ChE curriculum

Energy balances
on transient processes

Francisco Ruiz-Beviá, M. Dolores Saquete, Ignacio Aracil, and M. Francisca Gómez

T
University of Alicante, P.O. Box 99 E-03080 • Alicante, Spain
his study proposes a set of experiments that will prove ances, along with the material balances, are essential tools
useful to students in understanding and putting into for the study of any basic operation of chemical engineer-
practice basic concepts involved in energy balances. ing, and therefore the fundamental concepts of material and
These experiments, which form part of a course of Chemical energy balances are usually incorporated in the first year of
Engineering Laboratory I, are very simple from the point of all chemical engineering curricula. This is reflected in intro-
view of the required equipment and operations that students ductory chemical engineering textbooks (e.g., Himmelblau,[2]
must carry out, as well as with regard to the concepts in- Henley and Rosen,[3] and Reklaitis[4]), which have followed
volved. The first part of the experiments consists of heating the guidelines established by the pioneer book by Hougen and
and later cooling a mass of water contained in a vessel by Watson, Material and Energy Balances,[5] published in 1943.
means of an electrical resistor of known power, which is first The journal Chemical Engineering Education has also shown
connected and then disconnected from a power source (batch
process). The results of this experiment necessarily have to be Francisco Ruiz-Beviá is professor emeritus of chemical engineering at
interpreted in terms of an unsteady-state energy balance. In the University of Alicante (Spain). He received his Ph.D. from Valencia
a second part, both heating and cooling of the water, by first University (Spain). He conducts research in phase equilibria and holo-
graphic interferometry applied to mass transfer.
connecting and then disconnecting the electrical resistor, are
carried out while at the same time allowing water to flow at
M. Dolores Saquete received her Ph.D. in chemical engineering from
Alicante University (Spain) in 2001. She is currently an associate profes-
a constant rate through the vessel. Under these conditions the sor at the University of Alicante (Spain). Her research interests include
temperature is fixed at the inlet and obviously changing at the phase equilibria and thermodynamic properties.

outlet. In this last case, a stationary state is eventually reached Ignacio Aracil received his Ph.D. in chemical engineering from the
University of Alicante (Spain) in 2008. He currently works as an assistant
and the difference of the enthalpy between the incoming and professor at that university. His research is mainly devoted to the study
exit water streams must appear in the energy balance. of waste treatments, including pollutant emissions derived from thermal
treatments of plastic wastes and process design to recover components
These experiments, and others involving unsteady-state of specific wastes such as waste ink and rice husk.
material balances, have been set up and tested in the laborato- M. Francisca Gómez received her Ph.D. in chemical engineering from
ries of the chemical engineering department at the University the University of Alicante (Spain) in 2008. She currently works as an

of Alicante, where they are used as a practical complement,

assistant professor at that university. Her research is mainly focused on
waste treatments, such as thermal treatment, to study the pollution in-
in the first year of the curriculum, to illustrate the concepts volved and different ways of taking advantage of this waste with minimum
developed in Chapter 11, “Balance on Transient Processes,” damage to the environment.

in the textbook by Felder and Rousseau.[1] The energy bal- © Copyright ChE Division of ASEE 2012

