Contents lists available at ScienceDirect

Chemical Engineering & Processing: Process Intensification

journal homepage: www.elsevier.com/locate/cep

HETP measurement using industrial-scale batch distillation

Nguyen Van Duc Longa,b,c,1, Dong Young Leec,d,1, Sun Yong Parke, Byeng Bong Hwange,Moonyong Leec,*a Department for Management of Science and Technology Development, Ton Duc Thang University, Ho Chi Minh City, Vietnamb Faculty of Applied Sciences, Ton Duc Thang University, Ho Chi Minh City, Vietnamc School of Chemical Engineering, Yeungnam University, Gyeongsan 712-749, Republic of KoreadHanbal Masstech Ltd, Golden Root Complex, Gimhae, Republic of KoreaeOunR2tech Co. Ltd, Pohang 37553, Republic of Korea

A R T I C L E I N F O

Keywords:Batch distillationContinuous distillationHETPStructured packingRandom packingRefrigerant reclamation

A B S T R A C T

The height equivalent to a theoretical plate (HETP), representing the mass transfer efficiency, allows designers todetermine the packed bed height. In this study, a novel methodology was proposed to evaluate the HETP usingan existing industrial-scale batch distillation when designing the height of an industrial-scale continuous dis-tillation column. Specifically, an existing batch distillation used for the reclamation of dichlorodifluoromethaneCCl2F2 (R-12 or Freon-12) from a waste refrigerant mixture was employed to determine the HETP value. Prior tothe simulation, a literature survey was investigated to determine the suitable property package for simulation.The simulation results were then compared with the experimental data to determine whether REFerence fluidPROPerties can be used as a suitable package for this study. Several batch distillation runs were carried out usingindustrial structured packing and random packing, which are candidates for the design and optimization of acontinuous distillation column in this specific separation. The batch distillation was then rigorously simulatedfor a comparison with the experimental data, enabling us to calculate the HETP value. The simulated resultsagreed well with experimental data, demonstrating that the proposed methodology is effective and easy to apply.

• A simple and effective methodology was proposed for estimating theHETP.

• An existing industrial-scale batch distillation was used.

• REFPROP method demonstrated good capability for predicting theVLE.

• Experimental results were in good agreement with the simulationresults.

• The effect of pressure can be counted.

1. Introduction

The distillation process, which is the most commonly employedprocess in the chemical and petrochemical industries, is normallyconducted in either continuous or batch mode [1]. Installation of theequipment can be achieved using a tray or packing, which is the mainaspect contributing to the performance of such equipment [2,3]. Lowerpressure drops can be achieved when using columns packed with

random and structured packing as compared to trayed distillationtowers. Their generally shorter height, mechanical simplicity, ease ofinstallation, and ability to be fabricated in a cost-effective manner fromcorrosion-resistant materials are other advantages of packed distillationcolumns [4].

Random packing is known for its easily installation, blocking re-sistance, and acceptable-efficiency devices for distillation columns [5].Currently, approximately 200 different types and sizes of randompackings are available from different manufacturers worldwide [6]. Thelarger the random packing, the higher the capacity, but at a cost oflower efficiency [7]. A smaller packing possesses a higher efficiency buta lower capacity and higher cost. The appropriate packing should beselected with the most economical balance between capacity and effi-ciency. Structured packing can be considered for a lower pressure drop.In addition, other advantages of a structured packing are usually rea-lized at a high surface area and high separation efficiency [8–11].However, they are considerably more expensive per unit volume thanrandom packing and are susceptible to corrosion [10].

The performance of packed distillation columns is frequently

https://doi.org/10.1016/j.cep.2020.107800Received 10 October 2019; Received in revised form 23 December 2019; Accepted 2 January 2020

⁎ Corresponding author. Tel.: +82 53 810 2512.E-mail addresses: nguyenvanduclong@tdtu.edu.vn (N. Van Duc Long), mynlee@yu.ac.kr (M. Lee).

