Vous êtes sur la page 1sur 30

FLOWSHEETING

&
MICROSOFT OFFICE
EXCEL

P. Ditl, J. Skrivnek

CONTENTS
INTRODUCTION
Decision-Making
Working Tools for Decision-Making
Complete System Characteristic

3
3
3
3

BASIC CONCEPTS
What Is Flowsheeting?
What Is a Balance?
What Is a Component?
What Is a Balanced System Boundary?
What Is a Limited Time of Balance?
What Is an Amount of Component?
What Is a Continuous Process?
What Is a Source Stream of Reaction?
Why Enthalpy Balance and not Heat Balance?
What is the Difference between Material Balance and Enthalpy Balance?
Enthalpy Balance
Formulation of the Problem Being Solved
How Many Equations Define the System Behaviour in Full?
Possibilities of Solving a Model

4
4
4
5
5
5
6
7
7
8
8
9
9
9
9

METHODS OF SETTING UP AND SOLVING A MODEL


Solving a System of Equations a Model
General Information
1. Process Basis and Data Available with Respect Thereto
2. Process Flow Sheet and Definition of Balance Blocks, Incidence Matrix
Flow Sheet
Incidence Matrix
3. List of Components in Streams and Their Definitions
4. Basis of Calculation and Conversion of Specified Flows and Compositions
Reference Stream Basis of Calculation
Conversion
5. Chemical Reactions, Determination of Number of Independent Chemical Reactions
Source Stream
Determination of Number of Independent Chemical Reactions
6. Specification Table and Matrix of Coefficients
Specification Table
Matrix of Coefficients
7. Formulation of Additional Relations
Types of Additional Relations
Enthalpy Balance
8. Conditions of System Solvability
9. Solving a System of Equations Using Methods of Linear Algebra, with the Advantage
of Using Excel
10. Table of Solutions Containing Complete System Characteristics

10
10
11
12
13
13
15
15
17
18
19
20
20
20
22
22
23
25
25
25
27
27

CONCLUSION

28

List of Projects Solved by the Authors


Enthalpy Data
Problems Solved
Problem 1: Solvay (Ammonia Soda) Process (problem_soda file)
Problem 2: Manufacture of Reduced Iron (problem_redukce file)
Problem 3: Manufacture of Styrene from Ethylbenzene (problem_styren file)
Problem 4: Catalytic Oxidation of SO2 to SO3 (problem_reactor file)

28

29
EXCEL file
EXCEL files

INTRODUCTION.
Decision-Making
Experience and practice have made us realise that mass and material balances represent effective
tools for decision-making. The concept of decision-making shall include the following spheres of
action:

Sizing lines and individual apparatus in designing.


Adjusting a line to the modified composition of materials.
Adjusting a line to modified requirements for quality (composition) of a product.
Diagnosing failures of operation of lines and locating causes of failure.
Checking or substituting measured quantities by calculating those quantities and taking
advantage of this principle to ensure easier control or to carry out process or guarantee
measurements.
Optimising operation of a line and searching for alternatives of operation using the method:
What will happen if.....?
Training and teaching operating personnel.

This enumeration is certainly not complete, however, it gives us an idea of the content of the concept
of decision-making.

Working Tools for Decision-Making


Decision-making that involves any one process requires a quantitative model. This decision-making
shall mean either proposing a new process, or modifying a process, or controlling an existing process.
Such a model should give us an opportunity to study behaviour of the system if the initial conditions
change. In other words, the model should enable one to study changes of the system behaviour after
modifying initial conditions, that is to search for an answer to the question: What will happen if. Such
a model constitutes a prerequisite for an objective quantitative economic analysis of the system. It is
necessary to bear in mind at all times that the quantitative model should be exact. That means we
receive one and only solution for the selected values of the designating parameters. When using
empirical coefficients, such as efficiency, yield, or adjustable constants guaranteed by the provider of
the equipment, maximum attention must be paid in order to define these coefficients accurately. The
use of empirical coefficients is thus limited to cases where those coefficients include deviations from
the ideal behaviour, express efficiency of the process, or adjust the theory to the real behaviour of the
equipment. These adjustable parameters often constitute the subject of know-how. In any case,
however, exact mass and enthalpy balances constitute the basis of the model. Thereby a quantitative
model of a process establishes a reliable foundation of the systems economic model, which is the
final tool for the management in making decisions.
The law of conservation of mass and the law of conservation of energy can be used as an exact
model for processes accompanied by a chemical reaction and heat exchange. On the basis of this
model, mass and enthalpy balances can be drawn up for the process in question and subsequently
complete system characteristic can be acquired.

Complete System Characteristic


Material balances are used for obtaining a complete characteristic for all streams that are present in
the process flow chart. This information is essential both for a designer and for the operation control
itself. A complete characteristic contains the following details:

Where the stream comes from and where it goes that defines the direction of the stream, i.e.
its orientation.
-1
-1
Size of the stream, expressed as weight per time unit (kg.s ) or moles per time unit (kmol.s ).
What components the stream contains and their part, expressed as a weight or mole fraction
of the component in the stream in question. The method of expressing the composition of the
stream should be invariant to temperature and pressure.
Temperature and pressure for each stream are added to the above-mentioned in the case of
enthalpy balance.

BASIC CONCEPTS
If we know the composition, temperature and pressure for a stream, we are able to determine:

any intensive physical property of the stream such as density, viscosity, or surface tension,
whether the stream consists of one or more phases and what these phases consist of.

In general we can say that all intensive physical properties represent a function of composition,
temperature and pressure. With liquid and solid phases the effect of pressure on the value of intensive
properties is mostly negligible. As to ideal gases, the change of enthalpy with pressure is null. A more
noticeable dependence of enthalpy on pressure is encountered only with streams the position of which
is in the proximity of a two-phase range. That leads to the following conclusions:

The composition of the stream must serve the purpose of mass balance.
The temperature of the stream must serve the purpose of enthalpy balance.
The pressure of the stream does not constitute the object of balance calculations. If we need
to know its value (for instance, in order to calculate in what phase the stream is whether in a
gaseous or liquid phase), its value must be determined.

What Is Flowsheeting?
Flowsheeting consists of a flow sheet of the process being analysed and the complete system
characteristic. It includes neither dimensions and structural design of the apparatus nor the planning of
the piping system. Flowsheeting consists in the following steps:

Drawing up a process flow sheet


Defining components in individual streams
Establishing balance equations

By working them out we obtain the missing information for complete system characteristic.
Flowsheeting is a working tool for decision-making.

What Is a Balance?
Any extensive property, that is a property, the numerical value of which changes with the size of the
system, can be balanced. This class includes mass, volume, internal energy, entropy, and so forth. On
the contrary, intensive properties are properties with their value being independent of the size of the
system, such as density, temperature, viscosity, specific heat, colour, hardness, and so forth.
Within the defined boundaries of the system being balanced and the limited time for which the
balancing is being performed, the following balance relation holds good for the extensive property
being balanced:
Input + Source = Output + Accumulation

Input is the quantity of the extensive property being balanced that passes into the system
through the balanced system boundaries within the limited time for which the balance is being
performed.
Source is the quantity of the extensive property being balanced that is generated within the
system boundaries in the course of the balanced period. Disappearance of the extensive
property is expressed as a source with a negative sign.
Output is the quantity of the extensive property being balanced that gets out of the system
through the balanced system boundaries within the limited time, for which the balance is being
performed.
Accumulation is the change in quantity of the extensive property being balanced within the
balanced system boundaries in the course of the balanced period.

The introduction of a source element enables us to balance all extensive properties, even those
properties to which the law of conservation does not apply (for instance, entropy, exergy, component
subject to a chemical reaction, and so forth). In order to apply a balance relation it is still necessary to
define in greater detail the concepts involved in the balance relation.

At the start of each balancing, three concepts must be defined clearly:

Extensive property being balanced


Balanced system boundaries
Limited time for which the balance is being performed.

What Is a Component?
An extensive property being balanced will be hereinafter referred to as a component. Components can
be

pure chemical compounds, such as methane, sulphuric acid, water, oxygen, to which a
specific molar mass can be attributed,
polymers chemically pure substances, whose molar mass is not clearly defined,
substances representing a group of chemical compounds, such as fats, R2O3, silicates,
hydrocarbons C4+,
substances that cannot be specified chemically, such as water-insoluble residue, substances
with density lower than water, substrate in bioreactors.
Also, a component may be defined as a particle with a size of 50 up to 100, or a particle
with other than white colour.

