100% found this document useful (1 vote)
6K views392 pages

Plant Design and Economics For Chemical Engineers Solution Manual - Peters, Timmerhaus

The document is a solutions manual for problems related to plant design and economics for chemical engineers, focusing on batch and continuous reactor systems. It includes detailed calculations for reaction rates, glucose concentrations, and reactor volumes, along with cost estimations for different reactor configurations. Key findings include optimal reactor arrangements and their impact on glucose conversion rates.

Uploaded by

kathleenyadan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
6K views392 pages

Plant Design and Economics For Chemical Engineers Solution Manual - Peters, Timmerhaus

The document is a solutions manual for problems related to plant design and economics for chemical engineers, focusing on batch and continuous reactor systems. It includes detailed calculations for reaction rates, glucose concentrations, and reactor volumes, along with cost estimations for different reactor configurations. Key findings include optimal reactor arrangements and their impact on glucose conversion rates.

Uploaded by

kathleenyadan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 392

Solutions Manual

to
accompany

Plant Design and Economics for


Chemical Engineers
Fifth Edition

McGraw-Hill Science
PROBLEM 13-8
Correction: Add, the ethanol yield is 0.47 kg ethanol/kg of glucose consumed. The
correct unit for the rate coefficient, k, is 1/s, not kg/m3·s

a) The time for a batch reaction is given by Eq. 13-8. Since Ni/VR is the reactant
concentration, glucose in this case, the equation can be written as
cglucose,f j=g

θ= ∫ d(cglucose)/(rglucose) ≈ Σ
j=1
(∆cglucose)/(rglucose)j,average
cglucose,i

where cglucose,f is the final glucose concentration = (0.05*10) kg/m3, cglucose,i the
initial glucose concentration=10 kg/m3, ∆cglucose the incremental change of glucose
concentration in the numerical integration (a negative quantity), and g=(cglucose,f –
cglucose,i)/∆cglucose. Because the rate is a complicated form in concentration, a
numerical integration is recommended. Any acceptable numerical integration
routine may be used. Here, integration was performed with a spreadsheet using the
trapezoidal approximation. The ∆-increment size was decreased in successive
integrations until the result does not change significantly with increment size. The
term (rglucose)j,average = (rglucose,j-1 + rglucose,j)/2 for integration by the trapezoidal
method.
The reaction rate is given by
(
rglucose= –(1.53*10-3) 1 – (cethanol/93) )0.52((cglucose)(cS.cerevisiae)/(cglucose + 1.7))
The concentrations are related as follows:
j j
cglucose,j = cglucose,i + ∑∆cglucose cethanol,j = –(0.47)∑∆cglucose,j
j=1 j=1

j
cS.cereisiaei,j = 0.01 – (0.06)∑∆cglucose
j=1

The reciprocal of rate as a function of cglucose and the value of the integral
(which is time) as a function of cglucose are shown on the next page.
The result for these conditions is
θ = 16.0 h ANSWER
PROBLEM 13-8 (continued-1)
RECIPROCAL RATE vs. GLUCOSE CONCENTRATION

25

20

15
(-1/r), h/(kg/m3)

10

0
0 2 4 6 8 10 12
GLUCOSE CONCENTRATION, kg/m3

TIME vs. GLUCOSE CONCENTRATION

18

16

14

12

10
TIME,h

0
0 2 4 6 8 10 12
3
GLUCOSE CONCENTRATION, kg/m
PROBLEM 13-8 (continued-2)
b) The PFR will have a volume of 7 m3. Eq. (13-13a) can be written in terms of the
total volumetric flowrate V, assumed constant, and glucose concentration as
cglucose,f

VR = (V) d(cglucose)/(rglucose)
cglucose,i

The integral must be reevaluated with the new initial concentration of cS.cerevisiae
(=0.9 kg/m3). The result is a value of 2.23 h.
Thus, the flow rate needed is
V = VR/(2.23) = (7 m3)/(2.23 h) = 3.1 m3/h ANSWER

c) The volume of a single CSTR for this reaction, assuming that the feed
concentration of cS.cerevisiae the same as in part b), is obtained from Eq. (13-18) in the
volumetric flowrate/concentration form as
VR = V(cglucose,f – cglucose,i)/(rglucose,f)
where rglucose,f is the rate of glucose reaction at the conditions leaving the reactor.

