View metadata, citation and similar papers at core.ac.uk brought to you by CORE
provided by RMIT Research Repository
Recovery and Regeneration of Carboxylic Acids from Aqueous Solutions Using Process Intensification
A thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy
Sumalatha Eda
M.Tech (Plant Design in Chemical Engineering)
B. Tech (Chemical Engineering)
School of Engineering
College of Science, Engineering and Health
RMIT University
August 2017
1 | Page Declaration
I certify that except where due acknowledgement has been made, the work is that of the author alone; the work has not been submitted previously, in whole or in part, to qualify for any other academic award; the content of the thesis is the result of work which has been carried out since the official commencement date of the approved research program; any editorial work, paid or unpaid, carried out by a third party is acknowledged and ethics procedures and guidelines have been followed.
Sumalatha Eda 31/08/2017
i | Page Learning gives Creativity, Creativity leads to thinking, Thinking provides knowledge, Knowledge makes you great -A.P.J. Abdul Kalam
ii | Page Acknowledgement
First, I would like to express my sincere thanks and gratitude to my supervisors
A/Prof. Rajarathinam Parthasarathy, Dr. Prathap Kumar Thella, and Dr. Satyavathi
Bankupally for their continuous support, invaluable guidance, and suggestions during my research work. I am always grateful to them for being such a tremendous source of inspiration, and truly appreciate the freedom given to me to explore new ideas in research and their confidence in my research skills. Their valuable encouragement and affection gave me the confidence to do good work.
I wish to acknowledge and thank my supervisor Associate Professor Rajarathinam
Parthasarathy at RMIT University, for his detailed and constructive comments, and for his important support throughout this work. His valuable advices, friendly help and extensive discussion around my work have been very significant for this study.
His wide knowledge and his logical way of thinking have been of great value for me.
I especially want to thank my supervisor at CSIR-IICT, Dr. Prathap Kumar Thella, whose support and guidance made my thesis work possible. He has been actively interested in my work and has always been available to advise me. I am very grateful for his patience, motivation, enthusiasm, and immense knowledge in process intensification that, taken together, make him a great mentor. Special thanks to Dr.
Satyavathi Bankupally for providing all-round support and guidance to complete my research work at CSIR-IICT, Hyderabad, India. It has been an honor to work with Dr.
Satyavathi. The joy and enthusiasm she has for her research was contagious and motivational for me, even during tough times in the PhD pursuit.
iii | Page Besides my supervisors, I would like to thank Prof. Suresh Bhargava and Dr. M.
Lakshmi Kantam for the establishing the RMIT-IICT Joint Research Center and for encouraging the students to continue their higher studies. I extend profound regards to all staff members of Chemical Engineering Division, IICT-Hyderabad. My cordial appreciation goes to my team members in the lab, P. Sudhakar, Alka Kumari, B.
Anoosha, K. Baby Jyothsna and Ch. Srinivas for their valuable assistance during experiments, assorted help and encouragement. I am also thankful to all my colleagues and lab mates in IICT, Nilesh Rane, B. Dharanija, K. Monika, Muqeet,
Ramesh Tangirala, Pavan Kumar, Ramesh Ajmeera, Vineet Aniya, Balaji and
Mallikarjun for their support and motivation during my research work.
I would be like to thank my RMIT-IICT friends Sayana Sree, Shravanthi Joshi,
Naresh, Kotaiah Naik, Ram Kumar and my seniors Dr. Shanthi Priya, Dr. Radha
Kumari, Dr. Jampaiah, Dr. Srinivasa Reddy and Dr. Vijay Kumar for making my stay a pleasant experience and their support at RMIT University, Melbourne. I would also like to thank staff members of School of Applied Sciences, RMIT University, Dr.
Emma Goethals, Dr. Lisa Dias, Dr. Ana Martins, and Piyumi Wickramaarachchi, and
Rawnaq who have been there to give advice and help in their respective roles and my friends.
My special thanks to my RMIT friends Pooja Takkalkar, Shwathy Ramesh,
Mohammad Farook, Veena, Nizamuddin, Ahmed, Sakin and Rajeev for your help at the end of the program.
I would also like to thank my mother, father and in-laws.Their love and everlasting support gave me the strength and encouragement to complete my PhD work
iv | Page successfully. Thanks to my sisters Lalitha and Vijayalatha for their utmost care and encouragement during my studies.
Last but not least many thanks to my husband Mr. Radha Krishna, who has always encouraged and being my side as a best friend throughout my life and career. Your love, patience, understanding and personal sacrifices made an everlasting impression on my life. This thesis dedicated to you.
Sumalatha Eda
v | Page Publications
Journal Papers
1 Sumalatha Eda, Alka Kumari, Prathap Kumar Thella, Satyavathi Bankupally,
Rajarathinam Parthasarathy, Recovery of Volatile fatty acids by reactive
extraction using tri-n-octylamine and tri-butyl phosphate in different solvents:
Equilibrium studies, pH and Temperature effect and Optimization using
multivariate Taguchi approach. The Canadian Journal of Chemical
Engineering 95(7), 2017, 1373-1387.
2 Sumalatha Eda, Prathap Kumar Thella, Satyavathi Bankupally, Sudhakar
Pabbu and Rajarathinam Parthasarathy, Recovery of Succinic acid by Reactive
Extraction using Tri-n-Octylamine in 1-Decanol: Equilibrium Optimization Using
Response Surface Method and Kinetic Studies, International Journal of
Chemical Separation Technology 1(2), 2016, 1-14.
3 Sumalatha Eda, Baby Jyothsna Kota, Prathap Kumar Thella, Satyavathi
Bankupally, Suresh K Bhargava, Rajarathinam Parthasarathy, Regeneration of
Levulinic Acid from Loaded-Organic Phase: Equilibrium, Kinetic Studies and
Process Economics, Chemical Papers 2017, 1-13, doi: 10.1007/s11696-017-
0188-6.
4 Sumalatha Eda, Anoosha Borra, Rajarathinam Parathasarathy, Satyavathi
Bankupally, Suresh Bhargava, Prathap Kumar Thella, Recovery of Levulinic acid
by Reactive Extraction using Tri-n-octylamine in Methyl isobutyl ketone:
Equilibrium and Thermodynamic Studies and Optimization using Taguchi
vi | Page Multivariate approach (Separation and Purification Technology, Under
revision: Manuscript No: SEPPUR_2017_2141)
Conference Proceedings
1 Sumalatha Eda, R. Parthasarathy, T. Prathap Kumar, Reactive Extraction of
Succinic acid from aqueous solutions using Tri-n-octylamine in 1-Decanol,
Equilibrium and effect of Temperature International Conference: APCChE
September 2015, RMIT University, Melbourne Australia.
Conferences
1. Sumalatha Eda, R. Parthasarathy, T. Prathap Kumar, B. Satyavathi, Recovery of
volatile fatty acids from fermentation waste by reactive extraction using tri-n-
octylamine and tri-butyl phosphate in different solvents, Effective Technologies
and Tools Research Centre (WETT), RMIT University, Melbourne, Australia,
2016.
2. Sumalatha Eda, Anoosha. B, Baby Jyothsna. K, Satyavathi. B, Prathap Kumar. T,
R. Parthasarathy, Reactive Extraction of Succinic Acid from Aqueous Solutions
using Tri-n-Octylamine in Kerosene+1-Octanol: Equilibrium, effect of temperature.
National symposium on Technosmania 2k16, University College of Technology
Osmania University, Hyderabad.
3. Sumalatha. Eda, P. Sudhakar, R. Parthasarathy, B. Satyavathi, T. Prathap
Kumar , Reactive Extraction of levulinic acid from aqueous solution using tri-n-
octylamine in methyl isobutyl ketone (MIBK) : Equilibrium and effect of
Temperature International conference on New Frontiers in Chemical, energy and
environmental Engineering (INCEEE-2015)
vii | Page Table of Contents
Declaration...………………………………………………………………………………….i
Acknowledgement…………………………………………………………………………..ii
List of Publications………………………………………………………………………….vi
Table of contents…………………………………………………………………………..viii
List of Figures……………………………………………………………………………….xv
List of Tables……………………………………………………………………………….xix
Nomenclature………………………………………………………………………………xxi
Abstract……………………………………………………………………………………….1
CHAPTER 1
General Introduction…………………………………………………………………………………..5
1.1 Introduction………………………………………………………………………………6
1.1.1 Background of the work……………………………………………………………6
1.1.2 Research gap………………………………………………………………...... 9
1.1.3 Research questions……………………………………………………………...... 9
1.1.4 Aims and objectives………………………………………………………………10
1.1.5 Thesis structure……………………………………………………………...... 11
CHAPTER 2
Literature Review……………………………………………………………………………………..13
2.1 Methods for carboxylic acids recovery………………………………………………14
2.1.1 Preparation….………………………………………………………………...... 14
2.1.2 Solvent extraction or liquid-liquid extraction……………………………………15
2.1.3 Membrane separation…………………………………………………………….15
2.1.4 Ionic liquid extraction………………………………………………………...... 17
viii | Page 2.1.5 Chromatographic methods……………………………………………………….18
2.1.6 Distillation…………………………………………………………………………..19
2.1.7 Reactive extraction………………………………………………………………..20
2.2 Kinetic studies literature………………………………………………………………38
2.3 Regeneration/Back extraction of carboxylic acids from loaded-organic phase………………………………………………………………………………………..40
2.4 Conclusions…………………………………………………………………………….42
CHAPTER 3
Recovery of Levulinic acid by Reactive Extraction using Tri-n-Octylamine in
Methyl-isobutyl Ketone: Equilibrium, Kinetics and Thermodynamic Studies and Optimization using Taguchi Multivariate Approach…………………………..43
3.1 Introduction………………………………………………………………………...... 44
3.2 Materials and methods………………………………………………………………..47 3.1.1 Chemicals…………………………………………………………………………..47
3.2.2 Methods……………………………………………………………………………48
3.2.2.1 Equilibrium studies……………………………………………………………..48
3.2.2.2 Taguchi design with mixed levels L18 for optimization……………………..49
3.2.2.2.1 Desirability approach……………………………………………………...51
3.2.2.3 Kinetic studies……………………………………………………………….....53
3.2.2.3.1 Specific rate extraction (RA)……………………………………………...53
3.3 Theory…………………………………………………………………………………..54
3.3.1 Mass action law model……………………………………………………………54
3.3.1.1 Physical extraction……………………………………………………………55
3.3.1.2 Chemical extraction…………………………………………………………..57
ix | Page 3.3.2 Specific interfacial area……………………………………………………………59
3.4 Results and discussion………………………………………………………………..61
3.4.1 Equilibrium studies………………………………………………………………..61
3.4.1.1 Physical equilibrium…………………………………………………………...61
3.4.1.2 Chemical equilibrium………………………………………………………….63
3.4.1.3 Loading ratio…………………………………………………………………...65
3.4.1.4 Thermodynamic studies……………………………………………………...67
3.4.1.5 Optimization of reactive extraction………………………………………….73
3.4.2 Kinetic Studies…………………………………………………………………….79
3.4.2.1 Determination of individual mass transfer coefficients…………………..79
3.4.2.2 Effect of agitation speed on specific rate of extraction…………………..82
3.4.2.3 Order of reaction with respect to initial levulinic acid and TOA
concentration……………………………………………………….…...….84
3.4.2.4 Determination of rate constant ( 2 )………………………………………...85
3.4.2.5 Criterion for reaction regime……………………………………………….86
3.5 Conclusions…………………………………………………………………………….87
CHAPTER 4
Recovery of Succinic acid by Reactive Extraction using Tri-n-Octylamine
(TOA) in 1-Decanol: Optimization of Equilibrium Studies using Response
Surface Method and Kinetic Studies………………………………………………….89
4.1 Introduction……………………………………………………………………………..90
4.2 Materials and methods………………………………………………………………..93
4.2.1 Material……………………………………………………………………………..93
4.2.2 Methods…………………………………………………………………………….94
x | Page 4.2.2.1 Equilibrium studies…………………………………………………………..94
4.2.2.2 Response surface methodology (RSM) and experimental design……..95
4.2.2.3 Kinetic studies………………………………………………………………..95
4.2.2.4 Specific rate of extraction ( ) ……………………………………………...96
4.3 Results and Discussion……………………………………………………………….97
4.3.1 Effect of process variables on extraction efficiency ( ) ……………………….97
4.3.2 Effect of succinic acid and TOA concentrations on % ………………………98
4.3.3 Effect of temperature (T) with succinic acid and TOA concentrations on %
…………………………………….…………………………………….……..99
4.3.4 Experimental design with response surface methodology……………….....101
4.3.4.1 CCD analysis………………………………………………………………..102
4.3.4.2 Process optimization……………………………………………………….108
4.3.5 Kinetic studies……………………………………………………………………108
4.3.5.1 Determination of individual mass transfer coefficients………...... 108
4.3.5.2 Effect of agitation speed on specific rate of extraction ( ) …………...111
4.3.5.3 Order of reaction with respect to initial succinic acid and TOA
concentration…………………………………………………..………...... 112
4.3.5.4 Determination of rate constant ( 2) ……..………………………………...113