1 These two authors contributed equally to this work.

Chemical Engineering & Processing: Process Intensification 148 (2020) 107800

Available online 07 January 20200255-2701/ © 2020 Elsevier B.V. All rights reserved.

T

expressed in terms of the height equivalent to a theoretical plate(HETP) [12]. HETP representing the mass transfer efficiency is anempirical but extremely practical parameter [13]. The number of the-oretical stages required for a specific separation and the HETP for aparticular type of packing are both used to determine the actual heightof the packing required to achieve the desired separation. Note thatheight of transfer unit (HTU) can also be considered to estimate thepacked-height, although this HETP approach is usually preferred. Thisis because the HETP approach is suitable for multicomponent systems,can use the stage-by-stage computer programs, and enables easiercomparison with plate columns, while the HTU approach is morecomplex and more difficult to use [8].

Three approaches are commonly used for HETP prediction, namely,mass transfer models, rule of thumb, and data interpolation [8]. Themass transfer models first determine the height of a transfer unit (HTU),which is then used to calculate the value of the HETP [14,15]. Theaccuracy of the model clearly depends on the accuracy of the correla-tions used for predicting the basic parameters, i.e., the mass-transfercoefficients for the gas and liquid phases, respectively, and the effectiveinterfacial area [16]. In addition, because there are only a few variablesthat significantly affect a random packing HETP, and owing to theunreliability of even the best mass transfer model, a rule of thumb forthe HETP competes successfully with mass transfer models [2,8,17].Furthermore, an interpolation of experimental HETP data is anothermeans of obtaining the designed HETP values [18].

2. Problem statement

The HETP is essential for designers to estimate the height of apacking column. HETP is determined as

HETP = H/n (1)

where H is packed bed height and n is the number of actual stages.Despite the significant amount of data available, no method yet

exists for HETP prediction with a high degree of confidence [17]. TheHETPs are different for different packing sizes and types and for dif-ferent chemical systems [19]. The HETP of structured packing is alsoaffected by the pressure (high pressure), vapor, and liquid load [8].Thus, it would be better if the HETP can be estimated experimentallyfor the same mixture and similar operating conditions.

In this study, a novel methodology is proposed to estimate the HETPused to design the packing height of a continuous distillation unit usingthe existing industrial-scale batch distillation unit. An existing in-dustrial-scale batch distillation used for reclaiming dichlorodi-fluoromethane CCl2F2 (R-12 or Freon-12), which is the most commonrepresentative chlorofluorocarbon (CFCs), from a waste refrigerantmixture was used to estimate the HETP value of the desired packing.Prior to the simulation, a literature survey was carried out to determinethe suitable property package required for simulating the refrigerantmixtures. Based on a comparison between data from simulated Pxydiagram and the experimental data, the thermodynamic model forevaluating the vapor-liquid equilibrium (VLE) behaviors can be se-lected. An experimental operation using the existing industrial-scalebatch distillation unit was carried out, and the obtained results werecompared with the simulation results to determine the HETP value oftwo industrial structured packings and random packing NMTP 15.

3. Methodology

3.1. Materials

An illustrative example is the purification of R-12 from a wasterefrigerant mixture. A mixture including R-12 and 1,1,1,2-Tetrafluoroethane CH2FCF3 (R-134a) was obtained from OunR2tech(Korea). The waste refrigerant was recovered in a liquid state and

injected into the evaporator; the liquid refrigerant was then evaporatedin the evaporator to treat the impurities such as oil and water in themixture.

3.2. Apparatus

3.2.1. GC apparatusThe samples collected for the batch distillation run were analyzed

through a GC analysis, which was carried out using an Agilent 490equipped with a flame ionization detector (FID) and fitted with a 20m×0.25mm column programmed at 120 °C. The sample injection vo-lume is 10 μL. Helium was used as the carrier gas under a pressure of0.6 MPa and a flow rate of 0.5ml/min.

