Chemical Engineering Department

Introduction

COURSE CODE: KMÜ 241/22

COURSE TITLE: MATERIAL AND ENERGY BALANCES I

COURSE TEXTBOOK: Felder, R. M. and R. W. Rousseau, "Elementary Principles of

Chemical Processes", 3rd Edition, John Wiley&Sons, Inc., New York, 2000.
Instructor (s) Prof. Dr. Tülay Durusoy
The course aims to teach the students how to draw process flowcharts for single and
multiple-unit processes, how to perform degree-of-freedom analysis and derive and
Course objective solve material balances on various nonreactive and reactive systems, and how to
report engineering calculations and problem solutions in a professional manner.

Material balances for batch and continuous processes.

Course Content Industrial applications of material balances.


 1 Apply knowledge of basic science and engineering fundamentals to solve
material balances
2 Draw and label the flowcharts of processes
3 Carry-out degree-of-freedom analyses of reactive and nonreactive processes
4 Analyze, write and solve material balance equations for single unit and multiple
Learning outcomes unit processes with and without chemical reaction.
5 Identify the physical and/or chemical data required in the calculation of
material balances using tables and diagrams
6 Gain ability to scale the flowcharts of the processes
7 Develop an appreciation for the importance of environmental safety in the
process industries.

Course outline weekly
Weeks Topics
Week 1 Introduction to chemical engineering calculations. Units and dimensions. Conversion of units. Systems of units.
Force and weight. Numerical calculation and estimation.
Week 2 Validating results. Sample mean. Sample variance. Sample standard deviation. Dimensional homogeneity and
dimensionless quantities. Process data representation and analysis.
Week 3 Processes and process variables. Mass and volume. Flow rate. Chemical composition.
Week 4 Pressure. Temperature. Fundamentals of material balance calculations. Process classification. The general
balance equation. Material balances on continuous steady-state processes. Integral material balances on batch,
semibatch and continuous processes.

Week 5 Material balance calculations. Flowcharts. Flowchart scaling. Basis of calculation. Quiz 1.
Week 6 Degree-of-freedom analysis. General procedure for single-unit process material balance calculations. Material
balances on multiple-unit processes. Homework 1.
Week 7 Recycle and bypass.
Week 8 Midterm.
Week 9 Balances on reactive systems. Stoichiometry. Limiting and excess reactants.
Week 10 Fractional conversion, and extent of reaction. Chemical equilibrium. Multiple reactions, yield, and selectivity.
Quiz 2.

Week 11 Balances on molecular and atomic species. Independent equations, independent species, and independent
reactions. Molecular species balances.
Week 12 Atomic species balances. Extent of reaction. Product separation and recycle. Homework 2.
Week 13 Purging. Combustion reactions. Theoretical and excess air. Quiz 3.
Week 14 Material balances on combustion reactors. Preparation to final exam.
Week 15 Preparation to final exam.

Assesment methods
Num Percen
Course activities
ber tage
Attendance 0 0
Laboratory 0 0
Application 0 0
Field activities 0 0
Specific practical training 0 0
Assignments 2 10
Quiz 3 20
Presentation 0 0
Project 0 0
Seminar 0 0
Midterms 1 30
Final exam 1 40
Total 100
Percentage of semester activities
6 60
contributing grade succes
Percentage of final exam contributing grade
1 40
succes
Total 100

