Louisiana State University

LSU Scholarly Repository

Honors Theses Ogden Honors College

11-2022

Comparison of Two Capital Cost Assessment Methods for A

Styrene Process Design
Marian K. Luzier

Follow this and additional works at: https://repository.lsu.edu/honors_etd

Part of the Chemical Engineering Commons

Recommended Citation
Luzier, Marian K., "Comparison of Two Capital Cost Assessment Methods for A Styrene Process Design"
(2022). Honors Theses. 884.
https://repository.lsu.edu/honors_etd/884

This Thesis is brought to you for free and open access by the Ogden Honors College at LSU Scholarly Repository. It
has been accepted for inclusion in Honors Theses by an authorized administrator of LSU Scholarly Repository. For
more information, please contact [email protected].
Comparison of Two Capital Cost Assessment Methods for A Styrene Process Design

by

Marian K. Luzier

Undergraduate honors thesis under the direction of

Dr. Adam Melvin

Department of Chemical Engineering

Submitted to the LSU Roger Hadfield Ogden Honors College in partial fulfillment of
the Upper Division Honors Program.

November, 2022

Louisiana State University

& Agricultural and Mechanical College
Baton Rouge, Louisiana
 2

Table of Contents
Chapter 1: The Objective for Chemical Engineering Students in a Process Design Course is
to Craft a Manufacturing Process. 3

Chapter 2: The Manufacturing Process Designed During the Fall 2022 Semester Produced
Styrene Monomer. 5

Chapter 3: A Capital Cost Assessment Comparison Was Completed to Determine the Best
Method to Use in Future Semesters. 11

Works Cited 27
 3

Chapter 1: The Objective for Chemical Engineering Students in a Process Design Course is
to Craft a Manufacturing Process.
Chemical engineering process design involves selecting and integrating a series of
processing steps in order to design a manufacturing process than produces a product at desired
specifications. [1] Process design endeavors give chemical engineers the opportunity to utilize
their creative skillset; these projects are often the most rewarding and satisfying tasks they will
undertake in their career. [2] At Louisiana State University (LSU), the process design course, CHE
4172 or “Plant Design”—as it is more commonly known to students, is a one-semester senior-level
course. The course is conducted five days per week with two parts—lectures and recitations.
Throughout the course of the semester, the students complete the design of an entire manufacturing
process. Each semester, a different manufacturing process is designed. Prior to the conclusion of
the semester, the process design is presented to industry judges. Students work in teams of four to
complete the process design project. Throughout the design process, students gain critical
experience with various aspects of the design process, such as equipment sizing, economic
analysis, utility optimization, and capital cost assessment. Alongside technical skills, careful
participation in such a project gives students the opportunity to heighten interpersonal skills such
as communication, presentation skills, technical writing skills, and problem solving. CHE 4172 is
unique in comparison to courses in other disciplines in that there is no solution in mind when the
problem is presented. While students are given the necessary tools to achieve the objective, they
are empowered to brainstorm and problem solve to present a viable solution. Each semester, an
industrial partner is selected to familiarize students with the manufacturing process chosen for
design. The industrial partner operates a similar process in their own facility. This partner offers a
presentation detailing the process as it operates in their own facility and, if possible, a facility tour
to the students. These interactions with the industrial partner occur early in the semester so students
have a better understanding of the process design they will complete. This industrial partner also
participates in the project judging at the end of the semester.
Throughout the semester, various engineering topics are covered in the lecture section of
the course. Some of these topics include process design strategy, separations, environmental
considerations, ethics, piping, heat exchanger design, and pressure relief device (PRD) design.
Students are tested frequently on course material and design progress through weekly quizzes and
recitations; this ensures that students are on track with the course material and design. Basics of
the design and key parameters are given in the Project 1 Assignment Document—provided by
LSU Professional in Residence Barry Guillory. [3] No two teams’ design will look exactly alike
as teams may choose differing equipment operating specifications, cross exchange networks, and
design alternatives to minimize capital and operating costs. Designs are completed using Aspen
Plus V12 Process Simulation software—or “Aspen” as it is commonly known. Aspen is a computer
software that quantitatively models a manufacturing process; it has various functions such as
process simulation, optimization, design specification, sensitivity analysis, and economic analysis,
amongst others. [4] Aspen is used commonly in the LSU Chemical Engineering curriculum
because it offers an approach to process modeling that is quicker and more realistic than typical
written calculations. At the beginning of the semester, students are tasked with familiarizing
themselves with the type of manufacturing process chosen. This is usually accomplished through
analyzing literature detailing the importance of the process, available commercial processes, safety
concerns, and design considerations. Students also calculate the required feed rate for the desired
product production rate so this can be later implemented in their Aspen Plus simulation.
 4

Throughout the semester, more challenging topics are covered that are appropriate to the level of
the design and offer additional opportunity for optimization.
As well as completing quizzes and recitations, each team submits three reports. The first
progress report for the Fall 2022 offering of CHE 4172 was due on October 17th, 2022. This report
consists of the basics of the design. Here, students are tasked with completing an operating, error-
free process design that meets the desired product purity specification and overall reactant
conversion. Waste disposal costs and by-product credits should also be evaluated in this report.
For this report, students complete an economic assessment which assesses aspects of the design
such as the raw material cost, the total capital investment (TCI), utility costs, and yearly product
sales. A discounted cash flow table is used to determine the product price. For the second report,
more thorough design is needed in order to decrease the capital cost and offer a more competitive
product price. Students explore design alternatives to improve the profitability of their system—
hoping to offer the product at a more competitive price. A safety analysis is also completed on the
design for the final report.
As a component of the economic analysis, a capital cost assessment is conducted to
determine the TCI of the designed process. The TCI includes all of the costs to build a new facility,
such as the costs for equipment purchase and installation, spare equipment costs, initial catalyst
costs, costs for plant startup, and working capital. [5] Historically, the capital cost assessment has
been completed using LSU Professor Armando Corripio’s Equipment Costs Excel Spreadsheet.
This spreadsheet includes cost calculations for trayed columns, packed columns, heat transfer
equipment, vessels, reactors, compressors, expanders, and pumps. The spreadsheet utilizes
calculations from Product and Process Design Principles by Seider, W.D., Seader, J.D., Lewin,
D.R., and Wildago, S.; Conceptual Design of Chemical Processes by J.M. Douglas, and Capital
Cost Estimating by K.M. Guthrie. [6] Students use the costs calculated from this spreadsheet in
their intermediate progress reports, final report, and final presentation. In the semesters preceding
Fall 2022, students and faculty often received feedback from the industrial judges that the
equipment costs and therefor capital cost assessments were largely undervalued—often times as
much as 10 times less than industry values. This problem has continued to progress as there is no
intermediate feedback from industry professionals to aid students in fixing their capital cost
assessments prior to the final presentation. Due to the competitive nature of the industry, prices
for similar equipment are rarely or never available online for students to access and include in their
assessments. These two situations hinder students from checking their capital costs assessments
prior to the final presentation.
Aspen Plus V12 APEA software is a viable alternative to the traditional spreadsheet
method that has been used. The APEA software easily calculates equipment and utility costs and
indexes them to the current year. [3] During the Fall 2022 semester, the CHE 4172 utilized this
method to calculate the capital costs for their design projects. This thesis will evaluate the
differences in the two capital cost assessment methods—Method 1: Aspen APEA Economic
Analysis and Method 2: Equipment Costs Excel Spreadsheet. It is expected that Method 1: Aspen
APEA Economic Analysis will give a more appropriate capital cost assessment due to its rigorous
sizing ability, individualized material specification factors, and updated information database. The
Product and Process Design Principles text provides additional reasoning as to why one should
expect Method 1 to provide better cost estimates. It concludes that Method 1 will provide more
accurate estimates as Method 2’s cost estimates are based on correlations formulated from data
that is only accurate to ±25%. [5] Providing students with a tool that provides accurate capital cost
assessments is crucial as it will allow them to produce an accurate economic analysis for their
 5

process designs. Furthermore, when students produce accurate capital cost assessments they will
complete their senior-level courses with the confidence and understanding to recognize feasible
capital cost assessment values. Providing accurate cost assessments is crucial as it allows industry
judges to be confident in LSU’s ability to produce competent chemical engineers.
Recommendations will be provided as to which method will be best to move forward with in future
semesters.

