Thesis Armghan.pdf

Optimization of a Cryogenic Air Separation Unit for an IGCC Power Plant

By Armghan Ahmed

Thesis submitted to faculty of engineering at PIEAS in partial fulfilment of requirements for the degree of M.S. Mechanical Engineering

Department of Mechanical Engineering Pakistan Institute of Engineering and Applied Sciences Nilore, Islamabad, Pakistan

(October, 2017)

Department of Mechanical Engineering, Pakistan Institute of Engineering and Applied Sciences (PIEAS) Nilore. Islamabad 45650, Pakistan

Declaration of Originality I hereby declare that the work contained in this report and the intellectual content of this report are the product of my own work. This report has not been previously published in any form nor does it contain any verbatim of the published resources which could be treated as infringement of the international copyright law. I also declare that I do understand the terms ‘copyright’ and ‘plagiarism,’ and that in case of any copyright violation or plagiarism found in this work, I will be held fully responsible of the consequences of any such violation.

Signature: _______________________________

Name: __________________________________

Date: ____________________

Place: ____________________

ii

Certificate of Approval This is to certify that the work contained in this thesis entitled

“Optimization of a Cryogenic Air Separation Unit for an IGCC Power Plant” was carried out by Armghan Ahmed Under my supervision and that in my opinion, it is fully adequate, in scope and quality, for the degree of MS Mechanical Engineering from Pakistan Institute of Engineering and Applied Sciences (PIEAS).

Approved By:

Signature: ________________________ Supervisor: Dr. Muhammad Zaman

Verified By:

Signature: ________________________ Head, Department of Mechanical Engineering Stamp:

iii

Dedication “This work is dedicated to my father who passed away during my MS”

iv

Acknowledgement

I would like to express my sincere gratitude to my supervisor Dr. Muhammad Zaman as well as our Head of Department Dr. Rizwan Alim Mufti who gave me the opportunity to do this wonderful project, which helped me in doing a lot of research and I came to know about so many new things, I am really thankful to them. Secondly I would like to thank my parents and friends who helped me a lot in finalizing this project within the limited time frame.

Armghan Ahmed

v

Table of Contents 1 Introduction ................................................................................................................. 1 1.1

Motivation ....................................................................................................... 1

1.2

Air Separation Unit (ASU).............................................................................. 2

1.3

ASU Technologies .......................................................................................... 2

1.3.1

Cryogenic Distillation .............................................................................. 2

1.3.2

Pressure Swing Adsorption (PSA) ........................................................... 3

1.3.3

Ion Transport Membrane ......................................................................... 4

1.4 2

Background and Problem Definition ..................................................................... 7 2.1

3

Thesis Layout .................................................................................................. 5

Literature Review ............................................................................................ 7

2.1.1

Process Configurations ............................................................................ 7

2.1.2

Thermodynamic Analysis ...................................................................... 16

2.1.3

Economic Analysis ................................................................................ 17

2.2

Open Issues ................................................................................................... 17

2.3

Problem Definition ........................................................................................ 18

2.4

Objectives ...................................................................................................... 18

Methodology ........................................................................................................ 19 3.1

Configuration Selection................................................................................. 19

3.2

Operating Conditions .................................................................................... 19

3.3

Solution Approach......................................................................................... 20

3.4

Process Description ....................................................................................... 21

3.5

Exergy Analysis ............................................................................................ 23

3.6

Economic Analysis ........................................................................................ 24

3.7

Optimization Problem Formulation .............................................................. 25

vi

4

Results and Discussions ....................................................................................... 28 4.1

Simulation of base case and validation of results ......................................... 28

4.2

Exergy Analysis ............................................................................................ 33

4.3

Cost Analysis................................................................................................. 34

4.4

Parametric Study ........................................................................................... 36

4.4.1

Stages of distillation columns ................................................................ 36

4.4.2

Split Ratio .............................................................................................. 39

4.4.3

Pressure Ratio ........................................................................................ 40

4.5 5

Optimization .................................................................................................. 44

Conclusion and Recommendations ...................................................................... 46 5.1

Conclusions ................................................................................................... 46

5.2

Recommendations ......................................................................................... 47

References .................................................................................................................... 48

vii

List of Figures Figure 1 Conventional double-column ASU [3] ............................................................ 2 Figure 2 Adsorption-based air separation process [2] ................................................... 4 Figure 3 Ion transport membrane air separation process [2] ......................................... 5 Figure 4 Process Flow sheet of two-column ASU [4] ................................................... 8 Figure 5 Process Flow sheet of three-column ASU [4] ................................................. 8 Figure 6 Flow sheet of single-column ASU [5] ............................................................. 9 Figure 7 Single-column ASU with LNG integration [7] ............................................. 10 Figure 8 Double-column air separation unit [8] .......................................................... 11 Figure 9 Proposed process with flash separator [3] ..................................................... 12 Figure 10 Stage column- Equilibrium based model .................................................... 21 Figure 11 Flow sheet of Base Case .............................................................................. 22 Figure 12 Optimization scheme ................................................................................... 26 Figure 13 O2 purity VS Stages ..................................................................................... 37 Figure 14 O2 in N2 Stream VS Stages .......................................................................... 38 Figure 15 Total annualized cost VS Stages ................................................................. 38 Figure 16 O2 purity VS split ratio ................................................................................ 39 Figure 17 Total annualized cost VS split ratio............................................................. 40 Figure 18 O2 purity VS PR of COMP3........................................................................ 41 Figure 19 O2 purity VS Pressure Ratios ...................................................................... 42 Figure 20 O2 in N2 Stream VS Pressure Ratios ........................................................... 43 Figure 21 Total annualized cost VS Pressure ratios .................................................... 43 Figure 22 Total annualized cost VS stages of LPC ..................................................... 44

viii

List of Tables Table 1 Summary of literature ..................................................................................... 13 Table 2 Exergy Destruction in Components ................................................................ 17 Table 3 Compressor specifications .............................................................................. 20 Table 4 Feed Air composition and Flow rate............................................................... 20 Table 5 Stream Properties ............................................................................................ 29 Table 6 Exergy Destruction of components................................................................. 33 Table 7 Utilities cost .................................................................................................... 34 Table 8 Cost of unit operations .................................................................................... 35 Table 9 Summery of Cost Analysis ............................................................................. 35 Table 10 Pressure ratios of COMP1 and COMP2 ....................................................... 41 Table 11 Final conditions for Compressors ................................................................. 45 Table 12 Final conditions for Columns ....................................................................... 45 Table 13 Cost comparison with base case ................................................................... 45

ix

Abstract Oxygen is one of the most important industrial gases and is an essential part of most of the processes which involves combustion or chemical reactions. In the integrated gasification combined cycle (IGCC) power plant, gasification of coal is done by reacting it with the controlled amount of pure oxygen. Oxygen used in the process is highly pure (usually >97%). Therefore separation of oxygen from air is a vital process in IGCC plant. Significantly higher operating cost is required for the production of pure oxygen. The cost of air separation unit (ASU) contributes more than 15 % in the overall operating cost of the IGCC power plant. Operating cost of the ASU can be reduced by optimizing its design and operation. This study involve the optimization of various design and operating parameters of ASU by minimizing the total annualized cost, without compromising the desired product purity. Number of stages for the high pressure column, number of stages for low pressure column, pressure ratios and split ratio were used as the optimizing parameters. ASPEN PLUS® software was used for simulation and optimization of these parameters, and around 1.35 MUSD/ year were reduced compared to base case.

x

1 Introduction Industrial gases have vast applications in the chemical and power industry being used as feed materials for synthesis processes, these gases improve the economy of processes and plants, and ensure the quality of products and safety of the plant. Oxygen is used in medical procedures and other industrial processes such as metal manufacturing processes, glass, ammonia, oxy-fuel combustion and the gasification process for power generation. The major use of nitrogen is as an inert blanking gas and as a reactant in chemical processes. Argon finds its use as an inert shielding gas in welding, in growing silicon crystals as well as for electronics like the light bulb. So to separate different components of air, different methods are utilized. There are three methods of air separation usually used: the cryogenic distillation process, the pressure swing adsorption (PSA) process and the Ion transport membrane process. The gasification-based power plants like integrated gasification combined cycle power plant (IGCC) are equipped with oxygen-blown gasifiers which are better than air-blown reactors due to its high cold gas efficiency. Oxygen production is of huge importance for IGCC power plants, the technologies to produce pure oxygen are becoming mature day by day but it still requires large energy consumptions and high investment costs, and there is a great scope for improvement in this regard. This section includes the motivation for this study, different ASU technologies and layout of the complete thesis.

1.1 Motivation A large amount of plant energy is consumed for production of pure oxygen from air separation unit in an IGCC power plant, about 15% of the overall cost of IGCC power plant. About 80% of in-house consumption is due to air separation unit i.e. about 12% of gross power is consumed by ASU [1]. Therefore, performance of IGCC power plant may be improved by optimizing the operating conditions and configuration of the air separation unit (ASU). Optimizing the cost of ASU will decrease the overall cost of IGCC power plant. This study is focused to reduce the cost of ASU without compromising the product purity and integrity of the system.