When defining a component it is necessary to define in a precise manner the method through which
the component is determined. Differences in methods used for determining a component can lead to
significant discrepancies. For instance, it is not enough to define a component as water content in a
solid phase, it being loose material. Water can be on the surface of particles; in the case of porous
particles it can be bound through adsorption to the inner surface of particles; it can be water of
crystallisation if the particles are of crystalline nature, and in the case of some substances it can be
chemically bound water. Based on the method of determining the water content, results can differ
even in the order of magnitude.
This applies also to a component defined as a particle size. If we determine the proportion of particles
of a specific dimension by measuring the particles largest dimension under the microscope, by
separating on sieves, or by measuring the distribution of the particles using the settling method, we
will arrive at different results.

What Is a Balanced System Boundary?


A boundary of the system being balanced must be defined in order that a component can enter or get
out of the system only by crossing the systems boundaries. System boundaries do not have to be
identical with the apparatus housing. A system can also be an imaginary volume defined within the
apparatus. This method is very often used to define differential volume, it being the system with dx, dy
and dz dimensions in rectangular coordinates. Then the balance relation for the component leads to a
partial differential equation.
The system being balanced does not have to be identical with the individual units of apparatus. Units
of apparatus can be combined for the purposes of balancing. Admittedly, this reduces the number of
input data, required for the performance of the balance calculation, but the possibility of acquiring
further information is thereby lost. For instance, if we link, for the purposes of balancing, an
evaporating apparatus and crystalliser, we lose the information about the properties of the
concentrated solution entering the crystalliser, which is an information necessary for the structural
design of the evaporating apparatus. An opposite approach is possible as well. For instance, instead
of balancing the whole distillation column individual levels can be balanced.

What Is a Limited Time of Balance?


If we divide all elements of the balance relation by the limited time of balance, we will obtain the
components quantity per unit of time.
This quantity can be termed as stream. Then the balance relation for any one component can be
posted as follows:
Input Stream + Source per Unit of Time = Output Stream + Rate of Accumulation

Source of component per unit of time (rate of source) can be seen as an imaginary (fictive) stream that
enters into the system; however, it does not cross its boundaries at the same time. It can have both
positive a component is formed and negative - a component disappears - value.
Accumulation per unit of time can be designated as the rate of accumulation; it is the change in
quantity of the balanced component in the system with time. Therefore, the final form of the balance
relation reads as follows:
Input Stream of Component + Fictive Stream of Component = Output Stream of Component +
Accumulation Rate of Component.

What Is an Amount of Component?


In order to substitute into the above-mentioned relation we need to know the stream of the component,
which is the quantity of the component per unit of time. However, we do not usually have the
opportunity to measure the component stream directly. In most cases we measure the overall stream
and its composition. The component stream can be calculated from these two pieces of information.
If the component is a differential particle size, the size of the stream of this component can be
calculated by multiplying the overall flow of particles by the relative frequency of occurrence of the
particles within the monitored interval. When reducing the interval of the particle size to zero we get a
differential and integral description known in literature as population balance.
The numerical value of the flow of the component is required to be invariant to temperature and
pressure. This requirement excludes the overall stream to be expressed as volume per unit of time.
There remains the possibility of expressing the flows of components as mass or amount (number of
pieces) kilograms of flour and number of eggs per unit of time.
Furthermore, it is required to have the opportunity to convert mass to amount (number of pieces) and
vice versa. For it is common practice to weigh small screws instead of counting our thousands of
pieces. The approach is accurate only if every piece has exactly the same mass as the other ones.
This requirement is satisfied strictly only with molecules. Molecules are too small to be counted out.
However, we know that one kmol of any substance contains the same number of molecules,
26
-1
corresponding to the Avogadros number, which is 6.0221367102 kmol . If we know the molar mass
-1
for a component (kgkmol ), we can convert the amount of substance (kmol) to mass (kg) and vice
versa. If a component cannot be attributed molar mass, its quantity can be expressed only as mass in
kg.
Consequently, we shall express the fraction of the component present in the stream only as a weight
-1
-1
fraction or a mole fraction, depending on whether the overall stream is expressed in kmols or kgs .
It is common practice in industry and business to encounter the definition of composition in volume per
cent. The following example will show us what ambiguousness this can lead to.
The content of alcohol in rum commonly available in the market is declared on the label as 38 volume
per cent. Lets assume this denotes a solution created by mixing 38 volume parts of ethanol with 62
volume parts of water. If we perform this operation at different temperatures, solutions with various
alcohol content are formed as the following table shows.
Sample Problem: Declared composition of rum

38

volume per cent of ethanol

Composition of rum with 38 volume per cent in weight fractions at various temperatures
temperature

kg/m3
ethanol

water

m3/kg
ethanol

0
10
20

water

806.0
798.0
789.0

999.8
999.6
998.2

0.001241
0.001253
0.001267

0.001000
0.001000
0.001002

33.07
32.85
32.64

66.93
67.15
67.36

38.00
38.00
38.00

30
40
50

781.0
772.0
763.0

995.6
992.2
988.0

0.001280
0.001295
0.001311

0.001004
0.001008
0.001012

32.47
32.29
32.13

67.53
67.71
67.87

38.00
38.00
38.00

60

754.0

983.2

0.001326

0.001017

31.97

68.03

38.00

weight per cent


ethanol
water

volume per cent


ethanol

70

745.0

977.7

0.001342

0.001023

31.87

68.17

38.00

water
0C
999.8

0.001342

0.001000

31.35

68.65

38.00

Minimum alcohol content


temperature

ethanol
70C
745.0

The view of revenue offices on this case is not clear. The relative deviation of the highest and lowest
value is more than 5%. However, this implies for us that from now on we will use either the amount in
-1
-1
kmols or in kg s for expressing the size of the stream in balancing. This is in agreement with the
-1
expression of composition by mole fraction of kmolkmol in the first case or weight fraction of kgkg
1
in the second case.
The dimensions of flows selected in that manner agree with the SI system. Energy flows come out in
-1
enthalpy balances as kW, that is kJ s .
It is necessary to point out that there are also other possibilities of expressing the composition to make
it invariant to temperature and pressure for instance, a relative mole fraction, molar concentration,
and the like. However, it is not difficult to convert these values to mole or weight fractions.

What Is a Continuous Process?


The accumulation of a component in the system is the change in quantity of the property being
balanced (hold-up). If a process is run continuously in a steady state, the change in hold-up and its
composition with time is null, and therefore the rate of accumulation is also null.
The accumulation of mass is usually applied in batch systems. If we choose a limited time of balance
exceeding the time of a batch process cycle by a long way, we can accept the assumption of a zero
rate of accumulation. If we choose no matter how long balance period, the system accumulation will
be given by the level of the system hold-up at the end of the balance period minus the system hold-up
at the start of the balance period. This can be the difference between the maximum and the minimum
hold-up at the most. If we choose the balance period to be just as long so that the maximum
accumulation possible is negligible compared with the input and output streams for the balance period,
we can regard also the periodical batch processes to be continuous and deem the rate of
accumulation to be zero.
Complete system characteristic does not say anything about the dimensions of individual pieces of
equipment, and therefore the relations we use for balancing do not contain any parameters of the
equipment. However, in order to calculate the rate of accumulation it is necessary to know the hold-up,
which constitutes a parameter of the equipment. Therefore, our concern will be focused on continuous
and periodical batch processes, where the accumulation rate can be ignored.

What Is a Source Stream of Reaction?


Just as in the case of accumulation, it is necessary to know the hold-up of the equipment in which the
chemical reaction takes place in order to calculate the source of a component directly. The rate of
generation of the component being balanced is the amount of component generated at the given place
in the unit of capacity of the system per unit of time. That means that an intensive property is involved.
Since the source in balances is an extensive property, we have to integrate the intensive property
through the capacity of the system being balanced. In the case of an ideal stirred tank, where
composition and temperature do not constitute functions of the place, we simply multiply the
mentioned intensive property by the capacity of the hold-up, which is, however, a parameter of the
equipment. In order to avoid this difficulty, we can substitute the source by a fictive stream of reaction.
For every independent reaction there will be one fictive stream, which will be denominated as nrk for
balances of amount of substance and as mrk for mass balances, where k is the serial number of
reaction in case that more independent reactions are taking place in the block being balanced.
Instead of posting mole and/or weight fractions into the fictive stream column in the specification table
for k, we

enter the components stoichiometric coefficient into the components row, with a positive sign
if the component is a product of the reaction (second term on the right side of the
stoichiometric equation), or with a negative sign if the component is a reactant (first term on
the left side of the stoichiometric equation) in the case of material balance;

multiply, in addition, the stoichiometric coefficient (with the appropriate sign) by the
components molar mass in the case of mass balance.

