1.0 Introduction
The minor equipment that is designed in this project is lean/rich MEA heat exchanger E-114. This heat exchanger is a counter flow shell and tube heat exchanger and is designed to heat up the rich MEA stream flowing from the CO2 absorber to the stripper. The principle that is applied is heat exchange between cold stream and hot stream which in this case the heat energy is transferred from the lean MEA stream to the rich MEA stream. Apart from this, the chemical engineering design for this heat exchanger includes the determination of its dimensions and heat exchange coefficient as well as pressure drop. The mechanical design covers the design of pressure vessel, head, supports and piping. In addition, the operating design which includes the commissioning, start-up, shutdown and maintenance procedures, process control, and HAZOP study is considered. 2.0 Process Description

Figure 2.1 Schematic of rich/lean MEA heat exchange process flow sheet
The lean/rich MEA heat exchange process is presented in Figure 2.1. The MEA-2 stream containing rich CO2 is flowing from CO2 absorber and enters the heat exchanger to be heated up from 61°C to 80°C by MEA-7 before entering the stripper. The MEA-7 is then cooled down from 105°C to 84°C when pass through the heat exchanger and recycle back to the CO2 absorber. The cold stream in this case is MEA-2 and MEA-3 while the hot stream is MEA-7 and MEA-8.
3.0 Chemical Engineering Design
3.1 Design Methodology
The rich/lean MEA heat exchanger is a counter flow shell and tube heat exchanger. The chemical engineering design methodology for this heat exchanger includes the following steps of Kern’s method according to Sinnott (2005): (a) assume overall heat transfer coefficient , U; (b) select number of shell and tube passes, calculate ΔTlm, correction factor, F and ΔTm; (c) determine heat transfer area; (d) decide tube size, type and arrangements; (e) Calculate number of tubes; (f) Calculate shell diameter; (g) Estimate tube-side heat transfer coefficient; (h) Decide baffle spacing and estimate shell-side heat transfer coefficient; (i) Calculate overall heat transfer coefficient and (j) Estimate tube and shell side pressure drop. Coolant is the contaminated solution of MEA with CO2 which is corrosive; hence it should flow through tube-side.

3.2 Overall heat transfer coefficient determination
a) Assume overall heat transfer coefficient, U
The hot and cold fluid pass through the heat exchanger is dilute stream with large amount of water, hence, its overall heat transfer unit is initially assumed as water, U = 1000 W/m2.°C.
b) Select number of shell and tube passes, calculate ΔTlm, correction factor, F and ΔTm
The heat exchanger is cross-flow with 1 shell pass and 2 tube passes. The log mean temperature can be obtained from equation 3.1.
∆Tlm=Th,i-Tc,o-(Th,o-Tc,i)ln⁡(Th,i-Tc,oTh,o-Tc,i) (3.1)
R=Th,i-Th,oTc,o-Tc,i 3.2 S=Tc,o-Tc,iTh,i-Tc,i (3.3)
Where
* Th,i = 104.65°C is the inlet hot fluid temperature; * Th,o = 104.65°C is the outlet hot fluid temperature; * Tc,i = 104.65°C is the inlet cold fluid temperature; * Tc,o = 104.65°C is the outlet cold fluid temperature.
The ΔTlm = 23.87°C, R = 1.08 and S = 0.43 is calculated from the equation 3.1, 3.2 and 3.3 respectively. In order to obtain the true mean temperature difference, the correction factor has to be found by using Figure 12.29 in Sinnott (2005) or equation 3.4.

(3.4)