4.3.5.5 Criteria for reaction regime…..…………………………………………….114
4.4 Conclusions…………………………………………………………………………..115
CHAPTER 5
Regeneration of Levulinic acid from Loaded-Organic phase: Equilibrium,
Kinetic Studies and Process Economics…………………………………………..116
5.1 Introduction……………………………………………………………………………117
xi | Page 5.2 Theory…………………………………………………………………………………121
5.2.1 Determination of liquid-liquid interfacial area…………………………………121
5.3 Experimental studies………………………………………………………………...123
5.3.1 Chemicals………………………………………………………………………...123
5.3.2 Equilibrium studies……………………………………………………………....124
5.3.2.1 Equipment…………………………………………………………………...124
5.3.2.2 Experimental methods………..…………………………………………….124
5.3.3 Kinetic studies…………………………………………………………………….126
5.3.3.1 Experimental…………………………………………………………………127
5.4 Results and discussion………………………………………………………………128
5.4.1 Equilibrium studies………………………………………………………………128
5.4.1.1 NaOH method…………………………………………………………………128
5.4.1.2 Temperature swing method………………………………………………….130
5.4.1.3 Diluent swing method………………………………………………………...131
5.4.1.4 TMA method………………………………………………………………….132
5.4.1.5 Mass action law model………………………………………………………135
5.4.2 Kinetic studies for TMA method………………………………………………..137
5.4.2.1 Effect of stirrer speed on specific rate of extraction ( )………………138
5.4.2.2 Effect of phase volume ratio () on ( )………………………………139
5.4.2.3 Order of reaction with respect to succinic acid and TMA
concentrations…………………………………………………………….140
5.4.3 Evaporation of TMA……………………………………………………………..142
5.4.4 Economical evaluation of scale up of reactive extraction process…………143
5.4.4.1 Capital investment…………………………………………………………..144
xii | Page 5.4.4.2 Operating and maintenance cost………………………………………….144
5.5 Conclusions…………………………………………………………………………..147
CHAPTER 6
Recovery of Multi-acids (Volatile Fatty Acids) by Reactive extraction using Tri- n-Octylamine and Tri-butyl Phosphate in Different Solvents: Equilibrium
Studies pH and Temperature Effects and Optimization using Multivariate
Taguchi Approach………………………………………………………………………149
6.1 Introduction……………………………………………………………………………150
6.2 Materials and methods………………………………………………………………153
6.2.1 Chemicals and reagents………………………………………………………..153
6.2.2 Preparation of VFA solution…………………………………………………….153
6.2.3 Instrumentation…………………………………………………………………..154
6.2.4 Reactive extraction experiments……………………………………………….154
6.2.4.1 Equilibrium studies…………………………………………………………..155
6.2.5 Optimization of reactive extraction process: Taguchi L36 mixed design…...156
6.2.5.1 Desirability approach……………………………………………………….160
6.2.6 Analytical method………………………………………………………………..161
6.3 Results and discussion………………………………………………………………162
6.3.1 Effects of diluents and extractants……………………………………………..162
6.3.2 Effect of extractant concentration……………………………………………...170
6.3.3 Effect of temperature……………………………………………………………173
6.3.4 Effect of pH……………………………………………………………………….177
6.3.5 Optimization of reactive extraction process: Taguchi L36 mixed design…...181
6.4 Conclusions…………………………………………………………………………..190
xiii | Page CHAPTER 7
Conclusions……………………….……………………………………………………………………191
7.1 Overall conclusions…………………………………………………………………..192
7.2 Recommendations for future work……………………………………………….195
References……………………………………………………………………………………..197
xiv | Page List of Figures
2.1 Total process of reactive extraction process, Blue region-the work carried out
in the report………………………………………………………………………..21
3.1 Physical extraction mechanism………………………………………………….62
3.2 Chemical extraction mechanism at equilibrium………………………………..63
3.3 Influence of temperature on distribution coefficient, degree of extraction of
levulinic acid with TOA (0.229 kmol m-3) in MIBK at 293 K…………………..65
3.4 ∗ 2 (2 ‒ ) versus , plots for the reactive extraction of levulinic acid using
0.229 kmol m-3 TOA in MIBK at different temperatures……………………....67
3.5 Determination of apparent enthalpy and entropy of reaction for the extraction
of levulinic acid with 0.458 kmol m-3 TOA in MIBK……………………………73
3.6 Main effect plots ( / ratios) for (a) distribution coefficient, (b) extraction
efficiency, % …………………………………………………………………….76
3.7 Percentage contribution of factors for (a) distribution coefficient ( ) (b)
extraction efficiency (% E)……………………………………………………….78
3.8 Variation of acid concentration in the aqueous phase with time……………..79
3.9 Determination of mass transfer coefficient for acetic acid in
kerosene/water……………………………………………………………………82
3.10 Effect of speed of stirring on specific rate of extraction...... 83
3.11 Effect of initial levulinic acid concentration on the specific rate of extraction at
0.678 kmol m-3 TOA, at 1.16 rev s-1...... 85
3.12 Effect of TOA concentration on the specific rate of extraction at 0.3 kmol m-3
initial levulinic acid, 1.16 revs s-1 speed of agitation…………………………..85
3.13 Determination of second order reaction rate constant, k2……………………86
xv | Page 4.1 Procedure chart for response surface methodology…………………………..92
4.2 Shaking incubator…………………………………………………………………93
4.3 Schematic diagram of experimental set up…………………………………….96
4.4 Effect of succinic acid and TOA concentrations on extraction efficiency
(% )………………………………………………………………………………...99
4.5 Effect of TOA concentration and temperature (K) on extraction efficiency
(% )………………………………………………………………………………100
4.6 Effect of succinic acid and temperature (K) on extraction efficiency (% ).101
4.7 RSM Model predicted v/s experimental degree of extraction………………106
4.8 Determination of mass transfer coefficient for acetic acid in
kerosene/water…………………………………………………………………..110
4.9 Effect of speed of stirring on specific rate of extraction……………………..112
4.10 (a) Effect of initial succinic acid concentration on the rate of extraction,
0.2291 kmol m-3 TOA, at 1 rev s-1……………………………………………...113
(b) Effect of TOA concentration on the rate of extraction at 0.5 kmol m-3 initial
succinic acid, 1.16 revs s-1 speed of agitation………………………………..113
4.11 Determination of second order reaction rate constant, 2 …………………..114
5.1 Process flow sheet for the recovery of carboxylic acids…………………….119
5.2 Schematic diagram of experimental setup: R-Jacketed reactor, F-Feed inlet,
S-Stirrer, M-Motor, WS-Water source, WI-Water inlet, WO-Water outlet, M-
Motor, SP-Sampling point, TI-Temperature indicator………………………..127
5.3 Recovery of levulinic acid from loaded-organic phase using NaOH method, 1
% = 0.0578, 2 % = 0.115, 3 % = 0.235, 4 % = 0.251 and 5 % = 0.313 kmol
m-3 of levulinic acid in organic phase………………………………………….129
5.4 Recovery of levulinic acid from loaded-organic phase using temperature
xvi | Page swing, 1 % = 0.0578, 2 % = 0.115, 3 % = 0.235, 4 % = 0.251 and 5 % =
0.313 kmol m-3 of levulinic acid in organic phase…………………………....131
5.5 Recovery of levulinic acid from loaded-organic phase using diluent swing.132
5.6 Transport mechanism of levulinic acid in reactive extraction and regeneration
for TMA method. Red-levulinic acid, Blue-TOA, Green-TMA………………133
5.7 Recovery of levulinic acid from loaded-organic phase using TMA method, 1
% = 0.0578, 2 % = 0.115, 3 % = 0.235, 4 % = 0.251 and 5 % = 0.313 kmol
m-3 of levulinic acid in organic phase………………………………………….134
5.8 Variation of acid concentration in the aqueous phase with time……………138
5.9 Effect of stirrer speed rate on specific rate of extraction…………………….139
5.10 Effect of volume ratio on specific rate of extraction………………….140 ( )
5.11 Effect of levulinic acid concentration in aqueous phase on specific rate of
extraction ( )……………………………………………………………………141
5.12 Effect of TMA concentration in aqueous phase on the specific rate of
extraction( ) …………………………………………………………………....141
5.13 Determination of reaction rate constant, k1.5………………………………….142
5.14 Schematic diagram for scaled-up reactive extraction process to recover
levulinic acid from the aqueous stream……………………………………….143
6.1 A plot of distribution coefficient , extraction efficiency (% ) (a) for 1-
decanol and MIBK, (b) 10, (c) 20, and (d) 30 % composition of extractants in
1-decanol and MIBK diluents. Temperature = 303.15 K, pH = 3.5………...163
6.2 Distribution coefficient ( ) and extraction efficiency (% ) for different
organic phases. (a) TOA+1-decanol, (b) TBP+1-decanol, (c) TOA+MIBK, and
(d) TBP+MIBK……………………………………………………………………171
xvii | Page 6.3 Distribution coefficient ( ) and extraction efficiency (% ) for individual acids
(AA, PA, and BA) at different temperatures (a) TOA+1-decanol, (b)TBP+1-
decanol, (c) TOA+MIBK and (d) TBP+MIBK…………………………………174
6.4 Overall extraction efficiency (% E) and overall distribution coefficient ( ) at
different temperatures (a) TOA+1-decanol, (b) TBP+1-decanol, (c)
TOA+MIBK, and (d) TBP+MIBK……………………………………………….177
6.5 Extraction efficiency (% E) and distribution coefficient ( ) for individual acids
(AA, PA, and BA) at different pH (a) TOA+1-decanol, (b) TBP+1-decanol, (c)
TOA+MIBK, and (d) TBP+MIBK……………………………………………….179
6.6 Overall extraction efficiency and overall distribution coefficient at different pH
(a) TOA+1-decanol, (b) TBP+1-decanol, (c) TOA+MIBK, and (d)
TBP+MIBK………………………………………………………………………..181
6.7 Main effect plots ( / ratios) for (a) distribution coefficient ( ) (b) extraction
efficiency (% )…………………………………………………………………..183
6.8 Percentage contribution of factors for (a) distribution coefficient ( ) and
extraction efficiency (% )………………………………………………………184
6.9 3-D surfaces showing effect of combination of factors on responses. (a-c)
distribution coefficient ( ), (d-f) extraction efficiency (% )………………189
xviii | Page List of Tables
1.1 List of carboxylic acids process development status…………………………….7
2.1 Ionic liquids used in extraction process to separate various substances…….18
2.2 The mechanism involved in the amine extractants……………………………..25
2.3 The available literature on some of the carboxylic acids recovery using
reactive extraction process………………………………………………………33
3.1 Properties of chemicals……………………………………………………………48
3.2 L18 Factors and levels by Taguchi mixed design………………………………51
3.3 Mixed array (three factors at different levels and results………………………52
3.4 Effect of temperature on for the extraction of levulinic acid using
MIBK…………………………………………………………………………………62
3.5 Partition ( ) and dimerization coefficient ( ) values for the extraction of
levulinic acid using MIBK at different temperatures (293-333 K)……………...63
3.6 Variation of loading ratio (z) at various concentrations of TOA at 293 K……..66
3.7 Effects of temperature (293-333 K) on the reactive extraction of levulinic acid
using TOA in MIBK…………………………………………………………………69
3.8 Gibbs free energy at different temperatures…………………………………….68
3.9 Response table for signal-to-noise ratios ( / ; larger is better)………………76
3.10 ANOVA for distribution coefficient ( ) and extraction efficiency (% ) in 18
array design…………………………………………………………………………77
3.11 Multivariate optimisation of and % ………………………………………….78
3.12 Diffusion and Mass transfer coefficients at 293 K………………………………82
4.1 Selected process parameters and their levels for succinic acid extraction…102
4.2 Experimental design values with the responses………………………………104
xix | Page 4.3 Different R2 values for reactive extraction of succinic acid…………………..103
4.4 Analysis of variance (ANOVA) for response surface quadratic model……...107
4.5 Verification of experimental results at optimum process variables………….108
4.6 Diffusion and Mass transfer coefficients at 303 K……………………………..111
5.1 Properties of Chemicals………………………………………………………….124
5.2 List of equipment and their capital costs in the reactive extraction plant with a
capacity of 2 m3 h-1 to recover levulinic acid…………………………………...144
5.3 Operating cost of the reactive extraction plant with a capacity of 2 m3 h-1 for
the recovery of levulinic acid…………………………………………………….145
5.4 Evaluation of profitability of the reactive extraction plant with a capacity of 2 m3
h-1 for the recovery of levulinic acid……………………………………………..147
6.1 36 Factors along with their levels by Taguchi mixed design……………….157
6.2 36 mixed array (levels of five different factors and obtained results)……….157
6.3 Individual distribution coefficients of AA, PA and BA, overall distribution
coefficients and extraction efficiency % at different pH and temperatures..164
6.4 Response Table for signal-to-noise ratios( ) - Larger is better…………..160
6.5 ANOVA for distribution coefficient ( ) and extraction efficiency (% ) in 36
array design………………………………………………………………………..186
6.6 Multivariate optimization of and %E and confirmation of the optimum
condition……………………………………………………………………………188
xx | Page Nomenclature
Chapter 3
Distribution coefficient in diluent
Distribution coefficient
Concentration of acid (kmol m-3)
% Extraction efficiency
Specific rate of extraction (kmol m-2 s-1)