Sample Problem:
If we take the stoichiometric equation of formation of sodium hydrogen carbonate in manufacturing
soda
NaCl + NH3 +CO2 + H2O = NaHCO3 + NH4Cl
as an example, the column for the fictive stream will read as follows:
for molar balance:

for mass balance:

components nr

components

1 H2O

-1

1 H2O

-18.02

2 NaCl

-1

2 NaCl

-58.44

3 CO2

-1

3 CO2

-44.01

4 NH3

-1

4 NH3

-17.03

mr

5 NaHCO3

5 NaHCO3

84.01

6 NH4Cl

6 NH4Cl

53.49

Why Enthalpy Balance and Not Heat Balance?


The change in energy of 1 kg of water upon being heated by 0.5C corresponds to the change in the
-1
mass flow rate of the same by 65 ms or raising by 213 m. Such changes in kinetic and potential
energy can be hardly encountered in a chemical plant, the exception thereof being power-producing
machinery (turbines). For the purposes of balancing we can therefore ignore potential and kinetic
energy of streams and substitute heat balance by enthalpy balance.
This has the same advantage as disregarding the accumulation rate. In order to calculate kinetic and
potential energy of a stream at the given place we need to know the diameter of the piping and its
height above the ground, which are parameters of equipment that do not belong into the system
characteristic. We do not need this information to calculate the enthalpy of a stream, and therefore we
replace the overall heat balance by enthalpy balance. The mistake that can be thereby caused is
negligible in the absolute majority of cases.

What Is the Difference between Material Balance and Enthalpy Balance?


For a continuous process or a batch process occurring periodically, material balance of a component
has the following form:
Input Stream of Component + Fictive Stream of Component = Output Stream of Component +
Accumulation Rate of Component.
That applies to every component, which means that for a balanced system we have as many material
balances as there are components present in the system.
-1

-1

The dimension of individual parts of the balance equation is kgs , or kmols .


The system being balanced can interchange energy with the surroundings not only through the
material streams, but also through the exchange across the system boundaries heat exchangers, or
stirrers as the case may be, and so forth.
In that case, the enthalpy balance for a continuous process or a batch process occurring periodically
will have the following form:
Input Stream Enthalpy + Efficiency of Source = Output Stream Enthalpy + Interchange of Energy with
Surroundings

Individual parts of the balance equation are expressed in kW.


Removal of energy into the surroundings has a plus sign, while power supply has a negative sign. The
system in which the interchange of energy with the surroundings is zero, is called adiabatic.
Only one enthalpy balance holds good for the balanced block.

Enthalpy Balance
To draw up an enthalpy balance physical and chemical data of individual components must be
available. Various databases can serve as sources of data. Data from individual databases may not
be quite consistent and therefore it is practical to use only one database if possible and not to take
data for various components from different databases. A database adapted for Excel from the book by
R.C. Reid, J.M. Prausnitz, and T.K. Sherwood - The Properties of gases and liquids (Mc Graw Hill
1982) is used for the purposes of teaching at the Czech Technical University.
The database of the National Bureau of Standards or the DECHEMA database can be recommended
for industrial applications. The respective Internet addresses are provided at the end.

Formulation of the Problem Being Solved


In order to comply with the law of conservation of mass and enthalpy the following equations must
hold true for every block of the flow sheet:

material balance for every one component (in order words, the number of balance equations
agreeing with the number of different components present in the streams, which pass into or get
out of the given block we take into account both real and fictive streams);

summation equations for every one stream (in other words, the total of weight or mole fractions of
all components in the given stream must equal one);

other equations, if any, follow from the laws that determine the relations among the composition,
pressure, and temperature of individual streams. These are the laws of thermodynamics, chemical
and phase balance, or laws of kinetics of processes. These relations are almost always non-linear;

other equations issue from specifications given by technological regulations (such as yield,
selectivity) or in data provided by the producer of the equipment (such as efficiency of machines);

in the case of changes in temperature one enthalpy balance is added.

The first two groups form a system of linear equations.


Enthalpy balance is non-linear, which is given by the fact that molar heats represent functions of
temperature.
This set of equations forms a mathematical model, for which methods of solving must be established.

How Many Equations Define the System Behaviour in Full?


A system is defined in full if the sum of balance equations, summation equations, equations for
physical laws, specification equations, and enthalpy balance is equal to the number of unknowns in
the equations.

Possibilities of Solving a Model


The issue of simplicity of information transmission plays an important role with the model used. There
is specialised software available for these models, such as CHEMCAD, PRO, or MAX. In addition,
advantage can be taken of mathematical software, such as MATLAB, Mathematica, MATHCAD, and
so forth.
This software has two big disadvantages firstly, it is very expensive, and secondly, to become
proficient in the software requires great efforts and a lot of time. It is necessary to bear in mind that a
model set up by means of certain software will be used from a certain level of management upwards.
At this level it is possible to encounter the willingness to sacrifice the financial means to buy the

software, however, it is an absolute illusion to expect that someone would be willing to sacrifice the
time needed in order to become familiar with the software.
However, there is software available with the following advantages:

It is installed in every computer in an enterprise one can assume that Microsoft Office is
installed in every PC.
Almost everyone, who is able to use a computer, has a good command of it. And if the
manager is not willing to communicate with a computer, his secretary is bound to have a
good command of this programme.
Operating records and source data are stored by means of this programme. Transmitting
source data into our model is therefore possible without even one number being retyped
using a keyboard.
All and any economic and statistical analyses are performed by means of this programme.
Transferring data for these analyses is also performed through floppy disks, over the net and
through other channels, but not by typing the data on the keyboard.
Tables and diagrams obtained using this programme can be, without the slightest difficulty,
transferred into the most widespread Word processor.

We are talking about Excel. Working with Excel may not be as elegant as working with special
programmes, but the above-mentioned advantages are significant to such an extent that for the time
being Excel cannot be replaced with any other tool exception perhaps being Lotus II, but it is fully
compatible with Excel.

METHODS OF SETTING UP AND SOLVING A MODEL


The methods of solving have been drawn up to be generally valid and uniform for all types of balance
cases. Exactly the same procedure is used for all cases, ranging from the problem of what ratio to use
to mix two streams of water with different temperatures so that the resulting stream can have the
temperature required, to systems containing dozens of blocks and streams.
These uniform methods have facilitated the elaboration of an exact method of solving, which is
available to the engineering public at large and which consists of the following steps:
The following is required for any process we wish to balance:
1.
2.
3.
4.
5.
6.
7.
8.
9.

Process Basis and Data Available with Respect Thereto.


Process Flow Sheet and Definition of Balance Blocks, Incidence Matrix.
List of Components in Streams and Their Definitions.
Basis of Calculation and Conversion of Specified Flows and Compositions.
Chemical Reactions, Determination of Number of Independent Chemical Reactions.
Specification Table and Matrix of Coefficients.
Formulation of Additional Relations.
Conditions of System Solvability.
Solution of a System of Equations Using Methods of Linear Algebra, with the Advantage of
Using Excel.
10. Table of Solutions Containing Complete System Characteristic.
Comments:
Once the balance sheet has been drawn up and once it has been established which components are
present in individual streams, the rest of the balancing is, more or less, a matter of routine for there is
an exact procedure that leads step by step straight to the ultimate goal.
When strictly following the stated procedure the result may be incorrect only if the streams (both real
and fictive) and components have not been defined correctly. An operations engineer, who has a
perfect command of the process and knows the methods of measuring and analyses in detail, is the
most suitable person to set up the balance programme in Excel (or to take part in this task).
Putting together a list of components and defining these components is essential to satisfy the
summation equation. The requirements for perfect acquaintance with the process are focused in this
very issue.