Vol. 46, No. 4, Fall 2012 231

its interest in this topic, e.g., in the two-part article written
by Bullard and Felder,[6,7] “A Student-Centered Approach to
Teaching Material and Energy Balances.”
The experiments relating to unsteady-state energy balances
presented here have also been studied by other authors in
CEE, such as Condoret[8] and Luyben.[9] Condoret, in his
article “Teaching Transport Phenomena Around a Cup of
Coffee,” included in the category of ChE Class and Home
Problems, presents the problem like this: We put a cup of cof-
fee on a table. Its initial temperature is around 80 ˚C. What
is the temperature of the coffee after 10 min, for instance?
The paper solves the problem using a simulation model of
the cooling process; the model takes into account the heat
loss at vessel wall of the cup, the heat loss by heat transfer
only at the surface of the liquid, and the heat loss resulting
from evaporation. The paper also describes a simple lab ex-
Figure 1. Experimental apparatus.
periment using porcelain cups filled with water, a numerical
thermometer, a balance, and a stopwatch. The didactic value
Table 1
of the paper comes from showing how using heat transfer
Details of the Experimental Apparatus
coefficients makes it possible to model and predict the simple
Parameter Value
experiment of cooling a cup of hot coffee. Another experiment
relating to unsteady-state energy balances is described by P (W) 110
Luyben[9] in the article “The Devil’s in the Delta,” included in Mw (kg) 1.75
the ChE Laboratory category. The process consists of a stirred Mv (kg) 1.63
vessel, 1 m diameter, containing 785 kg of water. The rpm’s Mr (kg) 0.10
of the agitator can be varied to see the effect on the inside
Cpw (J/kg ˚C) 4180
fluid coefficient. A spiral coil is wrapped around the outside
Cpv (J/kg ˚C) 460
of the vessel. The liquid in the vessel is initially at ambient
temperature. It is heated by introducing steam at the top of Cpr (J/kg ˚C) 836
the coil. When the water of the vessel reaches about 80 ˚C m (kg/s) 4.0·10-3
the steam is shut off and cooling water is introduced. The
didactic value of this paper comes from providing a clear dis- also insulated at the base to isolate it from the magnetic
tinction of the three “deltas” as used in chemical engineering; stirrer, and in addition, has a methacrylate cover that blocks
the author refers to them as “In Minus Out” Delta, “Driving evaporation. Therefore, water mass remains constant and
Force” Delta, and “Time” Delta. As the author mentions in evaporative cooling is avoided. Table 1 shows the power of
the paper conclusions, although the distinction of the three the resistor (P), the overflow capacity of the vessel (mass
deltas is obvious to the experienced engineer, they are often of water, Mw), the mass of the vessel (Mv) and the resistor
misapplied by young students. (Mr), as well as the heat capacities Cpw, Cpv, and Cpr, of the
water, vessel, and resistor, respectively. It also contains the
EXPERIMENT value of constant flow rate of water that enters the vessel
Apparatus (m). Three thermocouples interfaced with a PC measure and
record the temperature of the water in the inlet (Tin) and at
The experimental apparatus used is shown in Figure 1. It
two other points in the vessel (T1 in the uppermost part of
consists of a vessel with water inlet and outlet, equipped with
the vessel and T2 in the lowermost). The extent to which the
an electrical resistor heater and a magnetic stirrer to ensure
vessel is well stirred depends upon how close T1 and T2 are.
that the temperature is uniform throughout the vessel. The
The continuous monitoring and recording of temperatures by
water, fed to the apparatus at a constant flow rate, comes
the thermocouples interfaced with the PC makes it possible
from a tank that is kept at a constant water level. Water exits
to extract numerous experimental data points.
the vessel at the same flow rate as in the feed. In addition
to the experimental apparatus just described, a stopwatch, The experimental method consists of the following steps.
beakers, and a balance will prove useful for measuring the Without water circulation:
flow rate properly. Experiment A1: The vessel, full of water, is warmed up to 45 ˚C
The vessel consists of a stainless steel cylinder of 10 cm by connection of the resistor to the power source (nonstation-
internal diameter, 23 cm length, and 1.63 kg weight. It is ary regime).
232 Chemical Engineering Education
 Figure 2 shows the experimental data obtained in the form
of temperature T of the water in the vessel vs. time t. The mea- A1 - Heated by resistor with no water inlet
sured values of temperatures T1 and T2 are practically equal, 50
with less than two-tenths of a degree difference between them,
which indicates that the vessel is perfectly stirred. These cir- 45 T exp (ºC)
cumstances—equal temperatures and good stirring—together T calc eq.(3) (ºC)
with the fact that water has a high heat capacity, ensure that T recon eq.(10) (ºC)
40
the heat generated by the resistor is quickly transferred to the
mass of water, preventing the resistor surface temperature

T (ºC)
from reaching higher than 100 ˚C, thus avoiding local boiling. 35

The average value of T1 and T2 has been used to represent

the experimental data graphically, and is shown as the dashed 30
line in Figure 2.
25
Experiment A2: When the temperature reaches 45 ˚C, the
resistor is disconnected from the power source and the evolu-
tion of temperature with time is studied as an experiment of 20
water cooling under nonstationary conditions. 0 500 1000 1500 2000

Figure 3 shows (dashed line) the experimental data obtained t (s)

in the form of temperature T of the water in the vessel vs. time t. Figure 2. Experiment A1. Water heating to 45 °C.
With water circulation:
Experiment B1: The water in the vessel is pre-heated to 45 ˚C. Fig. 2 46
A2 - Cooled with no water inlet
The experiment begins when a valve is opened to allow water
to flow through the vessel at a constant rate without disconnect- 44
ing the resistor from the power source. This implies a cooling 42 T exp (ºC)
experiment, and the time and temperatures are recorded until 40
a steady state temperature (constant temperature) is reached.
T calc eq.(8) (ºC)
38
The flow rate of water that enters the vessel is measured when
T (ºC)