VR = (3.1 m3/h)(0.5 – 10 kg/m3)/(–1.79 kg/m3·h) = 16.6 m3 ANSWER

From Fig. 13-15, choosing a glass-lined, jacketed, agitated reactor, the cost is
estimated to be
$ 120,000 ANSWER

For the case of three equal volume reactors, an iterative solution is required,
wherein reactor sizes are represented as rectangular areas on a reciprocal rate vs.
concentration graph. The outlet concentration is varied until the three reactor
volumes (rectangular areas on the graph) are equal. See Ex. 13-7. The result, as
shown on the graph on the following page, is that the value of the integral for each
of the three reactors is 0.90 h. Thus, the volume of each reactor is
VR = V*(0.90) = (3.1)(.90) = 2.8 m3
and the total volume of the 3 reactors = 3*2.8 = 8.4 m3 ANSWER
From Fig. 13-15, again choosing the glass-lined reactor, the cost is estimated to be
$ 50,000 each, or $150,000 for the three ANSWER
Even though the total volume is just half of that for a single CSTR, the cost
is greater for the three reactors. This is a consequence of the economy of scale for
such reactors. There are other factors, such as flexibility of multiple reactors that
would influence such a selection. If a material other than glass-lined was chosen,
the cost figures would be different, but the relative values would be very similar.
PROBLEM 13-8 (continued-3)

RECIPROCAL RATE vs. GLUCOSE CONCENTRATION

0.6

0.5
-RECIPROCAL OF RATE

0.4

0.3

0.2

Reactor 3 Reactor 2 Reactor 1


0.1

0
0 1 2 3 4 5 6 7 8 9 10
GLUCOSE CONCENTRATION
PROBLEM 13-9
This solution uses the reciprocal rate vs. concentration curve developed in part b) of
problem 13-8 and shown below. The term “best arrangement” is used to mean the
combination of the given reactors that gives the greatest degree of conversion for
the given reactor sizes and types, and the flowrate and conditions of part b),
problem 13-8.
Each of the reactor types and volumes suggested in parts a) through d), will
be matched with the reciprocal rate vs. concentration curve to find the lowest
concentration achievable.
This reaction is autocatalyzed, and the reaction rate initially increases as the
glucose concentration decreases, because the biomass concentration increases.
Eventually, the glucose concentration decreases sufficiently to cause the reaction
rate to decrease; the increase of ethanol concentration contributes somewhat to the
decrease also. Two guidelines are helpful. For PFR combinations, the order or
sequence of PFR reactors makes no difference, because the reactor volume is
proportional to the area under the reciprocal rate vs. concentration curve. So this
area and the resulting outlet concentration are fixed, regardless of the order. The
order of CSTRs is important, however, because the conditions at the reactor outlet
determine the reaction rate in the reactor. Starting at the initial glucose
concentration, as long as the reciprocal rate is decreasing (rate is increasing), the
conversion achieved with a given volume of CSTRs will be greater than that for the
same volume of PFRs, and the larger CSTR should be be used first. These
guidelines are illustrated by the following solution.
The following solution is based on the conditions for part b) of problem 13-
8. A graph of reciprocal rate vs. concentration is shown in the solution for problem
13-8, part b), and the value of the integral vs. concentration from that problem is
shown on the following page.

a) Eq. (13-13a) for PFR, in concentration terms, is


cglucose,f

VR = (V) d(cglucose)/(rglucose); thus, Intergal = VR/V = (2+1)/3.1 = 0.968
cglucose,i

For two PFRs in series in either order, the value of the integral is fixed, so the result
is independent of the arrangement. The only variable left is cglucose,f and this is read
from the abcissa of the graph on the following page for the ordinate equal to 0.968.