1

1.2 Air Separation Unit (ASU) An air separation unit separates atmospheric air into its primary components, typically oxygen and nitrogen, and some amount of argon. Separation is carried out either by liquefying the air and separating different components at their unique boiling points or by utilizing the adsorbing capacity of some materials to separate these components.

1.3 ASU Technologies There are different methods for separation of air but the three main methods which are commercially used include the cryogenic distillation process, the pressure swing adsorption (PSA) process and the ion transport membrane process.

1.3.1 Cryogenic Distillation When high purity products are needed along with high flow rates then cryogenic distillation process is used. The process of distillation can fractionate air by taking advantage of the different liquefaction temperatures of O2 and N2. This process can produce large oxygen flow rates with purity up to 100% and, among all the air separation technologies, this is the most efficient and the most cost-effective in terms of energy consumption per unit of product eventually when there is requirement of high flow rates of oxygen product [2].

Figure 1 Conventional double-column ASU [3]

2

A flow diagram of conventional double-column ASU is shown in Figure 1 in which atmospheric air is first compressed in a multistage compressor to a pressure of 6.5 atm. The compressed air is then split into two unequal fractions. Smaller fraction of air is further compressed to 8 atmospheric in a second compressor. Now these two streams i.e. larger stream of 6.5 atm pressure and smaller stream of 8 atm, are passed through a counter-current heat exchanger (also termed as cold box) where incoming air exchange heat with the cold purified products coming out from the distillation column to lower the temperature to cryogenic conditions. The larger fraction is then fed to the high pressure column while the smaller fraction is sent to the turbine. The turbine is used to recover some of the energy in order to compensate for high energy consumption of compressors. The air is throttled from 8 atm. to 1.3 atm. in the turbine. The exit stream of turbine is then used as one of the feed stream for low pressure column (LPC). In high pressure column (HPC), the larger fraction is processed to produce nitrogen enriched stream from top of the column and oxygen enriched stream from bottom of the column. The nitrogen enriched air from HPC is used as a reflux in Low pressure column while oxygen enriched air from HPC is introduced in low pressure column as a second feed stream apart from one coming from turbine. The final products from low pressure chamber are nitrogen enriched stream (>90%) from top of the column and oxygen enriched stream (>90%) from bottom of the column.

1.3.2 Pressure Swing Adsorption (PSA) This process is used when a nitrogen flow is needed between 10 and 100 m3/h with a purity of 98-99.5 vol% [2]. Process of adsorption is based on the ability of any natural or synthetic material to adsorb nitrogen or oxygen but usually nitrogen is preferred. Thus, for this process of air separation, nitrogen molecules adsorbs more rigorously than oxygen or argon molecules. Zeolite material is used in case of nitrogen. As air is passed through a bed of adsorbing material, nitrogen is adsorbed and an oxygen-rich stream exits at the other end of the bed. In case of oxygen carbon molecular sieves are used which have really small pore sizes, of the same order of magnitude as the size of air molecules. Due to slightly smaller size of oxygen molecules than the nitrogen molecules, they diffuse more into the cavities of the adsorbent than nitrogen and causing the separation. Thus, carbon molecular sieves are chosen for oxygen and zeolites for nitrogen. 3

Figure 2 Adsorption-based air separation process [2]

A schematic of adsorption based air separation process is shown in figure 2. First air is filtered and pressurized according to requirement, air than enters a vessel which contains the adsorbent. Nitrogen molecules are adsorbed and an oxygen rich stream is produced until the bed become saturated with nitrogen. At this situation, the feed air is switched to a fresh vessel of adsorbent and regeneration of the used bed begin simultaneously. Regeneration is accomplished either by heating the bed or by reducing the pressure in the bed, which reduces the nitrogen holding capacity of the adsorbent at equilibrium condition. The faster cycle time and simplification of operation associated with pressure reduction method usually makes it the process of choice for air separation. However, significant research is needed to explore adsorbent material with low energy requirements.

1.3.3 Ion Transport Membrane The membranes in ion transport membrane (ITM) process are solid inorganic oxide ceramic materials which produce oxygen by the passage of oxygen ions through the crystal structure of the ceramic membranes. Oxygen molecules are converted into oxygen ions at the surface of the membrane and then transported through the 4

membrane with the help of an applied electric voltage or by partial pressure difference of oxygen, which converts these ions into oxygen molecules again after passing through the membrane material.

Figure 3 Ion transport membrane air separation process [2] Figure 3 shows the process flow sheet of ion transport membrane process in which air is compressed first and then heated to operating temperature by exchanging heat with the hot process streams along with auxiliary heat addition. In general, the heating of air can be done by either exchanging heat indirectly and/or direct firing of the fuel. The oxygen ions are passed through the ITM at very high flow rates and at the permeate side of the membrane highly pure oxygen is produced. The oxygen stream is compressed to desired pressure for use in gasification or other applications [2].

1.4 Thesis Layout This thesis comprises of five chapters. In this chapter (chapter 1), introduction of the project is discussed that includes the motivation for this project along with air separation technologies. The next chapter includes literature review of different analysis done for various ASU configurations along with some details of the analysis performed. It also discusses some related open issues in this field of study and objectives of this thesis.

5

The third chapter describes the methodology used for this study which includes configuration, model, tool and operating conditions used in this study. It also gives the process description of air separation unit. Fourth chapter is of results and discussions, it includes the results of the exergy and economic analyses, parametric trends for product purity and cost, and constraint for optimization constraints and its results. Finally in the last chapter, the conclusions are drawn from the thesis results and recommendations for the future work are given.

6

2 Background and Problem Definition In literature we came across various technologies for air separation, all of them have their advantages and disadvantages. In adsorption process size of bed is the factor that controls the capital cost and rate of production is directly related to bed volume, due to which capital costs increases rapidly compared to cryogenic process. Similarly, for ion transport membrane process, production rate is less compared to cryogenic process. This chapter discusses the various configuration of the process found in the literature, open issues in this field along with problem definition and objectives of the thesis.

2.1 Literature Review In this study we will consider cryogenic distillation of air as the technique to separate air because of its large utilization in industry, high purity and large quantities of nitrogen and oxygen associated with it, which is required in the Integrated Gasification Combined Cycle (IGCC) power plant. The literature can be divided into two parts, the type of configuration used in the process and the type of analysis performed on it. Two type of analysis that were found in literature are thermodynamic analysis and economic analysis. Thermodynamic analysis is done to find exergy destruction in components, exergy efficiency and specific energy consumption. Cost analysis is done to estimate equipment cost, operating and capital costs.

2.1.1 Process Configurations There are several configurations of cryogenic air separation unit which can be found in the literature. Qian Fu et al. [4], Hua Zhou et al. [5], Mehdi Mehrpooya et al. [6] and Zheng Jieyu at al. [7] discussed single column configurations in their study. Armin Ebrahimi et al. [8], Zeinab A. M. Khalel et al. [3], L.V. van der Ham [9, 10], Mehdi Mehrpooya et al. [6] and Stefanie Tesch et al. [11] all discussed double column configuration with some variation in their study. Other configuration discussed in literature are conventional double column with heat recovery cycle, conventional with LNG as heat sink and single-column ASU using LNG cold energy. 7

L.V. van der Ham et al. [9] discussed the ASU configurations having two and three distillation columns respectively and did the exergy analysis of the both the processes. A schematic for two-column ASU is shown in Figure 4 and for three-column ASU in Figure 5. He found out that addition of a third column can reduce the exergy destruction of the distillation section by 31% due to low separation capacity requirements by the addition of an extra distillation column and it has overall 12% less exergy destruction than the two-column configuration.

Figure 4 Process Flow sheet of two-column ASU [4]

Figure 5 Process Flow sheet of three-column ASU [4]

8

The study suggested that the three-column design did better than the two-column design; their exergy destructions amounted to 4.12 and 4.67 kJ/mol air for two and three column respectively, which means that the three-column design destroyed 12% less exergy than two column design. It also shows that almost half of the exergy destruction was caused by the compressor after-coolers; the distillation sections 27% and the MHXs causes 16% of total exergy destruction. It did not cover the cost effect of the two processes because adding an extra distillation comes with more capital and maintenance costs thus overall cost of the system may increase. This study indicate the three-column design is better in terms of losses but the economics involved was a major aspect which is not covered. Hua Zhou et al. [5] suggested a single-column cryogenic air separation process, and simulated the different configurations of single-column ASU and compared them with conventional double column design.