3.2.2. Batch distillation systemBatch processing is commonly used in the pharmaceutical, bio-

chemical, and specialty chemical industries. This process is preferableto continuous distillation when small quantities of chemicals and bio-chemicals need to be separated [20,21]. The most prominent char-acteristic of batch distillation is its flexibility, allowing one to deal withuncertainties in the feedstock or product specifications. Another ad-vantage of batch distillation is that it is able to separate a multi-component mixture with only a single column.

In this study, an industrial-scale batch distillation used for R-12reclamation was employed for determining the HETP. After the re-frigerant was separated and purified from other heavy componentsusing the evaporator, the components were fed into the batch distilla-tion (shown in Fig. 1). An industrial-scale batch distillation unit con-structed at OunR2tech was tested using wire gauze structure packingHMP 500WG or wire gauze structure packing HMP 700WG, orrandom packing NMTP 15 as suggested by the vendor. Table 1 showsthe existing hydraulics and product specifications of the industrial-scalebatch distillation column. In this system, the control panel allows es-tablishing and controlling the operating parameters such as the hotwater temperature, cold water temperature, and collector level duringthe entire experiment.

3.3. Simulation

Aspen Hysys was used to select suitable property package through acomparison with the experimental data in the Pxy equilibrium phasediagram. Furthermore, to estimate the HETP value of the specifiedpacking, an Aspen Batch Modeler V10 was employed to simulate thebatch distillation, the results of which were compared with those of anexperiment on a real industrial-scale batch distillation.

4. Results and discussion

4.1. Phase equilibrium

Because a property package was required for the simulation of R-12and R-134a, a literature survey was carried out. As a result, theREFerence fluid PROPerties (REFPROP) method was found, whichprovides the thermodynamic and transport properties of industriallyimportant fluids and their mixtures with an emphasis on refrigerantsand hydrocarbons [22,23]. The literature data [24] on a Pxy equili-brium phase diagram were used to compare with the simulation resultsusing REFPROP in Aspen Hysys DB. As shown in Fig. 2, the Pxy curvesmatched the experimental data closely. Thus, REFPROP was selected asthe property package for the prediction of the vapor–liquid equilibriumand all simulations conducted in this study. All simulations were ap-plied using Aspen Hysys and Aspen Batch Modeler V10 simulators.

A mixture of R-12 (boiling point of -29.8 °C) and R-134a (boilingpoint of -26.3 °C) has an azeotrope with a composition of approximately60 mass% R-12, and an azeotropic temperature of approximately-34.5 °C. Although it is easier to separate at 1 atm (1.013 bar), the

N. Van Duc Long, et al. Chemical Engineering & Processing: Process Intensification 148 (2020) 107800

2

temperature is then too low, as shown in Fig. 3. Thus, the columnshould be operated at a higher operating pressure. A column pressure of8 bar was selected to allow the available heating and cooling water tobe applied together. Please note that the cooling water and hot wateravailable together are at 13 °C and 48 °C, respectively.

4.2. Batch distillation operation

The feed contains 1280 kg of 91.31 wt% R-12 and 8.69 wt% R-134a

(shown in Table 2). The required product purity is 99.5 wt% R-12 in thegenerator or reboiler. In addition, 320 kg for the collector, which canbalance the reboiler duty and recovery of R-12, was selected. Since thedensity of this mixture is 1242 kg/m3, the volume of liquid in the col-lector was 0.2576m3, which is equivalent to the 0.407m height of theliquid in the collector. This value was used to set the collector levelcontroller. The existing batch distillation unit has the following reflux

Fig. 1. (a) Existing industrial-scale batch column and (b) simplified flowchartillustrating the existing batch column for R-12 reclamation.

Table 1Hydraulics, energy performance, and product specifications of existingindustrial-scale batch distillation columns.

