM HHVZa1FCCb2VMc2 HHVZa2FCCb3VMc3 HHVZa3FCd3Cb2VMc2 HHVZa4FCd4Cb1VM HHVZa1FCCb2VMc2Cd2Ash HHVZaCb(FCCVM)CcAsh Criteria for selection

491

Assuming fuel HHV to be a linear function of its volatile, non-volatile and inorganic constituents Assuming fuel HHV to be a linear function of its volatile, non-volatile and inorganic constituents and combining the effect of ratio of non volatility to volatility Based on Kucukbayraks criteria [2] of assuming fuel HHV to be a polynomial function of its volatile, non-volatile and inorganic constituents Based on Corderos criteria [3] Assuming nonlinear effect of a volatile matter on HHV (Based on Kucukbayraks criteria [2]) Assuming nonlinear effect of a volatile matter on HHV Assuming nonlinear effect of a volatile and a non volatile matter on HHV Assuming nonlinear effect of a non volatile matter on HHV Assuming nonlinear effect of a volatile matter on HHV Assuming combined effect of a volatile and a non volatile matter on HHV (Based on Jimenezs criteria [6])

HHV, MJ/kg, dry basis; FC, %wt, dry basis; VM, %wt, dry basis.

2.3. Step 3: selection of suitable forms of correlations The basis of the equation should be such that there is no reason to restrict its applicability to coal/lignocellulosic material but it should be equally valid for any organic substance. Heating value is proportional to the carbon content and hydrogen content present in the sample. Hence, HHV is assumed to be the function of xed carbon (FC, wt%) and volatile matter (VM, wt%). Also, the diluent effect of mineral matter/ash is considered in the development of the correlation. Table 3 presents the different forms of expressions along with the criteria for their selection. An exhaustive computer algorithm has been developed based on generalized method of least squares to evaluate the constants of these assumed algebraic expressions. It is worth to note that the algorithm is capable to incorporate any number of variables and data points. Using this algorithm and 450 data points the constants of these 10 algebraic expressions were evaluated. 2.4. Step 4: selection of the best correlation To facilitate the selection of correlation, the average absolute and bias errors for each of these 10 correlations as derived above were computed as follows  n   HHVc K HHV  1X  100% Average Absolute Error Z  n iZ1  HHV (1) Average Bias Error Z 1 n
n  X iZ1

3.74% and bias error of 0.12% with respect to measured values of HHV HHV Z 0:3536FC C0:1559VM K0:0078ASH MJ=kg (3)

From the data in Table 2, there is a good relation between the determined and the calculated HHV and the Eq. (3) has given acceptable approximations to the measured values.

3. Validation of the correlation To conrm the validity of this equation, a variety of various samples were examined. Table 2 shows the results obtained which demonstrate that the equation provides a simple and rapid way to obtain good estimates of heating values of lignocellulosics and other materials. Validation of the correlation developed under the present work has been carried out by comparison of computed and measured values of HHV over 550 data points which include 450 data points used for the derivation and 100 data points, specially used for the validation of the correlation. The measured and computed values of HHV data have been presented graphically in Figs. 17. Fig. 8 presents this comparison for a whole range of fuels. The error band of G5% has been also shown in these gures.

HHVc K HHV 100% HHV

4. Comparison with other equations from the literature The validation of present correlation has further been established by comparing its predictions with the predictions of few published correlations. However, it is to be noted here that the published correlations are valid only for the fuels for which they are derived. Hence, this comparison is purely for academic interest.

(2)

The correlation giving minimum error levels over 450 data points was selected as the best one. The following is the correlation, which turned out to be the best in this manner with an average absolute error of

492

J. Parikh et al. / Fuel 84 (2005) 487494

Fig. 1. Comparison between measured and predicted HHV (coal).

Fig. 3. Comparison between measured and predicted HHV (wood).

To compare the present equation with existing ones, four equations have been selected from the literature (Table 1). Figs. 17 show the heating values calculated from these equations for all the different group of samples used for the present work.

It is observed that the average absolute error of Kucukbayrak [2], Cordero [3], Demirbas [5] and Jimenez [6] for the entire spectrum of fuels considered herein are in the order of 17.91, 6.66, 13.81 and 10.18%, respectively,

Fig. 2. Comparison between measured and predicted HHV (shell/seed).

Fig. 4. Comparison between measured and predicted HHV (stalks).

J. Parikh et al. / Fuel 84 (2005) 487494

493

Fig. 7. Comparison between measured and predicted HHV (char). Fig. 5. Comparison between measured and predicted HHV (hull/husk/dust).

while the bias error for the same correlations are found to be 9.74, 6.18, K7.23 and 3.02%, respectively. The comparison was also made for different categories of fuels from which the above correlations were derived and it was observed that irrespective of the fact whether a fuel falls under the classication of coal, biomass, refuse or chars,

the values of average absolute and bias error with the present correlation are either minimum or comparable to minimum. It must, however, be emphasized that when taken over a whole range of fuels (Fig. 8), the proposed correlation gives the absolute and bias errors which are the lowest as compared to any published correlation. With these minimum values of average absolute and bias error as 3.74 and 0.12%, respectively, the validity and merits of the proposed correlation stands established. It is remarkable in this comparison to look at the chart where the new equation (present work) leads to signicantly lower deviations in spite of the fact that some of these data points were among those used by these authors to t their equation and the present work also used the same analytical and experimental heating values reported by these authors. The most signicant difference of practical interest arises from the fact that the derived equation is covering a wide range of carbonaceous materials. While the other reported

Fig. 6. Comparison between measured and predicted HHV (waste/industry waste/refuse/MSW).