Figure 6 Flow sheet of single-column ASU [5] The flow sheet of a single column ASU is shown in Figure 6 and the tool used for simulation was ASPEN PLUS 7.0; the result showed that the proposed design had 6% more exergy efficiency than the conventional ASU which is equivalent to 23% increase in the exergy efficiency [5]. But this extra efficiency comes at the cost of product purity which is of great importance in our case. Another study using single-column configuration was done by Zheng Jieyu at al. [7] in which LNG cold energy was utilized. Liquefied natural gas (LNG) is transported all around the world as a clean and high efficient source of energy. With a storage 9

temperature of 110K, LNG requires re-vaporization before transmitting it to the costumers, which is generally done by transferring heat with the seawater or ambient air, almost 230 kWh/t of valuable cold energy gets wasted. Since air separation process (ASP) has minimum temperature of 100K, so integration with ASU it is highly suitable for full recovery of LNG cold energy.

Figure 7 Single-column ASU with LNG integration [7]

The tool used for simulation was ASPEN HYSYS®; the result suggested that the proposed process was more efficient in terms of producing pure liquid products with a NLIQ at around 0.218kWh/kg [7]. However, the flow rate of air is greater than the traditional process, which requires a longer start-up time. The total exergy efficiency was found 50% higher than the conventional process. The study also suggested that with elevated pressure of LNG the power efficiency can improve further. Armin Ebrahimi et al. [1] did an exergetic and economic assessment on a two-column air separation unit. The schematic of his proposed process is shown in Figure 8. He discussed the effect of size of system in terms of oxygen production rate (oxygen production per second) on return period and exergy destruction in various components of the system. 10

The study suggested that the return period is less than 5 years for oxygen production rate of 9kg/sec and higher. It also showed that the exergy loses are highest for LPC and HPC, and exergy destruction was lowest for pump and separator. The air compression process which have compressors and intercoolers; and the distillation columns (HPC and LPC) are the two biggest irreversibilities in the given system with 34% and 53% respectively.

Figure 8 Double-column air separation unit [8]

Armin Ebrahimi et al [12] also did a study using LNG as a heat sink and compared the result with study above. Before integration with LNG stream, the sizes of 9kg/sec of oxygen and more had return period of less than 5 years. But after integration, the period of return is economically feasible for sizes 5 kg/sec of oxygen and more. The integration with LNG increases the cold energy utilization of the system and help in achieving the desired cryogenic condition at low product rates. ASU cycle power requirement after integration with LNG stream decrease by 2137 kW which is 8.04% of total amount of required power for this cycle. The showed that integrated systems have low energy consumption compared to conventional systems. Zeinab A. M. Khalel et al. [3] added a flash separator in the conventional doublecolumn cryogenic air separation process. The flash separator was used in the replacement of the turbine, which was recovering some portion of the energy 11

consumption in the double-column air separation process. The proposed process diagram is shown in Figure 9. The flash separator helped in to areas; throttling and separation of the feed.

Figure 9 Proposed process with flash separator [3]

The results suggested that the proposed design was better than the conventional design in two aspects: (1) the specific energy consumption was lower than that of the conventional system by 20% and (2) the pure O2 production was increased up to 14% and pure N2 was increased up to 5%. The other feature that made the flash separator more reliable compared to turbine was not having any moving parts [3]. A brief summary of various configurations found from the literature is discussed in the Table 1 below which explains these configurations and the corresponding study performed on it. It also indicates different modifications in the conventional process like double-column ASU integrated with LNG stream (as heat sink), ASU with heat recovery cycle etc. and the respective study performed on them.

12

Table 1 Summary of literature SR#

Author

1

Armin Ebrahimi et

conventional cryogenic

>size of system (oxygen

period of return(less than 5 years

this study shows closeness to

al.

ASU

production per sec)

for oxygen production of

literature calculations at production

>exergy destruction in

9kg/sec or more)

of 9kg/sec of oxygen

(2015)

ASU SYSTEM

Parameter studied

Performance indicator

Remarks/Findings

components 2

Zeinab A. M.

Cryogenic Air Separation

>specific energy

> the SEC is lower than that of

Since flash separator contains no

Khalel et al.

Process with Flash

consumption

the conventional by 20%

moving parts, it Is more reliable

(2013)

Separator

>production rate

>the pure O2 production is

than the turbine

increased by 14.3% and pure N2 is increased by 5.4% 3

4

L.V. van der Ham

> ASU with additional

>Exergy efficiency of ASU The addition of an extra heat

the most significant change occurs

(2012)

heat exchanger

with HI stages

exchanger to capture maximum

with the addition of heat exchanger

>ASU with Heat

>Exergy efficiency of

cold energy of products

than the heat integrated stages.

Integrated stages

ASU with additional HEX

increases efficiency up to 30%

L.V. van der Ham

>ASU with two

>exergy destruction per

total exergy destruction for two

The exergy destruction caused by

et al.

distillation columns

feed's amount (kJ/mol)

column design was 4.67kJ/mol

the compressor and after-coolers is

(2010)

>ASU with three

>Exergy efficiencies of

while for three column design,it

almost half of the system. So by

distillation columns

two designs

is 4.12kJ/mol

using this heat the efficiency will increase

13

Table 1 (contd.) SR# 5

6

7

8

Author

ASU SYSTEM

Parameter studied

Performance indicator

Remarks/Findings

Armin Ebrahimi,

Cryogenic Air

System's size in terms

after integration with LNG stream, the

ASU cycle power requirement after

Masoud

Separation Unit (ASU)

of oxygen production

period of return is economically

integration will decrease 8.04% of

Ziabasharhagh

with Liquefied Natural

per sec

feasible in sizes 5 kg oxygen/sec and

total amount of power required.

(2017)

Gas (LNG) as heat sink

Mathew Aneke et

Cryogenic ASU with

Power Consumption

>Conventional ASU has specific power

In Comparison to the conventional

al. [13] (2014)

heat recovery ORC

in (kWh/kg) for each

consumption for oxygen 0.357, N2

3 stage water cooled process, the

system (using single

pure product

0.421 and argon 19.55 kWh/kg

C3WHR reduces the overall net

stage or multi-stage

>CASU with ORC system with 3 stage

power consumption by almost 11%

compression)

compressors has values of 0.316, 0.373

while the C1WHR achieves only

and 17.346 kWh/kg respectively

0.14% reduction.

more

Mehdi Mehrpooya cryogenic air separation

>energy consumption

Energy consumption of the process was

The proposed process with LNG

et al. (2014)

process with LNG cold

of the process

decreased by about 38% compared to

integration was better in terms of

energy utilization (single

>Energy and Exergy

the conventional cryogenic ASU. The

energy consumption and exergy

and double column

Efficiency of the

energy and exergy efficiencies were

efficiency of is also high compared

configuration)

process

increased by 59% and 67%.

to traditional process.

Qian Fu et al. [4]

Single column Air

Specific energy

The energy consumption of the

Single column ASU consumes less

(2014)

Separation Unit

consumption

proposed cryogenic ASU was decreased energy than conventional double by 31% in contrast to the conventional

column ASU but has less

process, when producing O2 with low

production rate and product purity.

purity (95 mol%)

14

Table 1 (contd.) SR# 9

10

Author

ASU SYSTEM

Parameter studied

Performance indicator

Remarks/Findings

Hua Zhou et al.

Single-Column

Exergy efficiency of

The simulations shows that the exergy

It has high exergy efficiency and

(2012)

Cryogenic Air

different Process

consumptions can be kept up to 23% in

low cost than the same stages of

Separation Process

Designs

contrast to the conventional process

double-column process.

Qian Fu et al.

>Conventional LP-ASU

energy consumption of Both proposed processes has

From all the proposed processes,

[14]

>Conventional EP-ASU

the processes

comparable energy consumption but

the elevated-pressure ASU has

>Proposed single-

9.7% and 11.1% lower than that of the

lowest energy consumption.

column LP-ASU

conventional EP-ASU and conventional

>Proposed single-

LP-ASU process, respectively

(2015)

column EP-ASU 11

Stefanie Tesch et

ASU integrated with

>Exergy Analysis

The system without a N2 liquefaction

Integration with LNG stream

al.

LNG regasification

>Cost Analysis

block has lower exergetic efficiency.

improves the energy consumption

(2017)

(with or without nitrogen

The cost of system with N2 liquefaction

of the system and system with N2

liquefaction block)

block is high because of additional

liquefaction block has high cost.

compressor and heat exchanger. 12

Zheng Jieyu at al.

Single-column

>specific energy

The specific power consumption of the Compared with the conventional

(2014)

cryogenic air separation consumption

suggested process is 0.218 kWh/kg, and ASU application, this process

process with LNG cold >Exergy efficiency

the total exergy efficiency of the system utilizes a nitrogen compressor

energy utilization

is 0.575. The system uses 39.1% lower acting as a heat pump, which energy and has 50.5% higher efficiency recuperates both the latent and than the conventional process.

15

sensible heat in the system.