The result is, cglucose,f = 5.6 kg/m3 ANSWER


PROBLEM 13-9 (continued-1)
TIME vs. GLUCOSE CONCENTRATION FOR CONDITIONS OF 13-8-b

2.5

2
TIME, h (value of integral)

1.5

0.5

0
0 1 2 3 4 5 6 7 8 9 10
GLUCOSE CONCENTRATION, kg/m3
PROBLEM 13-9 (continued-2)
b) For CSTRs, Eq. 13-18 in concentration form, gives the volume is by
VR = V(cglucose,f – cglucose,i)/(rglucose,f)
A value of cglucose,f is assumed, the corresponding (1/rglucose,f) is found from the
graph below, and iteration is continued until the foregoing equation is satisfied.
Since sequence matters for the two unequal CSTR case, both combinations must be
tried to find the one that gives the lowest cglucose,f . The better combination, the 2-
m3 CSTR followed by the 1-m3, is shown on the graph. In this case, the first
reactor reduces the glucose concentration to 6.9 kg/m3 and the second to 5.3 kg/m3.
So the final result is, a bit lower than the value obtained with the 2 PFRs in part a).
cglucose,f = 5.3 kg/m3 ANSWER

For comparison, if the 1-m3 is placed first, cglucose,f = 5.5 kg/m3, so the result is
better with the bigger reactor first.

RECIPROCAL RATE vs. GLUCOSE CONCENTRATION

0.6

0.5
-RECIPROCAL OF RATE

0.4

0.3

0.2

0.1
1 m3 2 m3

0
0 1 2 3 4 5 6 7 8 9 10
GLUCOSE CONCENTRATION

c) If the choice is between the four reactors given, it is obvious that the 2-m3 PFR
and the 2-m3 CSTR should be used. From the above graph and the results in parts
a) and b), it is clear that putting the 2-m3 CSTR first, followed by the the 2-m3 PFR
PROBLEM 13-9 (continued-3)
will achieve the lowest value of cglucose,f . The value is obtained by starting with the
6.9 kg/m3 from the 2-m3 CSTR, finding the time on the first graph corresponding to
that concentration (0.7h), and then adding to that the space time of the 2-m3 PFR
(0.65 h) and then, from the same graph, finding the concentration corresponding to
the total time (1.35 h). The result is
cglucose,f = 3.7 kg/m3 ANSWER

Note that some might interpret the statement of this part to mean the use of one
PFR and one CSTR of equal, but unspecified, volumes, to reach the original goal of
95 % conversion. The result for this case is a volume of 3.3 m3 for each reactor.

d) If all four reactors available reactors are combined, it can be deduced from the
results above that it is best to put the 2-m3 CSTR first, the 1-m3 CSTR second and
to follow these with the 2 PFRs in either order. The result (obtained as in part c) is
cglucose,f = 1.2 kg/m3 ANSWER

e) The glucose conversion = [(cglucose,i – cglucose,f )/ cglucose,i](100), and the values


are as follows:
a) 44%
b) 47%
c) 63%
d) 88%
PROBLEM 13-10
Corrections: The heat of reaction is 89.98 MJ/kg mol CH3OH, not kJ/kg mol.
The symbol p i represents fugacity, in kPa, of component i, which equals the
partial pressure, pi = Pyi, in kPa, when the fugacity coefficient is 1.

a) At equilbrium, the numerator of the reaction rate expression is equal to zero, or,
Keq = (pCH3OH)/(pCO)(pH2)2 = P(yCH3OH)/(P3(yCO)(yH2)2)
The mol fractions can be expressed in terms of the conversion of CO, X, starting
with 1 mol of CO and 2 moles of hydrogen, as follows:
CO + 2H2  CH3OH
(1-X) (2-2X) (X) (moles). At any conversion, the total number
of mols is the sum and equals (3-2X). So the mole fractions are given by,
yCO = (1-X)/(3-2X); yH2 = (2-2X)/(3-2X); and yCH3OH = X/(3-2X)
Substituting these terms into the equilibrium expression gives,
Keq = [X/(3-2x)]/[P2(4(1-X)3/(3-2X)3)], or 4KeqP2 = X(3-2X)2/(1-X)3
This latter equation can be solved by entering the known values and iterating on X
to find the conversion at equilibrium.
The result is Xeq = 0.610 ANSWER

b) A packed-bed reactor is specified, and is best modeled as a plug-flow reactor.