Chapter 2: The Manufacturing Process Designed During the Fall 2022

Semester Produced Styrene Monomer.
For the Fall 2022 semester, LSU CHE 4172 students designed a styrene
monomer manufacturing process. As described in the Project 1 Assignment
Document, the styrene is to be provided to a hypothetical polystyrene unit
within the same facility. [3] Styrene is an important product in the chemical
industry and is most commonly found in plastics. [8] Styrene is crucial in the
production of “polystyrene, ABS, SAN, styrene–butadiene latex, SBR, and Figure 1. Molecular
unsaturated polyester resins.” [7] Styrene possesses many qualities that make Structure of a
it desirable such as its colorlessness, stability, and low cost. [8] Commercial Styrene Molecule:
processes for styrene were developed in the 1930s, but World War II sparked This figure was
the large-scale production of styrene due to the large amounts of styrene- reproduced from the
National Center for
butadiene rubber needed to support the war efforts. [8] Styrene production has
Biotechnology
been steadily increasing since then with significant demand in South America, Information. [9]
Eastern Europe, and the Middle East. [7] A styrene molecule is shown in
Figure 1 and contains a benzene ring and is made entirely of carbon and hydrogen atoms. [9]
𝐸𝐵𝑍 ⇋ 𝑆𝑇𝑌 + 𝐻+ Rxn. (1)
𝐸𝐵𝑍 → 𝐵𝑍 + 𝐶 = Rxn. (2)
𝐸𝐵𝑍 + 𝐻+ → 𝑇𝑂𝐿 + 𝐶𝐻1 Rxn. (3)
Styrene is produced from a feed of ethylbenzene (EBZ) by three reactions. The first reaction
produces the product, styrene (STY), and hydrogen (𝐻+ ). [3] Two side reactions (rxns. 2 and 3)
produce by-products benzene (BZ), ethylene (C=), toluene (TOL), and methane (𝐶𝐻1 ). [3] The
design will be optimized to achieve an EBZ conversion of 65% while reducing the amount of by-
products produced. [3] STY is produced as the main product, but there is opportunity to utilize
some by-products advantageously.

Economics
Designing a chemical manufacturing plant requires exploring every option to save money and
increase profits while also accounting for safety and environmental protection. The prices for feed,
product, and by-products as given in the Project 1 Assignment Document are shown in Table 1.
[3] The polystyrene facility to which the styrene is provided currently purchases their styrene from
another supplier for 0.70 $ per pound. [3] The objective of the process design project was to design
a process that can offer styrene at a more competitive price. The purchase price of EBZ and
injection steam is also shown in Table 1. As specified in the Project 1 Assignment Document,
unpurified hydrogen can be sold as a by-product and combustible gas wastes (containing only
carbon, hydrogen, and oxygen) can be sent to an on-site steam boiler in exchange for a fuel credit.
[3] Any waste streams containing greater than 10 weight% water must be disposed of properly,
incurring a cost that is a combination of the waste water processing, organic separations, and
organic disposal costs. [3] As shown in Table 2, there are twelve utilities that can be used to
 6
Chemical Species: Price:
Styrene 0.70 $/lb
Ethylbenzene 0.55 $/lb
Injection Steam 0.004 $/lb
Hydrogen 4.00 $/1000 SCF
Combustible Gas Wastes Fuel Gas Credit
Waste Water Processing Cost 41$ / 1000 m3
Organic Separations Cost 0.15 $/lb
Organic Disposal Cost 3.51 $/GJ
Table 1. Economic Costs for the Styrene Process. This data in this table was reproduced from the Project
1 Assignment Document. [3]

operate the equipment. The most economical utility that will achieve the objective should be
chosen to lessen overall utility costs for the proposed process. An Aspen simulation must be
completed for the process prior to economic evaluation. Details of the Aspen process design
simulation will be described in the next two sections and the economic evaluation will be presented
in Chapter 3.

Table 2. Utility Costs and Properties for the Styrene Process. This table was reproduced from the
Report 1 Assignment supplied by Mr. Guillory to the CHE 4172 class. [3]

The Process
The styrene process as modeled in Aspen is shown in Figure 3. The process consists of two mixers,
two plug flow tubular catalytic (PFTR) reactors, one flash drum, two pumps, two compressors,
two distillation columns, three furnaces, three heaters/coolers, and two HeatX heat exchangers.
The annual production rate of styrene is 525,742 metric tons/year which requires 559,338 metric
tons/year EBZ and 186,244 metric tons/year of high-pressure injection steam. Unreacted EBZ is
recycled to decrease raw material costs to the system. A mixer (MMEIF) is used to mix the
recycled and fresh EBZ. The injection steam is mixed with the fresh and recycled EBZ feed in
another mixer (ME1F). The injection steam works to heat the feed to achieve the intended
conversion. The mixed streams then pass through a heater (E1XA), a heatX heat exchanger
(E13X), and a furnace (E1XB) in order to be heated to a high enough temperature prior to entering
the reactors. The EBZ is then reacted in two reactors (R1 and R2) to produce styrene and other by-
 7

E7 KC1
E3 E4
MME1F
KC2
E1XA E1XB E2
C1
E13X R1 R2 D1 P1
ME1F C2
E6

PC2B
Figure 3. Styrene Process as Modeled in Aspen Plus. This figure was reproduced from the Aspen
simulation created by our team.

products. E2 is an intermediate furnace that heats the stream to reach the desired EBZ conversion
in R2. The outlet of the second reactor is then cross-exchanged with the inlet to the reactors for
cooling. Two more heat exchangers (E3 and E4) are used to cool this stream prior to entering the
flash drum (D1). D1 is used to separate the lighter components (H+ , C =, and CH1 ) from the other
components. The lighter components are sent to an on-site steam boiler for a fuel gas credit which
requires them to leave the process at greater than or equal to 25 psia in order to gain the fuel credit.
[3] A compressor is shown on this stream (KC2) that is not necessary for the process as the stream
is at 30 psia. This is included for capital cost comparison carried out in this thesis because
preliminary results indicated that Method 2 may not correctly cost compressors of smaller and
larger sizes (data not shown). Water is also separated from the other components in D1 and is
treated as waste water. A pump (P1) is not necessary for this stream but is included for capital cost
comparison purposes. This additional analysis is necessary because pumps and compressors are
sized similarly in Method 2 and there is reason to believe that Method 2 may also incorrectly cost
pumps of smaller and larger sizes. The bottom stream from D1 (D1B) contains mainly STY, EBZ,
TOL, and BZ. This stream is heated in heat exchanger E6 and is then sent to the first column (C1).
In C1, TOL and EBZ are the light and heavy components, respectively. A majority of the TOL
and BZ are produced in the distillate of C1 while the STY and EBZ are sent to the second column
(C2). The distillate of C1 is heated and compressed in order to be sent to the on-site steam boiler
for a fuel credit. C2 separates EBZ and STY with EBZ leaving in the distillate to be recycled as
feed. A pump is utilized on the bottoms of C2 to achieve the appropriate pressure prior to being
sold as product. The subsequent section will offer more detailed analysis of the process used to
design each type of equipment in the Aspen simulation. Pertinent technical details, including
design assumptions, will also be provided.