2.1.2 Thermodynamic Analysis The purpose of doing thermodynamic analysis is to find irreversibilities in the system. It also helps in finding errors in the system during the development phase of process flow sheet. Thermodynamic analysis includes the exergy analysis of each component and whole of the system. The equilibrium model is used to solve each entity. The heat and mass balance equation utilized by the software are given below. Mass Balance 𝑭𝒋 𝒁𝒊𝒋 + 𝑳𝒋−𝟏 𝒙𝒊,𝒋−𝟏 + 𝑽𝒋+𝟏 𝒚𝒊,𝒋+𝟏 − 𝑳𝒋𝒙𝒊𝒋 − 𝑽𝒋𝒚𝒊𝒋 = 𝟎

(1)

i = 1…NC (Component Number); j = 1…N (Number of stages) Heat Balance 𝑭𝒋 𝑯𝑭𝒋 + 𝑳𝒋−𝟏 𝑯𝒍,𝒋−𝟏 + 𝑽𝒋+𝟏 𝑯𝑽,𝒋+𝟏 − 𝑳𝒋𝑯𝒍𝒋 − 𝑽𝒋𝑯𝑽𝒋 + 𝑸𝒋 = 𝟎

(2)

𝑯𝒍𝒋 = 𝒉(𝑻𝒋 , 𝑷𝒋, 𝒙𝒋 ); 𝑯𝑽𝒋 = 𝒉(𝑻𝒋 , 𝑷𝒋, 𝒚𝒋 ); 𝑯𝑭𝒋 = 𝑯𝑭(𝑻𝒇 , 𝑷𝒇, 𝒛𝒋 ) Equilibrium Expression 𝒚𝒊𝒋 = 𝑲𝒊𝒋 𝒙𝒊𝒋

(3)

𝑲𝒊𝒋 = 𝑲(𝑻𝒋 , 𝑷𝒋, 𝒙𝒊𝒋 )

(4)

The method used to solve the equilibrium model is Peng-Robinson equation of state, which are applied on the system to calculate stream properties. The data for all the streams is obtained from the software and then evaluated to find exergy destruction across each component of ASU. The expression for exergy and exergy destruction are given in the equation 5 and 6 respectively. 𝒆𝒙,𝒊 = (𝒉𝒊 − 𝒉𝟎 ) − 𝑻𝟎 (𝒔𝒊 − 𝒔𝟎 ) 𝑻

∆𝑬 = ∑ (𝟏 − 𝑻𝟎 ) 𝑸𝒋 + (∑𝒊 𝒎̇𝒊 𝒆𝒙,𝒊 )𝒊𝒏 − (∑𝒊 𝒎̇𝒊 𝒆𝒙,𝒊 )𝒐𝒖𝒕 𝒋

(5) (6)

The equations for exergy destruction across some important components such as compressors, heat exchanger and distillation columns are mentioned in the Table 2.

16

Table 2 Exergy Destruction in Components Components

Exergy Destruction (kW)

COMP1

WCOMP1-(Ex(2)-Ex(1))

COMP2

WCOMP2-(Ex(4)-Ex(3))

COMP3

WCOMP3-(Ex(8)-Ex(7))

HEX1

((Ex(22)-Ex(21))+(Ex(25)-Ex(24))-((Ex(6)-Ex(12))+(Ex(9)-Ex(10))

HEX2

((Ex(15)-Ex(14))+(Ex(18)-Ex(17))-(Ex(23)-(Ex(24))

HPC

QCond–(Ex(14)+Ex(17))-(Ex(11)+Ex(13))

LPC

QReb–(Ex(20)+Ex(23))-(Ex(16)+Ex(19))

Pump

Wpump-(Ex(21)-Ex(20))

2.1.3 Economic Analysis Economic Analysis is essential in determining the total annualized cost of the system. For that all the information is gathered from Aspen Process Economic Analyzer (APEA) to determine the operating cost, utility costs, equipment cost, and capital cost etc. The Economic Analysis gives detailed information of each and every equipment and provides total operating cost, utilities used with total utility cost, equipment size with total equipment cost and total capital cost.

2.2 Open Issues During the literature review it was observed that a lot of different configurations of cryogenic air separation unit (ASU) have been presented. All of them have their pros and cons, like using a single-column configuration will be better for economic point of view but if the desired outcome is high product purity and quality then it has room for improvement. Similarly double and triple column configurations have their strength and weaknesses. Process efficiency is another open area where every configuration is open for improvement. So improving the efficiency of any configuration would be a big achievement. Integration with another system can improve the power consumption, in doing so, may improve the overall efficiency. 17

Most of the published work is about thermodynamic analysis and very few authors have presented economic analysis. Therefore, economic analysis is another open field due to limited available literature. So a cost analysis of a suitable configuration would be a good addition to literature.

2.3 Problem Definition The main objective of this study is to improve the cost effectiveness of a doublecolumn air separation unit by changing operating conditions. Other objectives are the development and simulation of double column ASU using ASPEN PLUS® software, comparison of simulation with literature, exergy analysis of individual components, and economic analysis.

2.4 Objectives From the understanding of open issues in the field of air separation and problem definition, following objectives are set for this thesis. 

Suitable choice of process configuration should be done according to product purity and flow rate requirements



Development of flow sheet for the selected configuration in ASPEN PLUS® with the operating conditions close to literature (base case)



The results of the base case are validated with literature for the authenticity of our system



Exergy analysis is performed on each and every equipment to find the biggest irreversibilities of the process



Cost analysis of the base case is done to calculate the total annual cost of the system



Effect of number of stages of distillation columns on product purity and annual cost of the system



Effect of split ratio on product purity and annual cost of the system



Effect of pressure ratios on product purity and annual cost of the system



Optimization of double-column configuration of ASU by changing all the parameters to obtain minimum annual cost for the system without compromising the process constraints

18

3 Methodology Since the chosen technology is cryogenic distillation and the boiling points of oxygen and nitrogen are -183ºC and -196ºC respectively, achieving this much negative temperature, a heat sink with much less temperature is required. So the pressure of the feed air is increased to achieve high boiling points and to exchange heat with the incoming cold product streams. In this chapter the methodology to solve the problem is discussed which contains configuration selection, operating conditions, flow sheet development, process description and scheme for thermodynamic and economic analyses.

3.1 Configuration Selection The criteria to select the configuration is high product purity, high flow rate and the economics involved in the process. Since all three parameters are directly related with number of distillation columns used in the system as increasing the number of column increases product purity and cost. Single column configuration is better in terms of process economics and energy consumption but it does not produce required product purity. Three column configuration is better in terms of exergy destruction in the system but adding extra column increases cost of the system a lot. There is not much difference between two and three column configurations in terms of production rate and product purity. So due to large utilization in the industry and the requirement of our process a double-column system was chosen.

3.2 Operating Conditions The general operating conditions were found from the literature and further modified during the optimization of the system which was the primary objective of the thesis [8]. The power rating of the IGCC power plant depends upon the amount of pure oxygen required for gasification process, which is produced by ASU, is estimated to be about 600MW.

19

The operating pressure ratios for compressors and their respective adiabatic efficiencies are given in Table 3. Table 3 Compressor specifications Compressors

Pressure Ratio

Adiabatic Efficiency

Compressor1

2.645

80%

Compressor2

2.593

76%

Compressor3

1.836

72%

Air composition and conditions are found in the literature, and for sake of simplicity the mole fraction are rounded. The composition of inlet air along with its conditions such as mass flow rate, temperature and pressure are given in Table 4 [8]. Table 4 Feed Air composition and Flow rate Mole Fraction% Ar

O2

N2

1

21

78

Flow rate (kg/hr)

Temperature oC

Pressure kPa

360000

25o

101.3

The number of stages for HPC and LPC are 50 and 60 respectively [8].

3.3 Solution Approach The complete flow sheet was developed, simulated and validated against the available information from the literature [8]. For this purpose ASPEN PLUS® 9.0 was used, the capital and operating cost was evaluated by Aspen Process Economic Analyzer (APEA).Equilibrium based model was used for distillation columns, equations for the model are discussed in section 2.1.2. The equilibrium model for distillation column is shown in Figure 10. The model state that for any stage of the distillation column the liquid and vapor phase are at equilibrium and the composition remain constant for a particular stage.

20

Figure 10 Stage column- Equilibrium based model Selection of the right property model is an important step in Aspen Plus® simulations. The property model is used by Aspen Plus® simulator to calculate the thermodynamic properties of the components. Peng-Robinson equations of state was used to calculate the properties and vapor liquid equilibrium (VLE) relationships.

3.4 Process Description The developed flow sheet of the process is shown in Figure 11 in which feed air is entering in the system at 1 and compressed in a two stages of compressors followed by intercoolers to achieve desired pressure around 100psi. The compressed air is then split into two fractions. The temperature of incoming air is than reduced with the help of cold purified products coming out from the distillation column in a multi-stream heat exchanger. Both the fractions are then fed to high pressure column (HPC) at different stages according to its composition. HPC will generate rectified streams, the stream enriched with nitrogen from top of the column and oxygen enriched stream from bottom of the column. The nitrogen enriched air from HPC is used as a reflux in low pressure column while oxygen enriched air from HPC is introduced in low pressure column as feed stream. 21

22

N2

O2 Figure 11 Flow sheet of Base Case

Before entering in low pressure column these stream further exchange heat with the N2 enriched product steam from the low pressure column to achieve temperature as low as possible. The final products from low pressure column (LPC) are nitrogen enriched product from top of the column and oxygen enriched product from bottom of the column.