The catalyst-to-feed ratio is given by Eq. 13-15. In order to find the ratio, the
integral of (dXiz)/(-ric) must be evaluated between the limits Xio = 0 and Xiz = 90
percent of the equilibrium value from part a (0.9*0.61=0.549).
The reaction rate for CO is written as
-rCO = {[4P3(1 - X)3/(3-2X)3] - (P)(X)/[(3 - 2X)(3.18*10-7)]}
[23400 + 126(P)(1-X)/(3-2X) + (47)(P)(2-2X)/(3-2X)]2
Because the rate rate expression is a complex function of X, analytical integration
is impractical and a numerical evalution is easier. Any numerical integration
method can be used. Here, the value of –rCO has been calculated for values of X
incremented from 0 by a constant until X= 0.549 has been passed. The area under
the curve of 1/(-rCO) vs. X was obtained by the trapezoid rule, and the calulation
was repeated with a smaller increment size until the value of the integral does not
change significantly. Graphs of 1/(-rCO) vs. X and the value of the integral vs. X
are shown on the second following page. Interpolating for X= 0.549 gives a value
for the integral equal to 7.2.
By Eq. 13-15, this is the value of Wc/Fio, i.e.
Wc/Fio = 7.2 kg catalyst per kg mol CO/min ANSWER
PROBLEM 13-10 (continued-1)
c) For a methanol production rate of 50 metric tons/h, the CO feed rate to the
reactor must be
(50,000 kg CH3OH/h)(1 kg mol/32 kg)(1 kg mol CO/0.549 kg mol CH3OH)
(1h/60 min) = 47.43 kg mol CO/min. And therefore the reactor volume (filled
with catalyst) is
(7.2 kg cat./kg mol CO/min)( 47.43 kg mol CO/min)/(653 kg cat./m3 reactor vol.)
= 0.523 m3 reactor volume ANSWER

Because the large heat of reaction must be removed, it is expected that a


shell-and-tube reactor will be needed, with the catalyst packed inside the tubes.
The heat duty for the reactor is
(0.43 kg mol CH3OH/s)(89,980 kJ/kg mol) = 38,650 kJ/s
The tube number, length and diameter must be selected so as to provide
sufficient volume for the reaction and sufficient area for the heat transfer. This is
done as illustrated in Ex. 13-8. Using 0.0508-m diameter and 6.09-m tubes from
the example, the number of tubes needed is
(0.523 m3)/(0.00956 m3 per tube) = 55
The available surface area is then,
(0.973 m2/tube)(55 tubes) = 53.5 m2
The heat transfer capability must be checked. Assume a ∆Tlog mean of 50ºC and a
heat transfer coefficient of 0.25 kW/m2٠K. The heat transfer rate is then
Q = (0.25)(53.5)(50) = 2140 kW= 669 kJ/s
Clearly in this case, heat transfer dominates the sizing of the reactor. The area, A,
required to handle the heat removal is
A = (38,650)/(.25)(50) = 3090 m2
and the number of tubes required is 3090/(0.973) = 3176. From Fig. 14-17, it is
found that the maximum size for a standard floating head shell-and-tube
exchanger is 1000-m2 area, with a base cost of $70,000. Therefore, three such
exchangers would be reasonable. With carbon steel, 690 kPa design pressure, and
an adjustment for tube diameter of ~1.75/0.91 (from Fig. 14-21) and 0.96 for tube
length, the base cost is $70,000 for each exchanger.
The total cost of the exchangers is estimated to be
(3)($70,000)(1.75/0.91)(0.96) = $388,000 ANSWER
It would be expected that the same amount of catalyst as calculated in part
b) would still be used. It would be mixed with sufficient inert catalyst support to
pack the tubes full.
PROBLEM 13-10 (continued-2)
R a t e I nt e gr a t i o n

45

40

35

30

25

20

15

10

0
0 0.1 0.2 0.3 0.4 0.5 0.6

x , c onv e r s i on

Value of integral

5
Integral

0
0 0.1 0.2 0.3 0.4 0.5 0.6
x, conversion
PROBLEM 13-11
The time for a batch reaction is given by Eq. 13-8. Since Ni/VR is the reactant
concentration, such as lactose in this case, the equation can be written as
clactose,f j=g