36
it exits by means of the “bucket and stopwatch” method.
34
Experiment B2: While maintaining water circulation the re- 32
sistor is disconnected. The temperature then decreases until
a different stationary state is reached.
30
28
Experiment B3: The resistor is connected without varying the 0 5000 10000 15000 20000
water flow rate, thus raising the temperature to the stationary t (s)
state temperature that was reached at the end of experiment B1.
Figure 3. Water cooling with resistor switched off.
Figure 4 represents the data obtained
in experiments B1, B2, and B3 as three
different but connected portions of a Fig. 3B - Heated and cooled under water flow
dashed line. 50
45
DISCUSSION 40

Without water circulation:

35
30
Experiment A1 is a process of heat-
T (ºC)

25
ing in the nonstationary regime where 20
the temperature of the mass of water 15
Mw, the mass of the vessel Mv, and the 10 T exp (ºC) T calc eq.(12) (ºC)

mass of the electrical resistor Mr are 5

raised thanks to the difference between 0
the heat received from the electrical 0 1000 2000 3000 4000 5000 6000
resistor of power P and the heat lost t (s)
by convection Ql through the walls of
Figure 4. Experiments B1-3. Cooling and heating with flow.
the vessel.
Vol. 46, No. 4, Fall 2012 233
Fig. 4
 Experiment A1 can be modeled by means of the following 0,0
energy balance -0,2
0 5000 10000 15000 20000

dT
(M W Cp + M V Cp + M r Cpr )
W v
dt
= P − Q1 (1) -0,4
T exp (ºC)
-0,6
The mechanical work is not included in this energy balance
T calc eq.(7) (ºC)
-0,8
because in the present case it arises only from the magnetic
-1,0
stirrer and it can be considered negligible.
-1,2
Neglecting the losses of heat through the walls, Eq. (1)
simplifies to -1,4

dT -1,6
(M W Cp + M V Cp + M r Cpr )
W v
dt
=P (2) -1,8

Integration then leads to t (s)

P
T = T0 + t ( 3) Figure 5. The slope of the straight line allows calculation
( W p
M C + M
W V C p + M r C pr )
v of the parameter UA.
where T0 is the initial temperature of the water in the vessel ture5 at the end of experiment A1, 45.1 ˚C, while the ambient
Fig.
at the beginning of experiment A1, 22.9 ˚C in this case. temperature is 25.3 ˚C.
Notice that Figure 2 contains the plot of the simplified model Figure 5 is another plot of the experimental data of the
embodied in Eq. (3) (equation of a straight line). It can be cooling process but of the form
observed that the calculated temperatures (represented by the T − Tam b
thin continuous line) are a little higher than the experimental 1n
T0 − Tam b
ones (dashed line), but the difference does not become greater
than 2 ˚C until only after 500s. A small improvement can be as a function of time t.
obtained if Eq. (1) is used instead, i.e., by taking into account According to Eq. (7), the slope of the straight line in Figure
the heat lost by convection Ql through the walls of the vessel. 5 is -8.50 ·10-5 s-1, that leads to a value for the parameter UA
This lost heat can be expressed as of 0.693 W/˚C. Since the external surface of the vessel A is
Q1 = UA ( T − Tam b ) (4) known, 0.07235 (m2), the value for the global coefficient of heat
transfer U can be calculated (9.59 W/m2 ˚C). In this cooling
where experiment A2, there are different heat transfer mechanisms,
A= External surface of the vessel (m2) namely: internal liquid convection in series with conduction in
U= Global coefficient of heat transfer ( W/m2 ˚C) the stainless steel wall and external transfer to ambient air. The
internal individual coefficient of heat transfer is high (forced
T= Water temperature (˚C) convection since the water is stirred). Also the heat transfer by
Tamb= Ambient temperature (˚C) conduction in the stainless steel wall is high. The external coef-
The experimental value of product UA can be determined ficient (free convection to air) is low, however, and therefore
from the data of experiment A2, where the vessel is cooled. this last mechanism is probably the limiting one. In fact, the
Applying an energy balance in this case, in which electrical value of the overall coefficient computed, U= 9.59 W/m2 ˚C, is
energy P is not supplied, yields close to the value of a usual external heat transfer coefficient.
dT Eq. (7) written in the following format
(M W Cpw + M V Cp + M r Cpr )
v
dt
= −Q1 (5)
− UA t
T − Tam b ( M C p +M C p +M rC pr )
which combined with Eq. (4) results in =e W w V v (8)
T0 − Tam b
dT
(M W Cpw + M V Cp + M r Cpr )
v
dt
= −UA ( T − Tamb ) (6) permits comparison using Figure 3 of the calculated (continu-
ous line) and experimental (dotted line) values of the cooling
and the integration leads to experiment A2. It can be observed that there is good agree-
T − Tam b UA ment between the experimental and calculated data.
1n =− t ( 7)
T0 − Tam b (M W Cpw + MV Cp + M r Cpr ) In the same way, once the value of UA is known, the
variation of temperature with time in experiment A1 can be
v