10

Solving a System of Equations a Model


In general, tasks are not noticeably complicated from the mathematical point of view. Solving a system
of equations, most of which are linear, is always involved.
It is important to solve the essential question of whether the system has been overdetermined, that is
whether also data that should be the result of the calculation are available for the system
characteristic, or whether the system has been determined insufficiently, that means that not enough
information is available to enable us to determine the complete system characteristic.
The method of how to proceed from process description to solvable mathematical model has certain
logic, which must be adhered to when computing with a calculator as well as when using programmes
such as PRO, MAX, or CHEMCAD. The mentioned programmes obtain the required information from
the user using various menus. Formation of and recording in what is known as a specification table is,
in fact, making a list of all equations that are available to us to solve the problem.
The methods for EXCEL have exactly the same logical structure as the mentioned programmes, with
the difference consisting in that we know exactly why we take respective steps. Although this method
is more laborious, we know the logic and subject-matter of the procedure, and therefore we are able to
eliminate any contingent mistake or error in the specification right at the beginning.
A system can be solved using a calculator only in the cases where it is possible to apply the solution
through iterative method by means of substitution.

11

The following part demonstrates in detail on sample problems the application of the general methods
discussed above. Stress is put on the uniform procedure, which must be unconditionally observed for
success of the solution to be guaranteed.

General Information
In describing the methods we will proceed by providing the reader with a standard text and with a
solution in EXCEL of the problem being described. This file has been provided with quite a detailed
description and numerous commentaries, so that the standard text can serve, after the initial working
through, only as reference. In addition, series of problems solved in EXCEL have been attached so
that the user can review the methods on other problems as well. This file contains problems ranging
from the most simple to the larger-scale ones. Commentaries on these problems do not constitute part
of the standard text anymore; the commentary, with which the EXCEL file has been provided, should
be sufficient for the user.
Lets discuss the individual phases of the procedure in detail.

1. Process Basis and Data Available with Respect Thereto


It is demonstrated on the Solvay process. The following table contains a typical process description:

Manufacture of Soda Bicarbonate Column


Description of Technology
A saturated solution of NaCl contains

35.6600 g NaCl/ 100g H2O.

It reacts with the yield of


with a gas mixture consisting of

100.0000 %
32.0000 volume % NH3
and 68,0000 volume % CO2.

Solid sodium bicarbonate (NaHCO3) and mother liquor


containing, on the top of NH4CL, also
are formed in the reaction.
The plant processes

7.5300 g NaHCO3/100g H2O


100.0000 tons NaCl / day.

This description, which corresponds to a typical textbook example, is quite insufficient from the point of
view of industrial application.
However, lets pay attention to the most relevant characteristic of the process description, which are
the cells marked in colour. An essential rule applies to the methods:
The content of the cells can be changed (by entering a value using the keyboard) only in the fields
distinguished with colour.
Lets take notice of the cell giving the content of carbon dioxide in the entrance gas. If the entrance
gas contains only two components and if the content of ammonia has been entered, the content of
carbon dioxide cannot be entered since the percentage of its content must correspond to 100 the
percentage of ammonia.
It is very important to obey this rule. If it is disobeyed, the programme will not present a correct result
should any parameter be changed in a proper manner. It is very difficult to try to find, especially with
larger files, the cell into which a numerical value was entered by mistake instead of allowing the
operation with other cells. The greatest danger arises if there are more persons with access to the
programme. If another person different from the author uses the programme, the following precaution
must be taken:
To lock all cells apart from the ones marked with colour (i.e. into which data can be entered using
k eyboard).

12

In describing the technology it is necessary to have the opportunity to check the accuracy of such
description. All numerical entries must come from somewhere. The source must be obvious with every
numerical entry. It must be clear whether the figure in the cell has been taken from literature, operating
records, whether it is the required value, or an assumption or estimate.
Lets see to what extent the above-mentioned description meets these requirements:
It has been stated that the saturated solution contains 35.66 g NaCl / 100 g H2O. However, it has been
mentioned neither from which monograph this information was taken nor for which temperature this
information holds true.
It reacts with the yield of 100%. It has not been stated whether this is required or assumed. Only one
component can actually disappear from the reacting system. If NaCl disappears completely, it means
that the relief gases will contain a negligible amount of NH3. In this case it is necessary to distinguish
precisely whether those 100 % constitute an assumption or requirement.
The composition of the gas mixture the information that the entrance gas contains only carbon
dioxide and ammonia and that there is 38 volume per cent of ammonia must come from some
source Encyclopaedia by Kirk-Othmer, operating records from another soda processing plant,
assumption of a designer, estimate but in any case this source must be provided.
As to solubility of NaHCO3, it must be clear for what temperature this solubility holds true, whether the
influence of NH4Cl being present is considered, or whether a saturated or supersaturated solution is
taken into account.
The amount of processed salt can be limited by the sales of the salt, transmission capacity of traffic
roads, permissible quantity of salts in a watercourse, investment means available, and so forth.
This determination of the origin of the data is not an end in itself. For purposes of optimising it is
necessary to know which parameters can be changed and to what extent. It is possible to change the
proportion of ammonia in the entrance gas to quite a considerable extent, but it is not possible to
change the solubility more than as adequate to the permissible temperature range.

3. Process Flow Sheet and Definition of Balance Blocks, Incidence Matrix


Flow Sheet
A flow sheet is the next step following the process description. This phase gives a lot of freedom to the
user. The nodes being balanced can be drawn as rectangles or as schematic drawings of the pieces
of apparatus, which they represent, streams connecting individual nodes can have any thickness,
equipment with one input stream and one output stream, with which the amount and composition of
the stream does not change, can be left out from the flow sheet. The examples of such equipment
include, for instance, an intermediate storage tank or a pump. Individual nodes (units of apparatus)
can be interconnected for the purposes of balancing into greater units-blocks and these can be
subsequently marked in various ways. However, it is necessary to observe certain rules:
The nodes being balanced are designated using alphanumerical signs visibly different from the
numbers assigned to the streams.
Standard types of nodes:
1.

Blender - at least two streams entering

2.

Divider output streams are of the same composition and of the same phase at least two
streams exiting

13

3.

Separator - output streams are of various composition and possibly of different phase as well

4.

Reactor, bioreactor at least one entering and one exiting stream

Streams are indicated by numbers. In compliance with the practice established in designing the
following indications are recommended:
2
2

Streams must be provided with arrows so that it can be unequivocally determined where the stream
proceeds from and where it passes into orientation of the stream.
Streams cannot be connected or disconnected in any point. Such points must be drawn in the flow
sheet as the nodes being balanced (blenders or dividers).
In other respects, all and any tools can be used in order to draw flow sheets, be it a set of rectangles
connected by streams, represented by connected arrows perpendicular to one another, or more
elegant flow sheets taking advantages of the possibilities offered by AUTOCAD.
The following drawing can serve as an object example of a flow sheet. It was drawn up using EXCEL.
One can see the differentiation of the direction of streams and application of the Divider balance
nodes where the streams split and the Blender balance nodes where the streams join.

Example of a Flow Sheet:

K
R

5
6
D

2
2
8
9

With flow sheets that consist of more balance nodes, an incidence matrix must be set up.

14

Incidence Matrix
An incidence matrix provides information about what streams enter into the pieces of apparatus and
what streams exit from the pieces of apparatus.
The number of columns of an incidence matrix exceeds the number of nodes being balanced by one.
This one extra balance node is the surroundings. The surroundings is formally regarded as a node
being balanced. The number of rows agrees with the number of streams.
For every stream in its respective row we enter -1 into the cell in the column corresponding to the
node from which the stream exits and +1 in the column corresponding to the node into which the
stream passes. We post zeros into the remaining cells in the row.
If we mark the
Reactor column
Condenser column
Divider column
Blender column
Surroundings

with
with
with
with
with

R
K
D
S
O,

then the incidence matrix for the given flow sheet will read as follows:
stream
1
2
3
4
5
6
7
8
9
10
11

incmat
R
1
-1
-1
0
0
0
0
0
1
1
-1

K
0
0
1
-1
-1
0
0
0
0
0
0

D
0
0
0
0
1
-1
-1
0
0
0
0

S
0
0
0
0
0
0
1
1
-1
0
0

O
-1
1
0
1
0
1
0
-1
0
-1
1

The incmat i;k element of the incidence matrix is the value for stream i in block k.
We set up a matrix by first filling all cells with zeros. Then we proceed stream by stream i.e. row by
row and we mark in each row from which node the stream exits (we enter 1 into the respective cell)
and into which node the stream passes (we enter 1 into the respective cell).
Thereafter we perform a check by establishing in each column the number of streams passing into the
node the number of cells containing 1 and the number of streams exiting from the node the
number of cells containing -1. This check must be performed carefully because an error may have
fatal consequences.
The incidence matrix will facilitate our further work when setting up a matrix of coefficients.