Signal to noise ratio
Response characteristics
D Desirability responses
Overall desirability
Specific interfacial area (m-1)
time (s-1)
Partition coefficient
Dimerization coefficient
Stoichiometric (m:n) equilibrium complexation constant
Loading ratio
N Number of droplets formed during splitting
Speed of agitation (rev s-1)
H Height of the liquid in the reactor (m)
Droplet radius (m)
xxi | Page Dispersed phase volume (m3)
Droplet surface area (m2)
Droplet volume (m3)
Radius of reactor (m)
Liquid volume (m3)
∆ Enthalpy of reaction (kcal mol-1)
∆ Entropy of reaction (cal mol-1 K-1)
Gas constant (kcal K-1 mol-1)
T Temperature (K)
∆ Gibbs free energy (kcal K-1 mol-1)
0 -3 Initial acetic acid concentration (kmol m )
Acetic acid concentration (kmol m-3)
Acetic acid mass transfer coefficient (m s-1)
jth component diffusion coefficient (m2 s-1)
2 Second order rate constant
Levulinic acid mass transfer coefficient (m s-1)
[ ]0 Tri-n-octylamine initial concentration (kmol m-3)
Greek letters
Ρ Density (kg m-3)
Υ Interfacial tension (N m-1)
Φ Dispersed phase volume ratio
ɣ Kinematic viscosity (m2 s-1)
xxii | Page Subscripts
Aq Aqueous phase
Org Organic phase
Superscript
* Equilibrium
Chapter 4
Concentration of succinic acid (kmol m-3)
T Time (s)
A Specific interfacial area (m-1)
Specific interfacial area (kmol m-2 s-1)
T Temperature (K)
Acetic acid concentration (kmol m-3)
0 -3 Initial concentration of acetic acid (kmol m )
Mass transfer coefficient of acetic acid (m s-1)
Mass transfer coefficient of succinic acid (m s-1)
Diffusivity (m2 s-1)
ɣ Kinematic viscosity (m2 s-1)
Subscripts
In Initial
Aq Aqueous
Org Organic
Superscripts
* Equilibrium
xxiii | Page Chapter 5
Acid concentration (kmol m-3)
Speed of agitation (rev s-1)
ℎ Height of the liquid in the reactor (m)
Droplet radius (m)
Dispersed phase volume (m3)
Droplet surface area (m2)
Droplet volume (m3)
Radius of reactor (m)
Specific rate of extraction (kmol m-2 s-1)
Number droplets formed during splitting
Specific interfacial area (m-1)
Levulinic acid
Acid dissociation constant
TOA dissociation constant
TMA dissociation constant
% Recovery efficiency
[ ] TMA concentration (kmol m-3)
Greek letters
Density of liquid (kg m-3)
Interfacial tension (N m-1)
dispersed phase volume ratio
Subscript
aq Aqueous
org Organic
xxiv | Page Superscript
* Equilibrium
Chapter 6
Concentration of acid (kmol m-3)
Distribution coefficient
% Degree of extraction
Response characteristics
Desirability of response
Overall desirability
Signal to noise ratio
Subscripts
Iith desirability
Aq Aqueous phase
Org Organic phase
Min lower tolerance
Max upper tolerance
Superscripts
* Equilibrium
R Desirability
xxv | Page Abstract
Recovery and regeneration of value added carboxylic acids using process intensification, i.e. reactive extraction is a promising technique to recover carboxylic acids from dilute fermentation broths. The study is divided into three parts. In the first part of the work, the reactive extraction of levulinic and succinic acids are studied individually using tri-n-octylamine (TOA) in methyl isobutyl ketone (MIBK) and 1- decanol, respectively. In the second part of the work, the regeneration of levulinic acid from the loaded-organic phase using 4 different techniques is studied. The third part of the work is the recovery of multi-acids (volatile fatty acids) by reactive extraction.
Equilibrium and kinetic studies were conducted in the reactive extraction of levulinic acid. Physical equilibrium studies of levulinic acid were carried out with MIBK at various temperatures (293-333 K). Partition ( ) and dimerization ( ) coefficients were estimated at different temperatures to represent the physical equilibrium. The
distribution coefficient ( ) at physical equilibrium was found to be very low.
Chemical equilibrium studies conducted at different concentrations of TOA showed that the highest values of the distribution coefficient ( ) and the extraction efficiency
(% ) achieved 58.0 and 98%, respectively for 0.1 kmol m-3 of levulinic acid and
0.678 kmol m-3 of tri-n-octylamine at 293 K. Chemical equilibrium was also found to lead to the formation of 2:1 complexes. Taguchi mixed design multivariate approach
( (61 32)) was used to optimise the process variables and the ratio (larger-is 18 better) criterion was adopted to optimise the performance parameters. The optimum
-3 combination of variables was found to be an acid concentration ( 1) of 0.3 kmol m ,
-3 a TOA concentration ( 2) of 0.678 kmol m and a temperature ( 3) of = 293 K.
1 | Page Kinetic studies performed at optimum parameters showed that the overall order of the reaction was second order with respect to levulinic acid and tri-n-octylamine.
Based on the Hatta number the reaction regime was found to be instantaneously occurring in the film of the organic phase.
The reactive extraction of succinic acid using TOA in 1-decanol at different temperatures (298-333 K) was studied by employing a response surface methodology to explore the reaction kinetics in a stirred cell. Extraction efficiency (
% ), a response function, was optimized by using three process parameters: initial succinic acid concentration ( , ), TOA composition ( ), and temperature ( ).
In conjunction with response surface methodology, a central composite design consisting of twenty experimental runs was also employed to optimise the reactive extraction of succinic acid. A statistical second order polynomial quadratic model predicted an extraction efficiency (% ) of 93.75% with the optimum parameter values of 0.2 kmol m-3 succinic acid concentration, TOA composition of 33 (v/v), and a temperature of 305.5 K. The actual extraction efficiency obtained at optimum conditions was 91%. Kinetic studies were carried out to analyze the process. An interfacial area ( ) correlation was derived based on the droplet radius. The mass transfer coefficients of succinic acid, TOA, and the 1:1 acid-amine complex in 1- decanol were determined using a water/acetic acid/kerosene system. Based on the
Hatta number, the reaction regime was found to be instantaneously occurring in the film.
The second part of the study, the regeneration of carboxylic acids from the loaded- organic phase, is an essential step to complete the reactive extraction process. A study on the regeneration of levulinic acid from the loaded-organic phase (MIBK
2 | Page +TOA +levulinic acid) was carried out using various techniques including the NaOH, temperature swing, diluent swing, and tri-methylamine methods. Equilibrium data obtained show that among all the methods, the recovery of acid is the highest for the tri-methylamine method when the molar ratio of tri-methylamine to levulinic acid concentrations is greater than 1. Kinetic studies performed for the tri-methylamine method showed that there are no changes in the specific rate of extraction with
changes in stirrer speed rate and phase volume ratio ( ), and the overall order of reaction is 1.5. Based on the effects of stirrer speed and phase volume ratio on the specific rate of extraction, the reaction was found to occur in the fast regime. Also, about 80% of acid was recovered by the evaporation of the tri-methylamine phase at
104–140 0C. A detailed economic evaluation for the recovery of levulinic acid using reactive extraction with a feed rate of 2 m3 h-1 shows that the payback period for recovering the capital investment is 0.49 years.
The third part of the study, the recovery of volatile fatty acids (multi-acids) from fermentation broth has been investigated by adopting an intensified approach using extractants TOA and tri-butyl phosphate (TBP) dissolved in 1-decanol and MIBK.
The effects on the distribution coefficient ( ) and extraction efficiency (% ) were studied by varying the operating conditions of temperature (293.15–323.15 K), pH
(2.5, 3.5, and 4.5), and composition of extractant (10, 20, and 30 %). Taguchi ( 36 ) orthogonal design with five factors, (diluents, extractant type, composition, temperature, and pH), was employed for the multivariate optimisation of the reactive extraction of volatile fatty acids. In the Taguchi approach, a “larger is better” criterion was adopted to maximise and % . Statistical analysis indicated that the degree of influence on by the experimental variables follows the following trend:
3 | Page extractant type ( 2 ) > pH ( 5 ) > diluent type ( 1 ) > temperature ( 4 ) > extractant concentration ( 3 ). The trend for % observed is as follows: extractant type (2 ) > pH ( 5 ) > temperature ( 4 ) > extractant concentration ( 3 ) > diluent type ( 1 ). The combination of optimum parameters were obtained as 1 = 1-decanol, 2 = tri-n- octylamine, 3 = 20 %, 4 = 293.15 K, and 5 = 3.5. A confirmation run was conducted using these parameters and and % values from this run were determined to be 8.65 and of 89.64 %, respectively, which were very close to the predicted values = 10.36 and % = 91.26 %.
4 | Page Chapter 1 General Introduction
5 | Page 1.1 Introduction
1.1.1 Background of the work
Carboxylic acids are organic acids containing a carboxylic group (-COOH). The common formula of a carboxylic acid is R-COOH. Carboxylic acids are very valuable in the food and chemical industry as they are regularly used in large-scale production of food and chemicals. Different carboxylic acids are produced from renewable resources by biotransformation or fermentation of cells using pure cultures
(Goldberg, Rokem & Pines 2006; Sauer et al. 2008; Straathof, AJ 2013). Volatile fatty acids (acetic, butyric and propionic acid) are derived from the waste streams of anaerobic slurries. Based on the demand, of renewable resources, the production of some carboxylic acids such as itaconic acid, citric acid, D-gluconic acid, lactic acid, and 2-keto-L-gulonic acid have been commercialized. Currently, due to an increasing demand for succinic acid, it is produced at commercial levels by many manufacturers
(McCoy 2009). Other acids currently produced at commercial levels recently are propionic acid (Della Pina, Falletta & Rossi 2011), adipic acid (Beardslee &
Picataggio 2012), and acrylic acid. Table 1.1 shows a list of carboxylic acids and the stage at which their commercial manufacture is currently at.
6 | Page Table 1.1 List of the carboxylic acids whose process development is underway currently
Name of Molecular Development of Application Literature the acid structure production process
Acetic Industrial Vinegar (Xu, Shi & Jiang
2011)
Propionic At design stage Preservative (Liu et al. 2012)
Butyric At design stage Flavoring agent (Dwidar et al.
2012)
Succinic Industrial Polymers (McKinlay,
Vieille & Zeikus
2007)
Levulinic At research Precursor for (Girisuta,
stage pharmaceuticals Janssen &
Heeres 2007)
Acrylic At research Polymers (Straathof, AJ et
stage al. 2005)
Pyruvic At research Chemical (van Maris et al.
stage 2004)
Lactic Industrial Food, Polymers (Miller et al.
7 | Page 2011)
Fumaric Industrial Food, Polymers (Straathof, AJJ
& van Gulik
2012)
Malic Research Chemicals (Zelle et al.
2008)
Itanoic Industrial Polymers (Klement &
Büchs 2013)
Glutaric Research Polymers (Otto, Yovkova
& Barth 2011)
Citric Industrial Food (Soccol et al.
2006)
Adipic At design stage Polymers (Polen, Spelberg
& Bott 2013)
The production cost of carboxylic acids mainly depends on the feedstock cost and the downstream processing. The downstream processing cost is usually around the
30-40% of the production cost (Straathof, 2011). To minimise the costs of downstream processing, recent research has been targeted towards developing an effective method for recovering carboxylic acids from the fermentation broth.
Carboxylic acids can make up to 10% of fermentation broth. There are many methods available to recover carboxylic acids from fermentation broth such as
8 | Page solvent extraction, precipitation, membrane separation, ionic liquids extraction,
Chromatography, distillation and reactive extraction.
1.1.2 Research Gap
Although a significant amount of research has been carried out to date on the reactive extraction of carboxylic acids, there is still lack of information on many aspects of the process. In particular, there is a significant lack of knowledge on the regeneration of carboxylic acids from the loaded organic phase and the scale-up of the selected recovery process. Additionally, there is not much information on the recovery of mixtures of multiple acids from the fermentation broth. The majority of the available literature is on the recovery of a single acid. Therefore, it is worthwhile to study the recovery of carboxylic acids using reactive extraction, a process intensification method, which is economically attractive compared to other separation methods.