Here, T0 is the initial temperature of the water at the begin- recalculated by means of Eq. (9), which includes the losses
ning of experiment A2, and corresponds to the final tempera- of heat through the walls.
234 Chemical Engineering Education
 dT In the same way, integration of Eq. (14) leads to an equa-
(M W Cp + M V Cp + M r Cp
W v r
) dt = P − UA ( T − T ) (9) am b tion similar to Eq. (12) in which P is absent and in which T0
must be the initial temperature of water when experiment B2
An expression that once integrated leads to begins. Therefore, the final temperature of the stationary state
P − UA ( T − Tam b ) UA at the end of experiment B1 must be used. Figure 4 shows the
1n =− t (10 ) calculated values (continuous line) of experiment B2, which
P − UA ( T0 − Tam b ) ( M W Cp + M V Cp + M r Cp )
W v r exhibit good agreement with the experimental ones (dashed
line). For longer time periods, when the stationary state is
Figure 2 shows the variation of water temperature with time
reached and dT/dt = 0 , the following equation will hold true
calculated using Eq. (10). It can be observed that this simulation
with Eq. (10) (bold solid line), which does take into account 0 = mCpw ( Tin − T) − UA ( T − Tam b ) (15)
the losses of heat through the walls of the vessel, represents
a very slight improvement on the simulation done using Eq. (3) (thin continuous line), and allows calculation of the final
which neglected those heat losses. This small improvement suggests that the Q1 term in temperature of experiment B2.
Eq. (1) can be neglected to a first approximation to give Eq. (2). Indeed, the effect of heat Since in this experiment Tin =
losses on the temperature is greatest toward the end of the experiment when ∆T reaches Tamb, according to Eq. (15) the final
its highest value, around 10 W, which compared to a power of 110 W of the electrical temperature should be equal or
resistor means a deviation of around 10% when neglecting heat losses through the walls. close to Tamb. In the present case,
With water circulation this corresponds to a final tempera-
ture of 20.5 ˚C. (Tamb=21.0 ˚C)
The process that takes place in experiment B1 can be modeled by means of Eq. (1)
extended to include the term mCpw (Tin – T). This term represents the change in enthalpy In experiment B3 the resistor
of the water flowing through the vessel between the inlet and outlet. In the experiments is reconnected while maintaining
where water is circulating through the vessel, T corresponds to the outlet water tem- water circulation. Therefore, the
perature if the vessel is well stirred. same differential Eq. (11) and the
same integrated Eq. (12) are valid,
dT
(M W Cp + MV Cp + M r Cpr ) dt = mCp ( Tin − T) + P − UA ( T − Tam b )
W v W
(11) the only difference being that T0
now corresponds to the final tem-
which after integration leads to: perature of experiment B2. Figure
4 shows the calculated values and
mCp Ten + P + UATam b − ( mCp + UA) T (mC + UA)
1n W W
=−
pW
t (12) the experimental ones plotted on the
mCp Ten + P + UATam b − ( mCp + UA) To
W W
(M W Cp + M V Cp + M r Cpr )
W v same graph. The good agreement of
the data is again evident. The final
In this case, T0, the initial temperature of water at the beginning of experiment B1, has a temperature of the stationary state
value of 45.5 ˚C. The temperature inside the vessel decreases to a constant value (station- will be the same in experiment B3
ary state). as in B1, which is given by Eq. (13).
Once a stationary state has been achieved, the temperature will not vary with time In experiment B3 the temperature of
and therefore dT/dt = 0 . Upon substitution of this into Eq. (11), the temperature of the the stationary state is 26.8 ˚C, very
stationary state is easily obtained: similar to the 26.9 ˚C reached in B1.
0 = mCp W
(T en − T) + P − UA ( T − Tam b ) (13) It should be pointed out that in
the beginning of each experiment a
Putting known data values into Eq. (12), the variation of temperature T with time t is small period of time passes where
obtained. These calculated values for experiment B1 are plotted alongside the experi- there is some inertia due to the initial
mental ones in Figure 4. The good agreement between the experimental (dashed line) connection or disconnection of the
and calculated (continuous line) data is evident. resistor. What happens during this
In the experiments involving water circulation the ambient temperature is 21.0 ˚C. time has not been taken into account