 CHAPTER 1:
What Some Chemical Engineers Do
for a Living?
You have only a limited idea of what
chemical engineering is or what
chemical engineers do.
No universally accepted definition of
chemical engineering exists, and
almost every type of skilled work
you can think of is done somewhere
by people educated as chemical
engineers.
We will therefore abandon the idea
of formulating a simple definition
and instead take a closer look at
what recent graduates did, either
immediately after graduation or
following a well-earned vacation.
Consider these examples and see if
any of them sound like the sort of
career you can see yourself pursuing
and enjoying.
• About 45% of the class went to
work for large chemical,
petrochemical,…, or textile
manufacturing firms.
• Another 35% went to work for
government agencies and design and
consulting firms and for companies
in fields such as microelectronics and
biotechnology that have not
traditionally been associated with
chemical engineering.
• About 10% of the class went directly
into graduate school in chemical
engineering. The masters degree
candidates will get advanced education
in traditional chemical engineering
areas. The doctoral degree candidates
will get advanced education and work
on major research projects, and in four
to five years most will graduate and
either go into industrial research and
development or join a university faculty.
• The remaining 10% of the class went into
graduate school in an area other than
chemical engineering, such as medicine,
law, and business.
• Several graduates went to work for
companies manufacturing specialty
chemicals-pharmaceuticals, paints and
dyes, and cosmetics, among many other
products. Some went to work for
companies that manufacture integrated
semiconductor circuits.
• Some joined companies that
manufacture polymers. One is
working on the development of
membranes for desalination of
seawater and for gas separations;
another is developing membranes to
be used in hollow-tube artificial
kidneys.
Even the chemical engineering graduates
who go into a traditional chemical
manufacturing process end up performing
a wide variety of different tasks.
Consider the fallowing example, and see if
any of the problems described seem to
present the sort of challenge you can see
yourself taking on and enjoying.
• A chemist in your company's research and
development division has discovered that if
he mixes two reactants in a certain
proportion at an elevated temperature, he
obtains a product significantly more
valuable than both reactants. The company
contemplates manufacturing the product
using a process based on this reaction.
• At this point the matter becomes an
engineering problem or, more precisely,
hundreds of engineering problems.
1. What type of reactor should be used?
How large?
Made of what?
Does it have to be heated?
Does the reaction supply its own heat, so that
heating is needed only for startup?
If so, can the reactor "run away" and possibly
explode?
Should control measures be introduced to
prevent this?
2. Where should the reactants be obtained?
In what proportions should they be fed to the
reactor?
3. Should the reactor effluent, which contains
the product and unconsumed reactants, be said
as is, or should the product be separated from
the reactants and the latter be sent back to the
reactor?
if separation is desirable, how can it be
accomplished?
Heat the mixture and draw off and
condense the vapor, which will be richer in
the more volatile substances than the
original mixture?
Add another substance that extracts the
product and is immiscible with the
reactants, and then separate the two
phases mechanically?
if all of the process materials are gases at
the reaction temperature, can the mixture
be cooled to a temperature at which the
product condenses but the reactants do
not, or vice versa, or if they are liquids can
the mixture be cooled to a temperature at
which the product crystallizes?

if one of these alternatives is chosen,

what kind of equipment is needed?
Are controls needed to keep the operation
of the process within rigid limits?
4. How should the reactant and product streams
be moved to and from the reactor and any
heating, cooling, and separation equipment
involved in the process?
What kinds?
How big?
In pipes made of what?
5. Is enough known about the reaction system
to be able to answer all of these questions, or
should additional laboratory studies be carried
out?
Can the laboratory data be used directly to
design the industrial plant, or should a smaller
pilot plant be constructed first to test the
design?
6. What can possibly go wrong with the
process, and what can be done if and when it
does?
7. Are waste products produced by the process?
Are they potentially harmful if released untreated
info the environment?
What should be done to reduce pollution hazards?
8. Chemically treat the wastes?
Dump liquid and solid wastes into containers,
seal and cart them out to sea?
Disperse gases in the atmosphere with a high
stack?
Precipitate solids electrostatically from gas
exhausts?
9. How much of the process should be
automated, and how should the automation be
done?
How much will all of this cost?
For how much can the product be said, and to
whom?
How much money will the process net each
year?
Is it enough to make it worthwhile?
10. Once the plant has been built, what
procedure should be followed for
startup?
11. Six months later when startup has
been achieved, why is the product not
coming out the way it did in the laboratory?
Is it an equipment malfunction or a change
in conditions somewhere between the
laboratory and industrial process?
How can we find out?
What can be done to correct the problem?
Is it necessary to shut down the operation
for modifications?
12. Is it significant or just a coincidental series of
bad breaks that there have been three
explosions and four fires within six months in
the reactor unit?
In either case, how do we stop them from
recurring?
13. All sorts of other things are going wrong
with the process operation.
Why weren't they on the list of things that could
possibly go wrong?
What can be done about them?
14. When the process finally starts working
perfectly and the next day an order comes
down to change the product specifications,
how can it be done without redesigning the
entire process?
The different careers and tasks described in
this lecture are clearly too diverse to fall into
a single category.
They involve disciplines including physics,
chemistry, biology, environmental science,
medicine, applied mathematics, statistics,
computer science, economics, management
and information science, research, design,
construction, sales and service, production
supervision, and business administration.