∆Tm=Ft∆Tlm (3.5)
The calculated true mean temperature difference is 21.25°C.
c) Determine heat transfer area
A=qU∆Tm (3.6)
Where
* q = 8.81×106 W is the heat duty of the heat exchanger obtained from Hysys; * U = 1000 W/m2°C is the initial guess of heat transfer coefficient.
Heat transfer area obtained is 414.62 m2.
d) Decide tube size, type and arrangements
The standard tube size with outer diameter 20 mm, inner diameter 18 mm and length 4.88m is used in this heat exchanger. The tube is made of stainless steel due to high corrosive MEA and CO2. The area of one tube is 0.31 m2.
e) Calculate number of tubes
By dividing heat transfer area with a single tube’s area, the number of tubes which is 1350 can be obtained.
(f) Calculate shell diameter
The tube arrangement is triangular pitch and the bundle diameter, Db = 985 mm can be calculated from equation 3.7.
Bundle diameter, Db=DoNK11n1 (3.7)
Where K1 and n1 are obtained from Table 12.4 (Sinnott 2005)
In order to estimate the shell diameter, bundle diametrical clearance is taken from Figure 12.10 in Sinnott (2005) using split-ring floating head type. The bundle diametrical clearance is therefore 79 mm. The shell diameter can then be known by totaling the bundle diameter and bundle diametrical clearance and gives 1.06 m.
(g) Estimate tube-side heat transfer coefficient
Tube side coefficient is calculated using equation 3.8. hidik=jhRePr0.33μμw0.14 (3.8)
Where
* di = 0.018 m is the inner diameter; * k = 0.561 W/m2.°C is the fluid thermal conductivity; * µ/µw is neglected as its value is insignificant small; * Re = 82418 is Reynolds number calculated from (ρudi/µ where ρ = 1046 kg/m3, u = 2.81 m/s and µ = 0.00064 Ns/m2); * Pr = 4.4 is Prandtl number calculated from (Cpµ/k, where Cp = 3845 J/kg°C) * jh = 0.003 is heat transfer factor taken from Figure 12.23 in Sinnott (2005) using L/D = 271.
The tube side coefficient obtained is 12600 W/m2°C.
(h) Decide baffle spacing and estimate shell-side heat transfer coefficient
The cross sectional flow area of shell can be achieved using equation 3.9 As=pt-doDslBpt (3.9)
Where
* pt = 0.025 m is the tube pitch taken from table 12.4 in Sinnott (2005); * lB = 0.36 m is the baffle spacing.
Shell side coefficient can later be calculated from equation 3.11.
Equivalent diameter, de=1.10dopt2-0.917do2 (3.10) hsdek=jhRePr13μμw0.14 (3.11)
Where
* de = 0.014 m is the equivalent diameter gained from equation 3.0; * k = 0.586 W/m2.°C is the fluid thermal conductivity; * µ/µw is neglected as its value is insignificant small; * Re = 260000 is Reynolds number calculated from (Gde/µ where G = 5810 kg/m2s and µ = 0.00032 Ns/m2); * Pr = 2.27 is Prandtl number calculated from (Cpµ/k, where Cp = 4207 J/kg°C) * jh = 0.001 is heat transfer factor taken from Figure 12.29 in Sinnott (2005) by assuming 25% baffle cut.
The shell side coefficient obtained is 14100 W/m2°C.
(i) Calculate overall heat transfer coefficient
1Uo=1ho+1hod+doln⁡(dodi)2kw+dodi×1hid×dodi×1hi (3.12)
Where
* kw = 16 W/m°C is the thermal conductivity of stainless steel; * hod = hid = 3000 W/m2°C is the fouling coefficient for aqueous salt solution.
The overall transfer coefficient, U = 1080 W/m2°C which is above the initial estimate of 1000 W/m2°C can be calculated from equation 3.12. This implies that U = 1000 W/m2°C is satisfactory. However, the pressure drops are needed to be checked.