The temperature at the water inlet is 20.4 ˚C in this case. because the energy balances in this
On the other hand, Eq. (13) produces a calculated temperature for the stationary state, case would not correspond exactly to
26.8 ˚C, that is very close to the experimental one. those (the equations) presented here.
In cooling experiment B2, the resistor is disconnected while maintaining water flow.
Therefore, this can be modeled by a modified Eq. (11), in which the term P, the power Experiences gained
of the electrical resistor, does not appear: by the students
dT The entire experiment, consisting
(M W Cp + M V Cp + M r Cpr )
W v
dt
= mCp W
(T
in − T) − UA ( T − Tam b ) (14 ) of two sessions lasting three hours
each, is conducted in pairs by the
Vol. 46, No. 4, Fall 2012 235
students. During the first session, the students observe the are low. The trickiest part of the apparatus is the electric re-
process without water circulation, whereas during the second sistance heater, which must be connected and disconnected at
one, the process with water circulation is studied. Students the right time. In order to prevent this part being broken due
gain ample practical experience, e.g., on measuring tempera- to forgetfulness, an electrical safety switch must be installed.
tures by thermocouples interfaced with a PC, on control and In their reports, students are asked to give an assessment of
measurement of flow rates. Most students find the module the experimental module both in terms of its pedagogical
effective as an introduction to the concept of unsteady-state value and the operation of the equipment. Most students give
process, of which most of them have only theoretical back- very positive feedback. Finally, at the end of the course, each
ground knowledge. In addition to this, the concept of overall student is expected to make an oral presentation of the experi-
heat transfer coefficient (U) is introduced and its experimental ment on his/her own in front of lecturers and other students.
value is obtained during the experiments.
After the experimental part, the students, still working in REFERENCES
pairs, are expected to submit a report containing all the results 1. Felder, R.M., and R.W. Rousseau, Elementary Principles of Chemical
obtained including a discussion that compares experimental Processes, 3rd ed., Wiley, New York (2000)
data with those calculated using the theoretical equations. In 2. Himmelblau, D.M., Basic Principles and Calculations in Chemical
Engineering, 1st ed., Prentice-Hall, Inc., Englewood Cliffs, NJ, (1962),
this way they test the potential of theoretical models to predict 7th ed. with J.B. Riggs, (2003)
experimental results. Occasionally something is bound to go 3. Henley, E.J., and E.M. Rosen, Material and Energy Balance Computa-
wrong during experimentation (random fluctuations in the tions, Wiley, New York, (1969)
flow rate that is not constant during the experiment, errone- 4. Reklaitis, G.V., Introduction to Material and Energy Balances, Wiley,
New York, (1983)
ous measurement of the flow rate by the student, erroneous
5. Hougen, O.A., and K.M. Watson, Chemical Process Priciples- Part I.
temperature readings caused by improper positioning of the Material and Energy Balances, 1st ed., Wiley, New York, (1943)
thermocouples or by the magnetic stirrer that is not work- 6. Bullard, L.G., and R.M. Felder, “A Student-Centered Approach to
ing properly, etc.) and therefore the experimental data end Teaching Material and Energy Balances. 1. Course Design, Chem.
up not fitting the theoretical models perfectly. In this case, Eng. Ed., 41(2), 93 (2007)
students also learn the importance of handling and taking 7. Bullard, L.G., and R.M. Felder, “A Student-Centered Approach to
Teaching Material and Energy Balances. 2. Course Delivery and As-
care of the experimental details in order to obtain valid and sessment, Chem. Eng. Ed., 41(3), 167 (2007)
reliable experimental results that are predicted by theoretical 8. Condoret, J.S., “Teaching Transport Phenomena Around a Cup of
models. Regarding safety aspects, the experimental set-up is Coffee,” Chem. Eng. Ed., 41(2), 137 (2007)
very simple and safe, without apparent danger in operation 9. Luyben, W.L., “The Devil’s in the Delta,” Chem. Eng. Ed., 41(1), 7
(2007) p
for students. The product flowing is water and temperatures

236 Chemical Engineering Education