In this study, forward and backward extractions of two carboxylic acids, levulinic and succinic acids, from the loaded-organic phase were studied. In addition, the recovery of multi-acids (volatile fatty acids,i.e., acetic, butyric and propionic acids) from the loaded organic phase was studied in this work. An optimization study was conducted using a Taguchi approach and a response surface method to determine the process variables that will lead to higher distribution coefficient and extraction efficiency. An economic analysis was carried out on the process involving the recovery of levulinic acid using reactive extraction.
9 | Page 1.1.3 Research Questions
This thesis aims to address the following research questions:
How efficient is a TOA+MIBK system in extracting levulinic acid from aqueous
solutions and what are the design parameters?
How efficient is a TOA+1-decanol system in extracting succinic acid from
aqueous solutions and what are the design parameters?
Which technique is the most efficient in regenerating the levulinic acid from
the loaded organic phase among the TOA+1-decanol diluent swing,
temperature swing, NaOH, and tri-methylamine (TMA) methods?
Which system is the most efficient in recovering the multi-acids from aqueous
solutions among the TOA+1-decanol, tri-butyl phosphate (TBP)+1-decanol,
TOA+MIBK and TBP+MIBK systems?
1.1.4 Aims and objectives
The aim of this work is to study the efficiency of the reactive extraction process in recovering the carboxylic acids from aqueous solutions using different extractants and diluents
Objectives:
To study the efficiency of reactive extraction (process intensification) in
recovering the following acids from aqueous solutions:
o Levulinic acid
o Succinic acid
To study the efficiency of regeneration/back extraction of levulinic acid from
the loaded-organic phase
10 | Page To study the efficiency of the recovering a multi-acid mixture containing
acetic, propionic and butyric acids from aqueous solutions using reactive
extraction (process intensification).
1.1.5 Thesis Structure
This thesis has been divided into seven chapters as shown below.
Chapter 1: A general introduction on the importance of carboxylic acids is presented. It also presents the background of the research area, research gap, research questions, aims and objectives of the thesis.
Chapter 2: In this chapter, a detailed review of the literature on the recovery of carboxylic acids is presented. It discusses various processes that have been used to recover carboxylic acids and explains how reactive extraction is one of the most promising techniques for the recovery of carboxylic acids.
Chapter 3: This chapter deals with levulinic acid recovery from aqueous solutions using (TOA) in MIBK with various initial concentrations of levulinic acid and TOA at different temperatures. It discusses how the process variables used in this work are optimised using the multivariate Taguchi approach with 3 parameters and at 3-levels.
It explains the kinetic studies carried out using stirred cells at optimised conditions.
Finally, it discusses the determination of the reaction regime in the reactive extraction process using the Hatta number.
Chapter 4: This chapter describes the recovery of di-carboxylic acid (succinic acid) from aqueous solutions using the TOA in 1-decanol with different initial concentrations of succinic acid and TOA in 1-decanol at different temperatures. It discusses the response surface methodology that has been used to optimise the
11 | Page process variables used in this work. It also explains the kinetic studies carried out at different concentrations of acid and extractant to find out the order and rate constant of the reaction. Finally, it discusses the determination of the reaction regime in the reactive extraction process using the Hatta number.
Chapter 5: This chapter discusses the back extraction of levulinic acid from the loaded organic phase (levulinic acid+TOA+MIBK). It discusses the equilibrium studies conducted using the NaOH, TMA, temperature swing and diluent swing methods to determine the most efficient method to regenerate levulinic acid. It discusses the kinetic study carried out using the TMA method. It also discusses the process economics for determining the feasibility of total recovery (forward + backward) of levulinic acid from the loaded organic phase.
Chapter 6: This chapter discusses the recovery of multi-acids (acetic, propionic and butyric acid) by the reactive extraction method using TOA and TBP in 1-decanol and
MIBK solvents from the loaded organic phase. It describes the use of a multivariate
Taguchi approach in optimising the process variables to enhance the distribution coefficients and extraction efficiency.
Chapter 7: This chapter presents the overall conclusions based on the results obtained from various sections of the work. It also presents recommendations for further work.
12 | Page Chapter 2 Literature Review
13 | Page 2 Literature Review
2.1 Methods for carboxylic acids recovery
2.1.1 Precipitation
Precipitation is an important industrial scale method to recover organic acids from the fermentation broth, which has been used for the recovery of citric acid and lactic acid since the last century. Precipitation recovers carboxylic acids very easily from the fermentation broth, which makes it very competitive compared to other separation techniques. The precipitation method separates carboxylic acids in four steps. The first step is filtration of the fermentation broth, removing any impurities, before adding CaCO3 and Ca(OH)2 to the left over liquor, in the presence of agitation. The second step is the filtration of the precipitated calcium salt of carboxylic acids. The third step is mixing the calcium salt with concentrated sulfuric acid to free the target carboxylic acid. After this, the acid is purified using further refining methods (Lee, SC & Kim 2011; Wasewar, KL 2005). Heading and Gupta
(Heding & Gupta 1975) and Pazouki and Panda (Pazouki & Panda 1998) recovered citric acid using the precipitation method with 100 % yield and optimised the conditions of the process as 50 0C and 20 min. Min et al. (1998) reported that the molar ratio of calcium lactate to sulfuric acid is a significant parameter to isolate lactic acid with precipitation. The lactic acid yield was 92% at optimum conditions
(Pazouki & Panda 1998). Some researchers (Berglund, Yedur & Dunuwila 1999;
Datta, R 1992; Datta, R et al. 1992) patented their work on carboxylic acid separation from fermentation broth using the precipitation method. Succinic acid has also been recovered using an ammonium precipitation method. Di-ammonium succinate was generated using an ammonia-based titration agent; afterwards, the
14 | Page precipitated salt was treated with sulfuric acid to produce succinic acid. The obtained succinic acid yield was 93.3% with the ammonium precipitation method. Nowadays the precipitation method is a well-known process for the recovery of carboxylic acids from fermentation broths with the advantages of high selectivity and high product purity. The main disadvantages on the industrial scale are the formation of calcium hydroxides and carbonates in equal amounts to carboxylic acid and the consumption of sulfuric acid.
2.1.2 Solvent extraction or liquid-liquid extraction
Liquid-liquid extraction or solvent extraction is a low energy separation technique
(Kurzrock & Weuster-Botz 2010), which can be used as an in-situ product removal process. Acetic acid can be separated from fermentation broth using diethyl ether, ethyl acetate, and diethyl ether-hexane mixture; it has been observed that ethyl acetate can recover more acid than other solvents (Jipa et al. 2009). The recovered products are purified with film distillation followed by precipitation of calcium salts to meet the standards of food and medical applications. Chen et al. (2012) developed a new process for lactic acid recovery by adopting a simple calcium salt precipitation and a short-path distillation using the butyl alcohol. The novel approach was successful in producing lactic acid of higher quality that can be obtained without applying a vacuum. Compared to the earlier purification process, the efficiency in terms of the yield and purity of lactic acid was enhanced to 61.73% and 91.6%, respectively (Chen, L et al. 2012).
2.1.3 Membrane separation
Membrane separation is one of the most important processes for carboxylic acid recovery from the fermentation broth. The membrane, a thin artificial barrier, allows
15 | Page the solute or solvent through to reach the required physical separation with a high yield and purity. The common membranes used in carboxylic acids recovery are ultrafiltration, microfiltration, reverse osmosis, nanofiltration (NF), electrodialysis, and pervaporation (Cheng et al. 2012; Mao et al. 2014; Vane 2005). NF has also been used in lactic acid recovery from clarified broth. The lactate ions and the non-ionized lactic acid ratio can affect the permeation and rejection through the NF membrane.
Hence, pH is a significant parameter for determining the effectiveness of NF in separation (Gonzalez et al. 2008). Pervaporation is a fermentation technology that can be used for near-boiling point substances. The pervaporation flux is still less than 2,000 gm-2h-1 which is a usual requirement to improve the process. In membrane technology, a large gap between pervaporation and other membrane technologies is evident (Gonzalez et al. 2008; Marszałek & Kamiński 2012; Vane
2005). Since the early 1970s, electrodialysis has been used for citric acid separation from the fermentation broth. The bipolar membrane is considered as an advantageous feature of electrodialysis. Habova et al. (2004) established a two- stage method for lactic acid separation with specific energy consumption of 1.5 kWh per kg. Though, due to the high consumption of energy, this method is still limited in its applications (Hábová et al. 2004). Acetic acid and succinic acid can also be separated from the fermentation broth by using the emulsion liquid membranes. Lee et al. (2011) reported that the efficiency of the acetic acid extraction was higher than
95% with an enrichment ratio of 3. Membrane technologies are a highly efficient separation processes, albeit more energy intensive than other technologies.
However the membrane itself is very cost effective.
16 | Page 2.1.4 Ionic liquid extraction
Ionic liquids (ILs) are a mixture of organic salts such as imidazolium, (Lateef,
Gooding & Grimes 2012; Wang et al. 2005), quaternary phosphates (Marták &
Schlosser 2007; Oliveira et al. 2012), or quaternary ammonium salts (Kulkarni et al.
2007; Mikkola, Virtanen & Sjöholm 2006), they are non-flammable and non-volatile and are able to withstand high temperatures (Domańska & Lukoshko 2013). ILs are chemically stable with a low viscosity and high density compared to other organic solvents. Lateef et al. and Sun et al. reported that ILs are more effective extracting solvents than organic solvents (Lateef, Gooding & Grimes 2012; Sun, Luo & Dai
2011). Lactic, malic, and succinic acids can be separated from aqueous solutions using phosphonium-based ILs (Oliveira et al. 2012). ILs efficiency in recovering short chain organic acids from aqueous solutions has been shown to reach 73% (Oliveira et al. 2012). Martak and Schlosser (2007) showed that industrially treated phosphonium-based ILs have better extracting ability than other organic solvents in lactic acid separation. Wang et al. (2005) showed that the properties of aqueous solutions such as pH, alkyl chain length and anion nature have a strong effect on the
IL/water partition coefficient. Even though ILs have many advantages in the separation of carboxylic acid, the cost is the main obstacle (Wang et al. 2005).
17 | Page Table 2.1 Ionic liquids used in extraction process to separate various substances.
Ionic liquids Separated substances References
Imidazolium–based Lactic acid, tryptophan, (Lateef, Gooding &
ionic liquids phenylalanine tyrosine and Grimes 2012; Wang et
succinic acid al. 2005)
Phosphonium-based Lactic, malic and succinic (Marták & Schlosser
ionic liquids acids 2007; Oliveira et al.
2012)
Quarternary ammonium Organic and rare earths (Kulkarni et al. 2007;
ionic liquids Mikkola, Virtanen &
Sjöholm 2006)
Pyrrolidinium-based Acetic acid and aromatic (Domańska & Lukoshko
ionic liquids hydrocarbons 2013)
2.1.5 Chromatographic method
The chromatographic method depends on the adsorptive or ion-exchange properties of the resin used. It is a conventional process for the final purification of carboxylic acids from aqueous solutions. In this method, the properties of the resins that are used for purification are physically and chemically stable and should be insoluble in acids and organic solvents. This process is economically favorable as there is no phase transition (Davison, Nghiem & Richardson 2004; Park, C et al. 2014). There are two kinds resins used; ion exchange resins and macroporous resins (Bishai et al.
2015). Polymers that have substituted basic or acidic functional groups can be used
18 | Page as resins. The exchange capacity is determined by particle size, exchange ions and degree of cross-linking (Guo, Chang & Hussain 2009). Ion-exchange resins normally used in bio-separation result in a low energy consumption process (Boonkong et al.
2009). The performance of macroporous resins depends on van der Waals forces, dipole ion interactions and hydrogen bonding.
Zhou et al. (2011) used polyresin to separate lactic acid from the fermentation broth and achieved purity and yield of 88 % and 95 %, respectively. Tong et al. (2004) reported that weak anion-exchange resins could be used for lactic acid purification.
They showed that the recovery yield could be increased when the pH of the fermentation broth was kept between 5.0 and 6.0 and achieved purity and yield of
92.2% and 82.6%, respectively (Tong, W-Y et al. 2004). Davison et al. (2004) reported a resin that can be recycled more than 10 times with a stable capacity for the recovery of succinic acid using 25 different sorbents at either neutral or acidic conditions. They showed a succinic acid yield of up to 95% (Davison, Nghiem &
Richardson 2004; Kumar, S & Babu 2008).
Chromatography does not produce by-products and has high yields. However, large quantities of excess mother liquor are produced, and a large quantity of salt is consumed in the elution process. Furthermore, the resin capacity will diminish over time (Aljundi, Belovich & Talu 2005).