3.5 Exergy Analysis Conventionally, energy analysis based on first law of thermodynamics has been used to evaluate the energy efficiency as a first step prior to detailed design of processes and power plants. However, it became a challenge due to increased complexity of the processes and power plants, competitive market and environment related regulations, thus superior analysis was essential to guarantee the optimal uses of energy resources without violating environment related constraints. Therefore, an advanced methodology was developed by combining both the first and second law of thermodynamics simultaneously [15]. In this analysis, the thermodynamic losses usually known as irreversibilities are evaluated based on 2nd-law rather than 1st-law. This type of analysis is usually known as 2nd-law analysis, exergy analysis, irreversibility analysis or availability analysis. Exergy analysis was successfully applied for the investigation and optimization of process and power plants [15-18]. Exergy analysis of an individual component of the process or the complete plant provides essential information about the wasted part of supplied exergy to a specific process or plant. This gives the portion of inefficiency for an individual unit or the complete plant. This analysis will help to identify the components which contribute most to the inefficiency of the plant. Exergy analysis gives key information to be used for the design and operation of plant [16]. A component of the air separation unit which has high exergy destruction high exergy loses and the improvement in the design and operation of this unit will have the maximum contribution for the improvement of the whole process or plant. The simulation of ASU process was run using Aspen Plus® V9.0 software to calculate the stream properties, flow rates and compositions. These results were utilized for the calculation of exergy at each point on the flow sheet. After collecting the exergy for all the streams, exergy destruction was evaluated using Table 6 and equation 6.

23

𝑇

∆𝐸 = ∑ (1 − 𝑇0 ) 𝑄𝑗 + (∑𝑖 𝑚̇𝑖 𝑒𝑥,𝑖 )𝑖𝑛 − (∑𝑖 𝑚̇𝑖 𝑒𝑥,𝑖 )𝑜𝑢𝑡 𝑗

..... (6)

Since exergy analysis can only inform about the inefficiencies of the individual components and the complete plant. The actual cost of the product must involve detailed design of all the equipment and process configuration. However major focus of optimization can be derived from the results of the exergy analysis.

3.6 Economic Analysis The capital investment and annual operating cost for the processes is calculated using Aspen Process Economic Analyzer (APEA) engine integrated in Aspen Plus®. There are four steps involved in an APEA economic analysis: Setting up the model for process economics, mapping unit operations to process equipment, sizing process equipment, evaluating cost and investment metrics. Step-1 involves collecting all of the information that will be needed for the analysis and creating a model that will be utilized for the economics evaluation. Information for units of measurement, currencies, operating hours per year, operating life of the plant, feed and product prices, and process utilities needed is collected and compiled in this step. Step-2 involves mapping of Aspen Plus® models in Aspen Process Economic Analyzer (APEA) which generates cost estimates based on various parameters of the Aspen Plus® models and streams. Mapping is basically the translation of a block, which is a set of equations, to real process equipment. Step-3 is sizing each piece of equipment according to the design standards and codes. The design parameters like dimensions of a pressure vessel, heat transfer area of heat exchangers etc. are calculated in this step. Step-4 Once the mapping, equipment sizing and customization steps are complete, the cost engine evaluates the cost of the process and its equipment. The Aspen Process Economic Analyzer (APEA) gives the operating, utility and capital costs. To compare these cost on same level it is converted into total annualized cost of the system. The total annualized cost (TAC) of the process can be calculated by using the following equation:

24

Total annual cost = Total Operating Cost + Annulized Capital Cost The capital cost can be annualized using; Annulized Capital Cost =

𝑖(𝑖 + 1)𝑛 × Total Capital Cost (𝑖 + 1)𝑛 − 1

Where 𝒊 is the fractional discount rate per year and n is the plant lifetime. The factor 𝒊(𝒊+𝟏)𝒏 (𝒊+𝟏)𝒏 −𝟏

is called fixed capital charge factor, hence

Annulized Capital Cost = Fixed Capital Charge Factor × Total Capital Cost

For 𝒊=0.125 and n=20 years, fixed capital charge factor comes out to be 0.114 for process of air separation [19].

3.7 Optimization Problem Formulation Since the integrated multi-column and multistage ASU is relatively a complex process. Presence of design details, multi stream heat exchanger and recycles futher complicates the ASU. Therefore, rigorous optimization is not used in this work. The optimization problem formulation is minx(f(x)) Where f(x) is the annual cost of the plant “x” is set of decision variables (pressure, split ratio, stages of distillation column etc.) g(x)=0 h(x)≤0 Where g(x) is a vector of equality constraints (energy and mass balance, thermodynamic relations etc.) Whereas h(x) is vector of inequality constraints such as YO2,P1≥ 98% where YO2,P1 is percentage of oxygen in the product stream 1 YO2,P2≤ 5% where YO2,P2 is percentage of oxygen in the product stream 2 [20, 21] 25

Input and operating conditions

Exergy analysis Important units of the process (HPC, LPC) NLPC, NHPC Effect of number of stages of LPC and HPC on product purity and total annualized cost

Evaluation of constraints

Yes

Constraint violation No

PR, SR, NLPC, NHPC

(small range of design decision space)

Parametric effect

Yes

Constraint violation No

Optimization results

Figure 12 Optimization scheme 26

The overall strategy of optimization is given as a flow chat in Figure 12 which shows that starting with the exergy analysis of the system gives the important units of the system which are the HPC and LPC. So the focus of optimization should be on them. After that the number of stages are lowered to the extent that they do not violate the constraints of the problem, which gives a small range of design decision space for number of stages. After that parametric effect of various parameters like pressure ratio, split ratio, number of stages of LPC and HPC was found. Lastly, all the parameters are manipulated at constant purity of 98% to find the optimized operating conditions.

27

4 Results and Discussions The sections discusses the simulation results of the base case which includes stream properties and exergy destruction across each component of the system. After that the effect of number of stages of distillation columns on product purity and annual cost are evaluated, which lead to small range of design decision space of 23-27 total number of stages. After that effect of split ratio and pressure ratios on product purity and annual cost were evaluated. And lastly, system was optimized by manipulating all these parameters at constant product purity of 98% to achieve lowest annualized cost of the system.

4.1 Simulation of base case and validation of results Peng-Robinson equation of state is used to calculate material properties. The properties include temperature, pressure, molar enthalpy, mass enthalpy, molar density, mass density, enthalpy flow, mole flows, mole fractions, volume flow, exergy flow rate, mass exergy and molar exergy of each and every material stream in the flow sheet. These properties were utilized in the exergy analysis to find exergy destruction across each component. The stream properties are shown in Table 5 , and the important properties like mole fraction, exergy flow rate, mass and molar exergies of each stream are written in bold face. The highest exergy flow rate is 3.91×104 kW which is of stream 15. The results of the stream properties were validated against the literature which showed almost 99% similarity with the literature. The results of the base case were validated which means that our system is valid and prepared for further manipulation.