3.3 Tube and shell side pressure drop estimation
(j) Estimate tube and shell side pressure drop
Tube side pressure drop
∆Pt=Np8jfLdiμμw-m+2.5ρut22 (3.13)
Where
Np = 2 is the number of tube passes; jf = 0.003 is firction factor taken from Figure 12.24 in Sinnott (2005) using Re = 82418; µ/µw is neglected as its value is insignificant small; ut = 2.81 m/s is the fluid velocity.
The calculated pressure drop at tube side is 73.7 kPa.
Shell side pressure drop
∆Ps=8jfDsdeLlBρus22μμw-0.14 (3.14)
Where
* jf = 0.03 is firction factor taken from Figure 12.30 in Sinnott (2005) using Re = 260000 and assume 25% baffle cut. µ/µw is neglected as its value is insignificant small;
The calculated pressure drop at shell side is 4248 kPa which is too high. This could be reduced by doubling the pitch halves the shell side velocity, which reduces the pressure drop by a factor of about (1/2)2. This gives the pressure drop at shell side of 1062 kPa which is acceptable. On the other hand, this will reduce the shell side coefficient by a factor of (1/2)0.8 and results to hs = 8112 W/m2°C. However, the overall coefficient is reduced to 1030 W/m2°C which is still above the initial estimate of 1000 W/m2°C.
4.0 Mechanical Design
4.1 Materials of Construction
Material selection is the fundamental considerations in engineering design as significant time and cost can be saved in design work and design errors can be avoided. Material selected lean/rich MEA heat exchanger should form a well-coordinated and integrated entity in which it not only can meet the requirements of its functional utility, but also safety and product purity. The lean/rich MEA heat exchanger is required to handle highly corrosive materials such as CO2 and MEA. Thus, stainless steel 304 which is highly heat and corrosion resistant is chosen for it in order to ensure the product purity and safety.

4.2 Design of Thin-walled Vessels Under Internal Pressure
4.2.1 Vessel Thickness
In order to resist internal pressure, the minimum thickness of the cylindrical shell lean/rich MEA heat exchanger has to be determined from equation 4.1. e=PiDi2Jf-Pi+corrosion allowance (4.1) where * Pi = 3.3 N/mm2 (at design pressure) is the internal pressure; * Di = 1060 mm is the internal diameter; * J =1 is the welded joint factor where double-welded butt joint and 100 per cent degree of radiography is used as the joint type for the vessel; * f = 145 N/mm2 (for stainless steel at design temperature 100°C) is the design stress; * corrosion allowance = 4.0 mm as lean/rich MEA heat exchanger is required to handle high corrosive MEA and CO2 thus high corrosion allowance is needed.
The total vessel thickness is 16.2 mm.

4.2.2 Head and Closure
Ellipsoidal heads, which are shown in Figure 4.1, are chosen as the covers for the vessel since they are the most economical closure for the operating pressure above 15 bars which is at 33 bars in this process. Sinnott (2005, 819) states that for most standard ellipsoidal heads, they are made with a major and a minor axis ratio of 2 : 1 that ease the calculation of the minimum thickness of head in equation 4.2. e=PiDi2Jf-0.2Pi+corrosion allowance (4.2) where the required values are taken from Section 4.2.1 in which the values are used to calculate the vessel thickness.
The total head thickness is 16.1 mm.

Figure 4.1 Ellipsoidal head (Sinnott 2005, 817)
4.3 Saddle Support
Saddles are chosen for the support for lean/rich MEA heat exchanger as they are suitable for cylindrical vessels. The saddles will support the horizontal vessel at two sections as illustrated in Figure 4.2. The saddles are designed with the consideration of the load imposed by the weight of the vessel and contents as well as the weight of liquid. The total weight load can be determined from equation 4.3, 4.4 and 4.5.

Figure 4.2. Horizontal cylindrical vessel on support (Sinnott 2005)
a) Weight of Shell
Weight of the cylindrical vessel’s shell with ellipsoidal heads and uniform wall thickness can be determined from the equation 4.3.
Wv=CvπρmDmgHv+0.8Dmt×10-3 (4.3) where * Wv is the total weight of the shell; * Cv is a factor of the weight of nozzles, manways, internal supports and etc. (1.08 for vessels with only a few internal fittings); * Hv = 2.44 m is the length of the vessel; * t = 16.2 mm is the wall thickness; * ρ is the density of vessel material (8030 kg/m3 for SS 304 from Society for Amateur Scientists (2004)); * Dm = 1.08 m is the mean diameter of vessel.
The weight of the shell is 15.5 kN.
b) Weight of tubes
Wt=πdo24×L×N-πdi24×L×N×ρ×g (4.4)
Where
* do = 0.02 m is the tube outer diameter; * di = 0.018 m is the tube inner diameter; * L = 4.88 m is the tube length; * N = 1350 is number of tubes. * ρ = 8030 kg/m3 is the density of tube material
The weight of the tube is 31 kN.
c) Weight of liquid
WL=πd24×L×ρ×g (4.5)
Where
* d = 1.06 m is the vessel diameter; * L = 2.44 m is the vessel length; * ρ = 1000 kg/m3 is the liquid density.