The single feature they have in common is that

chemical engineers can be found doing them.
 CHAPTER 2
Introduction to Engineering
Calculations
2.1 Units and Dimensions
A measured or counted quantity has
a numerical value ( 2.47) and
a unit (kilograms).
It is useful in most engineering calculations
-and essential in many-
to write both the value and the unit of each quantity
appearing in an equation.
• 2 meters
• 1/3 second
• 4.29 kilograms
• 5 gold rings
A dimension is a property that can be measured,
such as length, time, mass, or temperature or
calculated by multiplying or dividing other
dimensions.

length
→ (velocity)
time
3
(length) → (volume)
mass
3
→ (density)
(length)
Units can be treated like algebraic variables
when quantities are added, subtracted,
multiplied, or divided.
The numerical values of two quantities may
be added or subtracted only if the units are
the same.
3 cm – 1 cm = 2 cm
3x – 1x = 2x
But
3 cm – 1 mm =?
3x-y=?
Numerical values and their corresponding units may
always be combined by multiplication or division.

3 N x 4m = 12 N.m
5.0 km/2.0 h = 2.5 km/h
3m x 4m = 12 m2
6g/2g =3 (3 is a dimensionless quantity)
2.2 Conversion of Units
A measured quantity can be expressed in terms
of any units having the appropriate dimension.
A particular velocity may be expressed in ft/s,
miles/h, cm/year, or any other ratio of a length
unit to a time unit.
You can convert a quantity expressed in one set
of units into its equivalent in other
dimensionally consistent units using conversion
factors.
(new unit/old unit)
For example, to convert
36 mg to its equivalent in grams, write

But,

is clearly wrong. More precisely, it is not what you intended to calculate

 EXAMPLE 2.2-1
Convert an acceleration of 1 cm/s2 to its
equivalent in km/yr2.

A principle illustrated in this example is that raising a quantity

(in particular, a conversion factor) to a power raises its units to the same
power. The conversion factor for h2/day2 is therefore the square of the
factor for h/day:
2.3 Systems of Units

• Base Units: mass, length, time

• Multiple Units: which are defined as multiples or fractions of
base units such as minutes, hours, and milliseconds, all of which are
defined in terms of the base unit of a second. It is simply more convenient
to refer to 3 yr than to 94 608 000 s

• Derived Units: obtained in one of two ways;

a) by multiplying and dividing base or multiple units; cm2, ft/min
b) as defined equivalents of compound units; 1 erg= 1 g.cm/s2
In 1960 an international conference formulated a system of metric units that
has gained widespread acceptance in the scientific and engineering
community. It is known as the
Systeme Internationale d’Unites, or the SI system for short.
Multiple Unit Preferences
tera (T)=1012
giga (G) =109 An SI convension is to use
blanks rather than commas
mega (M)=106 when writing large numbers:
kilo (k) =103 34 607 for 34,607
centi (c) =10-2
milli (m) =10-3
micro (m)=10-6
nano (n) =10-9
 EXAMPLE 2.3-1
Convert 23 lbm ft/min2 to its equivalent in kg cm/s2.
2.4 Force and Weight
According to Newton’s second law of motion,
force is proportional to the product of mass and
acceleration (length/time2)
 Natural units of force are;
kg.m/s2 = 1 newton (N)
g.cm/ s2 = 1 dyne
lbm.ft/ s2
In the American engineeering system, the
derived unit (called a pound-force (lbf)) is
defined as the product of a unit mass and
the acceleration of gravity at sea level and
45o latitude, which is 32.174 ft/ s2
1 lbf = 32.174 lbm.ft/ s2
To convert a force from a defined unit
(such as newtons) to a natural unit kg.m/s2,
a conversion factor (usually denoted by the symbol
gc) must be used.

for the SI system gc = 1 (kg.m/s2)/N

for the cgs system gc = 1 (g.cm/s2)/dyne
for the AE system gc = 32.174(lbm.ft/ s2)/lbf

The equation that relates force in defined

units to mass and acceleration is
F=m.a/gc
The weight of an object is the force exerted on
the object by gravitational attraction. Suppose
that an object of mass m is subjected to a
gravitational force W (W is by definition the
weight of the object) and that if this object
were falling freely its acceleration would be g.
The weight, mass and free-fall acceleration of
the object are related by
W=mg/gc

At sea level on the earth, g = 9.8066 m/s2

g = 980.66 cm/s2
g= 32.174 ft/s2
2.3 Numerical Calculation and
Estimation
Both very large and very small numbers are
commonly encountered in process calculations.

A convenient way to represent such numbers is

to use scientific notation, in which a number is
expressed as the product o another nuber and
a power of 10.