Detailed Design
The purpose of this section is to detail the design processes for each type of equipment and present
some of the important technical aspects of the design. This section will discuss how each piece of
equipment was designed in the Aspen software.

Reactors-R1, R2
The adiabatic, catalytic, PFTR reactors are designed to reach a 65% conversion of EBZ overall,
and 35% EBZ conversion in the first reactor. [3] The reactor diameter for each reactor is specified
(6 ft), but the Aspen Plus Design Specification tool (commonly known as a design spec) is utilized
for each reactor to determine the reactor length needed to reach the intended conversion. [3]
 8

Further analysis is completed with a sensitivity analysis for each reactor. Each sensitivity analysis
varies the pressure and temperature in each reactor. This works with the design specs to determine
the optimal reactor length for each temperature and pressure combination. Reactors should be
designed to have a length less than 50 feet, so the user chooses the best temperature and pressure
from the sensitivity analysis results that meets this criteria. [3] The reactor pressure is set to this
selected pressure and the heat exchanger, prior to each reactor, heats the inlet to the selected
temperature. The reactor pressures, lengths, and inlet temperatures are shown in Table 3.

Reactor Inlet Temperature (°F) Reactor Pressure (psia) Reactor Length (ft)
R1 1142.2 3.4 38.2
R2 1200.0 3.0 47.8
Table 3: Designed Characteristics for Reactor Blocks in Aspen Plus. The data in this table was collected
from the Aspen simulation created by our team.

Distillation Columns-C1, C2
Designing distillation columns in Aspen is most easily begun by using
DSTWU blocks. A DSTWU block is a shortcut distillation modeling
method that uses the Winn, Underwood, and Gilliland correlations to
provide the user with details for more rigorous design. [10] DSTWU
blocks can only be used for columns with one feed stream and two
product streams. [10] The user inputs the heavy key component, light
key component, and desired purities. A -1.2 reflux ratio is chosen as an
initial value in Aspen where the -1.2 reflux ratio indicates a reflux ratio
1.2 times the minimum reflux ratio. Condenser and reboiler pressures
are also chosen. The DSTWU is simulated with the results including
the number of stages, actual reflux ratio, distillate to feed (DtoF) ratio,
and feed stage among others. These results are transferred to a RadFrac
block for more rigorous design as Radfrac blocks utilize rigorous
fractionation calculations for rating and design. [10] Design specs are Figure 4 Aspen
added to the RadFrac block to specify the purities for the heavy and Hydraulic Analysis of
light key components. Once the intended purities are achieved, the Column 2 Internals. This
Aspen Column Internals tool is used to specify the tray or packing type, figure was reproduced
column diameter, and other internal column specifications. The Aspen from the Aspen Plus
Auto Section tool is used to section the column. Column parameters model created by our
team.
are adjusted until the hydraulic plots across the column are completed
without errors or warnings. As shown in Figure 4, a column which has internals that are
hydraulically suitable will show all stages as blue. This approach was used to design each of the
distillation columns with column 1 being designed prior to column 2.

Column Reflux Ratio DtoF Ratio Number of Feed Stage Condenser Reboiler
Stages Pressure Pressure
(psia) (psia)
C1 112.7 0.007 23 11 7 26
C2 20.0 0.343 111 61 15 21
Table 4: Designed Characteristics for Distillation Columns in Aspen Plus. The data in this table was
collected from the Aspen simulation created by our team.
 9

The designed parameters for both distillation columns are displayed in Table 4. Both columns 1
and 2 are designed as trayed columns with sieve trays. Each column has two sections. The overall
tray efficiency is chosen as 70% and the tray spacing for each section is chosen as 2 ft. [3] Other
characteristics of the column internals are displayed in Table 5. Both columns utilize the CW20
utility in the condenser and the STM160 utility in the reboiler to achieve the appropriate reflux
and boilup rates.

Column Section Start Stage End Stage Number of Passes Diameter (ft)
C1 1 2 10 1 23.4
2 11 22 2 17.1
C2 1 2 80 2 21.9
2 81 110 2 20.7
Table 5: Designed Column Internals for Distillation Columns in Aspen Plus. The data in this table was
collected from the Aspen simulation created by our team.

Pumps-P1, PC2B
Pump design in Aspen is completed by choosing the desired discharge pressure of the pump. P1
discharges at 220 psia and PC2B discharges at 25 psia. All pumps are assumed to have 80% pump
efficiency and 95% driver efficiency. [3] These efficiencies are included in the Aspen pump
design. All pumps use the ELECT utility.

Compressors-KC1, KC2
Compressors are designed similarly to pumps needing only the discharge pressure to be specified.
KC1 discharges at 25 psia and KC2 discharges at 250 psia. All compressors are assumed to have
80% polytropic efficiency and 95% mechanical efficiency. [3] These efficiencies are included in
the Aspen compressor design. All pumps use the ELECT utility.

Flash Drums-D1
A flash drum (D1) is used to separate the water and lighter components from the STY, EBZ, TOL,
and BZ. H+ is the main component being removed from the drum and it can be sold as a by-product
without purification. Since purification is not necessary, distillation is not required and therefore a
flash drum can achieve the needed separation at a much lower cost. The flash drum is operated at
85 ºF and 30 psia and there is no pressure drop across the flash drum. The flash drum is assumed
to be 50% full of liquid at any time. [3] The residence time cannot be inputted directly into the
Aspen Flash Drum block, but it is utilized in both equipment costing methods. Consistent with the
default in Aspen APEA, the flash drum is chosen to have a residence time of 5 minutes.

Furnaces-E2, E1XB, E7
There are three furnaces in the styrene process, E2, E1XB, and E7. Furnaces are utilized when
Exchanger Inlet Temperature (°F) Outlet Temperature (°F) Pressure (psia)
E2 912.0 1200.0 2.4
E1XB 891.0 1142.0 3.3
E7 128.0 200.0 7.0
Table 6: Designed Characteristics for Furnaces in Aspen Plus. The data in this table was collected from
the Aspen simulation created by our team.
 10

heating at high temperatures is needed or when temperature crossover is a concern. Furnaces are
simple to design as only the outlet temperature and pressure need to be specified. The designed
outlet temperatures and pressures are shown in Table 6. The furnaces are modeled as heater blocks
in Aspen and use the FURNACE utility. E7 and E1XB are modeled as furnaces due to their high
temperatures. These furnaces are vital as they ensure that the stream reaches the correct
temperature prior to entering the reactors so the appropriate conversions can be achieved. E7 is
designed as a furnace so it can achieve the desired outlet temperature without creating any pressure
drop. E7 is essential as it heats up the stream to KC1 so this stream can be properly pressurized
prior to being sold for a fuel gas credit.

Heaters/Coolers-E1XA, E3, E4
There is one heater, E1XA, and two coolers, E3 and E4, in the styrene process. These
heaters/coolers must have straight heat duty-temperature (TQ) curves. [3] The heaters/coolers are
designed exactly like the furnaces with only the outlet temperature and pressure needing to be
specified. There is one additional step when designing heaters and coolers. After the inputs are
specified, the Hcurves tool in Aspen should be used to populate the TQ curves to determine their
linearity. If the TQ curve is like Figure 5, the heater/cooler needs to be separated into two separate
Aspen heater blocks with the first exchanger’s outlet temperature designated as the temperature at
which the two separate slopes meet (Figure 5, red circle). The two new heater/coolers should have
TQ curves like Figure 6. The inlet/outlet temperatures, pressures, and utilities of each of the

New intermediate
temperature

Figure 5. Example of a Non-linear TQ Curve. This figure was reproduced from

the Aspen Plus simulation created by our team.