The weight of the liquid is 21.3 kN.
Hence, the total weight for the vessel is 67.8 kN by totaling up the weight of shell, tube and liquid.
The saddles design based on the total weight loads is illustrated in Figure 4.3. The design is according to the dimensions of typical “standard” saddle designs (Sinnott 2005). In addition, the bolt diameter is 20 mm and bolt holes are 25 mm.

Figure 4.3. Saddles design

4.4 Piping
The optimum pipe diameter and pipe wall thickness for the pipes that connected to the lean/rich MEA heat exchanger is determined from the equation 4.6 and 4.7 respectively (Sinnott 2005). In addition, safe working pressure is obtained from equation 4.8. The material selected for the pipes is stainless steel 304 due to the present of corrosive MEA and CO2 in the streams. The streams that connected to the lean/rich heat exchanger are MEA-2, MEA-3, MEA-7 and MEA-8 which are specified in PFD in Appendix in major report. The optimum diameter, nominal diameter and pipe wall thickness as well as the safe working pressure obtained are listed in Table 1. d,optimum=260G0.52ρ-0.37 (4.6) t=Pd20σd+P (4.7)
Ps=(schedule no.)×σs1000 (4.8) where * P = 2 bars for MEA-2 and MEA-3 while P = 30 bars for MEA-7 and MEA-8, is the internal pressure; * σdis the design stress of SS304 (145 N/mm2 at design temperature 100°C); * Schedule no. = 40 for common pipes.

Table 4.1.Pipe dimensions Streams | d, optimum (mm) | Nominal diameter (mm) | Pipe wall thickness (mm) | Safe working pressure (bar) | MEA-2 | 506 | 500 | 0.35 | 58 | MEA-3 | 509 | 500 | 0.33 | 58 | MEA-7 | 480 | 500 | 4.97 | 58 | MEA-8 | 479 | 500 | 4.93 | 58 |
Nominal diameter is cited from Selmon Company (n.d.)

5.0 Specification Sheet

Figure 1. Specification sheet for lean/rich MEA heat exchanger
6.0 Mechanical Drawing

Figure 6.0 Mechanical drawing of lean/rich MEA heat exchanger (E-114)

7.0 Operational Design
7.1 Commissioning, Start-up, Shutdown and Maintenance * Commissioning
Lean/rich MEA heat exchanger must be operated under conditions that do not exceed the limits specified in specification sheet. Since the inlet and outlet nozzles of the heat exchanger are interchangeable, proper design and care must be taken. The heat exchanger also needs to be checked for all of its connections before start up.

* Start-Up and Shut-Down Procedures
In starting up lean/rich MEA heat exchanger, first vent all connections and cold stream is flowing in gradually until the cold side of the heat exchanger is fully flooded before the vents are closed. Later, flowing in the hot stream steadily until the heat exchanger is flooded. Finally, all vents are closed and the heat exchanger is ready for its operating conditions. All start-up procedures must be carried out in a gradual method in order to prevent temperature shock from occurring as well as to avoid pulsations which could lead to leaking (Yula Corporation n.d.).

In shutting down lean/rich MEA heat exchanger, hot stream should be shut off first before the cold stream. However, there might have certain circumstances that the cold stream should be shut off first. In that case, the hot stream should also be shut off first by by-passing the heat exchanger. All fluids are required to be drained after shut-down to prevent freezing and corrosion (Yula Corporation n.d.).