Fig. 8. Comparison between measured and predicted HHV (whole range).

494

J. Parikh et al. / Fuel 84 (2005) 487494 [4] Fernandez P, Diaz RM, Xiberta J. Fuel 1997;76(10):9515. [5] Demirbas A. Calculation of higher heating values of biomass fuels. Fuel 1997;76(5):4314. [6] Jimenez L, Gonzalez F. Study of the physical and chemical properties of lignocellulosic residues with a view to the production of fuels. Fuel 1991;70:94750. [7] Raveendran K, Ganesh A. Heating value of biomass and biomass pyrolysis products. Fuel 1996;75(15):171520. [8] ASTM D 2013-86, Standard method of preparing coal samples for analysis, in gaseous fuels; coal and coke, section 5, vol. 05.05, Annual book of standards; 1989. p. 226. [9] ASTM D 3172-73(84), Standards method of proximate analysis of coal and coke, in gaseous fuels; coal and coke section 5, vol. 05.05, Annual book of ASTM standards; 1989. p. 299. [10] ASTM D 3173-87, Standards test method for moisture in the analysis sample of coal and coke, in gaseous fuels; coal and coke, section 5, vol. 05-05; 1989. p. 300. [11] ASTM D 3175-89, Standards test method for volatile matter in the analysis sample of coal and coke, in gaseous fuels; coal and coke, section 5, vol. 05-05; 1989. p. 305. [12] ASTM D 3174-89, Standards test method for ash in the analysis sample of coal and coke, in gaseous fuels; coal and coke, section 5, vol. 05-05; 1989. p. 302. [13] ASTM D 2015-85, Standards test method for gross caloric value of coal and coke by the adiabatic bomb calorimeter, in gaseous fuels; coal and coke, section 5, vol. 05-05; 1989. p. 251. [14] IGT. Chicago, Illinois, Preparation of coal conversion systems technical data book, Project 8964 nal report, Contact No. E (4918)-1730, Available From NTIS, Springeld, Virginia, USA: US Department of Commerce; 1976. [15] Bliek A. Mathematical modeling of a co-current xed bed coal gasier. PhD Thesis, Eindhoven, The Netherlands: Twente University; 1984. [16] Grover PD, Iyer PVR, Rao TR. Biomass-thermochemical characterization, 3rd ed. IIT-Delhi and MNES; 2002. [17] Rose JW, Cooper JR. Technical Data on Fuel, 7th ed. British National Committee of the World Energy Conference, Edinburgh; 1977. [18] Spiers HM. Technical data on fuel, 6th ed. New York: Wiley; 1962. [19] Rossi A. Fuel characteristics of wood and non-wood biomass fuels. In: Tillman DA, Jahn EC, editors. Progress in biomass conversion, vol. 5. New York: Academic Press; 1984. p. 69. [20] Jenkins BM, Ebeling JM. Correlation of physical and chemical properties of terrestrial biomass with conversion: symposium energy from biomass and waste IX IGT; 1985. p. 371. [21] Jenkins BM. Downdraft gasication characteristics of major California residue derived fuels. PhD Thesis, Davis: University of California; 1980. [22] Tillman DA. Wood as an energy resource. New York: Academic Press; 1978. [23] Risser PG. Agricultural and forestry residues. In: Soffer SS, Zaborsky OR, editors. Biomass conversion processes for energy and fuels. New York: Plenum Press; 1981. p. 2556. [24] Grover PD, Anuradha G. Thermochemical characterization of biomass for gasication, Report on physico-chemical parameters of biomass residues, IIT-Delhi; 1988. [25] Boley EC, Landers WS. Entrainment drying and carbonization of wood waste, Report of investigation 7282, Bureau of Mines, Washington; 1969.

equations as seen from the comparison in chart cannot be extended for such a wide range.

5. Utility of the proposed correlation Having established the validity and merits of the proposed unied correlation, few of its applications have been summarized below: (i) The correlation can be used for HHV computation of any solid fuel and biomass material, from its proximate analysis. (ii) In performance modeling exercise of combustion, gasication and pyrolysis process. (iii) The correlation provides the facility of using HHV as an algebraic expression in terms of fuel constituents, which in turn is useful in studying the inuence of proximate analysis of a fuel on process performance.

6. Conclusions The present correlation has been derived based on a large number of data points having widely varying proximate composition and encompassing all categories of solid carbonaceous materials. This means that, within the specied range of FC, VM and Ash as wt%, and within the stated absolute error limit of 3.74%, the present correlation may be accepted as general correlation for estimating HHV on dry basis. Also, it must be noted that the equation compares very well with the other reported equations in the literature. The main advantage of this correlation is, based on only proximate analysis data, it provides a rapid, easy and economical estimation of the HHV. This may be of particular interest in the contexts where more sophisticated and expensive equipments for experimental HHV measurement or elemental analysis are not always available.

References
[1] Channiwala SA, Parikh PP. A unied correlation for estimating HHV of solid, liquid and gaseous fuels. Fuel 2002;81:105163. [2] Kucukbayrak S, Durus B, Mericboyu AE, Kadioglu E. Estimation of caloric values of Turkish lignites. Fuel 1991;70:97981. [3] Cordero T, Marquez F, Rodriquez-Mirasol J, Rodriguez JJ. Predicting heating values of lignocellulosic and carbonaceous materials from proximate analysis. Fuel 2001;80:156771.