Figure 6. Example of a Linear TQ Curve. This figure was reproduced from the
Aspen Plus simulation created by our team.
 11

heaters and coolers are presented in Table 7. The least costly utility should be chosen that can
accomplish the needed heat transfer.

Exchanger Inlet Temperature Outlet Temperature Pressure (psia) Utility

(°F) (°F)
E1XA 170.0 200.0 15.0 STM160
E3 230.0 172.0 30.0 CW10
E4 172.0 165.0 30.0 CW20
Table 7: Designed Characteristics for Heaters and Coolers in Aspen Plus. The data in this table was
collected from the Aspen simulation created by our team.

HeatX Heat Exchangers-E6, E13X

HeatX heat exchanger blocks provide a more rigorous approach to heat transfer because they can
estimate sensible heat and fouling factors. [10] Two HeatX blocks are used in the process, E6 and
E13X. Not only can HeatX blocks appropriately handle non-linear TQ curves, but HeatX blocks
can be used for cross exchange. Cross exchangers are a valuable resource that can save money in
utility and capital costs. Designing HeatX blocks requires more inputs than heaters, coolers, and
furnaces. When used in shortcut mode, the user specifies the hot inlet-cold outlet temperature
difference and the minimum temperature approach. The minimum temperature approach for all
HeatX blocks in the styrene process is 10 °F. [3] A constant overall heat transfer coefficient (Uo)
is also entered. E13X uses a Uo of 153.8 Btu/hr-sqft-R while E6 uses a Uo of 40.8 Btu/hr-sqft-R.
These values are estimated by the user based on the characteristics of the streams. E6 uses STM160
as the hot utility. The cold-side fluid is vaporized; the inlet temperature is 85 °F and the exit
temperature is 344 °F. E6 was modeled as a HeatX exchanger due to the vaporization that occurs.
E13X is a cross-exchanger. The cold-side fluid is heated from 200 °F to 891 °F; it enters as a
vapor-liquid mixture and it is completely vaporized in the exchanger. The hot-side enters as a
vapor at 1008 °F and exits as a vapor at 230 °F. HeatX blocks have the potential for very in-depth
design, having many options relating to exchanger geometry.

Chapter 3: A Capital Cost Assessment Comparison Was Completed to Determine the Best
Method to Use in Future Semesters.
A capital cost assessment is a crucial part of a manufacturing process design. This study will
compare two distinct capital cost assessment methods—Method 1: Aspen APEA Economic
Analysis and Method 2: Equipment Costs Excel Spreadsheet analysis. Capital cost assessments
entail various aspects of economic analysis, but the calculated sum of the installed costs of the
equipment is indicative of which capital cost assessment method is the most feasible. Therefore,
this thesis will compare the calculated installed equipment costs using each method. Method 1
computes installed equipment costs that are indexed to the current year, 2022 [3]. These values are
displayed in USD ($). It utilizes data directly from the Aspen simulation to calculate the installed
costs. The APEA tool automatically chooses to map each piece of equipment to a predesigned
format. Items can be remapped if different analysis is needed. Method 2 uses a CE inflation
index—or Chemical Engineering Plant Cost Index—factor to index equipment cost values to the
current year. [5] CE inflation index values are not widely published, but CPI values are. Since they
can be assumed to follow the same trend, CPI values are used to proportionally determine the CE
inflation index for 2022. Product and Process Design Principles: Synthesis, Analysis and
Evaluation gives the most recent economic values from 2013; the CPI value is 233 and the CE
 12

inflation index is 567. [5] The CPI index in September 2022 was 296.808. [11] This is proportional
to a CE inflation index value of 722.275. Method 2 is relatively simple to use—it requires the user
to model the equipment as a block in Aspen and use some of the results as inputs in the spreadsheet.
Method 2 uses correlations based on published equipment cost data. [5] These correlations use a
size factor that is unique to each type of equipment to determine the basic purchase cost of the
equipment. [5] The inputs required for these size factors may be heat duty, temperature, column
diameter, etc. depending on the type of equipment. Often factors are used for material type, design
type, design pressure, etc. [6] These factors are included in the spreadsheet. The base correlation
is based on common material specifications and equipment designs so these additional design
factors are used to adjust the base equipment cost for unique materials or equipment designs. [5]
Method 2 reports installed costs in thousand USD (k$) so each value will be multiplied by 1000
for comparison with the Method 1 values. Seven types of equipment were studied for capital cost
comparison—compressors, pumps, flash drums, heaters, HeatX heat exchangers, furnaces, and
distillation columns as described above. A recommendation regarding which method should be
used will be given for each type of equipment along with any other ancillary recommendations
that are discovered during the comparison. Two types of equipment shown in Figure 3 will not be
economically evaluated in this study—mixers and reactors. While mixers are shown as individual
blocks in Aspen, they do not represent physical equipment but rather the mixing of two streams in
piping. The reactors will not be evaluated due to the complexity of reactor sizing and evaluation.
Each reactor is unique and due to the various types
of reactors, costing one type of reactor would not
be representative of either method’s ability to
effectively cost all reactors.

Type 1: Compressors
Compressors are some of the easiest equipment to
evaluate economically using either method. The
styrene process requires one compressor—KC1—
to ensure the gases from C1 exit the system at a
pressure greater than 25 psia. Another compressor
was added to the system to study how the different
methods analyze compressors of different sizes.
This compressor is KC2 and is responsible for
raising the pressure of the D1 vapor stream to 250
psia. Using Method 1, APEA automatically maps
KC1 and KC2 to DGC CENTRIF- horizontal
centrifugal compressors. The driver power, gas
flow inlet rate, temperature, pressure, and other
factors are utilized from the Aspen simulation to
calculate the installed cost, as shown in Figure 7.
The user need only to specify the driver type for the
compressor—here MOTOR is chosen. For this
comparison, the casing material selection is not Figure 7. Method 1 Analysis for
pertinent, but the option is available if needed. Compressors. This figure was reproduced from
Using Method 2, both compressors are modeled by the APEA evaluation tool on the Aspen Plus
entering their total powers, in horsepower (HP), simulation created by our team.
 13

and compressor type factors. Both of these compressors are assumed to be centrifugal compressors
with motor drivers, so a design type factor of 1 is used. These calculations can be found in Figure
8. As shown in Table 8, the smaller compressor, KC1 (6.6 HP), is highly undervalued by Method
2 by as much as 90.7%. This is most likely because this compressor’s total power value is lower
than the range of total power values Method 2’s centrifugal compressor correlation is designed for.
KC1 has a total power of 6.6 HP while Method 2 is only designed to estimate costs for centrifugal
compressors that have total power values in the range of 200-30,000 HP. [5] Method 2 overvalues

Figure 8. Method 2 Analysis for Compressors. This figure was reproduced from the Equipment
Costs Excel Spreadsheet utilizing values from the Aspen Plus simulation created by our team. [6]

Equipment Name Installed Cost Installed Cost Over/Under

Method 1 [USD] Method 2 [USD] Evaluation (%)
KC1 $926,000 $86,000 -90.7%
KC2 $2,156,000 $3,394,000 +57.4%
Table 8: Installed Cost Comparison for Compressors. This table was creating using data from both the
Aspen APEA Economic Analysis Tool and the Equipment Costs Excel Spreadsheet.