2.1.6 Distillation
Distillation is a separation process based on the difference between boiling points of components in the fermentation broth (Errico & Rong 2012). In the early days of carboxylic acid recovery, process distillation was the preferred method of separation, along with crystallization. Distillation is an efficient process at low concentrations of
19 | Page carboxylic acids but is inefficient at higher concentrations, especially at the azeotropic point (Huang, H-J et al. 2008). Kumar et al. (2006) used a continuous distillation process for lactic acid separation with a single stage reactive distillation with methanol and reported that yield and selectivity were higher than the conventional batch processes and it also consumed less energy. Blahusiak et al.
(2012) reported an extractive distillation technique with ILs to separate the carboxylic acids from a fermentation broth. Butyric acid was recovered using a phosphonium IL in a two-stage short path distillation with the yield of 89%. In this process, the product is in the form of free acid rather than a mixture with salt (Tong, Y et al.
1998). Distillation is commonly used for the separation of various volatile constituents. Distillation is not considered an effective process for the carboxylic acids separation anymore because of the strong adsorb-electron effect of the carbonyl group, and the higher boiling point of carboxylic acids than water.
Furthermore, it is more energy intensive and less cost effective than other unit operations.
2.1.7 Reactive extraction
Of all the carboxylic acid recovery techniques, reactive extraction, a process intensification method which arises as an extension of the solvent extraction process, has been chosen as the focus of this study. In the solvent extraction process, the target solute is separated from the feed using another liquid to solvate the target solute. The first scientific application of liquid-liquid extraction was established circa 3500 BC to recover products from different natural resources. The production of waxes, pharmaceutical oils, and perfumes have been documented in a
Sumerian text dated 2100 BC. In the Medieval ages, extraction was performed in the
20 | Page Figure 2.1 Total process of reactive extraction process, Blue region-the work carried out in this work. presence of ethanol and was applied in hydrometallurgy to make use of mineral acids (Bart 2001). The design of an apparatus for reactive extraction was developed in 1891 based on the distribution law by Nernst. However, large scale liquid-liquid extraction didn’t begin until the early 1930’s. Lazar Edeleanu (1861-1941) was the first to apply this technique to separate aromatic sulphur from liquid kerosene using liquid sulphur dioxide as a solvating agent at a low temperature. (-6 to -120C). This process yielded kerosene as raffinate which could be used as fuel for residential lighting. In the modern world extractants are widely used throughout the process industry; from the extraction of metals in the mining industry to various environmental applications, to the extraction organic and carboxylic acids and intermediates of the
21 | Page organic synthesis process, and for separation and purification of pharmaceutical products.
Reactive extraction involves the combination of reaction and separation in a single unit process. Recent research is targeted towards the selection of a unit process that is cheap, simple, generates less waste and consumes less energy. Reactive extraction can effectively be utilised as a process intensification method to obtain significant improvements in the selectivity and yield of desired products. Solvent extraction and reactive extraction can be differentiated by whether or not the extractant reacts with the target solute in the organic phase. The reactant forms a complex with the target solute which is soluble in the organic phase. Kertes and King
(1986) categorised extractants into 3 distinct types of the extractant.
1. Higher molecular weight aliphatic amines.
2. Phosphoric solvents.
3. Carbon bonded oxygen based compounds.
Aliphatic amines and phosphoric solvents are reported as most effective extractants in the literature (Kertes & King 1986). Whilst extractants are very important in reaction, diluents also affect the level of extraction significantly. Water immiscible, non aromatic and polar solvents with high boiling points and intermediate molecular weights are generally considered to have high extraction efficiency and distribution coefficients for the recovery of carboxylic acids (Hartl & Marr 1993). Extractants can improve the physical properties of the solvent like viscosity, density, surface tension, etc. The stability of the solute-extractant complex is based on the strength of the extractant but can be improved by the choice ofthe solvent. Reactive extraction with suitable organic phase (extractant + solvent) has been shown to be an emerging
22 | Page field for the recovery of the carboxylic acids from dilute fermentation broth (Cascaval
& Galaction 2004; Hartl & Marr 1993; Kumar, S & Babu 2008; Wasewar, KL et al.
2002a, 2002b; Wennersten 1983). Reactive extraction has the following advantages over other separation processes:
1. Effective at high concentration of substrate in the fermentation broth.
2. Acid can be separated from acid-extractant complex, and the solvent can be
reused.
3. By using the process, pH can be controlled in the bio-reactor.
4. Acid can be recovered at a higher purity.
5. It reduces downstream costs.
6. Reactants are immiscible in the aqueous phase.
7. Products are immiscible in the reaction phase
8. Phase equilibrium can be positively influenced
Phosphoric extractants and carbon bonded oxygen extractants are non-reactive in nature, and they extract acid molecules by solvation compared to more reactive aliphatic amines. The main difference between the phosphorus-based solvents and carbon bonded oxygen based compounds is the strength of the solvation bonds. The bonds formed between carbon bonded oxygen compounds and acid are significantly weaker than that between phosphorus-based solvents and acid. Phosphorus based extractants offer a more specific extraction process, and the number of solvating molecules per extracted acid molecule can be found experimentally. Aliphatic amines react with the carboxylic acid molecules to form an acid amine complex by proton transfer or ion pair formation, which improves the distribution coefficient of the acid (Wardell & King 1978). The extraction efficiency of tertiary aliphatic amines is higher than that of primary and secondary aliphatic amines (Wennersten 1983).
23 | Page Primary amines tend to be more water soluble where as secondary amines tend to make the third phase at the interface which creates difficulty in the phase separation stage (Kertes & King 1986). Tertiary amines with long carbon tails are the most effective extractants for the carboxylic acids recovery (Kertes & King 1986). However they require large amounts of diluent due to their high viscosity and corrosive nature.
The alkalinity of amine compounds also heavily impacts the stability of the acid- amine complex formed in the process (Bizek et al. 1992). Table 2.2 represents the mechanisms employed by amine based extractants.
24 | Page Table 2.2 Mechanisms employed by amine based extractants.
Fermentation broth pH Carboxylic Amine Extractants Mechanism Remarks acid
+ - pH < pKa HA Primary, secondary and Ion pairing : BH A Ion pairing prevails over H- tertiary amines (B) bonding for strong basic B h-bonding : B:HA H- bonding between the amine hydroxyl group of the acid
Primary, secondary and H-bonding BH+X-:HA A-Weaker than X-, so anion tertiary ammonium saltb exchange will occur only for + - (BH+ X-) Anion Pairing BH A A-
Quaternary ammonium Ion pairing : Q+A- A-Weaker than X-, so anion hydroxide (Q+OH-) exchange will occur only for + - H- bonding Q X :HA A- Quaternary ammonium saltb (Q+ x-)
pH>pKa A- Primary or secondary amine H-bonding B:A H-bonding between the (B) carbonyl group of the acid and the amine
25 | Page Tertiary amine (B) Nonea
Primary, secondary or Ion pair BH+A- tertiary ammonium saltb (BH+X-)
Quaternary ammonium Ion pair Q+A- hydroxide (Q+OH-)
Quaternary ammonium salt Ion pair Q+A- (Q+A-) aHydrophobic interactions occur in all cases and are not indicated in the table bSeveral inorganic anion may be used
26 | Page Typically, diluents are categorized into 2 types based on their properties; active diluents and inactive diluents. Active diluents are polar in nature because of their functional groups. They are the best solvating media for the formation of acid-amine complexes (Tamada & King 1990). Alcohols, ketones, halogenated aromatic compounds and chlorinated hydrocarbons are included in the active diluent category.
Inactive diluents such as alkanes and aromatic compounds have low distribution coefficients and extraction efficiency. Inactive diluents form the third phase in the reactive extraction process at high acid concentrations. However these diluents act as better stripping agents in the regeneration process (Han & Hong 1996). Reactive extraction takes place at the interface of the organic and aqueous phases where solute molecules transfer by a diffusion mechanism. The mechanism depends on a range of factors such as the concentration of acid in the aqueous phase, extractant concentration in the organic phase, the molar ratio of complexes formed, properties of the organic phase, type of solvent, temperature, and pH (Kahaya et al., 2001).
These properties must be optimised to obtain good distribution coefficients and extraction efficiency.
The complete reactive extraction process includes two stages as shown in Figure 2.1 the first stage is the extraction of solute from the aqueous solution to organic phase to produce a loaded-organic phase containing the solute-extractant complex and solute free raffinate. The second step is stripping, the regeneration of the solute from the loaded organic, not only to obtain the solute but also to recycle the organic phase which contains the extractant and the diluent. Tamada and King (1990) achieved the regeneration step using two different processes; temperature swing and diluent swing. Yabannavar and Wang (1991b) have studied the regeneration of lactic acid from the loaded organic phase by using HCl and NaOH. Poole and King (1991) used
27 | Page a TMA solution with low volatility for the separation of acid from the loaded organic phase.
Reactive extraction has abundant applications in the chemical and pharmaceutical industries and in hydrometallurgical and environmental science. Ion exchangers can either be cation exchangers or anion exchangers. In this process, extractants are used as ion exchangers in the diluent because they are immiscible in water and highly viscous in nature. The diluent makes the organic phase easier to handle (Bart
2001). The available literature explores a range of factors relating to the complex formation mechanisms, type of extractants and diluents, the effect of temperature and pH, the effect of extractant composition on distribution coefficient, the effect of modifiers, design parameters of extractors in kinetic studies and stripping the acid from the loaded organic phase (Kertes & King 1986; Keshav, Chand & Wasewar
2008; Keshav, Wasewar & Chand 2008b; Kumar, S & Babu 2008, 2009; Ma et al.
2006; Poole & King 1991; Tamada, Kertes & King 1990; Tamada & King 1990; Tong,
Y et al. 1998; Wasewar, KL et al. 2002a, 2002b).
Theoretical and experimental studies on carboxylic acid recovery from aqueous solution using reactive extraction have been conducted by several researchers.
Some researchers have described reaction mechanisms and different equilibrium models involved in the process (Poposka, Nikolovski & Tomovska 1998; Poposka et al. 2000; Wasewar, KL et al. 2002a, 2002b). Authors have reported on the carboxylic acid recovery from the medium of production by using extractive fermentation and investigated how different parameters affect the process (Gu, Glatz & Glatz 1998;
Poposka, Nikolovski & Tomovska 1998; Siebold et al. 1995). In addition to carboxylic acid recovery, King et al. conducted equilibrium studies on the extraction of aromatics and chlorinated hydrocarbons (Barbari & King 1982), aliphatic alcohols
28 | Page with lower molecular weight (Kertes & King 1986) and ammonia (Mackenzie & King
1985) from aqueous solutions. Wardell and King (Tamada, Kertes & King 1990) investigated formic and acetic acid extraction from aqueous solution using phosphoryl solvents and tertiary amines dissolved in different solvents. They found that the extraction worked well with phosphoryl solvents with high electronegativity.
However, the extractive performance of these solvents decreased with decreasing electronegativity. The study also showed the acid distribution improved with increasing solubility of the extractant in the diluent.
Kertes and King (1986) worked on improvements in fermentation technology and discussed the requirements commercialising the process. They studied the recovery of the different monocarboxylic acids (citric, fumaric, itanoic, lactic, maleic, propionic, pyruvic, succinic and tartaric acids) that are produced by the aerobic fermentation of glucose through the glycolytic pathway and glyoxylate bypass. In studying carbon bonded oxygen and phosphorus bonded solvents in the extraction of monocarboxylic acids, they showed that the initial pH of the solution and the pKa values of the acids were important parameters for effective extraction. Equilibrium data were generated using the mass action law and Nernst distribution law to reveal a correlation between dimerization and partition coefficients. In studying the use of aliphatic tertiary amines extractants, they found that they were much more effective than primary and secondary amines and that monocarboxylic acids are more easily extracted than di- carboxylic or tri-carboxylic acids.
Tamada et al. (1990) carried out further equilibrium studies on the reactive extraction of carboxylic acids (acetic, lactic, succinic, fumaric) with differing pKa values and explored how these values impact the extraction process as well as how functional groups other than the primary carboxyl group impact the process. They concluded
29 | Page that acids with higher pKa values were extracted in greater amounts. The extractant used was Alamine 336 dissolved in various active (1-octanol, DCM, chloroform,
MIBK, nitrobenzene) and inactive diluents (heptane). The diluents were chosen based on their differing chemical properties such as whether they were electron donating or accepting and whether they were polar or nonpolar, in order to study the effect of diluent-complex interactions and how solubility affects the extraction process. They observed that the solubility order of the acid-amine complex in the diluent was as follows: alcohol ≥ nitrogen ≥ halogenated hydrocarbons > ketone > halogenated aromatic compounds > benzene > alkyl aromatic > aromatic hydrocarbon.