* Maintenance
A permit-to-work is a mandatory requirement in the maintenance of HE-101. This is to ensure that responsible individuals are clear and aware of the potential hazards before they proceed with the maintenance works. Prior to a issuing a permit-to-work, certain measures must be taken (Skelton 1997 33-36), I. Isolate electrical power II. Isolate mechanical drives III. Isolate heating and cooling sources IV. Isolate sources of flammable and toxic gases V. Working area tested for breathable atmosphere

7.2 Process Control Design

Figure 7.1. Schematic of a control system for lean/rich MEA heat exchanger
One of the major disturbances in this process control system is changes in pressure. A cascade control is designed to reduce the effect of the disturbance. Liquid pressure is a manipulated variable in this control system and it responds directly to changes in the pressure of the MEA-7 and to the changes in the heat duty required to heat the rich lean (MEA-2). A change in MEA-7 pressure can cause a change in the pressure inside the exchanger. The pressure controller responds quickly and absorbs this upset efficiently.
Another disturbance in this control system is the changes in temperature of the inlet stream (MEA-). A combined feedforward and feedback controller is used to handle the disturbance, The inlet temperature of the MEA-2 is the input to the feedforward controller while the outlet temperature of the stream (MEA-3) is the controlled variable for the feedback controller. The outputs from the feedback and feedforward controllers are added together and give the setpoint to the pressure controller on MEA-7 (Riggs and Karim 2006). A pressure relief valve is also installed in case the pressure in the heat exchanger exceeds the safe operation limit.

7.3 Hazard and Operability (HAZOP) Study
Table 7.1 HAZOP study of lean/rich MEA heat exchanger. Guide word | Deviation | Possible causes | Consequences | Action required | NONE | NO FLOW | (1) No inlet gas or solvent flow | Production loss | (a) Install low flow alarm | | | (2) Line blockage or the isolation valve shut in error | Pump overheats | Covered by (a) | | | (3) Line fracture or rupture | Gas or solvent discharged to environment | (b) Regular inspection and maintenance of transfer lines(c) Design Plant emergency shutdown procedures | | | (4) Control valve failure | As for (1) | (d) Review operator reliability | MORE OF | MORE FLOW | (5) Control valve failure (6) Flow transmitter failure | Reduction in absorption efficiencyFlooding occur | (e) Install high level alarm | LESS OF | LESS FLOW | (7) Pipe leakage | As for (1) | Covered by (a) | | | (8) Flange leakage | As for (1) | Covered by (b) | | | (9) Line blockage | As for (2) | Covered by (a) | | | (10) Line fracture or rupture | As for (3) | Covered by (a) and (b) | MORE OF | MORE TEMPERATURE | (11) High feed temperature (12) Cooler failure | As for (1)Pressure increases in transfer line | (f) Regular inspection and maintenance on temperature indicator and cooler | LESS OF | LESS TEMPERATURE | (13) Weather conditions (14) Cooler failure | As for (1) | Covered by (f) | MORE OF | MORE PRESSURE | (15) Line blocked (16) Flooding | Line fracture/flange leak, Overpressure, explosion | Covered by (e) | LESS OF | LESS PRESSURE | (17) Pipe leakage | As for (1) | Covered by (a) and (b) | MAINTENANCE | NONE | (18) Equipment failure | Process stops | (g) Use right materials for pipes and fittings constructions |

8.0 Critical Reviews
The lean/rich MEA heat exchanger is a shell and tube heat exchanger. The advantages of using this type of heat exchanger is because of its configuration that gives larger heat exchange area in a small volume. Furthermore, it is also easily fabricated and cleaned, and can be constructed from a wide range of materials. Besides, the lean/rich MEA heat exchanger is designed as floating head type because this type of heat exchanger is more versatile and suitable for high-temperature and fouling liquids, and also easy to clean (Sinnott 2005).

References
Riggs, J. B. and M. N. Karim. Chemical and Bio-Process Control. 3rd ed. Pearson.
Selmon Company. n.d. Pipe Sizes and Dimensions. http://www.maselmon.com/ (accessed 28 October, 2010).
Sinnott,R. K., 2005. Coulson & Richardson’s Chemical Engineering. Vol. 6. Elsevier Butterworth-Heinemann.
Society for Amateur Scientists. 2004. Engineering Material Properties Arranged by Material. http://www.sas.org/engineerByMaterial.html (accessed 26 October, 2010).
Yula Corporation n.d. Operation and maintenance of industrial heat exchangers. http://www.yulacorp.com/downloads/INDUS_O_M.pdf (assessed November 1, 2010).