the larger compressor, KC2 (2293.4 HP), by 57.4%. From this data, the conclusion is that Method
2 is inconsistent for compressors of different sizes. Therefore, the recommendation for
estimating the installed cost for compressors is to use Method 1: Aspen APEA. This option is
optimal because not only does it provide the most reasonable estimate, it is the easiest to use. In
order for Method 2 to be effective for compressors of all sizes, additional factors would need to be
calculated and implemented for compressors of small and large sizes.
 14

Type 2: Pumps
Pumps are modeled similarly to compressors, needing very few user inputs for either method.
There is only one pump needed for the styrene process—pump PC2B. This pump is used to ensure
the bottoms stream of C2
leaves the system at a
pressure greater than 25 psia
(similarly to KC1). This
pump is relatively small,
compared to P1, as it only
needs to raise the pressure
of the stream by 4 psi.
Another pump was added to
the system to study how the
different methods analyze
pumps of different sizes.
This pump was P1 and was
responsible for raising the
pressure on the water stream
from the flash drum (D1) by
190 psi. C1 and C2 are
reflux pumps that are also
shown in Figure 9 which
Figure 9. Method 1 Analysis for Pumps. This figure was reproduced
from the APEA evaluation tool on the Aspen Plus simulation created will be addressed later when
by our team. costing the distillation
columns. The liquid flow,
pressure rise, and liquid density are used to calculate the fluid head, power, and installed cost of
PC2B and P1. An efficiency of 80% is used for both methods. [3] Comparing the fluid heads for
PC2B and P1 in Figures 9 and 10, it is obvious that both methods use a similar process to calculate
the fluid head. Method 1 automatically maps the pumps to DCP CENTRF—a centrifugal single or

Figure 10. Method 2 Analysis for Pumps. This figure was reproduced from the Equipment Costs Excel
Spreadsheet utilizing values from the Aspen Plus simulation created by our team. [6]
 15

multi-stage pump. User inputs for Method 1 include the casing material, here cast iron (CI) is used,
and driver type, MOTOR is chosen. For Method 2, the material factor is chosen as 1 for ductile
iron and the pump type design factor is chosen as 1 for a single stage pump based on volumetric
flow and fluid head. While Method 1 only allows the user to specify the driver type as MOTOR,
Method 2 allows the user to decide between different types of motors as shown in Figure 10. For
PC2B and P1, the electric motor type factor is chosen as 1 for an enclosed, fan cooled motor. While
the base options for the pumps are the same across the methods, Method 2 does offer a more
individualized approach with more extensive pump and motor options.

Equipment Name Installed Cost Installed Cost Over/Under

Method 1 [USD] Method 2 [USD] Evaluation (%)
PC2B $54,100 $21,000 -61.2%
P1 $58,500 $23,000 -60.7%
Table 9: Installed Cost Comparison for Pumps. This table was creating using data from both the Aspen
APEA Economic Analysis Tool and the Equipment Costs Excel Spreadsheet.

As shown in Table 9, regardless of the size of the pump, Method 2 undervalues installed costs of
the pumps. P1 requires 14 times the power of PC2B and has a pressure rise that is 47.5 times
greater than PC2B. P1 does have a lesser volumetric flow, but
its fluid head is nearly 37 times as large as PC2B. Regardless,
Method 2 undervalues each pump by approximately 60%. Since
it is evident that both methods calculate the fluid head in the
same manner, it can be concluded that the difference in installed
costs is not due to this calculation. Due to the similarity in the
fluid head calculations and the consistency in the difference
between the two methods, it can be concluded that the
undervaluation by Method 2 is most likely due to outdated
costing methods and factors. Since it is known that costs have
been undervalued in previous semesters while using Method 2,
the recommendation for estimating the costs of installed
pumps is to utilize Method 1: Aspen APEA analysis. While
certain aspects of pump and motor design cannot be chosen with
this method, it offers a cost analysis closer to what is expected.
This method is also simpler, with only two inputs needed from
users. Method 2 includes the cost for a spare pump and motor;
it is not evident whether Method 1 includes the price for this
spare equipment.

Type 3: Flash Drums

There is only one flash drum needed for the styrene process—
D1. The Aspen APEA analysis for D1 is shown in Figure 11. Figure 11. Method 1 Analysis
Flash drums are automatically mapped to a DVT cylinder, or for Flash Drums. This figure
vertical process vessel in APEA. APEA automatically retrieves was reproduced from the APEA
the inlet and outlet volumetric flows to calculate the vessel evaluation tool on the Aspen
volume. The user need only to specify the material. If a certain Plus simulation created by our
fluid volume percent is needed, it can be adjusted here. A 50 team.
 16

fluid volume percent is chosen for comparison with

Method 2 which has a default of 50 fluid volume
percent. There are other design options in the interactive
sizing tool in the Aspen APEA analysis tool. To
determine what default parameters were used by Aspen
APEA to size and cost the equipment, the user must first
size and evaluate the equipment costs in APEA. Next,
the user should change the last row in Figure 11, “Allow
Resize”, to “No”, then choose the interactive sizing tool.
The default sizing parameters used for the sizing and
evaluation will be displayed in the interactive sizing
window. The interactive sizing tool for flash drums
displays default parameters such as the inlet streams,
outlet streams, residence time, diameter, and process
vessel height to diameter ratio, amongst others. The
Figure 12. Method 2 Analysis for Flash default residence time is 5 minutes, or 0.083 hours. The
Drums. This figure was reproduced from default outlet stream is set to D1B. The calculated
the Equipment Costs Excel Spreadsheet capacity using these parameters is also shown in the
utilizing values from the Aspen Plus interactive sizing tool. Changing or adding outlet
simulation created by our team. [6] streams in the interactive sizing tool does not seem to
have any effect on the calculated capacity. Method 2
requires a few additional inputs. The liquid volumetric outlet flow can be found by adding the
volumetric flows of streams D1B and D1W. The liquid residence time and length-to-diameter ratio
are chosen by the user. The appropriate factors are then chosen for a carbon steel vessel that
operates at 30 psia. This analysis is shown in Figure 12. Comparing the two methods, Method 2
calculates a vessel volume that is roughly 100 cubic feet (ft3) larger. This discrepancy is most
likely due to a difference in how the inlet/outlet flows are used to calculate the liquid volume.
While the user can directly enter the outlet flow to be used in Method 2, there is no way to see
what flows Method 1 uses to size the vessel. Method 1 boasts a liquid volume nearly twice that of
Method 2. A quick calculation reveals that a drum with an outlet flow of 4483.8 ft3/hr and a
residence time of 5 minutes would create a liquid volume of 373.5 ft3 at any time—the same liquid
volume as calculated with Method 2. The reasoning as to why Method 1 calculates a liquid volume
nearly two times the expected amount is currently unknown. The differences in calculated liquid

Equipment Name Installed Cost Installed Cost Over/Under

Method 1 [USD] Method 2 [USD] Evaluation (%)
D1 $236,600 $169,000 -28.6%
Table 10: Installed Cost Comparison for Flash Drums. This table was creating using data from both the
Aspen APEA Economic Analysis Tool and the Equipment Costs Excel Spreadsheet.

volume are apparent in the installed costs for the drum as shown in Table 10. While Method 2
does estimate the installed cost as almost 30% lower than Method 1, the recommendation for
costing flash drums is to continue to use Method 2 until a better understanding of Method 1
 17

is gained. Without knowing which flows

Method 1 uses to calculate the fluid volume of
the flash drum, it is difficult to be confident in
Method 1’s sizing ability.