Tamada and King (1990) used mass action law and infrared spectroscopy to study the stoichiometry of acid-amine complexes formed in the process. Infrared spectroscopy revealed that ion pair bonding is responsible for the formation of 1:1 complexes and the first bond in 2:1 complexes, whilst hydrogen bonding is responsible for attracting the second acid molecule in 2:1 complexes. The final study conducted by Tamada and King (1990) looked into the tendency of Alamine 336, dissolved in various diluents, to co-extract water along with acid from the aqueous phase. They observed that the amine had no effect on the co-extraction of water.
However the choice of diluent and the type of acid being extracted did. The degree by which diluents contribute to water co-extraction in succinic acid extraction is as follows: 1-octanol > MIBK > nitrobenzene > methylene chloride > chloroform > heptane. In exploring the effect of the target solute on water co-extraction, they found that monocarboxylic acids carry less water than di-carboxylic acids. Their final study also examined how temperature impacts an Alamine 336 reactive extraction system. It was found that the distribution coefficient and the equilibrium constant
30 | Page decrease with increasing temperature. This follows as the system becomes less ordered with increasing temperature and entropy, thus less accommodating to the formation of stable complexes. The study also worked on the stripping phase of reactive extraction, employing both the temperature and diluent swing techniques.
Yang et al. (1991) were the first to work with quaternary amines for reactive extraction, utilizing Alamine 336 and Aliquat 336 (a quaternary amine derived from the methylation of Alamine 336) dissolved in either kerosene or 1-octanol to recover lactic, acetic, propionic and butyric acid. They concluded that Alamine 336 could only extract un-dissociated acids whereas Aliquat 336 can extract both dissociated and un-dissociated acids as it has a large organic cation and a chloride ion. This ultimately means that Aliquat 336 is a more effective extractant than Alamine 336. It is also stated that the extractive capacity of Alamine 336 is proportional to the polarity of the diluent used, whereas the extractive capacity of Aliquat 336 showed no correlation to the type of diluent used.
Prochazka et al. (1994) studied the extraction of lactic, malic, and citric acids using a tri-alkylamine mixture dissolved in 1-octanol and n-heptane. They found that all the systems in their study were affected by temperature and extractant composition.
Juang et al. (1997) studied succinic acid and tartaric acid extraction using the tertiary amine TOA dissolved in xylene. In comparing the equilibrium data of succinic and tartaric acids with TOA dissolved in xylene, they observed that the distribution coefficient increased with the concentration of TOA in the organic phase. Popaska et al. (1997) studied citric acid extraction using TOA in isodecanol/n-heptane and modeled equilibrium data as a function of acid concentration and amine concentration at 298 K to propose an appropriate mathematical model.
31 | Page Until now, most of the available literature has focused on the reactive extraction of single acids from an aqueous phase. Juang et al. (1997) studied the parameters of reactive extraction where more than one acid is present in the aqueous phase. They studied the separation mechanism of reactive extraction of citric acid and lactic acid with TOA dissolved in xylene using a supported liquid membrane for the reactive extraction. They stated that positive synergistic effects take place at low citric to lactic acid ratios (α). However as the ratio increases to α =2 it starts to impact negatively on the extraction process compared to a single acid system. They concluded that the presence of the second acid has a positive impact on the extraction of citric acid and a negative impact on the extraction of lactic acid.
Matsumoto et al. (2001) investigated a new synergistic extraction system for organic acids. They collected extraction equilibrium data for acetic, glycolic, propionic, lactic, succinic, fumaric, malic and itanoic acids using TOA and TBP dissolved in hexane.
They stated that when a mixture of extractants is used, a positive synergy is developed and a more effective extraction of all organic acids is achieved. The equilibrium studies using reactive extraction are listed in Table 2.3.
32 | Page Table 2.3 the available literature on some of the carboxylic acids recovery using reactive extraction process
Carboxylic Extractant diluent Parameters studied Observations Reference acid Citric, lactic Ti-iso- 1-octanol and heptane Type of acid and (Malmary et al. and malic octylamine modifier effect 2000) acids (TIOA) Nicotinic TOPO, TBPBenzene, heptane, Effect of extractant Solvation number and (Kumar, S, kerosene, 1-octanol, type and extractant equilibrium complexation Wasewar & Babu MIBK, diethyl ether, concentration, initial constants are determined 2008) decane, kerosene+1- acid concentration octanol, and heptane + effect of diluent type 1-octanol Levulinic n-Lauryl tri- Dimethyl phthalate, Effect of diluent and 1:1. 2:1 and 3:1 (Uslu, İsmail alkyl-methyl dimethyl adipate, amine concentration Complexes equilibrium Kırbaşlar & amine dimethyl succinate, constants determined Wasewar 2009) (Amberlite LA-2) dimethyl glutarate, and linear solvation diethyl carbonate, model proposed isoamyl alcohol, 1- hexanol, 1-octanol, 1- nonanol, 1-decanol,
33 | Page DIBK and MIBK Citric Tri- MIBK, 1-octanol, Effect of initial acid Among all the diluents 1- (Bayazit, Uslu & dodecylamine toluene, cyclohexane, concentration, octanol is the best diluent İnci 2009) (TDDA) and 1-octanol + MIBK, extractant to extract the carboxylic Amberlite LA-2 MIBK + toluene, 1- concentration and acids octanol + toluene, 1- effect of diluents octanol + MIBK, MIBK + toluene and iso- octane. Lactic TBP Dodecane Initial acid Equilibrium constants and (Labbaci et al. concentration, solvent extractant molecules 2009) composition, phase number in the reaction volume and pH Acrylic Amberlite LA-2Cyclohexane, 2- Effect of diluent type Equilibrium constants (Aşçı & İnci̇ 2010)
octanone, toluene, and extractant (K11, K12 and
MIBK, iso-octane, concentration K23)determined for proton hexane and 1-octanol and non-proton donating diluents Formic TDDA and TBP Ethyl valerate, diethyl Effect of initial acid Physical and chemical (Şahin et al. 2009) adipate, diethyl concentration, equilibrium were studied. sebacate 1-octanol extractant TDDA suggested to be
34 | Page and heptane concentration and the best extractant effect of diluents Itaconic TBP and Aliquat Sunflower oil Initial acid Non-toxic system (Wasewar, KL, 336 concentration in the proposed Shende & Keshav aqueous phase and 2011) extractant concentration in organic phase Propionic TBP Kerosene and 1- Aqueous and organic Solvent modifier has (Kumar, S, Datta & decanol phase compositions strong effect on Babu 2011) and temperature extraction efficiency Caproic TBP MIBK and xylene Phase composition MIBK is a better solvent (Wasewar, KL & than xylene Shende 2011) Acetic TOA DCM, butyl acetate, Amine composition in Acid-amine complex (Caşcaval, heptanes, and 1- organic phase and structure depends on the Kloetzer & octanol pH of the aqueous solvent polarity Galaction 2011) phase Picolinic TBP Sunflower oil Initial acid and amine Different models were (Waghmare et al. concentration developed to represent 2011) equilibrium data Acetic, TBP Cyclohexane, Equilibrium time Extraction and stripping (Ren et al. 2011)
35 | Page propionic, sulfonated, kerosene temperature and conditions were found butyric and and 1-octanol phase ratio valeric Picolinic Tri-alkylamine Tetrachloromethane, Initial acid pH and alkalinity of (Zhang et al. 2012) (N235), TBP kerosene and 1- concentration and extractant effects the octanol amine concentration distribution coefficient in organic phase Citric TOA Rice bran oil, Initial acid Total extraction constants (Keshav, Norge & sunflower oil soybean concentration and and association number Wasewar 2012) oil and sesame oil amine concentration in organic phase Pyruvic TOA 1-octanol Initial acid Estimation of distribution (Marti, Gurkan & concentration and coefficient Doraiswamy 2011) amine concentration 4- N,n- n-decane, decanol, n- Acid concentration Developed a Oxopentanoi dioctyloctan-1- decane + decanol, and type of diluents mathematical model (Kumar, S et al. c amine toluene, MIBK, based on Mass Action 2015) dichloromethane Law to estimate the (DCM) overall equilibrium
constants (K11 and K21) Levulinic Aliquat 336 Dodecane, benzene, Initial acid, extractant DCM yielded the highest (Datta, D et al.
36 | Page 1-octanol, MIBK and concentration, diluent extraction efficiency 2016) DCM type among the diluents, minimum solvent to feed ratio was calculated to determine the number stages required to get more extraction efficiency
37 | Page 2.2 Kinetic studies literature
Wasewar et al. (2002b) conducted equilibrium and kinetic studies on the reactive extraction of lactic acid using Alamine 336 in 1-octanol and used the mass transfer coefficient to estimate reaction regime. They derived a relation between mass transfer coefficient and speed of agitation and showed that overall rate of extraction with respect to lactic acid and Alamine 336 was 1. The first order rate constant k1 was determined to be 0.21s-1 and the specific rate of extraction remained constant with agitation speed. Phase volume ratio was also studied to confirm the reaction regime taking place. As no change in the specific rate of reaction was observed, the regime was confirmed to be fall within regime 3, that is to say, that extraction was accompanied by a fast chemical reaction occurring in the diffusion film.
Poposka et al. (1998) reported on the kinetic mechanism and mathematical modeling of citric acid extraction with TOA in isodecanol. The kinetic data was presented as formal elementary kinetic models and as suggested mechanisms for the formation of
1:1 and 1:2 acid-amine complexes. They reported that both the kinetic models define the limiting case for equilibrium. They also showed that the mechanisms were useful in the interpretation of how the participation of individual complexes depends upon the composition of the extraction mixture and on the initial concentration of acid present in the aqueous phase.
Jun et al. (2005) studied the kinetics of succinic acid extraction from aqueous solution with TOA in 1-octanol solution. The studies were conducted by using a stirred cell with a microporous hydrophobic membrane. They stated that chemical reactions at the interface, or near to the interface, control the overall reaction process. They reported the order of the reaction was 1 with respect to the succinic
38 | Page acid in the aqueous phase and 0.5 with respect to TOA in the organic phase. The dissociation reaction of the succinic-TOA complex was found to be second order with respect to the succinic acid-TOA complex in the organic phase and -2 order with respect to TOA in the organic phase with a rate constant of 1.44±1.4 × 10 -4 mol m-2 s-1.
Jun et al. (2007) also studied the effects of fermentation broth conditions on the kinetics of succinic acid extraction using TOA in 1-octanol solution in a stirred cell with a microporous membrane. By correlating the interfacial concentration of solutes in different systems, they were able to explore the effects of salts, pH, and the presence of contaminated acid on the intrinsic reaction kinetics at play. They stated that to achieve a significant separation of succinic acid from the fermentation broth, the pH of the aqueous phase must be below the pKa value of the acid and that the extraction rate can be improved by removal of impurities such as salts and contaminated acid from the fermentation broth.
Datta and Kumar (2010) studied the kinetics of reactive extraction of nicotinic acid by
TOA dissolved in MIBK. They reported the mass transfer coefficient (kL) of nicotinic acid in MIBK as 2.03 × 10 -5 m s-1. The reaction was found to have orders of 0.8 with respect to nicotinic acid and 0.5 with respect to TOA. The rate constants of the forward and backwards reactions were reported as 3.19 × 10 -3 (kmol m-3)-0.3 s-1 and
8.6 × 10 -5 (kmol m-3)0.91 s-1. Similar to that reported by Wasewar et al. (2002a), the agitation speed was found to have no effect on the extraction rate; however, an increase in phase volume ratio did lead to an increase in the extraction rate.
39 | Page 2.3 Regeneration/Back-Extraction of carboxylic acids from the loaded-organic phase
The regeneration/back extraction of carboxylic acids from the loaded-organic phase is an important step in the recovery of carboxylic acids from fermentation broths after the reactive extraction process. This process is also necessary to recycle the organic phase (solvent + extractant) for reuse in the reactive extraction process. Other techniques, such as distillation, cannot be utilized for the regeneration of viscous carboxylic acids (di-carboxylic and hydroxyl carboxylic acids). There is limited literature available on the regeneration of carboxylic acids from the loaded organic phase obtained by reactive extraction. Tamada and King (1990) studied two approaches (temperature and diluent swing) for the regeneration of fumaric, malonic, maleic, succinic, lactic and acetic acids from an organic phase containing Alamine
336 to the aqueous phase by a back extraction process (Tamada & King 1990).
They found the equilibrium characteristics could be changed by the swing processes.
In the temperature swing process, the forward extraction is processed at a lower temperature than regular extractions. The loaded-organic phase is then contacted with a fresh aqueous phase at a higher temperature to separate the acid into the aqueous phase and to obtain an acid-free organic phase. In the diluent swing method, the organic phase composition is changed after the reactive extraction process. The altered organic phase is contacted with a fresh aqueous phase to regenerate the acid-free organic phase. They stated that the combination of the two approaches is leading the way to decrease the cost of the process and to obtain more favorable equilibria.