123 000 000 = 1.23x108 (or 0.123x109)

0.000028 = 2.8x10-5 (or 0.28x10-4)
The significant figures of a number are the digits
from the first nonzero digit on the left to either
the last digit (zero or nonzero) on the right if
there is a decimal point, or the last nonzero digit
of the number if there is no decimal point.

2300 or 2.3x103 has 2 s.f.

2300. or 2.300x103 has 4 s.f.
2300.0 or 2.3000x103 has 5 s.f.
23 040 or 2.304x104 has 4 s.f.
0.035 or 3.5x10-2 has 2 s.f.
0.03500 or 3.500x10-2 has 4 s.f.
If you report the value of a measured quantity
with three significant figures, you indicate that
the value of the third of these figures may be off
by as much as one half.

Thus, if you report a mass as 8.3 g (2 s.g.) you

indicate that the mass lies somewhere between
8.25 and 8.35 g, whereas if you give the value as
8.300 g (4 s.f.) you indicate that the mass lies
between 8.2995 and 8.3005 g.
Note, however, that this rule applies only to
measured quantities or numbers calculated
from measured quantities.
If a quantity is known precisely (like a pure
integer (2) or a counted rather than measured
quantity (16 oranges)) its value implicitly
contains an infinite number of significant
figures.
5 cows really means 5.0000……. cows.
When two or more quantities are combined
by multiplication and/or division, the number
of significant figures in the result should
equal the lowest number of significant figures
of any of the multiplicands or divisors.
(3.57) (4.286) = 15.30102
15.3

(5.2x 10-4) (0.1635x107)/(2.67) = 318.426966

3.2x102 =320
The rule for addition and subtraction concerns
the position of the last significant figure in the
sum-that is, the location of this figure relative to
the decimal point.

When two or more numbers are added or

subtracted, the positions of the last significant
figures of each number should be compared. Of
these positions, the one farthest to the left is
the position of the last permissible significant
figure of the sum or difference.
 1530
-2.56
1527.44 = 1530

denotes the
last significant figure
of each number
 1.0000
0.036
0.22
+
1.2560 = 1.26
2.75x106 + 3.400x104

0.03400x106

2.75x106
0.03400x106
18.76
7
11.76 = 12
A rule of thumb for rounding off numbers
in which the digit to be dropped is a 5 is
always to make the last digit of the
rounded-off number even:

1.35 1.4
1.25 1.2
2.5b Validating Results
1. How do I get a solution?
2. When I get one, how do I know it’s right?
Back-substitution is straightforward: after you
solve a set of equations, substitute your solution
back into the equations and make sure it works.
Order-of-magnitude estimation means coming
up with a crude and easy-to-obtain
approximation of the answer to a problem and
making sure that the more exact solution comes
reasonably close to it.
The test of reasonableness: Applying the test of
reasonableness means verifying that the
solution makes sense. (a calculated velocity of
water flowing in a pipe is faster than the speed
of light or the calculated temperature in a
chemical reactor is higher than the interior
temperature of the sun)
The procedure for checking an arithmetic calculation
by order-of-magnitude estimation:
1. 27.36 20 or 30
63,472 6x104
0.002887 3x10-3
2. (36,720) (0.0624) (4x104) (5x10-2)
≈
0.00478 5x10-4
= 4x10(4-2+4) = 4x106 The correct solution is 4.78x106
3. 1 1
4.13 + 0.04762
≈ 4
The correct solution is 0.239
Example 2.5.1

!"# ,- ,
̇
𝑉= + x
(%.'())(+!.#) (%.'++)(+!.#) (-,.-,#")(+%)

Estimate 𝑉̇ without using a calculator.

The exact solution is 0.00230.

!"% ,% , !+% , ,
̇
𝑉= + x ≈ x ≈
+% +% -.,% +.,% +% -.+.,%! #.,%!

= 0.2x10-2 ≈ 0.002
The third way to check a numerical result -and
perhaps the first thing you should do when you
get one - is to see if the answer is reasonable.
If, for example, you calculate that a cylinder
contains 4.23x1032 kg of hydrogen when the
mass of the sun is only 2x1030 kg, it should
motivate you to redo the calculation.
2.5c Estimation of Measured Values:
Sample Mean
Suppose we carry out a chemical reaction of the form
A Products,
starting with pure A in the reactor and keeping the reactor
temperature constant at 45oC. After two minutes we draw a
sample from the reactor and analyze it to determine X, the
percentage of the A fed that has reacted.
In theory X should have a unique value however, in a real reactor
X is a random variable, changing in an unpredictable manner
from one run to another at the same experimental conditions.
The values of X obtained after 10 successive runs:

Run 1 2 3 4 5 6 7 8 9 10
X,% 67.1 73.1 69.6 67.4 71.0 68.2 69.4 68.2 68.7 70.2

• Why don’t we get the same value of x in each run?

o It is impossible to replicate experimental conditions
o Even if conditions were identical in two runs, we could not possible draw
our sample exactly t= 2.000… minutes both times
o Variations in sampling and chemical analysis procedures invariably
introduce scatter in measured values.
We might ask two questions about the system at this
point
• What is the true value of x?
The “true value”—that is, the value we would measure if
we could set the temperature exactly to 45.0000..
degrees, start the reaction, keep the temperature and all
other experimental variables that affect X perfectly
constant, and then sample and analyze with complete
accuracy at exactly t=2.0000...minutes. In practice there
is no way to do any of those things, however. We could
also define the true value of X as the value we would
calculate by performing an infinite number of
measurements and averaging the results, but there is no
practical way to do that either. The best we can ever do is
to estimate the true value of X from a finite number of
measured values.
• How can we estimate of the true value of x?
The Sample Mean
The most common estimate is the sample mean (or
arithmetic mean). We collect N measured values of
X (X1, X2, . . . , XN ) and then calculate

For the given data, we would estimate

The more measurements of a random

variable, the better the estimated value
based on the sample mean. However,
even with a huge number of
measurements the sample mean is at best
an approximation of the true value and
could in fact be way off.
2.5d Sample Variance of Scattered Data
Consider two sets of measurements of a random
variable, X—for example, the percentage conversion in
the same batch reactor measured using two different
experimental techniques. Scatter plots of X versus run
number are shown in the figure. The sample mean of
each set is 70%,
but the measured values
scatter over a much
narrower range for the
first set
(from 68% to 73%) than
for the second set
(from 52% to 95%).
In each case you would estimate the true value of X for the given experimental
conditions as the sample mean, 70%, but you would clearly have more confidence in
the estimate for Set (a) than in that for Set (b).
Range
R = Xmax –Xmin

In the first plot of Figure

the range of X is 5% (73% - 68%) and
in the second plot it is 43% (95% - 52%).

The range is the crudest measure of scatter: it involves only two of

the measured values and gives no indication of whether or not most
of the values cluster close to the mean or scatter widely around it.
The sample variance
The sample variance is a much better measure. To define it we
calculate the deviation of each measured value from the sample
mean, Xj - 𝑋! (j = 1,2,...,N), and then calculate

Sample standard deviation

The degree of scatter may also be expressed in terms of
the sample standard deviation, by definition the square
root of the sample variance:
Values of measured variables are often reported with
error limits, such as X = 48.2 ∓ 0.6.
This statement means that a single measured value
of X is likely to fall between 47.6 and 48.8.
The midpoint of the range (X = 48.2) is almost always
the mean value of the data set used to generate this
result; however, the significance of the given error
limits (∓ 0.6) is not obvious unless more information
is given.
The interval between 47.6 and 48.8 may represent
the range of the data set (Xmax - Xmin) or ∓ 0.6 might
represent ∓ sX, ∓ 2sX, or ∓ 3sX.
2.6 Dimensional Homogeneity and
Dimensionless Quantities
If the units are the same, it follows that the dimensions of
each term must be the same.
for example,
if two quantities can be expressed in terms of
(grams/second, grams/second)
both must have the dimension
mass/time
Every valid equation must be dimensionally
homogeneous: that is, all additive terms on both sides of
the equation must have the same dimensions
 ! ! !
V ( ) = Vo( ) + g ( !) t(s)
" " "

!"#$%& !"#$%& !"#$%&

%'(" %'(" %'("

This equation is dimensionally homogeneous,

since each of the terms v, v0, and gt has the same dimensions
(length/time).
On the other hand, the equation
v = v0+g is not dimensionally homogeneous (why not?) and
therefore cannot possibly be valid.
v = v0+gt is both dimensionally homogeneous
and consistent in its units, in that each additive
term has the units m/s.
If values of v0 , g , and t with the indicated
units are substituted into the equation, the
addition may be carried out to determine the
value of v.
If an equation is dimensionally homogeneous
but its additive terms have inconsistent units,
the terms (and hence the equation) may be
made consistent simply by applying the
appropriate conversion factors.
For example, suppose that in the dimensionally homogeneous
equation v = v0+gt it is desired to express the time (t ) in minutes
and the other quantities in the units given above. The equation
can be written as
! ! !
V ( ) = Vo( ) + g ( !) t(min)(60s/min)
" " "
= Vo+ 60 g t
Each additive term again has units of m/s , so the equation is
consistent.
The converse of the given rule is not necessarily true—an
equation may be dimensionally homogeneous and invalid.