Type 4: Heaters and Coolers

Heater blocks (used to model both heaters and
coolers) are utilized three places in the styrene
process design. These heaters are automatically
mapped to TEMA shell and tube exchangers in
Method 1. The heat transfer area is calculated in
APEA by a set of parameters retrieved from the
heater block. Other than choosing the shell and
tube side materials, no other user inputs are
needed to complete the costing by Method 1.
Default parameters related to the sizing of the
heat exchangers can be found in the interactive
sizing tool in APEA. The interactive sizing tool
for heaters displays default parameters such as
the overall heat transfer coefficient (Uo),
overdesign factor, tube thickness, tube pitch, and
tube outside diameter, amongst others. Many of
these parameters can be changed by the user if
needed. The final surface area calculated by
Figure 13. Method 1 Analysis for Heaters. This
APEA is also shown in the interactive sizing
figure was reproduced from the APEA evaluation window. As shown in Figure 14, costing heaters
tool on the Aspen Plus simulation created by our using Method 2 is more complex. Users must
team. input the heat duty calculated in the Aspen Plus
heater block (in BTU/hr), the hot side
inlet/outlet temperatures, and the cold
side inlet/outlet temperatures. For
hot/cold sides where a utility that
changes temperature is utilized, it is
assumed that the utility completely
reaches its exit temperature. A default
overall heat transfer coefficient value
must be chosen by the user based on
the characteristics of the process
fluids. There are estimated values
from which the user can choose in the
Method 2 spreadsheet. Also, material,
design, and pressure factors must be
selected by the user. Comparing the
heat transfer areas calculated by both
Methods 1 and 2, it is evident that
Figure 14. Method 2 Analysis for Heaters. This figure was Method 1 calculates a much greater
reproduced from the Equipment Costs Excel Spreadsheet
utilizing values from the Aspen Plus simulation created by our
team. [6]
 18

heat transfer area than does Method 2. This calls into question the heat transfer parameters used in
each calculation. It is clear which overall heat transfer coefficients were used in Method 2 as these
were user-selected. Determining which overall heat transfer coefficient values were used in

Method 1 is more difficult as Aspen determines the default heat transfer parameters used; these
parameters can be located in the interactive sizing tool, as discussed prior. The comparison of the
overall heat transfer coefficients chosen by Aspen in Method 1 and the ones selected by the user
in Method 2 are displayed in Table 11. The user-selected overall heat transfer coefficient values
from Method 2 are larger than those selected by Aspen. For comparison, the exchangers will be

sized and evaluated using both sets of parameters. The heat transfer areas for each heater block
using each analysis method and each method’s overall heat transfer coefficient values are shown
in Table 11. The installed costs for each heater block as calculated using Method 1’s overall heat
transfer coefficients are shown in Table 12. The installed costs for each heater block as calculated
using Method 2’s overall heat transfer coefficients are shown in Table 13. Analysis shows that
Method 2 estimates the installed price for the equipment 5-89% higher than Method 1. The
installed costs calculated from the two methods have the greatest difference for E3, the exchanger
with the largest heat transfer area. When both methods are evaluated with consistent overall heat
transfer values, the calculated heat transfer areas are similar. This indicates that both methods are
 19

accurate in determining the heat transfer area. Method 1

calculates heat transfer areas that are slightly higher which is
most likely due to an overdesign factor. An overdesign factor
is a contingency factor added to a design calculation to
account for uncertainties in a process design. [1] The benefit
of using Method 1 in costing heaters and coolers is that a
more accurate overall heat transfer coefficient is used for
costing. One potential issue with Method 2 is the assumption
that each utility fully reaches its potential exit temperature.
To determine the actual exit temperature, users would need
to do calculations by hand which can be a tedious and
confusing process. Users should be discouraged from using
heater blocks for several reasons. The costing for heater
blocks compared to other types of equipment has the most
uncertainty. While Method 1 calculates larger heat trasfer
areas in each scenario, its calculated installed prices are
consistently less than those using Method 2. This indicates
that there is a major discrepancy in the way these costs are
calculated. Heater blocks should also be avoided because
they are only effective and efficient when their TQ curves
are straight. This is a quality often forgotten by users. If
heater blocks must be used, the recommendation for
Figure 15. Method 1 Analysis for costing heaters and coolers is to use Method 1: Aspen
HeatX Heat Exchangers. This figure APEA. A better alternative will be discussed in the next
was reproduced from the APEA section—HeatX heat exchangers.
evaluation tool on the Aspen Plus
simulation created by our team.
Type 5: HeatX Heat Exchangers
HeatX heat exchanger blocks are utilized in two places in the
styrene process—prior to the reactors as a
cross exchanger (E13X) and on the feed to
C1 (E5). HeatX blocks are unique when
compared to heater blocks because the block
itself calculates the area. This area is then
used by both methods to calculate the
installed cost for the exchanger. Method 1
maps these cross exchangers to TEMA shell
and tube exchangers. As shown in Figure 15,
the user need only to input the material type
for the shell and tubes. The heat transfer area
Figure 16. Method 2 Analysis for HeatX Heat used for the costing in Method 1 is exactly
Exchangers. This figure was reproduced from the the heat transfer area calculated by the HeatX
Equipment Costs Excel Spreadsheet utilizing values
block without incorporating an overdesign
from the Aspen Plus simulation created by our team.
[6] factor in the costing process. As shown in
Figure 16, Method 2 follows a similar process. The heat transfer area is retrieved from the HeatX
block and utilized in the spreadsheet to calculate the installed cost. The user then determines the
material, design, and pressure factors. One of the most notable differences in these approaches is
 20

the number of shells calculated. Method 2 calculates the number of shells as 3 for both exchangers
while Method 1 does not distinguish the number of shells. As shown in Table 14, there is a

substantial difference in the cost estimates from the two methods. Method 2 calculates an installed
cost that is nearly twice as large as the estimate from Method 1. This is most likely due to the
number of shells calculated by Method 2. When a heat exchanger sized by Method 1 is specified
to have 3 shells, the price is significantly higher—over 1 million USD. Method 2 gives
unreasonable estimates for
HeatX exchangers, so the
recommendation for
costing HeatX heat
exchangers is to utilize
Method 1. While this
estimate is lower, it is more
consistent.

Type 6: Furnaces
There are three furnaces
utilized in the styrene
process—E2, E7, and
E1XB. Prior to cost
analysis, these furnaces are
modeled as heater blocks in
Aspen in order to calculate
Figure 17. Method 1 Analysis for Furnaces. This figure was reproduced
the heat duty. Since these from the APEA evaluation tool on the Aspen Plus simulation created by
heater blocks utilize the our team.
furnace utility, they must be
mapped to a furnace prior to
completing costing with
Method 1. These furnaces
were mapped to EFU
BOX—or a box type process
furnace. APEA retrieves the
standard gas flow rate and
duty from the heater block.
Users need only to specify
the material as shown in
Figure 17. There are limited
costing options for these box Figure 18. Method 2 Analysis for Furnaces. This figure was
type process furnaces—it is reproduced from the Equipment Costs Excel Spreadsheet utilizing values
not possible to utilize from the Aspen Plus simulation created by our team. [6]
 21

interactive sizing on these pieces of equipment. Method 2 is equally as simple, requiring only that
the user input the heat duty from the heater block and select the desired cost factors. As shown in
Figure 18, the costing spreadsheet includes an input for efficiency. The efficiency is used to
determine yearly utility costs but it does not affect the installed cost analysis. Comparing the heat
duties used in the two cost analysis methods, it is evident that the heat duties utilized by Method 1
are higher than those in Method 2. The heat duties used in Method 2 come directly from the heater
block, therefore, Method 1 must use an overdesign factor of roughly 1.1.