Tray type Packing

Bed height (mm) 2550 / 2668Column diameter (mm) 102Max flooding (%) 84Energy requirement of reboiler (W) 5500R12 purity (mass%) 99.5

Fig. 2. R-134a+R12 VLE phase diagram compared with values from the lit-erature at (a) -15 °C, (b) 5 °C, and (c) 25 °C.

N. Van Duc Long, et al. Chemical Engineering & Processing: Process Intensification 148 (2020) 107800

3

operating and vapor loading policies:Fig. 4 shows the operation procedure. Before starting the batch

distillation, the column and collector were prepared under vacuumconditions, whereas at the start, the generator was filled with a charge(1280 kg of liquid at ambient temperature (18 °C), 4 bar). The operationwas then started by opening the valve in the vapor line of the re-generator, followed by turning on the pumps for cooling and hot water.Initially, the heat transfer rate is high because the temperature differ-ence between the generator liquid and hot water is high. The liquidthen starts to boil. The concentration of the more volatile componentand the volume of the liquid decrease, and the column fills up with theliquid hold-up.

There was no reflux during the first period (approximately 4–5 h).

During this period, all vapor was contained in the collector after beingtotally condensed. When the height of the collector was 0.407m, whichoccurs on a case-by-case manner, as estimated above, the reflux pumpstarted to operate to generate a reflux flow returning to the column. Thecollector level was then controlled and maintained by the reflux pump(shown in Fig. 1). Thus, the top vapor flow rate is equal to the refluxflow rate. Another important variable during batch distillation is thevapor loading of the reboiler. The reboiler was operated under a con-stant duty cycle and was set at the highest capacity without flooding inthe column.

Because there is no liquid withdrawn from the column, the lightcomponent is gradually accumulated in the collector, while the con-centration of less volatile component increases in the generator. Thus,during the operation, the generator temperature increases. Samples ofthe generator were taken for further mass quantification of R-12 usingthe previously described analytical methods. As time proceeds, thematerial in the generator becomes less rich in the more volatile com-ponents, and its vapor flowrate reduces because the reboiler duty re-mains constant. This process is continued until the specifications for theresidue in the generator are satisfied. After achieving the target purityof R-12 in the generator, the batch distillation is stopped and all re-maining components inside the column and collector are removed.Vacuum conditions are then generated in the column and collectorbefore another batch can be started.

4.3. Batch distillation simulation

The batch distillation was then simulated using Aspen BatchModeler. Table 3 shows all parameters needed for simulation of theexisting batch distillation. All operating policies used in experimentscan be implemented in Aspen formats. REFPROP was used as theproperty package. The zero distillate rate can be set in Aspen BatchModeler. The operation includes two steps. The first step is a feedcharging step, while the heat is supplied in the second step. During thesecond step, the collector level of 0.407m was kept constant by settingin Aspen Batch Modeler. Several simulation runs were applied with

Fig. 3. Temperature vs. composition plot at (a) 1 and (b) 8 bar.

Table 2Feed mixture conditions.

Feed Conditions

Component Mass Fractions (%)R-12 91.31R-134a 8.69Feed weight (kg) 1280Temperature (°C) 18Pressure (bar) 4

Fig. 4. Operation procedure.

N. Van Duc Long, et al. Chemical Engineering & Processing: Process Intensification 148 (2020) 107800

4

different numbers of stages using the Aspen Batch Modeler.Notably, in the Aspen Batch Modeler, if a start-up of the batch

column or the use of an initial charge is required to simulate, a partialcondenser is needed because the column is initially filled with nitrogenor air, and it is unrealistic to completely condense air or nitrogen whichenters the condenser during the start-up.

4.4. Comparison between simulation and experiment

4.4.1. HMP 500WG structured packingIn the first case study, because wire gauze structure packing HMP

500WG shown in Fig. 5a is a candidate for continuous distillation inthis project, it was tested to evaluate the HETP used to estimate thedesigned column height. Structured packing has a well-defined geo-metrical structure, which consists of an alternating arrangement ofcorrugated sheets forming intersecting open channels for a vapor flow[13]. Wire gauze structured packing is generally used in medium- tosmall-diameter columns for separations requiring a maximum numberof theoretical stages at the minimum column height [25].