For example, if M is the mass of an object, then the equation

M = 2M is dimensionally homogeneous, but it is also obviously
incorrect except for one specific value of M.
EXAMPLE 2.6-1
Consider the equation
D (ft) = 3 t(s) + 4
1. If the equation is valid, what are the dimensions
of the constants 3 and 4?

For the equation to be valid, it must be

dimensionally homogeneous, so that each term
must have the dimension of length.
The constant 3 must therefore have the dimension
(length/time), and 4 must have the dimension
(length).
2. If the equation is consistent in its units, what
are the units of 3 and 4?
For consistency, the constants must be
3 ft/s and 4 ft.
3. Derive an equation for distance in meters in terms of
time in minutes.
Define new variables D’(m) and t’(min).
The equivalence relations between the old and new
variables are
-.!'%' /0
D (ft) = D’ (m) = 3.28 D’
,1
+% 2
t (s) = t’ (min) = 60 t’
, 134
Substitute these expressions in the given equation

3.28 D’ = (3) (60t’) + 4

D’ (m) = 55 t’ (min) + 1.22
The unit of 55 is m/min and 1.22 is m.
A dimensionless quantity can be a pure number
(2, 1.3, 5/2 ) or a multiplicative combination
of variables with no net dimensions:
# (%)
Dimensionless group
#! (%)

)* ,
! "# $ %( )
+ )*-
, Dimensionless group
(( )
)*.+
Exponents (such as the 2 in X2), transcendental
functions (such as log, exp ≡ e, and sin),
and arguments of transcendental functions
(such as the X in sin X) must be dimensionless
quantities.
For example, 102 makes perfect sense, but 102 ft
is meaningless,
as is log (20 s) or sin (3 dynes).
EXAMPLE 2.6-2
A quantity k depends on the temperature T in the
following manner:

156 !% %%%
𝑘( ! ) =1.2 x 105 exp (- )
71 .2 ,.)'( 8

The units of the quantity 20 000 are cal/mol, and T is in K

(kelvin). What are the units of 1.2 x 105 and 1.987?

Since the equation must be consistent in its units and exp

is dimensionless, 1.2 x 105 should have the same units as
k, mol/(cm3.s). Moreover, since the argument of exp must
be dimensionless, we can write
 796 , 156.:
20 000
156 8(:) ,.)'( 796

156 796
1.2 x 10 5
" and 1.987
71 .2 156.:
2.7 PROCESS DATA REPRESENTATION
AND ANALYSIS
The operation of any chemical process is ultimately based on the
measurement of process variables temperatures, pressures, flow rates,
concentrations, and so on.
It is sometimes possible to measure these variables directly, but as a
rule, indirect techniques must be used.
Suppose, for example, that you wish to measure the concentration, C,
of a solute in a solution.
To do so, you normally measure a quantity,
X (such as a thermal or electrical conductivity,
a light absorbance, or the volume of a titer)
that varies in a known manner with C,
and then calculate C from the measured value of X.
The relationship between C and X is determined in a separate
calibration experiment in which solutions of known concentration are
prepared and X is measured for each solution.
Consider a calibration experiment in which a variable, y , is measured
for several values of another variable, x :

y might be a reactant concentration or some other process variable

and x would be a readily measured quantity (such as conductivity)
whose value correlates with the value of y .
Our object is to use the calibration data to estimate the value of y for a
value of x between tabulated points (interpolation) or outside the
range of the table data (extrapolation).
A number of interpolation and extrapolation methods are commonly
used, including
two- point linear interpolation,
graphical interpolation, and
curve fitting.
Which one is most appropriate depends on the nature of the
relationship between x and y .
 Following figures show several illustrative (x, y) plots.

If the plot of a given data set looks like that shown in plot (a) or (b)
of this figure, a straight line would probably be fitted to the data
and used as the basis for subsequent interpolation or
extrapolation.
On the other hand, if the plot is distinctly curved as in plot (c ), a
curve could be drawn in by inspection and used as the basis of
interpolation, or straight-line segments could be fitted to
successive pairs of points, or a nonlinear function y(x) that fits the
data could be sought.
 2.7a Two-Point Linear Interpolation
The equation of the line through (x1, y1) and (x2, y2) on a plot of y
versus x is
#$#!
y = y1 + (y2-y1)
#"$#!