28

Table 5 Stream Properties Stream Name From

Units

To

1

COMP1

Phase Temperature F

2

3

4

5

6

COMP1

C1

COMP2

C2

SPLITER

C1

COMP2

C2

SPLITER HEX1

Vapor

Vapor

Vapor

Vapor

Vapor

Vapor

77

290.717

86

309.264

86

86

Pressure

psia

14.6923

38.8612

38.8612

100.767

100.767

100.767

Molar Enthalpy Mass Enthalpy Molar Density Mass Density Enthalpy Flow Mole Flows

Btu/lbmol

-3.50809

1489.42

53.6851

1614.15

39.5677

39.5677

Btu/lb

-0.121095

51.4131

1.85314

55.7185

1.36583

1.36583

lbmol/cuft 0.0025525 0.0048242 0.0066449 0.012198

0.017264

0.017264

lb/cuft

0.0739449

0.139756

0.192502

0.353381

0.500138

0.500138

Btu/hr

-96108.8

4.08E+07

1.47E+06

4.42E+07 1.08E+06

lbmol/hr

27396.3

27396.3

27396.3

27396.3

27396.3

18835

N2

lbmol/hr

21369.1

21369.1

21369.1

21369.1

21369.1

14691.3

O2

lbmol/hr

5753.23

5753.23

5753.23

5753.23

5753.23

3955.34

AR

lbmol/hr

273.963

273.963

273.963

273.963

273.963

188.35

Mole Fractions N2

0.78

0.78

0.78

0.78

0.78

0.78

O2

0.21

0.21

0.21

0.21

0.21

0.21

AR

0.01

0.01

0.01

0.01

0.01

0.01

1.07E+07

5.68E+06

4.12E+06

kW

0

10219.8

8317.5

18672.1

16451.5

11310.4

kJ/kg

0

102.198

83.175

186.721

164.515

164.515

kJ/kmol

0

2960.65

2409.56

5409.27

4765.95

4765.95

Volume Flow Exergy flow rate Mass exergy Molar exergy

cuft/hr

29

745256

2.25E+06 1.59E+06 1.09E+06

Table 5 Stream Properties continued Stream Name From

Units

To Phase Temperature F

7

8

9

10

11

12

SPLITER

COMP3

C3

HEX1

T1

HEX1

COMP3

C3

HEX1

T1

HPC

HPC

Vapor

Vapor

Vapor

Mixed

Mixed

Vapor

86

229.066

76.46

-262.84

-278.893

-263.38

Pressure

psia

100.767

185.008

185.008

172.595

95.5799

94.1295

Molar Enthalpy Mass Enthalpy Molar Density Mass Density Enthalpy Flow Mole Flows

Btu/lbm ol Btu/lb

39.5677

1036.39

-47.3085

-4041.81

-4041.81

-2485.47

1.36583

35.7748

-1.63303

-139.518

-139.518

-85.7955

lbmol/cu ft lb/cuft

0.017264

0.025019

0.032364

0.362314

0.160864

0.050767

0.500138

0.724788

0.937576

10.4961

4.6602

1.47071

Btu/hr

338753

8.87E+06

-405024

-3.46E+07

-3.46E+07

-4.68E+07

lbmol/hr

8561.35

8561.35

8561.35

8561.35

8561.35

18835

N2

lbmol/hr

6677.85

6677.85

6677.85

6677.85

6677.85

14691.3

O2

lbmol/hr

1797.88

1797.88

1797.88

1797.88

1797.88

3955.34

AR

lbmol/hr

85.6135

85.6135

85.6135

85.6135

85.6135

188.35

Mole Fractions N2

0.78

0.78

0.78

0.78

0.78

0.78

O2

0.21

0.21

0.21

0.21

0.21

0.21

AR

0.01

0.01

0.01

0.01

0.01

0.01

cuft/hr

495903

342197

264533

23629.6

53221

371008

kW

5141.08

7080.29

6756.57

16693.3

16275.1

19064.6

kJ/kg

164.515

226.569

216.21

534.185

520.804

277.303

kJ/kmol

4765.95

6563.65

6263.55

15475.2

15087.6

8033.39

Volume Flow Exergy flow rate Mass exergy Molar exergy

30

Table 5 Stream Properties continued Stream Name From

Units

14

15

16

17

18

19

HPC

HEX2

T3

HPC

HEX2

T2

To

HEX2

T3

LPC

HEX2

T2

LPC

Phase

Liquid

Liquid

Mixed

Liquid

Liquid

Mixed

Temperature F

-276.997

-293.08

-313.994

-289.45

-302.26

-319.483

Pressure

psia

94.1295

92.6791

15.374

74.9845

73.5341

15.374

Molar Enthalpy Mass Enthalpy Molar Density Mass Density Enthalpy Flow Mole Flows

Btu/lb mol Btu/lb

-4818.16

-5046.03

-5046.03

-4764.25

-4950.06

-4950.06

-162.574

-170.263

-170.263

-169.913

-176.54

-176.54

1.82768

0.0866978

1.61224

1.69459

0.103129

54.1664

2.56944

45.2062

47.5153

2.89167

lbmol/cu 1.72446 ft lb/cuft 51.1073 Btu/hr

-7.69E+07 -8.05E+07 -8.05E+07

-5.45E+07 -5.66E+07 -5.66E+07

lbmol/hr

15957.1

15957.1

15957.1

11439.2

11439.2

11439.2

N2

lbmol/hr

9962.12

9962.12

9962.12

11407

11407

11407

O2

lbmol/hr

5742.18

5742.18

5742.18

11.0553

11.0553

11.0553

AR

lbmol/hr

252.823

252.823

252.823

21.1407

21.1407

21.1407

Mole Fractions N2

0.624306

0.624306

0.624306

0.997185

0.997185

0.997185

O2

0.35985

0.35985

0.35985

0.0009664 0.0009664 0.0009664

AR

0.0158439 0.0158439 0.0158439

0.0018481 0.0018481 0.0018481

cuft/hr

9253.39

8730.8

184054

7095.23

6750.41

110921

kW

36850.2

39057.6

38624.9

27168.3

28584.6

28345.3

kJ/kg

618.433

655.478

648.217

672.258

707.303

701.381

kJ/kmol 18328.3

19426.2

19211

18849.7

19832.3

19666.3

Volume Flow Exergy flow rate Mass exergy Molar exergy

31

Table 5 Stream Properties continued Stream Name From

Units

20

21

23

24

N2

O2

HEX1

HEX1

LPC

PUMP

LPC

HEX2

To

PUMP

HEX1

HEX2

HEX1

Phase

Liquid

Liquid

Vapor

Vapor

Vapor

Vapor

Temperature F

-296.945

-296.681

-317.383

-281.875

71.524

71.524

Pressure

psia

15.374

69.6135

15.374

15.0114

14.6923

68.1677

Molar Enthalpy Mass Enthalpy Molar Density Mass Density Enthalpy Flow Mole Flows

Btu/lb mol Btu/lb

-5527.74

-5522.13

-2763.81

-2509.75

-41.4806

-57.7419

-172.74

-172.565

-97.5259

-88.561

-1.46371

-1.80442

2.23622

0.010499

0.008049

0.002579

0.012012

71.5599

0.297544

0.228115

0.073078

0.384399

lbmol/cu 2.23764 ft lb/cuft 71.6052 Btu/hr

-2.61E+07 -2.61E+07 -6.27E+07 -5.69E+07 -940708

-272429

lbmol/hr

4718.05

4718.05

22678.3

22678.3

22678.3

4718.05

N2

lbmol/hr

3.09E-14

3.09E-14

21369.1

21369.1

21369.1

3.09E-14

O2

lbmol/hr

4717.14

4717.14

1036.09

1036.09

1036.09

4717.14

AR

lbmol/hr

0.907779

0.907779

273.056

273.056

273.056

0.907779

Mole Fractions N2

6.55E-18

6.55E-18

0.942273

0.942273

0.942273

6.55E-18

O2

0.999808

0.999808

0.045686

0.045686

0.045686

0.999808

AR

0.0001924 0.0001924 0.01204

0.01204

0.01204

0.000192

cuft/hr

2108.49

2109.83

2.16E+06

2.82E+06

8.79E+06 392767

kW

11877.3

11877.7

15214.6

11051.8

-0.44592

2256.4

kJ/kg

624.366

624.383

187.888

136.481

-0.00551

118.614

kJ/kmol 19979.9

19980.5

5324.61

3867.77

-0.15606

3795.68

Volume Flow Exergy flow rate Mass exergy Molar exergy

32

4.2 Exergy Analysis The Table 6 gives the exergy destruction across different components which shows that the largest irreversibilities are due to the two distillation columns. Almost 70% of the total exergy destruction is caused by the two distillation columns. Large exergy destructions of LPC and HPC are due to mixing of various streams and phases in both columns such as feed, reflux, side streams and vapour-liquid contact and mixing at each stage of distillation columns. Then intercoolers and compressors are the bigger contributors with 12% and 8% respectively. It also shows that the pump has the lowest exergy destruction among all components as pumping in this case is related to single phase incompressible flow and there is no heat transfer. Exergy analysis gives the insight about the components of system which are prone to greater loses. So we should focus on these components of high irreversibilities. Hence, the main focus of optimization should be to improve the design/operation of HPC and LPC. Table 6 Exergy Destruction of components

Equipment

Exergy destruction(kW)

COMP1

5765

C1

9300

COMP2

6447

C2

10287

COMP3

562

C3

2370

HEX1

4044

HEX2

2782

T1

418

T2

233

T3

442

PUMP

6

LPC

49587

HPC

63866

Total

156112 33

4.3 Cost Analysis The cost analysis of the process gives the operating, utility, equipment, installation and capital costs. It also shows the detail of equipment sizes that should be required for the process. The important results of cost and utilities are shown below in Table 7, Table 8 and Table 9. For cost analysis all the required data about utilities cost, period of return etc. are set prior to running the analysis and some values are taken automatically by Aspen Process Economic Analyzer (APEA). These results are converted into total annualized cost to be further used in parametric study and optimization problem solution. Table 8 gives the details about each of the equipment size, installation cost and operation cost per hour. Table 7 shows the price of electricity to run different equipment, cooling water utilization in coolers and its cost per hour, and steam utilization in reboiler of distillation column and its per hour price. Table 9 shows the summary of all type of costs involved in the cost analysis which include total equipment cost, operating cost and total capital cost. This data is used to calculate total annualized cost for the base case which comes out to be 38.85 MUSD/year. This entire procedure of cost analysis to repeated for every data point during the parametric study and optimization problem solution.