The samples collected for batch distillation were analyzed using GC.Fig. 6 shows the history of the composition purity. For a given samepacked bed height, each kind of packing has its own HETP value ormass transfer efficiency, which leads to different number of stages anddifferent operating time to achieve target purity. Several simulationruns were then carried out with different numbers of stages using theAspen Batch Modeler. The experimental variations of the generatorcomposition and temperatures based on the operating time were com-pared with the simulation results. The number of stages having thepurity history and operation time matched well with that of real op-eration was finally selected. As the results indicate, the purity historycurve under eight stages matched with that of the real operation mostclosely. Thus, the HETP, which was calculated by Eq. (1), was ap-proximately 319mm under these operation conditions. Furthermore,the minimum number of stages can be found to be seven under theseconditions.

Fig. 7 shows the temperature histories at the generator of thecolumn for both the simulation and experimental results, which de-monstrate extremely good agreement. Furthermore, it took 28.7 and28.3 h during the simulation and real operation, respectively, toachieve 99.5 % R-12, which indicates that simulated results are inagreement with those of the experimental results.

4.4.2. HMP 700WG structured packingWire gauze structure packing HMP 700WG, shown in Fig. 5b, was

developed by Hanbal Masstech Co. with the aim to increase the effi-ciency, which can lead to a reduction of the HETP. Fig. 8 shows thevariation of the composition of R-12 during both the experiment andsimulation. The experimental curve is located near the curves when thenumber of stages is 9 and 10. Thus, the HETP is approximately267–296mm for this specific separation. It took 24.2 and 23.6 h during

the simulation and real operation, respectively, to achieve 99.5 % R-12.The real batch distillation column allowed an R-12 product to beachieved at the generator of the column.

Fig. 9 shows the temperature histories of the generator of thecolumn for both the simulation and experimental results, which haveextremely good agreement. It took 28.7 and 28.3 h for the simulationand real operation, respectively, to achieve 99.5 % R-12. The real batchdistillation column allowed R-12 to be obtained at the generator of thecolumn. In general, a good agreement was obtained between the

Table 3Parameters for the batch distillation model.

Species R-12 and R-134a

Number of stages VariedDistillate flow rate 0 kg/hrGenerator geometry diameter 1 m

length 3 mCollector geometry diameter 1 m

height 1 mPressure 8 barInitial conditions EmptyOperating stepsStep 1 Charging (1280 kg)Step 2 Duty 5 kW

Collector level controller 0.407m

Fig. 5. (a) Wire gauze structure packing HMP 500WG, (b) HMP 700WG and(c) random packing NMTP 15.

N. Van Duc Long, et al. Chemical Engineering & Processing: Process Intensification 148 (2020) 107800

5

simulation and experimental results, demonstrating the capability ofREFPROP for modeling the thermodynamic behavior of the VLE be-tween R-12 and R-134a. Note that the flooding is 87.1 % under theseoperating conditions (from the Aspen Batch Modeler).

4.4.3. NMTP 15 - random packingNMTP 15 packing (Fig. 5c), which is applicable in a broad range of

services, was also investigated to measure the HETP. NMTP towerpacking has a high void fraction providing a superior high liquid rate

Fig. 6. R-12 composition purity history (wire gauze structure packing HMP500WG).

Fig. 7. Bottom temperature history (wire gauze structure packing HMP500WG).

Fig. 8. Purity history of R-12 composition (wire gauze structure packing HMP700WG).

Fig. 9. Bottom temperature history (wire gauze structure packing HMP700WG).

Fig. 10. (a) Composition profile and (b) flooding profile at the end of simula-tion.