%"$%! %$%!
Slope = Slope =
#"$#! #$#!

&"$&! &$&!
=
'"$'! '$'!
y2
'$'!
y − y1= (𝑦2 − 𝑦1)
'"$'!

'$'!
y1 y = y1 +
'"$'!
(𝑦2 − 𝑦1)

x1 x2
2.7b Fitting a Straight Line
A convenient way to indicate how one variable depends on
another is with an equation:
Y=3x+4
Y=4.24(x-3)2 -23
y = 1.3 x 107 sin(2x)/(x1/2 + 58.4)

If you have an analytical expression for y(x) like those

shown above, you can calculate y for any given x or (with a
somewhat greater effort) determine x for any given y , or
you can program a computer to perform these calculations.
Suppose the values of a dependent variable y have been
measured for several values of an independent variable x , and a
plot of y versus x on rectangular coordinate axes yields what
appears to be a straight line. The equation you would use to
represent the relationship between x and y is then
y = ax+b
If the points show
relatively little scatter a
line may then be drawn
through them by
inspection, and if (x1, y1)
and (x2, y2) are two
points—which may or may
not be data points—on the
Slope a= #!"#" Intercept
! "!
line, then
! "
b = y1-ax1 = y2-ax2
EXAMPLE 2.7-1
Fitting a Straight Line to Flowmeter Calibration Data

Rotameter calibration data (flow rate versus rotameter reading)

are as follows:

1. Draw a calibration curve and determine an equation for V(R).

2. Calculate the flow rate that corresponds to a rotameter reading of 36.
EXAMPLE 2.7-1

R1=10 𝑉1̇ =20

R2=60 𝑉̇ 2 =101

Check at Point 2
aR2 + b = (1.62)(60) + 3.8 = 101 =V2
2) The flow rate that corresponds to
a rotameter reading of 36 is
V = (1.62)(36) + 3.8 = 62.1 L/min
2.7c Fitting Nonlinear Data
Fitting a nonlinear equation (anything but y ax b) to data is usually
much harder than fitting a line; however, with some nonlinear
equations you can still use straight-line fitting if you plot the data in a
suitable manner.
Suppose, for example, that x and y are related by the equation
y2 =ax3 +b.
A plot of measured y versus x data would clearly be curved; however, a
plot of
y2 versus x3 would be a straight line with a slope a and intercept b.

More generally, if any two quantities are related by an equation of the

form
(Quantity 1)= a (Quantity 2) +b
then a plot of the first quantity (y2 in the above example) versus the
second (x3) on rectangular coordinates yields a straight line with slope
a and intercept b.
Examples of plots that yield straight lines:

𝑦 = 𝑎 𝑥# + b 𝑦

𝑥#

$
𝑦# = +b
% 𝑦2

1(
𝑥

&
& '
= a (x+3) + b
'
(x+3)
 𝑠𝑖𝑛𝑦
#
𝑠𝑖𝑛𝑦 = 𝑎(𝑥 - 4)

(𝑥 # - 4)
Even if the original equation is not in a proper form to generate a
linear plot, you can sometimes rearrange it to get it into such a form:
1
1 1
𝑦= = 𝐶& 𝑥 − 𝐶# 𝑦
𝐶& 𝑥 − 𝐶# 𝑦
𝑥

('*&)#
('*&)#
𝑦 = 1 + 𝑥(𝑚𝑥 # + n)&/# = m𝑥 # + n %#
%#

𝑥#
If you have (x, y) data that you wish to fit with an equation
that can be written in the form f (x, y) = a g(x, y) + b,
1) Calculate f(x, y) and g(x, y) for each tabulated (x, y)
point, and plot f versus g .
2) If the plotted points fall on a straight line, the equation
fits the data.
Choose two points on the line
(g1, f1) and (g2, f2)
and calculate a and b.
2!"2"
a= 𝑏 = 𝑓1 − 𝑎𝑔1 𝑜𝑟 𝑏 = 𝑓2 − 𝑎𝑔2
3!"3"
EXAMPLE 2.7-2
A mass flow rate 𝑚(g/s)
̇ is measured as a function of
temperature T (0C).
T 10 20 40 80
𝑚̇ 14.76 20.14 27.73 38.47
There is reason to believe that 𝑚̇ varies linearly with the
square root of T : .
𝑚̇ = aT1/2 + b
Use a straight-line plot to verify this formula and determine
a and b.