Table 7 Utilities cost

Utilities Cost Name

Fluid

Electricity

Rate

Rate Units

Cost

Cost Units

27332

KW

2118

USD/H

Cooling Water

Water

0.55

MMGAL/H

66

USD/H

Steam @100PSI

Steam

54.44

KLB/H

443

USD/H

23.03

MUSD/year

Total Utilities Cost

34

Table 8 Cost of unit operations

Unit operation Name

Equipment Cost [MUSD]

Installed Cost [MUSD]

Utility Cost [USD/HR]

PUMP

7.20×10-3

4.98×10-2

0.87

COMP2

11.25

12.02

983

LPC

1.45

2.41

443

C2

.29

.46

31

COMP3

1.45

1.72

202.35

C3

.13

.26

6.68

HEX2

.44

1.30

0

COMP1

26.33

27.45

983

SPLITER

0

0

0

HEX1

3.03

4.58

0

C1

.74

1.08

28.3

HPC

.82

1.66

7.23

Table 9 Summery of Cost Analysis

Summary Equipment Cost [MUSD/Year]

45.92

Total Installed Cost [MUSD]

53

Total Raw Materials Cost [MUSD/Year]

0

Total Capital Cost [MUSD]

82.82

Total Operating Cost [MUSD/Year]

29.41

35

4.4 Parametric Study Effect of various design and operating parameters such as PR, SR and number of stages of distillation columns of LPC and HPC was evaluated on product purity and total annual cost. The most important parameters significantly effecting the product purity and total annual cost were identified and then this information was used for optimization.

4.4.1 Stages of distillation columns As presented in exergy analysis, the most of the exergy destruction was contributed by HPC and LPC. Hence, the improvement in its design or operating parameters will have significant effect in the improvement of the whole ASU. In this section, the effect of number of stages for HPC and LPC has been evaluated on product purity and annual cost, and presented in Figure 13 and Figure 15 respectively. Various combinations of both HPC and LPC stages have been considered and shown in those figures. Since the constraint for the minimum mole fraction of O2 in oxygen product stream was 98%. Therefore, the no. of stages of both distillation columns can be reduced as shown in Figure 13 and Figure 14. However, below the number of stages (HPC=11, LPC=12) the reduction of stages is not possible due to violation of second constraint which is the O2 losses in the nitrogen stream which is set on less than 5% as shown in Figure 14. As we have observed from the exergy analysis that the parameters of LPC and HPC will have significant effect on the overall cost as compared to other parameters. Therefore optimal number of stages of stages will be around the combination of HPC=11 and LPC=12, as other parameters may not have significant effect as was observed for number of stages. There was also a possibility of having different effect on cost and purity for different combination of LPC and HPC. But this data was helpful in finding an optimal range of stages around 11 to 14 for each column.

36

O2 purity vs Stages

1

0.996

O2 purity

0.992

0.988

0.984

0.98 LPC HPC

12 13 13 15 15 20 20 25 25 30 30 35 40 40 40 40 40 44 48 52 56 11 12 13 13 15 15 20 20 25 25 30 30 30 35 40 45 50 50 50 50 50

Number of stages

Figure 13 O2 purity VS Stages The number of stages can be further reduced to a limited range and violation of constraints can be avoided by the variation of other parameters such as SR, PR and calibrating the operating conditions of the heat exchangers. The base case had conservative approach; really high number of stages and high purity in oxygen stream. The number of

stages were lowered to the extent of not violating the

constraints for our study which were 98% oxyen in oxygen product stream and 5% oxygen in nitrogen product stream. After that other parameters were studied to determine their effect on purity and annual cost.

37

O2 in N2 Stream VS Stages 0.05 0.0495 0.049

O2 in N2 stream

0.0485 0.048 0.0475 0.047 0.0465

0.046 0.0455 0.045

LPC HPC

12 13 13 15 15 20 20 25 25 30 30 35 40 40 40 40 40 44 48 52 56 11 12 13 13 15 15 20 20 25 25 30 30 30 35 40 45 50 50 50 50 50

Stages

Figure 14 O2 in N2 Stream VS Stages

Total annualized cost VS Stages 39

Total annualized cost

38.8 38.6 38.4 38.2 38 37.8 37.6 37.4 37.2

37

LPC HPC

12 13 13 15 15 20 20 25 25 30 30 35 40 40 11 12 13 13 15 15 20 20 25 25 30 30 30 35

40 40 40 44 48 52 56 40 45 50 50 50 50 50

Stages

Figure 15 Total annualized cost VS Stages

38

4.4.2 Split Ratio After varying the number of stages to find the reduced search space for optimal number of stages, split ratio was changed to find the effect of split ratio on product purity and total annual cost of the system which is showed in Figure 16 and Figure 17, respectively. Firstly the split ratio was changed to find corresponding product purity, which showed that changing the split ratio has negligible effect on the product purity because the reason for splitting the feed stream and increasing its pressure was to extract maximum cold energy from the incoming products and had negligible effect on product purity as shown in Figure 16.

O2 purity vs Split ratio

0.99

0.985

O2 purity

0.98

0.975

0.97

0.965

0.96 0.2

0.24

0.28

0.32

0.36

0.4

Split ratio

Figure 16 O2 purity VS split ratio

But it had significant effect on the cost of the ASU system as shown in Figure 17. This is because increasing the split ratio will increase the flow rate of compressor 3 of the system, and thus increasing the compressor size and duty which causes the cost to increase. Therefore an optimum split ratio should as minimum as possible to achieve the desired effect, better cryogenic condition and not to increase the cost too much. It was noticed that decreasing the split ratio decreases the cost but to achieve the desired 39

cryogenic conditions after first heat exchanger the optimum split ratio selected was 31.25% because after that the cost of the system increases rapidly.

Total annualized cost vs Split ratio 42

Total annualized cost

41

40

39

38

37

36 0.2

0.24

0.28

0.32

0.36

0.4

Split ratio

Figure 17 Total annualized cost VS split ratio

4.4.3 Pressure Ratio The pressure ratio is an important parameter of the air separation unit and it has substantial effect on the cost and product purity of the system. Compressor 3 has the only function to raise the pressure of the split stream so that it can exchange heat with the product streams at higher capacity. The graph between pressure ratio of COMP3 and product purity in Figure 18 presented a negligible effect same as with split ratio. But the cost increases or decreases with increasing or decreasing pressure ratio, respectively. So an optimum split ratio and pressure ratio of COMP3 should be selected to achieve lowest temperature as the minimum requirement after the first heat exchanger and better utilization of the cold energy of the products. Both the split ratio and compression ratio should be kept as low as possible. The pressure ratio of compressor 3 was chosen to be 1.836 for our system same as was in base case to achieve better cryogenic conditions. 40

O2 purity vs PR of COMP3 0.99

0.985

O2 purity

0.98

0.975

0.97

0.965

0.96 1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Pressure ratio of COMP3

Figure 18 O2 purity VS PR of COMP3 The pressure ratios of COMP1 and COMP2 determine the pressure of the system, so the variations of their parameters will affect the whole ASU system i.e. the crossover temperatures of HEX1 and HEX2, pressure of the high pressure distillation column (HPC), cooler duties and mainly the product purity and cost of the system. The pressure ratios of COMP1 and COMP2 were changed simultaneously, the values were changed with increment and decrement of 15% with respect to base case values which are given in the Table 10 below. The graph between pressure ratios of COMP1 and COMP2 with product purity is shown in Figure 19. Another graph was plotted for O2 in nitrogen stream in Figure 20. Table 10 Pressure ratios of COMP1 and COMP2

SR#

Change

PR of COMP1

PR of COMP2

1

-15%

2.25

2.2

2

0

2.65

2.59

3

15%

3.04

2.98

41

O2 purity VS Pressure Ratios 0.99

0.985

O2 purity

0.98

0.975

0.97

0.965

0.96 -20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

Pressure Ratios

Figure 19 O2 purity VS Pressure Ratios The results in Figure 18 and Figure 19 showed that decreasing the pressure ratio increases the product purity. At 15% higher pressure ratio values than the base case values, product purity comes out to be 96.4% and at 15% lower values it is 98.6%. The feed conditions such as flow rate were not changed during any plot. Since decreasing pressure increases purity it should be superior to operate at low pressure but it also decrease the amount of product as well as the wastage of O2 which means the amount of oxygen in nitrogen also increases showed in Figure 19. So decreasing pressure will require high feed flow rate, high equipment size and thus increasing the cost of the system if compared for same amount of product. Figure 20 shows the effect of pressure on annualized cost of the system which shows that by increasing the pressure cost of the system increases and vice versa. Hence, optimum values of pressure ratios should be taken to keep the oxygen percentage less than 5% in the nitrogen stream to limit the wastage as low as possible