3. List of Components in Streams and Their Definitions


In creating a list of streams and list of components we proceed in the following manner:
List of Streams:
Column 1 contains the numbers of the streams.
Column 2 contains the working title of the stream - brine, feed, distillate, total solids, and so forth.
Column 3 contains the list of components present in the stream.

15

List of Components:
Column 1 contains the components serial number.
Column 2 contains the components name.
Column 3 contains, in the case of a defined chemical compound, the components molar mass.
The total of masses of all components in a sample of the stream must be equal to the total mass of the
sample.
Abidance by these definitions is necessary for successful balancing.
Every component is defined using the method through which its amount in the sample of the stream
has been determined. This seemingly simple operation is the most common source of errors. In
comparison with time we will need to look for an error, checking every step three times will actually be
a great saving. Remember to take components one after the other and refer to the list of streams to
see through which stream the component passes into the system and through which stream it exits
the system. It isnt a waste of time.
Some components present in the system cannot be balanced. To decide which components will be
balanced and which wont is the most difficult task in balancing.
Molar masses are to be taken from a database. These values are recommended not to be retyped
from the database, but to be copied using a box. Thereby we protect ourselves against error, which
could be made by retyping. In retyping we transfer rounded values and thus deprive ourselves of part
of the information.
The following examples demonstrate what such a list may look like:
The first example corresponds to the bicarbonate column, the flow sheet of which is shown on p. 12.
List of Streams:
1 salt brine

NaCl, H2O

List of
Components:
1
H2O

molar mass
kg / kmol
18.02

2 relief gases

CO2, NH3

NaCl

58.44

3 entrance gas

NH3, CO2

CO2

44.01

4 suspension

NH4CL, NaCl, NaHCO3, H2O

NH3

17.03

5 product

NaHCO3

NaHCO3

84.01

6 mother liquor

NH4CL, NaCl, NaHCO3, H2O

NH4Cl

53.49

The second example is from the balance sheet of a desulphurisation plant. To facilitate
comprehension we provide a flow sheet as well. Notice that one component is defined as gasses from
which soluble salts are generated in the solution after absorption. That should serve as a proof of the
statement mentioned above that to define components is not a simple and primitive task. The
definition of the composition of streams and the definition of components is, indeed, the most difficult
1 flue gas in

2 flue gas out

3 oxidation air

4 process slurry

5 process water
6 limestone slurry
7 flying ashes
7 flying ashes
11 fictiv

A
B
S
O
R
B
E
R

8 gypsum
9 recycle

12 fictiv

16

F
I
L
T
E
R

10 bleed

part of the whole work as far as requirements for invention and knowledge of engineers are
concerned.
List of Streams:
absorber
components
1
flue gas in
O2, N2, SO2, CO2, H2O, sgg
1
O2
2
flue gas out
O2, N2, SO2, CO2, H2O
2
N2
3
oxidation air
O2, N2, H2O
3
SO2
4
process slurry
gypsum, CaCO3, H2O, solute, solids
4
CO2
5
process water
H2O
5
H2O
6
limestone slurry
CaCO3, H2O, solids
6
CaCO3
7
flying ashes
solids
7
gypsum
8
gypsum product
gypsum, CaCO3, H2O, solids
8
solids
9
recycle
H2O, solute
9
solute
10
bleed
H2O, solute
10
sgg
11
fictive
gypsum, CO2, CaCO3, H2O, SO2, O2
12
fictive solute
solute, CO2, H2O, sgg, CaCO3
The definition of some components follows
Definition of Components:
gypsum CaSO4.2H2O
solids
material not soluble in diluted sulphuric acid
solute
dissolved material
sgg
solute generating gases -sum of NOx and Cl2
In the case of the PRO 2, CHEMCAD, and MAX programmes as well as in the case of the methods for
EXCEL, the preparation of a flow sheet, list of streams and list of components makes the biggest
requirements on the skills of an engineer using these tools. Maximum attention must be paid to these
steps, and it is necessary to advise in strong terms against any haste and superficiality. Creativity and
knowledge must be shown in these very points.
In the following procedure we are already bound by strict rules and we have no chance to apply an
individual approach.

4. Basis of Calculation
Compositions

and

Conversion

of

Specified

Flows

and

Balancing mass (kg) is the fundamental method of balancing.


Molar balance (kmol) is practicable only if molar mass can be attributed to all components being
balanced.
Mass balances are expedient for using the same expression of quantity as economic and production
records. We are interested to know the tons of sulphuric acid manufactured and the price for one
kilogram of soda. The amount of kilo-moles processed and the price for one kilo-mole are figures that
have no informative value in the world of economy. Economy requires kilograms, not kilo-moles.
Molar balances are expedient in the case where streams are gases. The composition of gases is
given in volume per cent, which represents mole per cent at the same time. This is the product of the
law of Avogadro which says that the number of molecules in an identical volume of any gas is, at the
given temperature and pressure, the same. Measurements of gas flow rate are performed by
measuring the volume flow rate, with temperature and pressure being measured at the same time, so
3
the state-of-the-art flow meters indicate the flow in Nm (standard cubic meters, which represents the
volume at the temperature of 273.15 K and pressure of 101.325 kPa). On the assumption of ideal
behaviour of the gas, volume can be converted to kmol using the equation of state for ideal gas:
PV = nRT
and vice versa even if we do not know the exact composition of the gas. However, to calculate mass
(kg) we must know the exact composition of the gas.
All relations provi ded to us by physical chemistry for calculating intensive properties, phase and
chemical balances, as well as relations for kinetics of mass transfer and for kinetics of chemical

17

reactions are expressed by mole fractions. Therefore, it is more expedient to carry out all calculations
as molar balances and to convert just the final results to mass.
At the same time, it is necessary to bear in mind that amount of substance may be converted to mass
at any time while converting mass to amount of substance is possible only if molar mass can be
attributed to all components.
The relationship between a mole fraction and a weight fraction is non-linear, as shown in the graph for
a two-component mixture of ethanol water.
1,20
1,00
0,80
0,60
weight fraction
of ethanol

0,40
0,20
0,00
0

0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9

mole fraction of ethanol


As a non-binding rule we can recommend molar balances (kmol), if possible, with processes with
mostly gaseous streams and with chemical reactions in gaseous phase or, as the case may be, with
processes in which separation processes, such as absorption and distillation, take place.
When taking decisions it is necessary to bear in mind that the flow and composition of gaseous
3
streams is given in Nm and mole fractions, as distinct from kilograms and weight fractions with all the
other streams.

Reference Stream Basis of Calculation


Balancing is practicable only if the magnitude of at least one stream is known in the system. If that is
not the case, we have to select one stream as a reference stream and attribute a numerical value to it.
It is completely arbitrary which stream we select as a reference stream and what numerical value we
attribute to it. It is sometimes expedient to attribute the value of 1 to the magnitude of the stream and
to select such a stream to be the reference stream to make the resulting balance directly indicate the
quantity of product made from one kilogram of material or how many kilograms of the respective
material are required to produce one kilogram of product. If the magnitude of at least one stream has
been specified, a reference stream does not need to be selected. The selection of a reference stream
will be illustrated in the sample problems.
The decision to balance amount of substance, or mass, also indicates the standard onto which the
data specified in the process description must be converted.
First of all, it is necessary to decide on how we shall denote the mass flow rate and the amount of
substance.