N. Van Duc Long, et al. Chemical Engineering & Processing: Process Intensification 148 (2020) 107800

6

capacity as well as a lower pressure drop compared to other randompackings of similar size [25,26]. This packing combines the advantagesof a saddle-shape packing with those of modern high-performance ring-type packings [27]. The vendor has suggested using this packing for theseparation of R-12 and R-134a. Thus, it was tested to measure the HETPfor this specific separation.

At the beginning of the operation, the generator is filled with R-12and R-134a. As soon as the operation starts, R-134a tends to move to-ward the collector, leading to a low composition in the generator,where R-12 is dominant during the final period. The pressure drop was0.1-0.2 bar. The heat loss was ignored because the batch column systemis well insulated. Very good agreement exists between the experimentaland simulation results when nine stages are applied, leading to an HETPof 296mm. Fig. 10 shows the composition and flooding profiles at theend of simulation. Note that, under atmospheric conditions, the HETPvalue from the vendor is 250mm (shown in Table 4) because, in ad-dition to the packing type and size, the HETP values are complexfunctions of the pressure, vapor and/or liquid flowrates, distribution,and physical properties, among other factors. For greater reliability, it istherefore recommended to estimate the HETP for a specific separationusing a rigorous methodology.

Because the batch distillation and designed continuous distillationhave almost same pressure to separate the same system, HETP esti-mated using batch distillation has counted the effects of pressure. Inother words, with this novel methodology, the effects of pressure can beeliminated when estimating HETP and the height of the continuouscolumn.

5. Conclusions

This study proposed a simple and effective methodology for esti-mating the HETP using an existing industrial-scale batch column. TheREFPROP method has demonstrated a good capability to predict theVLE of R-12 and R-134a. The proposed methodology was tested usingHMP 500WG and HMP 700WG structured packings and randompacking NMTP 15 through a study on the current industrial separationof R-12 and R-134a. Experimental results of a commercial batch dis-tillation column showed good agreement with the simulation results.This novel methodology can be accomplished easily and efficientlyusing Aspen Hysys, Aspen Batch Modeler, and an existing industrial-scale batch distillation unit. For reliability, it has been recommended toapply a rigorous methodology when estimating the HETP for a specificseparation. The effects of pressure, distribution and physical propertiescan be eliminated when estimating HETP and the height of the con-tinuous column.

Declaration of Competing Interest

The authors declare that they have no known competing financialinterests or personal relationships that could have appeared to influ-ence the work reported in this paper.

Acknowledgements

This work was supported by the Basic Science Research Programthrough the National Research Foundation of Korea (NRF) funded bythe Ministry of Education (2018R1A2B6001566), and by PriorityResearch Center Program through the National Research Foundation ofKorea (NRF) funded by the Ministry of Education(2014R1A6A1031189), and the R&D Center for Reduction of Non-CO2

Greenhouse Gases (201700240008) funded by the Ministry ofEnvironment as a “Global Top Environment R&D Program.”

References

[1] N.V.D. Long, T.N. Pham, M. Lee, Optimal retrofit and debottlenecking using novelmulti-effect dividing wall column, Chem. Eng. Process. 127 (2018) 65–71.

[2] E.E. Ludwig, third ed, Applied Process Design for Chemical and PetrochemicalPlants Volume 2 Gulf Professional Publishing, 1979.

[3] S. Pelkonen, A. Górak, A. Ohligschläger, R. Kaesemann, Experimental study onmulticomponent distillation in packed columns, Chem. Eng. Process. 40 (2001)235–243.

[4] M. Pilling, B.S. Holden, Choosing trays or packings for distillation, CEP (2009)44–50.

[5] X. Li, X. Yang, H. Li, Q. Shi, X. Gao, Significantly enhanced vapor-liquid masstransfer in distillation process based on carbon foam ring random packing, Chem.Eng. Process. 124 (2018) 245–254.