40 | Page Poole and King (1991) demonstrated the back extraction of lactic, succinic and fumaric acids from tertiary amine-acid complexes in the organic phase into a solution of volatile amines in the aqueous phase. The volatile amines extract acid from the loaded organic phase to the aqueous phase. The acid-amine complex in the aqueous phase is decomposed thermally to obtain the carboxylic acid final product, and the amine is vaporized, reabsorbed into the water and recycled for further back extractions. For this process, it is necessary to select a tertiary amine as a back extractant that is volatile and water soluble; ammonia or primary and secondary amines cannot be used for this purpose since they form stable complexes with carboxylic acids that don’t thermally decompose. Poole and King used TMA, a very volatile tertiary amine, as a back extractant. Pure TMA boils at 3 0C and thermally decomposes at 350ºC. They conducted equilibrium studies for the back extraction of acids from Almaine 336 + MIBK extractant into an aqueous solution of TMA and stated that for all acids examined, a complete recovery of acid was achieved when there is at least 1 mole of TMA for every mole of acid. The equilibrium concentration of TMA in the organic phase is less than 0.0005 wt% at this condition. If the stoichiometric ratio of TMA to acid is greater than 1, the equilibrium concentration of
TMA in the organic phase is much greater. Thermal decomposition experiments were performed by Poole and King (1991), and they stated that fumaric and succinic acids were obtained in crystalline form. However, for lactic acid, only 63% of the water and 62% of the TMA could be removed, due to the water solubility and tendency for self-association of lactic acid.
Keshav and Wasewar (2010) investigated the back extraction of propionic acid from the loaded-organic phase using various techniques like the temperature swing, diluent swing, NaOH and TMA methods and found that the TMA method yielded the
41 | Page highest regeneration efficiency. They also studied the kinetics of propionic acid back extraction using TMA and reported that the reaction occurred in a fast regime and there was no change in the specific rate of extraction with a change in agitation rate or phase volume ratio.
2.4 Conclusions
This chapter has explained the literature available in an effort to place this study within the intellectual tradition of an existing body of work. It is disclosed that there is a lack of literature available on regeneration/ back extraction, multi-acids extraction, and optimisation of reactive extraction studies. This study will, therefore, employ a forward extraction (reactive extraction), regeneration of carboxylic acids from loaded- organic phase and multi-acids (volatile fatty acids) recovery through reactive extraction. Optimisations studies of process variables were conducted using Taguchi multi variate approach and response surface method.
42 | Page Chapter 3 Recovery of Levulinic acid by Reactive Extraction using Tri-n-Octylamine in Methyl Isobutyl Ketone: Equilibrium, Kinetics and Thermodynamics Studies and Optimisation using Taguchi Multivariate approach
43 | Page 3.1 Introduction
Among the various fermentation products that are available nowadays, carboxylic acids are some of the most important and useful products (Kertes & King 1986).
Levulinic acid or 4-oxopentanoic acid is an organic acid containing a keto group. It is a brown colored semi-solid at room temperature and is soluble in water and polar organic solvents like chloroform, ethers, and alcohols. It is used as an acid agent in beverages and foods and a precursor to biofuels and rubbers. Levulinic acid is also used in the manufacture of perfumes, foods, fuels and printing inks (Kumar, TP et al.
2010). It can be produced using an acid catalysed chemical processes such as dehydration of glucose, which makes the production of levulinic acid, which is a thermodynamically stable molecule, attractive because a number of low-cost lignocellulosic feeds can be utilised (Serrano-Ruiz, West & Dumesic 2010).
Carboxylic acids are generally available in low concentrations in fermentation waste.
Several separation processes like stripping, adsorption, electrodialysis, direct distillation, solvent extraction, evaporation, chromatography, ultra filtration, reverse osmosis, and drying are used for the recovery of carboxylic acids, but each of these processes has its own advantages and disadvantages. Recent research in carboxylic acid recovery is targeted towards the selection of a separation process which takes less time and requires less energy. The process should also generate less waste. An intensified process such as reactive extraction meets most of these requirements.
The main advantage of using reactive extraction for the recovery of carboxylic acids is that it offers significant improvement in both the reaction and separation stages (Alter
& Blumberg 1981).
44 | Page Reactive extraction is a combination of chemical and physical phenomena leading to a higher distribution coefficient for the solute. Therefore, it is an attractive technique for the separation of valuable acids (Wardell & King 1978; Wennersten 1983). It involves the use of extractants, which are usually highly viscous, solubilised in diluents thereby improving their physical properties like viscosity and surface tension.
Extractants generally used in the separation of carboxylic acids are carbon bonded oxygen compounds, phosphorus compounds, and aliphatic amine compounds.
Among them, long carbon chain aliphatic primary amines, secondary amines, and tertiary amines are generally preferred. The tertiary amines extractants with long chain carbon tails have benefits over the other extractants because they are cheaper and have higher distribution coefficient values (Ricker, Pittman & King 1980).
In the reactive extraction process, the extractant present in the solvent phase reacts with the target solute in the aqueous phase leading to the formation of a solute- extractant complex which gets solubilised in the solvent phase. The diluents may be inactive or active. Active diluents include inert aliphatic hydrocarbons like vegetable oils (Keshav, Norge & Wasewar 2012; Pal & Keshav 2016), kerosene (Pal et al.
2015) and active functional groups. A majority of the literature on reactive extraction shows that the active diluents are more effective than inactive diluents because they lead to high extraction efficiency when mixed with the extractant (Bízek, Horáček &
Koušová 1993; Biźek et al. 1992; Heyberger, Procházka & Volaufova 1998; Pal &
Keshav 2014). A previous study has shown that the separation efficiency is influenced by several factors such as the nature of the target solute, the target solute concentration, extractant concentration and the type of solvent (diluent) used
(Tamada & King 1990). The active diluents are usually classified into three classes: diluents containing chlorine atoms (e.g., methylene chloride), carbon bonded oxygen
45 | Page donor diluents (e.g., MIBK) and the oxygen donor phosphorus-bonded diluents (e.g.,
TBP) (Han & Hong 1996).
In recent years, several studies carried out on the separation of organic acids from fermentation waste involved reactive extraction because the extractant used in the process can be recovered easily and recycled. Since reactive extraction is a cheap and simple process, it has a lot of scope to be used as an efficient separation process. However, the main challenge in the use of reactive extraction for the recovery of acids is the identification of an efficient and cheap extractant
There are many papers in the literature that study the extraction of various acids using reactive extraction. Baniel et al.(Baniel, Blumberg & Hajdu 1981) discussed the recovery of carboxylic acids from fermentation broth using an extractant that comprises an organic solvent with at least one secondary or tertiary amine dissolved in it. Other systems studied include the extraction of tartaric acid using tri-iso- octylamine, lactic acid using Alamine 336 in decanol, succinic acid using long chain tertiary amines in 1-octanol, citric acid & levulinic acid using tri-propylamine, and levulinic acid from aqueous solutions using TOA as an extractant and esters as diluents (Baniel, Blumberg & Hajdu 1981; Hong, YK & Hong, WH 2000; Poposka et al. 2000; Rani et al. 2010; Uslu 2009; Wasewar, KL et al. 2002a). Separation of levulinic acid from aqueous solutions has been discussed in several papers in which researchers have proposed mass action law and linear solvation models for equilibrium studies based on their experimental work (Eda, Sumalatha, Kota, et al.
2017; Eda, S et al. 2016). However, a systematic study employing a statistical experimental design technique is lacking in the literature on the recovery of levulinic acid from an aqueous solution using reactive extraction. The novelty of this study is
46 | Page the application of a multivariate Taguchi approach to optimise the process variables used in the reactive extraction of levulinic acid from an aqueous solution.
The present work focuses on the effective recovery of levulinic acid from aqueous solutions using TOA as the extractant and MIBK as the diluent. Levulinic acid concentration, extractant concentration, and temperature are found to be the major factors affecting the reactive extraction process. Hence, these parameters are
1 2 optimised employing Taguchi 18 (6 3 ) mixed level design to maximise the distribution coefficient ( ) and extraction efficiency (% ). The kinetics of the reactive extraction was also determined using the optimal parameters
3.2 Materials and Methods
3.2.1 Chemicals
TOA (98% purity), a tertiary amine, supplied by Sigma Aldrich chemicals was used as an extractant. MIBK (99% purity) supplied by S. D. Fine-Chemicals Ltd., Mumbai,
India was used as a diluent. Levulinic acid (LAH) (98%) supplied by Sigma-Aldrich, and ultra-pure water from a Millipore purification system were used to prepare the aqueous solutions of levulinic acid. Acetic acid, which is supplied by S. D. Fine-
Chemicals Ltd., Mumbai, India, was used as a solute in kerosene to determine the mass transfer coefficient. Reagent grade sodium hydroxide supplied by S. D. Fine-
Chemicals Ltd. Mumbai, India was used in the preparation of NaOH solution, which was used in the titration against the aqueous phase. Phenolphthalein indicator (pH
8-10) was supplied by Thermo Fisher Scientific India Pvt Ltd. The properties of all chemicals used in this work are shown in Table 3.1.
47 | Page Table 3.1Properties of chemicals
Name M. W. (g mol-1) S.G. B.P. (0C) M.S.
Levulinic acid 116 1.14 245-246
MIBK 100.16 0.802 117-118
TOA 353.67 0.809 365-367
Acetic acid 60.05 1.05 118.1 CH3COOH
M. W= Molecular weight, S. G = Specific gravity, B.P = boiling point, M.S= Molecular structure, MIBK=Methyl isobutyl ketone,
TOA= Tri-n-octylamine
3.2.2 Methods
3.2.2.1 Equilibrium studies
The equilibrium studies were conducted at different temperatures varying from 293 to 333 K using TOA concentrations varying from 0.229 to 0.678 kmol m-3. The initial concentration of levulinic acid in the aqueous phase was varied from 0.1 to 1.0 kmol m-3, which is the typical concentration range of carboxylic acids usually found in fermentation broths.
The physical equilibrium (Baniel, Blumberg & Hajdu 1981)was determined as follows: A 25 mL aqueous solution containing 0.1 kmol m-3 levulinic acid was added with an equal volume of MIBK (organic phase) at a constant temperature and mixed using a shaking incubator (Daihan Labtech India Pvt Ltd., New Delhi) operating at
200 rpm for 5 h. The two-phase mixture was allowed to settle for 3 h at a constant temperature to obtain two liquid layers as products. The aqueous phase was separated from the organic phase by decantation and titrated against fresh 0.1 N
48 | Page NaOH solution. Phenolphthalein was used as the indicator to find the concentration of levulinic acid in the aqueous phase. The concentration of the extracted levulinic acid in the solvent phase was calculated by mass balance
The chemical equilibrium study was conducted by mixing 25 ml aqueous solution containing 0.1 kmol m-3 of levulinic acid with 25 ml solution of the solvent (organic) phase containing extractant (TOA) in MIBK. The immiscible mixture was kept at a constant temperature in a shaking incubator that was operating at 200 rpm for 5 h.
The resulting two-phase mixture was allowed to stand for 3h at the same temperature. The resulting two liquid phases were then separated. The aqueous phase was titrated against fresh 0.1 N NaOH solution using phenolphthalein as the indicator to determine the levulinic acid concentration in the aqueous phase.
3.2.2.2 Taguchi robust design with mixed levels ( ) for optimization
Taguchi design is an optimisation technique in the design of experiments. It was proposed by Genuchi Taguchi to improve the performance quality and trustworthiness in identifying optimum process variables using the least number of trials that provides complete information about the effect of all the factors on the performance parameters (Huang, H et al. 2009; Zolgharnein, Asanjarani &
Shariatmanesh 2013). This method involves lowering the variation in a process using a robust design of experiments. It contains orthogonal arrays, i.e., the design is balanced such that all the factors and levels weigh equally. In this method, each factor can be evaluated independently, and the effect of one factor does not influence the performance of another factor. This method is best suited for an intermediate number of variables (3 to 50) when interactions between variables are fewer, and only some variables contribute significantly to response parameters.
49 | Page A special feature of Taguchi method is the multivariate and mixed-level design that assigns factors at multiple levels for a particular design. It involves a class of experimental designs consisting of a small number of orthogonal arrays (combination of control and noise factors) chosen based on the number of variables and levels.
This approach recommends transformation of collected data into a signal-to-noise (
/ ) ratio, which represents the variations of a performance parameter that helps to measure the sensitivity of the extent of deviation from the measured values
(Daneshvar et al. 2007; Elizalde-González & García-Díaz 2010).
1 2 In this study, Taguchi orthogonal array 18 (6 3 ) design consisting of 3 factors having mixed levels (one factor at six levels and two factors at three levels) was employed to examine the influence of process parameters namely levulinic acid concentration, temperature, and extractant concentration on the distribution coefficient ( ) and extraction efficiency (% ). Six levulinic acid concentrations (0.1,
0.3, 0.5, 0.7, 0.9 and 1.0 kmol m-3) were tested in the temperature range of 293-333
K, using extractant concentrations of 0.229, 0.458 and 0.678 kmol m-3. MINITAB17
Statistical software (free-trial) was used for generating the experimental design matrix involving the chosen factors at their respective levels as shown in Table 3.2.