Notation:

Amount of substance is usually designated in the same way as in the equation of state for
ideal gas PV=nRT, i.e. as n. Then the flow should be a dotted n, but EXCEL cannot do that,
-1
and therefore we shall indicate the flow in kmols as n as well.
Mass is usually designated as m. Since the same rules apply here as with the flow of amount
of substance, we have no other option than to designate the mass flow rate as m.
The letter x has been introduced in physical chemistry to denote a mole fraction. We shall
adhere to this established practice.
In order to denote a weight fraction we shall choose a letter from the end of the alphabet as
well. However, y could be confused with the symbol introduced for a mole fraction in vapour
phase and z clashes with the symbol for compressibility factor. These two properties will be

18

present in our calculations, and therefore it is necessary to look for another sign to denote the
weight fraction. The nearest letter from the end of the alphabet is w.
Designations of fundamental properties:
*1
Flow of amount of substance
n
recommended dimension
kmols
Mole fraction
x
-1
Mass flow rate
m
recommended dimension
kgs
Weight fraction
w
This notation is not binding; any other notation of fundamental properties can be used as long
as it is defined properly.

However, the indexing is binding.

Every flow designation has an index that is identical with the number of the respective stream.
Every designation of composition has two indexes:
The first index designates the stream
The second index designates the component.

Thus the mole fraction of the component 2 in the stream 3 is x3;2 x3;2 can be used as notation in
order to increase the speed of posting in EXCEL. Similarly, one can post n3 instead of n3.

Conversion
3

-1

-1

To convert flows from Nm s to kmols one needs to know:


Normal temperature
Normal pressure
Gas constant R

273.15 K
101.325 kPa
-1 -1
8.31451 kJkmol K

Then we can substitute values in the equation of state for ideal gas:
PV = nRT.
If the flow (production) is given in tons per year, this figure is not converted to the total number of
hours, but to the annual working capacity, which is 8,000 working hours per year as for continuous
chemical operations. This is the value recommended in case a more accurate figure cannot be
obtained.
In converting the composition in mole fractions to weight fractions we proceed as follows:
1.
2.
3.
4.
5.
6.

The components present in the stream are entered into the first column.
The components quantity mole fractions, mole per cent, any numbers designating the ratios
in the numbers of moles of the components - is entered into the second column.
Molar mass of the components is entered into the third column.
Scalar product of the second and third column is entered into the fourth column.
The fifth column is created by dividing the individual parts of the fourth column by the sum of
the fourth column. We get the weight fractions of individual components as a result.

This procedure can be tested on the following table:


NH3
CO2

[kmol] [kg/kmol]
32
17.03
68
44.01
[kmol altogether]
100

[kg]
[kg/kg altogether]
544.96
0.154046
w3;4
2992.68 0.845954
w3;3
[kg altogether]
3537.64

In converting weight fractions to mole fractions we follow the same procedure, only the fourth column
is created by dividing the second column by the third column.
We will demonstrate the procedure in calculating the composition of saturated solution of salt in mole
fractions. The solubility of salt is specified as 35.66 g NaCl / 100g H2O.

NaCl
H2O

[kg]
36,66
100
[kg ]
136,66

kmol
[kmol/kmol ]
0,6273
0,1016
5,5494
0,8984
[kmol ]
19 6,1767

kg/kmol
58,44
18,02

Important warning If properties taken from the cells marked in colour are present in the calculation
formulas for converting (for example, molar mass), we have to introduce the value of this property into
the formula by reference to the respective cell and not by retyping the value of such a property.

5. Chemical Reactions, Determination of Number of Independent Chemical


Reactions
Source Stream
The next phase is to determine which nodes function as chemical reactors, and to resolve how many
independent reactions take place therein. For the bicarbonate column, which is part of the Solvay
process, the flow sheet and description of the reaction will be as follows:
Flow Sheet Bicarbonate Column

2
1

3
4
5
filt

Reaction:
NaCl + H2O + CO2 + NH3

NaHCO3 + NH4Cl

For every independent reaction we have one stream, which is given by the product of the reaction rate
and the reaction volume, and we shall call it the source stream. The formation or disappearance of
-1
-1
individual components in the reactor through the reaction in kgs or in kmol s is given by the product
of the source stream and the respective stoichiometric coefficient (plus signs for products of reaction,
negative signs for reactants). In the event of more independent reactions the calculated properties
must be added up.

Determination of Number of Individual Chemical Reactions


The number of source streams present in the balance agrees with the number of independent
reactions taking place in the system. However, there is one problem, namely the word independent.
With more complicated reaction we are able to estimate, based on the fundamental knowledge of
chemistry, which chemical reactions could take place in a system. Almost all real processes do not
take place as a single reaction, but as a complex of reversible, parallel and consecutive elementary
reactions. However, we still need to establish which reactions are independent. This task is identical
with the task of determining the rank of a matrix of coefficients. Lets take the manufacture of nitric

20

acid by absorption of gases of nitrogen into water as an example. The following reactions may take
place in the system.
2 NO + O2 = 2 NO2
NO + NO2 + H2O = 2 HNO2
3 HNO2 = HNO3 + 2 NO + H2O
3 NO2 + H2O = 2 HNO3 + NO
2 NO2 + H2O = HNO2 + HNO3
2 HNO2 + O2 = 2 HNO3
HNO2 + NO2 = HNO3 + NO
We can set up a matrix of coefficients from the stoichiometric coefficients, which are used for
multiplying the formulas of compounds in the individual stoichiometric equations:
NO
NO2
O2
H2O
HNO2
HNO3
-2
2
-1
0
0
0
-1
-1
0
-1
2
0
2
0
0
1
-3
1
1
-3
0
-1
0
2
0
-2
0
-1
1
1
0
0
-1
0
-2
2
1
-1
0
0
-1
1
We modify this matrix using the method of Gaussian elimination to make the cells below the diagonal
show zero. The Gaussian method of elimination is described for instance by J. Holenda in Linear
Algebra (Holenda J. Linern algebra. Prague: Czech Technical University 1978).
With matrixes one is allowed to replace any row (column) with a linear combination of the initial row
(column) with any other rows (columns) of the matrix. Therefore, if we want to generate zeros below
the diagonal in the first column, we proceed as follows:

row 1 this remains the same,


row 2 we multiply the second row by 2 and add it up with the first row,
row 3 we add the third row up with the first row,
row 4 we multiply the fourth row by 2 and add it up with the first row,
row 5 and row 6 remain unchanged,
row 7 we add the row 7 up with the row 2.

We get the following matrix


generation of zeros below the diagonal in the first column
-2
0
0
0
0
0
0

2
4
2
-4
-2
0
-2

-1
-1
-1
0
0
-1
0

0
2
1
-2
-1
0
-1

0
-4
-3
2
1
-2
1

0
0
1
2
1
2
1

We proceed in an analogous manner in generating zeros below the diagonal in the second column
and in the third column:
generation of zeros below the diagonal in the second column
-2
2
-1
0
0
0
0
4
-1
2
-4
0
0
0
1
0
2
-2
0
0
-1
0
-2
2
0
0
-1
0
-2
2
0
0
-1
0
-2
2
0
0
-1
0
-2
2

21

generation of zeros below the diagonal in the third column


-2
2
-1
0
0
0
0
4
-1
2
-4
0
0
0
1
0
2
-2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
The elimination for the remaining columns does not take place since they already show zeros below
the diagonal. Our matrix has the rank 3, and consequently only three of the seven mentioned
stoichiometric equations are independent.
We can take the liberty to refrain from determining the rank of a matrix only if we are absolutely sure.
An error made by incorrect determination of the number of independent equations is very difficult to
find.

6. Specification Table and Matrix of Coefficients


Specification Table

A specification table is a method of entering input data to make it capable of being used in
setting up a matrix of coefficients of a linear system of balance equations.
If we have converted all available data to the chosen basis of calculation, we arrange the
available values of the system characteristic into a specification table.
The specification table will have as many columns as there are streams present in the system,
plus one column for every independent reaction that takes place in the system.
The first row is reserved for magnitudes of streams. On the top of that, the specification table
will have as many rows for each block as there are components present in the block.
Individual fields in the columns of streams contain a weight or mole fraction of the component
corresponding to the respective row.
If we dont know the value that corresponds to the respective field, we will enter an
alphanumeric sign indicating this property. If we dont know the mass flow rate of the stream 5,
we will enter m5 into the respective field. If we dont know the mole fraction of the component
3 in the stream 4, we will write x4;3 into the respective field.
If we know the value required, we shall introduce it into the respective field by reference to the
cell in the Conversion paragraph never by retyping!
If we know that there is only one component in the stream, we will enter 1 into the respective
field, if we know that there is no component present in the stream, we will enter 0 into the
respective field.
We post stoichiometric coefficients of components of the corresponding reaction into the
columns for chemical reactions. Stoichiometric coefficients for reactants (components entering
into the reactions, or components on the left-hand side of the stoichiometric equation) are
attributed a negative sign. Stoichiometric coefficients of components, constituting products of
reactions (components on the right-hand side of the stoichiometric equation) will have a
positive sign.
That holds good in case we balance amount of substance. If we balance mass, we have to
multiply the individual stoichiometric coefficients by molar masses of the respective
components.
The number of unknowns agrees with the number of different alphanumeric signs in individual
cells. Beware of duplications!
For every block being balanced the number of balance equations available to us agrees with
the number of components present in the streams passing into or out of the blocks.
Attention: the overall balance of a node or a block is given by the sum of the component
balances, and therefore it is a dependent equation.
Moreover, the sum of weight (or mole) fractions in every stream must equal one.