Equipment Method 1 Method 1 Installed Cost Method 1

Name Equipment Purchase Cost [USD] Installation Costs
[USD] [USD]
E2 $1,902,300 $9,124,300 $7,222,000
E1XB $1,689,100 $5,638,800 $3,949,700
E7 $12,800 $99,500 $86,700
Table 15: Method 1 Installation Cost Comparison for Furnaces. This table was creating using data from
both the Aspen APEA Economic Analysis Tool and the Equipment Costs Excel Spreadsheet.

Equipment Name Installed Cost Installed Cost Over/Under

Method 1 [USD] Method 2 [USD] Evaluation (%)
E2 $9,124,300 $2,615,000 -71.3%
E1XB $5,638,800 $2,299,000 -59.2%
E7 $99,500 $1,000 -99.0%
Table 16: Installed Cost Comparison for Furnaces. This table was creating using data from both the
Aspen APEA Economic Analysis Tool and the Equipment Costs Excel Spreadsheet.

Table 15 demonstrates the total equipment purchase costs and equipment installed costs estimated
using Method 1. The difference in these two values equals the cost to install the equipment, or the
installation cost of the equipment. As shown in Table 15, a large portion of the installed cost is
the installation cost. This could indicate that Method 2 does not correctly estimate the installation
costs as the Method 1 equipment purchase costs are comparable to the Method 2 installed costs.
This is most likely due to Method 2 using outdated methods to calculate installation costs. As
shown in Table 16, Method 2 highly undervalues the costs by anywhere from 59-71%. While this
is not unexpected, this differs from other types of equipment because Method 2 undervalues
furnaces that are both small and large in size—therefore it is completely ineffective for costing
furnaces. E7 presents the largest difference in the installed cost estimates; the installed cost for E7
for Method 2 is only $1,000. The correlation used for the installed cost estimate in Method 2 is
only designed for furnaces that have heat duties in the range of 10-500 million BTU/hr. [5] E7 has
a heat duty that does not fall in this range, likely causing the undervaluation by Method 2. Because
Method 2 highly undervalues furnaces of all sizes due to outdated installation cost estimates, the
recommendation for sizing furnaces is to use Method 1: Aspen APEA analysis. This method
produces costs that are closer to those expected, is easier to use, and is consistent. Users must take
caution when using this method, however, as they must remember to map furnaces appropriately.
 22

Type 7: Distillation Columns

Distillation columns are a type of equipment that is very
capital intensive. The styrene process uses two distillation
columns—C1 and C2—to achieve the desired molar purity of
styrene and to separate the ethylbenzene for recycle.
Distillation columns are much more difficult to model than
pumps and compressors as utilizing both economic
assessment methods requires more user inputs. For both
methods, the column is first modeled as a RadFrac distillation
column in Aspen. Both of these columns are trayed columns
and will be evaluated as such. Method 1 calculates the column
as five distinct parts—column, reboiler, condenser, reflux
pump, and flash drum. Here, the results from the RadFrac
blocks are automatically loaded into the Aspen APEA
software for sizing and evaluation. As shown in Figure 19,
there are only a few user inputs needed to complete the model
for the columns (and therefore trays). The bottom material and
top material are chosen as carbon steel (CS). The reboilers are
calculated as
separate pieces of
equipment (see
Figure 20).
Calculating the cost
of the reboilers
using Aspen APEA
is relatively simple
as all needed
quantities are
automatically
received from the
simulation
excluding the shell
and tube materials
materials which the
user selects.
Reboilers are sized
and evaluated in
APEA much like the
other heat exchange
blocks. Therefore,
the overall heat
transfer coefficients
Figure 19. Method 1 Analysis for Figure 20. Method 1 Analysis for
Trayed Distillation Columns. This Distillation Column Reboilers. This
figure was reproduced from the APEA figure was reproduced from the APEA
evaluation tool on the Aspen Plus evaluation tool on the Aspen Plus
simulation created by our team. simulation created by our team.
 23

Figure 22. Method 1 Analysis for Distillation

Column Flash Drums. This figure was
reproduced from the APEA evaluation tool on the
Aspen Plus simulation created by our team.
for the reboilers can be located in the interactive
sizing tool. The condensers are sized and evaluated
in APEA much like the reboilers (see Figure 21);
they are automatically mapped to TEMA shell and
tube exchangers by the Aspen APEA software. The
Figure 21. Method 1 Analysis for material here is also chosen as CS for the shell and
Distillation Column Condensers. This tube materials. The overall heat transfer coefficients
figure was reproduced from the APEA for the condensors are located in the interactive sizing
evaluation tool on the Aspen Plus simulation tool. The column flash drums are named as C2-cond
created by our team.
acc and C1-cond acc (see Figure 22). Users will only
need to update the material specifications if desired. These flash drums—along with the reflux
pumps shown in Figure 9—are costed as individual equipment, unlike in Method 2. The pumps
are mapped to DCP CENTRF. Like the other pumps, user inputs here include only the casing
material and driver type.

Method 2 calculates the columns as four distinct parts—column, trays, condenser, and reboiler. A
factor of 1.2 is applied to each of these parts to account for other equipment not included in this
analysis—such as reflux pumps and flash drums. One simply needs to sum the analysis of these
four parts in order to find the total installed cost for the column. As shown in Figure 23, after
simulating in RadFrac, the user inputs the number of ideal plates, column diameter, and efficiency
of the top and bottom sections to calculate the column’s installed cost. [6] These parameters are
found in the RadFrac results or specified in the Project 1 Assignment Document (as is the case for
efficiency.) [3] The design factors are then chosen; these design factors are displayed in Figure
24. A design pressure factor of 0.00 is chosen as the pressure is less than 50 psig in the column
and the material factor is chosen as 1 for carbon steel. A tray spacing factor of 1.00 is chosen for
 24

the 2’ tray spacing, the tray type factor is chosen as 0.00 for the sieve trays, and a tray material
factor of 0.00 is chosen for the carbon steel trays. Next, the user inputs the condenser and reboiler
heat duty, temperature, and utility temperature values which can be found in the RadFrac results.
An overall heat transfer coefficient
must be chosen based on the nature of
the column distillate to be condensed.
Since the column distillate consists of
light hydrocarbon liquids that are being
condensed with a cooling water
coolant, a Uo value of 200 BTU/hr-ft2-
°F is chosen. Once this is complete, the
user then specifies the design factors
for the condensor and reboiler. For the
condensor, the material cost factor is
chosen as 1.00 for CS shell and tubes,
a design factor of 0.8 for a fixed-head
exchanger, and a pressure factor of
0.00 since the pressure is less than 150
psig. For the reboiler, all of the same
factors are chosen except a design
factor of 0.85 is chosen as the reboilers
Figure 23. Method 2 Analysis for Trayed Distillation
are U-tube reboilers.
Columns. This figure was reproduced from the Equipment
Costs Excel Spreadsheet utilizing values from the Aspen
Plus simulation created by our team. [6] Comparison of column quantities
calculated by the two
methods reinforces the
accuracy of both
methods regarding
column/tray design
calculations. The
number of actual trays
calculated by both
methods is
approximately the same.
However, the heat
transfer areas in the
exchangers do not
follow the same trend.
The heat transfer area of
a heat exchanger—in
this case the reboiler or
Figure 24. User Inputs for Trayed Distillation Columns. This figure was condenser—is
reproduced from the Equipment Costs Excel Spreadsheet. [6] indicative of its
installation cost.
Method 2 greatly underestimates the heat transfer area for the reboilers. However, Method 2
calculates a heat transfer area for the C1 condenser that is greater than the heat transfer area for
 25