T 10 20 40 80
𝑚̇ 14.76 20.14 27.73 38.47
T1/2 3.162 4.472 6.325 8.944
 Since the plot is linear, the proposed
formula is verified. A line drawn through
the data points passes through the first
and last points, so these points may be
used to calculate the slope and intercept.

F = aT1/2 + b

("$(! "*.,-$.,.-/
𝑎= '/& = *.0,,$"../!= 4.10 g.s K
-1 -1/2
'/&
)& $)'

./!
𝑏 = 𝐹1 − 𝑎𝑇. = 14.76 − 4.10 3.162 = 1.80 𝑔/𝑠

𝐹 = 4.10 T1/2 + 1.80

Two nonlinear functions that are often occur in process
analysis are the exponential function
y = a ebx or y = a exp(bx), where

e= 2.7182818, and the power law, y = axb.

ln x = 2.302585 log10x
y=ax ln y = ln a + ln x
y = xb ln y = b ln x

ln 𝑦
𝑦 = 𝑎 exp(𝑏𝑥) ln y = ln a + b x
𝑥
𝑦 = a xb ln y = ln a + b ln x
ln 𝑦

ln𝑥
2.7d Logarithmic Coordinates
Suppose you wish to fit an exponential function
y= a exp(bx) to measured (x, y) data.
If there are many data points, calculating the logarithm of each
y value (needed to plot ln y versus x )
could take more time than the fitting procedure itself.

However, suppose that an

additional scale were drawn
parallel to the lny axis,
on which values of y were
shown adjacent to the
corresponding values of ln y on
the first scale.
Now, instead of having to
calculate ln y for each tabulated
y to locate the data points on
the graph, you can find the y
values on the second scale and
locate the points directly.
If the same type of scale (called
a logarithmic scale) were drawn
parallel to the other axis, you
could also plot ln x values
without having to calculate
them from tabulated x values.
A plot with logarithmic scales on both axes is called a log plot, and
a plot with one logarithmic and one rectangular (equal interval)
axis is called a semilog plot.
Log paper is graph paper with logarithmic scales on both axes, and
semilog paper has one logarithmic axis and one rectangular axis.
Most computer plotting packages allow selection of log or semilog
scales as alternatives to rectangular scales.

When you plot values of a variable y on a logarithmic scale you

are in effect plotting the logarithm of y on a rectangular scale.
Suppose, for example, that y and x are related by the equation
y=a exp (bx) (ln y= ln a+ bx).
To determine a and b you may plot y versus x on a semilog plot,
choosing two points (x1, y1) and (x2, y2) on the resulting line, or
you may plot ln y versus x on rectangular axes, drawing the line
through the corresponding two points (x1, ln y1) and (x2, ln y2).
• In either case, b and a are obtained as
In summary,
1. If y versus x data appear linear on a semilog plot, then ln y
versus x would be linear on a rectangular plot, and the data can
therefore be correlated by an exponential function y= a exp (bx).

2. If y versus x data appear linear on a log plot, then ln y versus

ln x would be linear on a rectangular plot, and the data can
therefore be correlated by a power law y= axb.
3. When you plot values of a variable z on a logarithmic axis
and your plot yields a straight line through two points with
coordinate values z1 and z2, replace z2 -z1 with
ln (z2 /z1) (= ln z2 - ln z1) in the formula for the slope.

4. Do not plot values of ln z on a logarithmic scale and expect

anything useful to result.
Example 2.7.3
A plot of F versus t yields a line that passes the points
(t1=15, F1=0.298) and (t2=30, F2= 0.0527) on
1) A semilog plot and
2) A log plot.
3) For each case, calculate the equation that relates F and t.
Semilog Plot lnF = bt + lna
F = a ebt
ln (F2/F1)
b = (ln F2-lnF1)/(t2-t1) = = - 0.1154
t2−t1
lna = lnF1 – bt1 = ln(0.298) + (0.1154)(15) = 0.5218
or
lna = lnF2 – bt2 = ln(0.0527) + (0.1154)(30) = 0.5218
a = e0.5218 = 1.685
F = 1.685 e(-0.1154t)
Log Plot lnF = b lnt + lna
F = a tb
ln (F2/F1)
b = (ln F2-lnF1)/ln(t2-t1) = = - 2.50
ln(t2−t1)
lna = lnF1 – b lnt1 = ln(0.298) + (2.50)(15) = 5.559
or
lna = lnF2 – bt2 = ln(0.0527) + (0.1154)(30) = 5.559
a = e5.559 = 260
F = 260 t -2.5