For illustration see the sample specification table provided for the case of bicarbonate column.

22

Specification Table
stream
1
-1

block

kgs
component

reaction

4,315

1,001157 m3

m4

m5

m6

mr

0.732
0.268

0
0.00

0
0

w4;1
0.02

0
0

0
0

-18.02
-58.44

3 CO2

w2;3

0.846

-44.01

4 NH3

w2;4

0.154

-17.03

5 NaHCO3

w4;5

84.01

6 NH4Cl

w4;6

53.49

1 H2O
2 NaCl

0
0

0
0

0
0

w4;1
0,02

0
0

w6;1
w6;2

0
0

5 NaHCO3

w4;5

w6;5

6 NH4Cl

0.00

w4;6

w6;6

column 1 H2O
2 NaCl

filter

Matrix of Coefficients
A specification table can be transformed into a matrix of coefficients and a right-hand-side vector of
the system of linear algebraic equations in the following manner:
We open a new clean sheet and transfer the reference to the top left corner of the
specification table into a suitable field (left margin). Only that one cell, not the whole matrix!
We place the cursor onto the bottom right edge of the cell (+ will appear) and drag it onto the
whole column and then the whole matrix. Thereafter we copy the whole specification table.
Having been copied and slightly adjusted, the specification table on the Sheet 2 looks as follows:
1
stream kgs 4.315

2
1.0012

3
m3

4
m4

5
m5

6
m6

reaction
mr

1 H2O
2 NaCl

0.732
0.268

0
0

0
0

w4;1
0.02

0
0

0
0

-18.02
-58.44

3 CO2

w2;3

0.846

-44.01

4 NH3

w2;4

0.154

-17.03

5 NaHCO3

w4;5

84.01

6 NH4Cl

w4;6

53.49

1 H2O
2 NaCl

0
0

0
0

0
0

w4;1
0,02

0
0

w6;1
w6;2

0
0

5 NaHCO3

w4;5

w6;5

6 NH4Cl

w4;6

w6;6

-1

First of all, we create a right-hand-side vector. We place it at the end as the last column,
designate it, for instance, as vps and post zeros into all cells of this column.
Then we transfer the numerical values from the row of streams into the right-hand-side vector.
We proceed in the following manner:
We create a new end row and post zeros into all fields.
We post 1 in the last row of the column wherefrom we want to transfer the numerical value of
the stream.
We drag and drop the numerical value (using an arrow) in the last row of the column of the
right-hand-side vector.
We enter the alphanumeric designation of the stream for the respective column into the
cleared cell in the row of streams.
We repeat this procedure for every stream for which we know the numerical value. In our case
we start with m1 and go over to m2.

23

The alphanumeric designation for concentration of components in streams is removed from the table
in the following manner:

Behind the column with the alphanumerical designation of the composition of the stream we
insert as many new columns as there are unknown concentrations present in the stream.
We fill the new columns with zeros.
We enter into the row of streams that the values in the column indicate the product of flow and
composition. Therefore, we will have five columns for stream 6 m6, mw6;1, mw6;2, mw6;5,
and mw6;6.
We replace the alphanumeric designation of the content of the component in the stream by
zero and enter 1 into the respective column in the same row.

After these steps have been taken, the matrix reads as follows:
m1

m2

mw2;3 mw2;4 m3

m4

mw4;1 mw4;5 mw4;6 m5

m6

mw6;1 mw6;2 mw6;5 mw6;6 mr

vps

0.73

-1

-18

0.,27

-0.02

-58.4 0

-1

0.85

-44

-1

0.15

-17

-1

84

-1

53.5

-1

0.02

-1

-1

-1

-1

4.31

Thereby we have created a matrix of coefficients for balance equations. Apart from balances of
components in individual nodes it must hold true that the sum of weight or mole fractions in every
stream equals 1.
This relation will be referred to as the summation relation and we will incorporate it into the matrix of
coefficients using the following procedure:

We create a new end row and insert zeros into all fields.
We choose one stream in which there are some unknown concentrations. We select the
blocks from which this stream exits or into which this stream enters. We enter the value of 1
the sum of absolute values of the content of the cells in the selected block - into the last row in
the column of this stream. If a known value is present in the stream for instance in stream 4,
a value a will appear in the end row, for which the following holds true:

0< a 1

If all concentrations in the stream are unknown, then a = 1.


We enter 1 into the end row in the columns for the flow-concentration product.
We repeat this procedure for all streams, in which there were some unknown concentrations.

Thereafter the table of coefficients reads as follows:


m1

m2

mw2;3 mw2;4 m3

m4

mw4;1 mw4;5 mw4;6 m5

m6

mw6;1 mw6;2 mw6;5 mw6;6 mr

vps

0.73

-1

-18

0.27

-0.02

-58.4 0

-1

0.85

-44

-1

0.15

-17

-1

84

-1

53.5

-1

0.02

-1

-1

-1

-1

24

4.31

-1

-0.98

-1

7. Formulation of Additional Relations


Types of Additional Relations
In addition to balance relations and summation equations there are usually also other relations that
must be satisfied. They are called additional relations. In most cases they are defined in the
description of technology. They can include physical and chemical data (phase and chemical
balances), or adjustable parameters of the process and equipment specified by manufacturers and
technology support (efficiencies, specific consumption, conversions of chemical reactions). These
relations can be both linear and non-linear (most relations of equilibrium). Heat balance with nonisothermal processes can also be included in additional relations. Enthalpy balance can serve as an
additional relation as long as it does not contain new unknowns. In that case we actually deal with
enthalpy balance and molar balance at the same time. In most cases, molar and enthalpy balances
can be solved independent of each other, with enthalpy balance being based on molar balance.
In our case, the additional relation is the concentration ratio of sodium hydrogen carbonate and water
in mother liquor. The description of technology suggests that this relation will have the following form:
mw6;5 = 0,0753 mw6;1
We create another (this time final) end row for this additional relation and it shall read as follows:
m1

m2

mw2;3 mw2;4 m3

m4

mw4;1 mw4;5 mw4;6 m5

m6

mw6;1 mw6;2 mw6;5 mw6;6 mr

vps

0.075

-1

Full comprehension of the recommended procedure will best be ensured by the complete Excel file
with commentaries clarifying the particular steps in greater detail. This file is entitled problem_soda.
The linear additional relation as the one stated above is rather an exception than a rule. Additional
relations arising from phase or chemical balances or from enthalpy balances are non-linear. A problem
of a non-linear additional relation is dealt with in the problem_redukce file. The Excel file has been
provided with abundant commentaries that should be sufficient in order to make one understand the
applied procedure. One non-linear relation can be managed using the Solver function. If there are
more non-linear additional relations, the problem must be dealt with numerically using the NewtonRaphson method. Excel suggests an elegant method of obtaining the value of Jacobian, which is
essential for the application of this method. The problem is solved in the problem_reactor file. The
commentaries provided in this file should be sufficient to make one understand the procedure.

Enthalpy Balance
The number of balance equations of components for each block agrees with the number of
components present in the given block.
We have one enthalpy balance for each block.
For the amount of matter and block l it will read as follows:

i=1

j=1

k=1

incmati;l n i x i; j h i; j + nrk hrk = Q


25

Index i indicates streams and index j indicates components.