the C1 condenser calculated by Method 1. This is most likely due to the low operating temperature
in this condenser; the correlations in Method 2 may not be equipped to properly estimate costs for
condensers with operating temperatures close to utility inlet/outlet temperatures.
Equipment Sub Equipment Installed Cost Installed Cost Over/Under
Name Name Method 1 [USD] Method 2 [USD] Evaluation (%)
C1 Column + Trays $1,999,300 $1,874,000 -6.3%
Reboiler $870,100 $731,000 -16.0%
Condenser $340,500 $1,757,000 416.0%
Flash Drum $216,900 -- --
Reflux Pump $100,800 -- --
TOTAL $3,527,600 $4,362,000 23.7%
C2 Column + Trays $7,009,100 $7,081,000 1.0%
Reboiler $1,063,100 $799,000 -24.8%
Condenser $250,700 $409,000 63.1%
Flash Drum $256,200 -- --
Reflux Pump $143,200 -- --
TOTAL $8,722,300 $8,289,000 -5.0%

Table 16: Installed Cost Comparison for Trayed Distillation Columns. This table was creating using
data from both the Aspen APEA Economic Analysis Tool and the Equipment Costs Excel Spreadsheet.

As shown in Table 16, Method 2 undervalues all pieces of equipment except for the condensers
and the columns/trays. Both methods are consistent in regards to the combined prices for the
columns and trays, with Method 2 being within 1.0-6.3% of Method 1’s estimate. However, the
condensers are overvalued by Method 2, and the C1 condenser is extremely overvalued—more
than 416%. This accounts for roughly half of the cost for this column. This condenser operates at
a much lower temperature—127.8 °F—which leads to a greater heat transfer area and therefore
increased equipment cost. Method 2 also indicates that the C1 condenser should be comprised of
four shells; this also contributes to the high installed cost estimate for the C1 condenser. While
Method 2 does provide an accurate estimate for the column and trays, it produces an inconsistent
overall estimate due to inconsistency in condenser estimates, underpricing of reboilers, and lack
of individual pricing for column flash drums and pumps. Therefore, the recommendation for
sizing distillation columns is to use Method 1: Aspen APEA analysis. Not only does this
method provide a more accurate estimate, it is more consistent and easier to use. No special
considerations need to be made for reboilers or condensers that operate at temperatures higher or
lower than that which is typical.

Discussion
Both installed cost calculation methods have their pros and cons. As shown in Table 17, Method
1 has the greatest total installed cost estimate, 22.1% greater than Method 2. Much of this
difference can be contributed to costs associated with units KC1, E2, E7, and E1XB. As evident
with small equipment such as KC1 and E7, Method 2 does not scale well. If users design smaller
equipment, it will impact their cost assessments greatly. E7, for example, is estimated by Method
2 to cost $1,000 dollars. It is obvious that this is not reasonable, however, this error could be easily
overlooked by users. Method 2 is more straightforward and simpler as users are in control of the
inputs used for costing and sizing. Therefore, it is easier to understand which factors affect the
 26

costing. This method is more tedious, however, as the inputs must be selected manually. Method
1 is more user-friendly and much quicker. Using Method 1 saves a notable amount of time as each
time a change is implemented in the Aspen Plus simulation, the Aspen APEA tool automatically
updates the economic analysis. When using Method 2, each time a change is implemented, each
equipment’s data must be located in the Aspen Plus simulation then reinputted into the Excel
spreadsheet. This leads to a lot of wasted time that could be saved by utilizing Method 1. This
saved time would allow students the opportunity to focus on perfecting other portions of their
project. Another benefit of using Method 1 is that a certain level of engineering judgement is

Method Method 1: Aspen Method 2: Difference Over/Under

APEA Spreadsheet Evaluation (%)
Total Installed Cost $32,086,200 $25,012,000 $7,074,200 -22.1%
Table 17. Overall Installation Cost Comparison. This table was creating using data from both the Aspen
APEA Economic Analysis Tool and the Equipment Costs Excel Spreadsheet.

programmed into the software. Users are more apt to choose equipment specifications that are not
reasonable such as designing a distillation column that is 23.42377861 feet in diameter or costing
a reactor that is 47.8270865 feet tall. Aspen recognizes that these sizes are infeasible and updates
them to more reasonable sizes such as 23.5 feet and 48 feet. This tool could lead to a better
understanding for users as it enhances their engineering judgement skills. Another benefit of using
the Aspen APEA analysis is that there are more individualized material factors. For example,
APEA has separate material factors for different variations of stainless steel whereas Method 2
only has one generic material factor for stainless steel. Therefore, Method 1 should be used in
future semesters as it produces a cost estimate closer to what is expected and is more
consistent than Method 2. It is recommended to continue to use Method 2 when sizing and
costing flash drums until a more thorough study is completed to achieve a greater understanding
of the calculations for either method. This conclusion reflects the underlying hypothesis that
Method 1 provides more accurate costing than Method 2. Method 1 offers rigorous sizing ability,
individualized material specification factors, and an updated information database providing
students with an up-to-date capital cost assessment.
 27

Works Cited

[1] Smith R. Chemical Process Design. Wiley Online Library.

https://onlinelibrary.wiley.com/doi/10.1002/0471238961.chemsmit.a01.pub2. Published
December 15, 2015. Accessed November 14, 2022.

[2] Towler G, Sinnott R. Chapter 1: Introduction to Design. In: Chemical Engineering Design:
Principles, Practice and Economics of Plant and Process Design. 3rd ed. Cambridge, MA:
Butterworth-Heinemann; 2022:3.

[3] Guillory B. Report 1 Assignment. October 2022.

[4] Al-Malah KIM. Preface. In: Aspen Plus: Chemical Engineering Applications. 2nd ed.
Hoboken, NJ: John Wiley & Sons, Inc.; 2023:21.

[5] Seider WD, Lewin DR, Seader JD, Widagdo S, Gani R, NG KAM. Chapter 16 Cost
Accounting and Capital Cost Estimation. In: Product and Process Design Principles:
Synthesis, Analysis and Evaluation. 4th ed. Wiley; 2016:426-497.

[6] Corripio A. Equipment Costs.

[7] Chen S-S. Styrene - chen - major reference works - wiley online library. Wiley Online
Library.
https://onlinelibrary.wiley.com/doi/full/10.1002/0471238961.1920251803080514.a01.pub2
. Published September 15, 2006. Accessed September 28, 2022.

[8] James DH, Castor WM. Styrene. In: Ullmann's Encyclopedia of Industrial Chemistry.
Weinheim: Wiley-VCH Verlag GmbH & Co. KGaA; 2005:1-17.

[9] Styrene. National Center for Biotechnology Information. PubChem Compound Database.
https://pubchem.ncbi.nlm.nih.gov/compound/Styrene. Published September 16, 2004.
Accessed September 28, 2022.

[10] AspenTech. Aspen Plus V12 Help. September 2020.

[11] United States consumer price index (CPI)October 2022 data - 1950-2021 historical. United
States Consumer Price Index (CPI) - October 2022 Data - 1950-2021 Historical.
https://tradingeconomics.com/united-states/consumer-price-index-cpi. Accessed November
8, 2022.