[6] J. Maćkowiak, J.F. Maćkowiak, Random packings, in: A. Gorak, Z. Olujic (Eds.),Distillation: Equipment and Processes, Elsevier, London, 2014.

[7] https://www.amacs.com/what-is-random-packing-and-how-is-it-used-in-process-plants/ (accessed 11 November 2018).

[8] H.Z. Kister, Distillation Design, McGraw-Hill, 1992, pp. 523–584.[9] Z. Olujic, A.F. Seibert, B. Kaibel, H. Jansen, T. Rietfort, E. Zich, Performance

characteristics of a new high capacity structured packing, Chem. Eng. Process. 42(2003) 55–60.

[10] Z. Olujic, Types of distillation column internals, in: A. Gorak, Z. Olujic (Eds.),Distillation: Equipment and Processes, Elsevier, London, 2014.

[11] N.V.D. Long, M. Lee, Advances in Distillation Retrofit, 1st ed, Springer, NewYork,2017.

[12] A.E. Orlando Jr., L.C. Medina, M.F. Mendes, E.M.A. Nicolaiewsky, HETP Evaluationof Structured Packing Distillation Column, (2009) 26 619 - 633.

[13] L. Spiegel, M. Duss, Structured packings, in: A. Gorak, Z. Olujic (Eds.), Distillation:Equipment and Processes, Elsevier, London, 2014.

[14] J.L. Bravo, J.R. Fair, Generalized correlation for mass transfer in packed distillationcolumns, Ind. Eng. Chem. Process Des. Dev. 21 (1982) 162–170.

[15] S. Piché, S. Lévesque, B.P.A. Grandjean, F. Larachi, Prediction of HETP for ran-domly packed towers operation: integration of aqueous and non-aqueous masstransfer characteristics into one consistent correlation, Sep. Purif. Technol. 33(2003) 145–162.

[16] G.Q. Wang, X.G. Yuan, K. Yu, Review of mass-transfer correlations for packedcolumns, Ind. Eng. Chem. Res. 44 (2005) 8715–8729.

[17] R.E. Strigle, Packed Tower Design and Applications - Random and StructuredPackings, 2nd edition, Gulf Publishing Company, Houston, 1987.

[18] J.S. Eckert, Tower packings - comparative performance, Chem. Eng. Progress 59(1963) 76–82.

[19] P.C. Wankat, Separation Process Engineering, Prentice Hall, 2007.[20] U. Diwekar, Batch Distillation - Simulation, Optimal Design, and Control, CRC

press, 2012.[21] A. Garcia, J. del C. Zavala-Loria, L.E. Vilchis-Bravo, J.A. Rocha-Uribe, Design of

batch distillation columns using short-cut method at constant reflux, J. Eng. (2013),https://doi.org/10.1155/2013/685969 Article ID 685969, 14 pages.

[22] Aspen Physical Property System, Aspen Technology, 2013.[23] https://wtt-lite.nist.gov/wtt-lite/help/properties/viscosity_gas/REFPROP.html (ac-

cessed 11 November 2018).[24] M. Kleiber, Vapor-liquid equilibria of binary refrigerant mixtures containing pro-

pylene or R134a, Fluid Phase Equilib. 92 (1994) 149–194.[25] http://www.koch-glitsch.com/Document%20Library/KGIMTP.pdf (accessed 11

November 2018).[26] Sv. Nakov, N. Kolev, L. Ljutzkanov, D. Kolev, Comparison of the effective area of

some highly effective packings, Chem. Eng. Process. 46 (2007) 1385–1390.[27] http://www.koch-glitsch.com/Document%20Library/KGMRP-02.pdf (accessed 11

November 2018).

Table 4HETP values.

Packing type New methodology Vendor

HMP 500WG structured packing 319mm 250-350 mmHMP 700WG structured packing 267−296mm 200-300 mmNMTP 15 random packing 296mm 250-350 mm

N. Van Duc Long, et al. Chemical Engineering & Processing: Process Intensification 148 (2020) 107800

7