θ= ∫ d(clactose)/(rlactose) ≈ Σj =1(∆clactose)j/(rlactose)j,average
clactose,i

where clactose,f is the final lactose concentration = 20 kg/m3, clactose,i the initial
lactose concentration = 50 kg/m3, and g = (clactose,i – clactose,f)/∆clactose; ∆clactose is
negative. Because the rate is a complicated form in concentration, a numerical
integration is recommended. Any acceptable numerical integration routine may
be used. Here, integration was performed with a spreadsheet using the trapezoidal
approximation. The ∆-increment size was decreased in successive integrations
until the result does not change significantly with increment size. The term
(rlactose)j,average = (rlactose,j-1 + rlactose,j)/2 for integration by the trapezoidal method.
The reaction rate is given by
(
rlactose=–(0.901) 1+(cpropionic acid/4.4214) )-1((clactose)(cP.acidipropionici)/(clactose + 32.5))
The concentrations are related as follows:
j j
clactose,j = clactose,i + ∑∆clactose; cpropionic acid,j = –(0.307)∑∆clactose;
j=1 j=1

j
cP.acidipropionici,j = 0.1 – (0.145) )∑∆clactose
j=1

The rate as a function of clactose and the value of the integral as a function
of clactose are shown on the next page.
The result for these conditions is
θ = 81.4 h ANSWER

The concentrations of lactose, P. acidipropionici and propionic acid are


shown vs. time in the graph on the second next page.
The required reaction time can be substantially reduced by increasing the
initial concentration of biomass, P. acidipropionici. For example, if the initial
concentration is 1 kg/m3, rather than 0.1 kg/m3, the time required is 44.4 h.
PROBLEM 13-11 (continued-1)
RATE VS. LACTOSE CONCENTRATION

0.6

0.5

0.4
-RATE, kg/h-m3

0.3

0.2

0.1

0
0 10 20 30 40 50 60

LACTOSE CONCENTRATION, kg/m3

VALUE OF INTEGRAL

90

80

70

60

50
TIME, h

40

30

20

10

0
0 10 20 30 40 50 60
3
LACTOSE CONCENTRATION, kg/m
PROBLEM 13-11 (continued-2)

CONCENTRATIONS VS. TIME

60

50
CONCENTRATION, kg/m3

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90

TIME, h

Lactose P. acidipropionici Propionic acid


PROBLEM 13-12
Clarification: The third sentence should read, “In the proposed recovery process,
each kg of catalyst-containing ceramic is mixed with two kg of aqueous sodium
cyanide solution, and the Pt is dissolved by complexing with the cyanide.”

a) Assuming complete Pt recovery and a 90% operating factor, the annual


production rate is
= (12 blocks/h)(0.05 kg-Pt/block)(24*365*0.9 h/y) = 4,730 kg-Pt/y

This small rate indicates use of a batch reactor. See Ch. 4.


ANSWER

b) There is no effect of pressure on the reaction, it is liquid phase. So, ambient


pressure, or a pressure somewhat lower to prevent cyanide leakage, is
recommended.
The temperature is selected to achieve the specified reactor space time of 4 h.
Using Eq. 13-10, with θ in s.

Xie Xie

θ = (Nio/VR)∫[dXi/(-ri)] = cio∫{dXi/[(cio)(k2)(1-Xi)]} = (1/k2)ln[1/(1-Xie)]


0
0

= 4*3600 = {ln[1/(1-0.9999)]}/[(5*107)exp(-13720/T)]

from which,
exp(-13720/T) = ln(1/0.0001)/[(5*107)(4*3600)] = 1.28*10-11

-13720/T = -25.08, and T = 547 K ANSWER

c) The reactor volume, for a space time of 4 h, is given by Eq. 13-7 as


VR = (Fi*νiF)θ
where (Fi*viF) is equal to the average volumetric flowrate of the reactor feed.
This is the flowrate of catalyst blocks plus cyanide solution. Assuming the
volume of this mixture is 0.003 m3/kg of ceramic catalyst (approximated as 1 kg
of ceramic per 2 kg of solution, thus ~3 kg/kg of ceramic), the reactor volume is
VR = (12 blocks/h)(5 kg ceramic/block)(0.003 m3/kg ceramic)(4h)
= 0.72 m3 ANSWER
PROBLEM 13-12 (continued-1)
d) The reactor must contain a hot solution of sodium cyanide/hydrocyanic acid,
as well as the abrasive ceramic material of the catalyst block. Either a steel or
glass-lined reactor might be used. Steel has the problem of being somewhat
subject to corrosion by weak acids, while a glass-lined vessel has the problem of
being subject to scratching and erosion from the ceramic catalyst support. Costs
are estimated from Fig. 13-15 to be
Glass-lined-reactor cost = $24,000
Steel reactor cost = $10,000 (TENTATIVE ANSWERS)
Further study, including more detailed corrosion and erosion information and an
alternative investment comparison, would be needed to make a definitive
recommendation.

You might also like