S is the total number of streams; C is the total number of components.
hi;j is the molar enthalpy of component j at the temperature and pressure of stream i.
nrk is the source stream of the k-th reaction and hrk is the reaction enthalpy of this reaction.
N is the total number of independent reactions.
Q is the energy exchanged with the surroundings.
All parts are expressed in kW.

Enthalpy balance for mass balance will have the following form:
S

h i; j

i=1

j=1

Mj

incmati;l mi w i; j

+ nrk hrk = Q
k =1

Mj is the molar mass of component j.

Consequently, molar enthalpy of a component at the given temperature and pressure is the
fundamental property for enthalpy balance. It is a relative property; its value is related to a specific
state for which we choose the enthalpy of a component to be zero. These conditions shall be referred
to as standard state conditions.
In our calculations this component will behave like ideal gas at the temperature of 298.15K. Pressure
does not need to be defined because with ideal gas enthalpy does not constitute a function of
pressure.
A component can be transferred from the standard state conditions into the final state using the
following steps:

By heating component j from the standard state temperature up onto the final temperature Ti
(K). The change in enthalpy is given by the following relation:

h j = CpA j Ti +

CpB j

Ti 2 +

CpC j

Ti 3 +

CpD j
4

Ti 4 H 0 j

where

H 0 j = CpA j 298,15 +

CpB j
2

298,15 2 +

CpC j
3

298,15 3 +

CpD j
4

298,15 4

The nomenclature is in compliance with the EnthalpyData database.

By condensing the component. The database provides evaporation heat Hvap (condensation
heat with the opposite sign) for boiling temperature. It can be converted onto any other
temperature using the following equation:

h vap = Hvap

1 T C

TB
1

TC

0 , 374

This calculation is approximate. For a more accurate calculation access must be found to a
database containing data for Cp in a liquid phase.

In order to include heat of admixture, heat of dilution, melting heat and heat of crystallization,
one must refer to specialised databases.

Heat of reaction hr can be calculated from standard heats of formation Hform and stoichiometric
coefficients ? k;j that have a negative sign for reactants and a positive sign for products of reaction. For
one reaction:
C

hr = j Hform j
j =1

26

Our EnthalpyData database defines Hform by stating that a component in a gaseous state is formed
from elements.
We provide, for information, the internet address of the DECHEMA database
and the internet address of the National Bureau of Standards database

www.ddbts.de

webbook.nist.gov

An illustration of the application of the methods of calculating enthalpy balances in EXCEL can be
found in the problem_styrene file. In this case, the problem is solved on the basis of both molar
balance and mass balance.

8. Conditions of System Solvability


Using Excel we are able to solve the following problems:
1.
2.
3.
4.
5.

We have more equations than unknowns, and the system is linear.


We have as many equations as unknowns, and the system is linear.
We have as many equations as unknowns, and one equation is non-linear.
We have as many equations as unknowns, and more equations are non-linear.
We have one equation less than unknowns, and the system is linear.

If our problem does not fit into any one of these categories, the problem must be redefined, i.e. other
additional relations must be found, some blocks in the block diagram must be consolidated, some of
the components must be disregarded, and so forth.

9. Solving a System of Equations using Methods of Linear Algebra, with the


Advantage of Using Excel
Problem 1
When solving Problem 1 we proceed as follows:

We multiply the transposed matrix of coefficients by the matrix of coefficients to get a square
matrix of the most expectation values of the matrix coefficients in the sense of the method of
minimum squares. The formula has the following form: PRODUCT OF MATRIXES
(TRANSPOSITION(A1:D18);A1:D18), where A1 is the cell in the top left corner of the matrix
of coefficients and D18 is the cell in the bottom left corner.
We multiply the transposed matrix of coefficients by the right-hand-side vector to get a
corrected vector with the same dimension as the square matrix.
We multiply the inversion of the square matrix by the corrected vector to get the vector of
solutions.

Problem 2
We solve this problem by multiplying the inversion of the square matrix of coefficients by the righthand-side vector.
The formula has the following form: PRODUCT OF MATRIXES(INVERSION(A1:H8);I1:I8), where A1
is the cell in the top left corner of the matrix of coefficients and H8 is the cell in the bottom right corner.
Column I1:I8 would correspond to the right-hand-side vector.
Problem 3.
In solving Problem 3 we proceed as follows:

We attribute a numerical value to one of the unknowns.


As a consequence of the incorrectness of the value fixed in this manner the non-linear
additional relation is not satisfied.
We keep modifying the fixed value until the non-linear relation is satisfied with the accuracy
required.
This procedure can be performed manually, however, using the Solver function is more
efficient. The procedure is described in detail in the problem_redukce file.

Problem 4

27

This Problem is solved using the Newton- Raphson method.


The procedure can be clarified by studying the problem_reactor file. The sheet entitled Directions
provides a detailed method of applying this method in the environment of EXCEL. The description of
the method can be found for instance in the book by M. Kubcek Numerical Algorithms of Solutions to
Chemical Engineering Problems (Kubcek, M.: Numerick algoritmy reen chemicko-inenrskch
loh. Prague: SNTL 1983).
Problem 5
In this case we cannot obtain a solution, but functionality. We choose one of the unknowns to be an
independent variable, and then we can attribute some numerical value to it. Now the system is
solvable. We can study how other variables (unknowns) change subject to the value of the
independent variable selected, whereby we can obtain valuable information.

10. Table of Solutions Containing Complete System Characteristic.


A table of solutions provides complete characteristic for each stream; that is the flow of components,
overall flow, temperature and pressure. The columns represent individual streams and the rows are
used for entering the mentioned parameters (see, for instance, the problem_soda file).

CONCLUSION
The proposed methods of balancing using EXCEL enable one:

To create balance models of lines, nodes and equipment, with the aim to answer the question:
What will happen if .
To serve as a guideline for interventions in the operation of the line, especially if parameters
change unexpectedly.
To acquire balance sheet values to elaborate a basic design.
To determine the minimum number of measured quantities needed for the purposes of
determining the complete system characteristic.
To evaluate and check industrial experiments and operating data.
To balance raw materials, products, and wastes for the purposes of economy and ecology of
an enterprise.
To make the training of the operating personnel easier.

28

Address for consultation, training and advisory services: see the main Web page.
List of Selected Projects Solved by the Authors:
Project 1: Desulphurisation Unit of the Mitsubishi Company
Project 2: Co-generation Plant of the koda Company
Project 3: Catalytic Reactor - Draslovka Koln
Project 4: Co-generation Plant with the Utilisation of a Heat Pump - Brun &Sorensen - Denmark
Project 5: Manufacture of Cole-seed Oil - Sindat Engineering
Project 6: Refining of Crude Spirit - Alcohol Distillery
Project 7: Low-pressure Manufacture of Nitric Acid - Sindat Engineering
Project 8: Manufacture of Pure Salts OSRAM Bruntl
Project 9: Enthalpy Balance of a Bakery Factory - Pragosoja Strncice
Project 10: Enthalpy Balance of a Boiler for Steam Production - Detmarovice Power Plant - Mitsubishi
Co.
Problems Solved:
Problem 1: Solvay (Ammonia Soda) Process (problem_soda file)
Problem 2: Manufacture of Reduced Iron (problem_redukce file)
Problem 3: Manufacture of Styrene from Ethylbenzene (problem_styren file)
Problem 4: Catalytic Oxidation of SO2 to SO3 (problem_reaktor file)

29

Declared composition of rum

38

volume per cent of ethanol

Composition of rum with 38 volume per cent in weight fractions at various temperatures
kg/m3

m3/kg

weight per cent

volume per cent

temperature
0
10

ethanol
806.0
798.0

water
999.8
999.6

ethanol
0.001241
0.001253

water
0.001000
0.001000

ethanol
33.07
32.85

water
66.93
67.15

ethanol
38.00
38.00

20
30
40

789.0
781.0
772.0

998.2
995.6
992.2

0.001267
0.001280
0.001295

0.001002
0.001004
0.001008

32.64
32.47
32.29

67.36
67.53
67.71

38.00
38.00
38.00

50
60
70

763.0
754.0
745.0

988.0
983.2
977.7

0.001311
0.001326
0.001342

0.001012
0.001017
0.001023

32.13
31.97
31.87

67.87
68.03
68.17

38.00
38.00
38.00

0.001342

0.001000

31.35

68.65

38.00

Minimum alcohol content


ethanol
water
temperature

70C
745.0

0C
999.8

30