42

O2 in N2 Stream VS Pressure Ratios 0.07 0.06

O2 in N2 Stream

0.05 0.04 0.03

0.02 0.01 0 -20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

15%

20%

Pressure Ratios

Figure 20 O2 in N2 Stream VS Pressure Ratios

total annualized cost vs Pressure ratios 45

40

total annualized cost

35 30 25 20 15 10 5 0 -20%

-15%

-10%

-5%

0%

5%

10%

Pressure ratios

Figure 21 Total annualized cost VS Pressure ratios 43

4.5 Optimization From the study of various parameters and their effect on product purity and annual cost, near to optimal ranges of parameter were found. For optimization various combination of number of stages of LPC and HPC were employed along with different operating conditions while product purity was kept constant. First of all flow sheet of the process was manipulated each and every time according to the values so that there are no errors left in the system. After that cost analysis was done for each step and the total annualized cost for every change in the parameters was calculated. For the already decided range of 11 t o14 stages for both columns the total annual cost was evaluated for various combinations for the number of stages as shown in Figure 22. The constraints of product purity if violated were made to move into the feasible region by the variation of other parameters such as PR and SR. The plot in Figure 22 covers all the combinations within the already selected optimum range of number of stages. The minimum annualized cost for ASU was found to be corresponding to 12 stages of LPC and 13 stages of HPC. The corresponding cost reduction was 37.509 MUSD/year.

total annualized cost vs stages of LPC Total annualized cost [MUSD/year]

40

39

38

HPC12

HPC13

HPC14

37

36 10

11

12

13

number of stages of LPC

Figure 22 Total annualized cost VS stages of LPC 44

14

15

The focus of the optimization was to reduce the cost of the system. Thus, the final operating conditions of the system in comparison to base case are shown in Table 11 and Table 12. Our system is optimized at the operating conditions shown in Table 11 and Table 12. The cost comparison with base case is discussed in Table 13 which shows that for optimized conditions the total annualized cost comes out to be 37.5 MUSD/year which is 1.35 MUSD/year lower than the base case price which would become a huge amount of 27 MUSD when considered for whole plant life of 20 years. Table 11 Final conditions for Compressors

Compressor

Pressure ratio (Optimized case)

Pressure ratio (Base case)

COMP1

2.5

2.625

COMP2

2.48

2.593

COMP3

1.836

1.836

Table 12 Final conditions for Columns Columns

Stages (optimized case)

Stages (base case)

LPC

12

60

HPC

13

50

Table 13 Cost comparison with base case

Performance Indicators

Optimized case

Base case

Operating cost (MUSD/year)

28.26

29.41

Capital cost (MUSD)

81.10

82.82

Total annualized cost (MUSD/year)

37.50

38.85

45

5 Conclusion and Recommendations Two step methodologies was used to optimize the cryogenic air separation unit. The first step was the exergy analysis, to find the most important units of the ASU and parameters. The second step was the optimization by parametric variation. The equipment contributing largest to the exergy destruction were the two distillation columns including low pressure column and high pressure column. This share was nearly 70% of the total exergy destruction for the complete ASU system. For this purpose, a double–column configuration was selected on the bases of required product purity and flow rate. Simulations were run using the ASPEN PLUS® 9.0 for stream properties and economic assessment of the system. Parametric study showed the dependence of product purity and cost on the number of stages (for LPC and HPC), split ratio and pressure ratio of the first and second compressor. The constraints for optimization were 98% purity of oxygen stream and oxygen leaving in the nitrogen stream not more than 5%. For optimization of the system different combinations of number of stages and operating condition were employed to obtain minimum total annualized cost.

5.1 Conclusions The findings of this thesis can be concluded as follows: 

HPC and LPC distillation columns share around 70% of the total exergy destruction of the complete ASU system.



Relatively small portion of the total exergy destruction was caused by intercoolers and compressors to be 12 % and 8 %, respectively.



The least contribution in the exergy destruction was found by the pump due to single phase flow, incompressible fluid, no changes in temperature, no heat transfer and no phase change.



Split ratio and pressure ratio of the compressors do not have any significant effect on the product purity. However, total annual cost was found to increase by increasing split ratio and pressure ratio.

46



The optimum numbers of stages for LPC were 12 and for HPC were 13 and the corresponding minimum cost was 37.5 MUSD/year which is nearly 4 % less than the base case.

5.2 Recommendations The following recommendations are given for future work: 

Cost comparison with other technologies of air separation should be performed.



Energy integration with the other units of IGCC power plant should also be considered for cost reduction of air separation.



Time based scheduling and optimization of the plant for flexible operation as the part of IGCC should also be considered.

47

References [1]

K. Roh and J. H. Lee, "Control structure selection for the elevated-pressure air separation unit in an IGCC power plant: self-optimizing control structure for economical operation," Industrial & Engineering Chemistry Research, vol. 53, pp. 7479-7488, 2014.

[2]

A. Smith and J. Klosek, "A review of air separation technologies and their integration with energy conversion processes," Fuel processing technology, vol. 70, pp. 115-134, 2001.

[3]

Z. A. M. Khalel, A. A. Rabah, and T. A. M. Barakat, "A New Cryogenic Air Separation Process with Flash Separator," ISRN Thermodynamics, vol. 2013, pp. 1-4, 2013.

[4]

Q. Fu, Y. kansha, C. Song, Y. Liu, M. Ishizuka, and A. Tsutsumi, "An Advanced Cryogenic Air Separation Process Based on Self-heat Recuperation for CO2 Separation," Energy Procedia, vol. 61, pp. 1673-1676, 2014.

[5]

H. Zhou, Y. Cai, Y. Xiao, Z. A. Mkhalel, B. You, J. Shi, et al., "Process Configurations and Simulations for a Novel Single-Column Cryogenic Air Separation Process," Industrial & Engineering Chemistry Research, vol. 51, pp. 15431-15439, 2012.

[6]

M. Mehrpooya, M. M. Moftakhari Sharifzadeh, and M. A. Rosen, "Optimum design and exergy analysis of a novel cryogenic air separation process with LNG (liquefied natural gas) cold energy utilization," Energy, vol. 90, pp. 2047-2069, 2015.

[7]

Z. Jieyu, L. Yanzhong, L. Guangpeng, and S. Biao, "Simulation of a Novel Single-column Cryogenic Air Separation Process Using LNG Cold Energy," Physics Procedia, vol. 67, pp. 116-122, 2015.

[8]

A. Ebrahimi, M. Meratizaman, H. Akbarpour Reyhani, O. Pourali, and M. Amidpour, "Energetic, exergetic and economic assessment of oxygen production from two columns cryogenic air separation unit," Energy, vol. 90, pp. 1298-1316, 2015.

[9]

L. V. van der Ham and S. Kjelstrup, "Exergy analysis of two cryogenic air separation processes," Energy, vol. 35, pp. 4731-4739, 2010. 48

[10]

L. V. van der Ham, "Improving the exergy efficiency of a cryogenic air separation unit as part of an integrated gasification combined cycle," Energy Conversion and Management, vol. 61, pp. 31-42, 2012.

[11]

S. Tesch, T. Morosuk, and G. Tsatsaronis, "Exergetic and economic evaluation of safety-related concepts for the regasification of LNG integrated into air separation processes," Energy, 2017.

[12]

A. Ebrahimi and M. Ziabasharhagh, "Optimal design and integration of a cryogenic Air Separation Unit (ASU) with Liquefied Natural Gas (LNG) as heat sink, thermodynamic and economic analyses," Energy, vol. 126, pp. 868885, 2017.

[13]

M. Aneke and M. Wang, "Potential for improving the energy efficiency of cryogenic air separation unit (ASU) using binary heat recovery cycles," Applied Thermal Engineering, vol. 81, pp. 223-231, 2015.

[14]

Q. Fu, Y. Kansha, C. Song, Y. Liu, M. Ishizuka, and A. Tsutsumi, "An elevated-pressure cryogenic air separation unit based on self-heat recuperation technology for integrated gasification combined cycle systems," Energy, vol. 103, pp. 440-446, 2016.

[15]

"The Exergy Method of Thermal Plant Analysis. Von T. J. Kotas. Butterworths, London 1985. XIX, 296 S., zahlr. Abb. u. Tab., geb., £ 45,–," Chemie Ingenieur Technik, vol. 59, pp. 365-365, 1987.

[16]

A. Bejan and G. Tsatsaronis, Thermal design and optimization: John Wiley & Sons, 1996.

[17]

M. Hazami, F. Mehdaoui, N. Naili, M. Noro, R. Lazzarin, and A. Guizani, "Energetic, exergetic and economic analysis of an innovative Solar CombiSystem (SCS) producing thermal and electric energies: Application in residential and tertiary households," Energy Conversion and Management, vol. 140, pp. 36-50, 5/15/ 2017.

[18]

V. Mrzljak, I. Poljak, and T. Mrakovčić, "Energy and exergy analysis of the turbo-generators and steam turbine for the main feed water pump drive on LNG carrier," Energy Conversion and Management, vol. 140, pp. 307-323, 5/15/ 2017.

[19]

H. Zhai, K. Kietzke, and E. Rubin, "IECM technical documentation: probabilistic comparative assessment using the IECM," ed: Carnegie Mellon University: Pittsburgh, PA, 2012. 49

[20]

A.

Jaya,

"AIR

SEPARATION

UNITS

(ENGINEERING

DESIGN

GUIDELINE) " 2013. [21]

R. J. Allam, "Improved oxygen production technologies," Energy Procedia, vol. 1, pp. 461-470, 2009.

50