Distribution coefficient ( )% and extraction efficiency ( ) were chosen as the response parameters. Since the target is to achieve high values for the distribution coefficient ( ) and extraction efficiency (% ), the larger-is-better criterion is opted for / ratio which is expressed as:
1 1 = ‒ 10 10 [ ∑ 2 ] (3.1) =1
Where is the number of runs and is the response parameter. The unit of / ratio is decibel (dB), which is often used in communication engineering. The /
50 | Page ratio values calculated using equation (3.1) for individual responses is reported in
Table 3.3.
Table 3.2 18 Factors and levels by Taguchi mixed design
Levels
Factors 123 456
-3 Levulinic acid ( 1 ) (kmol m ) 0.1 0.3 0.5 0.7 0.9 1.0
-3 TOA ( 2 ) (kmol m ) 0.229 0.458 0.678 ------
Temp ( 3 ) (K) 293 303 333 ------
3.2.2.2.1 Desirability approach
Desirability approach, which verifies the optimum performance of multiple responses, is an established and useful technique for the simultaneous optimisation of process variables. In this approach, each response is converted into desirability function that varies from 0 to 1 (lowest to highest). The multivariate optimisation was carried out to identify the combination of factors that maximises the overall desirability of the process. In this study, the desirability function for two response parameters ( and
% ) was calculated using the following equations for satisfying the larger-the-better
(LTB) criterion.
If ≤ =0
‒ If ≤ ≤ = (3.2) ( ‒ )
If ≥ = 1
51 | Page where indicates the response parameters, and are the lower and upper
ℎ tolerance limits of response parameters, respectively, is the desirability of response and is the desirability index. In a multi-response condition, the ideal case is achieved if the desirability of all responses equals one, and the case is unacceptable if the desirability of response reaches 0. Therefore, the overall desirability is computed as the geometrical mean of all the individual desirability ( ) according to equation (3.3) :
1 =( 1 2….. ) (3.3)
Where is the overall desirability and is the total number of responses. MINITAB
17 was used for the design of experiments, analysis of variance analysis (ANOVA) and optimisation of the extraction process. ANOVA was performed to study systematically the relative importance of factors, which indicate the contribution of each factor on responses relative to the total contribution.
Table 3.3 18 Mixed array (three factors at different levels and results)
Exp. No 1 2 3 / ratio (db) % / ratio (db)
1 1 1 1 4.69 13.42 82.44 38.32 0.55
2 1 2 2 10.9 20.74 91.60 39.23 0.95
3 1 3 3 4.36 12.78 81.36 38.20 0.52
4 2 1 1 2.27 7.12 69.46 36.83 0.30
5 2 2 2 6.70 16.52 87.02 38.79 0.70
6 2 3 3 3.32 10.42 76.84 37.71 0.42
7 3 1 2 1.65 4.34 62.29 35.88 0.21
8 3 2 3 2.39 7.56 70.53 36.96 0.31
52 | Page 9 3 3 1 11.83 21.45 92.21 39.29 1
10 4 1 3 0.80 -1.93 44.49 32.96 0.05
11 4 2 1 6.33 16.02 86.37 38.72 0.67
12 4 3 2 4.99 13.96 83.30 38.41 0.57
13 5 1 2 0.92 -0.72 47.84 33.59 0.08
14 5 2 3 1.64 4.29 62.08 35.85 0.21
15 5 3 1 5.72 15.14 85.13 38.60 0.63
16 6 1 3 0.56 -5.03 35.72 31.05 0
17 6 2 1 5.18 14.28 83.81 38.46 0.59
18 6 3 2 3.28 10.31 76.61 37.68 0.41
3.2.2.3 Kinetic studies
The kinetic experiments were carried out in a 500 ml jacketed glass reactor equipped with an overhead stirrer. The reactor was a cylindrical glass vessel of 0.1 m diameter and 0.13 m height with round bottom. The stirrer was a paddle impeller with four blades whose speed was controlled by a variable speed drive. The reactor vessel consisted of four necks with glass stoppers to facilitate the periodic withdrawal of liquid samples during the reaction. A water bath with a digital temperature controller was used for maintaining the temperature of the reactor within the range of ± 0.01oC of the set temperature by circulating water through the jacket.
3.2.2.3.1 Specific rate of extraction (RA) with different initial succinic acid and
TOA concentrations
To determine the specific rate of extraction, the reactor was initially charged with 150 ml of aqueous solution of levulinic acid and 150 ml of organic phase consisting of
53 | Page TOA in MIBK. The agitator speed was maintained at 1.16 rps (70 rpm) during the reaction. Samples of the aqueous phase were collected at regular intervals to find the concentration of acid transferred from aqueous phase to organic phase. The acid concentration in the aqueous phase was determined by titrating it against fresh 0.1 N
NaOH solutions whereas that in the organic phase was determined by mass balance. A selected few experiments were repeated to determine the reproducibility of the results. A similar procedure was followed for various levulinic acid and TOA concentrations at different impeller speeds. Only the initial rates of reaction were considered in the evaluation of reaction kinetics to avoid the problems due to reaction reversibility. The initial rate was determined from the linear plot of levulinic acid concentration in the aqueous phase ( ) versus time at time =0 as per equation (3.4),
1 = ‒ │t=0 (3.4)
-1 where is the specific rate of extraction, (m ) is the specific interfacial area.
3.3 Theory
3.3.1 Mass action law model
Guldberg and Waage proposed mass action law model in 1864 to analyse equilibrium data (Ferner & Aronson 2016). Kertes and King (1986) used this law for interpreting data for the separation of carboxylic acids from aqueous solutions by the reactive extraction process. In this law, component concentrations are assumed to be proportional to the activity of the components, and the proportionality constant is considered as an equilibrium constant in developing a mathematical model for
54 | Page representing the equilibrium of a reactive extraction process. In this work, mass action law was used for two categories. They are 1) physical extraction, in which only the diluent is used to extract acid from the aqueous phase, 2) chemical extraction, in which both diluent and extractant are used to extract acid from the aqueous phase into an organic phase.
3.3.1.1 Physical extraction
In physical extraction, the levulinic acid is extracted using only diluent, i.e., MIBK.
This process involves the dissociation of levulinic acid in the aqueous phase and the partition of un-dissociated levulinic acid between the two phases followed by the dimerization of un-dissociated acid in the solvent (organic) phase.
Partial dissociation of levulinic acid (LAH) relies on the strength of the acid in the aqueous phase, and it can be represented as:
‒ + [LAH]aq ↔LA +H (3.5)
The dissociation constant for this process can be written as,
[ ‒ ][ + ] = (3.6) [ ]
‒ + where [ ] , [ ] [ ] are the cation, anion forms of levulinic acid and concentration of levulinic acid in the aqueous phase, respectively.
The total concentration of levulinic acid in the aqueous phase and un- dissociated acid [ ] can be related as:
‒ =[ ]+[ ] (3.7)
From equations (3.6) and (3.7), the following equation for [ ] can be written:
55 | Page [ ]= (3.8) 1+ ( [ + ])
Un-dissociated levulinic acid partition between the aqueous (aq) and organic (org) phases is represented as
[LAH]aq↔[LAH]org (3.9)
The partition co-efficient (P) for this process can be written as,
[LAH]org P= (3.10) [LAH]aq
Where [LAH]org is the concentration of acid in the organic phase, [LAH]aq is the concentration of acid in the aqueous phase.
The un-dissociated acid in the solvent phase can be dimerised due to hydrogen bonding which is stronger than the solute-solvent interaction. Levulinic acid dimerisation form can be expressed as:
2[LAH]org↔[ ]2 (3.11)
The dimerisation co-efficient(D ) for this process can be expressed as:
[LAH]2,org D= 2 (3.12) [LAH]org
where [LAH]2,org is the concentration of levulinic acid in the organic phase and
2 [LAH]orgis the concentration of levulinic acid in the dimerization form.
In physical extraction, the distribution coefficient ( ) of levulinic acid is expressed as
, = (3.13) ,
56 | Page Where , and , are the concentrations of the levulinic acid in the organic and aqueous phases, respectively.
The physical distribution coefficient can be expressed in terms of partition coefficient ( ) and dimerization coefficient ( ) using equation (3.14)
2 +2 [ ] = (3.14) 1+ + [ ]
For the dilute concentrations of acid, the second term in the denominator of equation
(3.14) can be neglected, and the equation can be simplified as
2 = +2 [ ] (3.15)
The degree of extraction (% ) of levulinic acid can be calculated based on the distribution coefficient ( ) using the equation (3.16).
% = ×100 (3.16) (1 + )
Where is the distribution coefficient in chemical equilibrium studies.
3.3.1.2 Chemical extraction
The chemical extraction is based on the interaction of levulinic acid with TOA. It can take place in two different ways. It can be described as 1) hydrogen bonding of un- dissociated acid molecule and 2) ion pair formation, and they can be expressed using the equations shown below
+ ↔ ‒ (3.17)
‒ + ‒ + + + ↔ (3.18)
57 | Page The chemical extraction mechanism is based on the pH of the levulinic acid solution, pKa of the levulinic acid, the concentrations of the levulinic acid and TOA, and the
TOA basicity with respect to levulinic acid. The chemical equilibrium process can be expressed by the following equation.
+ ↔( ) (3.19)
According to the equation (3.19), ‘m’ molecules of levulinic acid react with ‘n’ molecules of TOA to form m:n complex. The reaction proceeds to reach equilibrium with an equilibrium complexation constant of .
[ ] = (3.20) [ ] [ ]
where [ ] = , and [ ] = , .
A correlation that relates the loading ratio (z) and concentration of levulinic acid in organic and aqueous phases can be derived as shown in equation (3.21).
, = (3.21) , 0
The loading ratio ( ) is defined as the ratio of levulinic acid concentration in the solvent phase and the initial concentration of extractant in the solvent phase at equilibrium. It relies on the bonding strength of the acid-extractant interaction, initial acid concentration in the aqueous phase and stoichiometry of the extraction process.
It has been noted that when the loading ratio is low ( <0.5), levulinic acid concentration in the solvent phase is very low, leading to the formation of 1:1 acid-
extractant complex. The formation of 1:1 complex was confirmed by plotting (1 ‒ ) against [LAH] and determining the slope of the straight line (which is equal to the
58 | Page complexation constant, 11 ) that passed through origin according to the following equation:
(1 ‒ ) = 11[ ]
(3.22)
If the concentration of acid is high ( > 0.5), the equation (3.22) can be written as
( ‒ ) = 1[ ] (3.23) where m is the number of levulinic acid molecules involved in the complex. The equation (3.23) represents the relationship between the loading ratio and m:1 equilibrium complexation coefficient ( 1 ). Equations (3.22) and (3.23) represent the relationship between the loading ratio and the concentration of the acid.
3.3.2 Specific interfacial area ‘a’
The liquid-liquid interfacial area ( ) is dependent on several factors such as interfacial tension, impeller speed, reactor and impeller diameter, liquid density and viscosity. The interfacial area can be determined as follows.
The average radius of droplets in liquid-liquid dispersion in a cylindrical reactor can be estimated using equation (3.24) (Eda, Sumalatha, Kota, et al. 2017; Starks
1999)
1 (2 3 ‒ 1)(n)Υ r = (3.24) d ρN2x2
59 | Page where n is the number of droplets formed during splitting, is the density of the liquid
(kg m-3), is the rotational speed of the stirrer (rps), is the distance of the shell of liquid from the center of the reactor andΥ is the interfacial tension.
For a cylindrical reactor with radius and height ℎ , operating at a stirrer speed of
(rev s-1), the volume of dispersed phase in a cylindrical shell can be estimated using equation (3.25),
dVd =2 (h) x.dx (3.25)
3 where is the volume of the dispersed liquid (m ), is the distance of the shell of liquid from the center of the reactor, is the thickness of the shell, and ∅ is the volume fraction of the dispersed phase in the continuous phase.
The total number of dispersed phase droplets in the cylindrical shell is calculated by dividing by the average droplet volume. When the total number of droplets is multiplied by the average interfacial area per droplet, we obtain equation (3.26) for the total interfacial area in the thin shell.
dV dV 3dV d d 2 d d(Area) = sdroplet = 4 4 rd = (3.26) vdroplet 3 rd 3πrd
3 where is the average droplet volume (m ) and is the average interfacial area per droplet (m2).
By substituting equations (3.24) and (3.25) for and respectively in equation
(3.26) and integrating the resulting equation between the limits = 0 and = , we get the following equation for the interfacial area.
Interfacial area =8.53ℎ 2 4 (3.27)
60 | Page By dividing the equation for the interfacial area (equation. 3.27) by the liquid volume (
), we get equation (3.28) for the specific interfacial area ( ) as, (Eda, Sumalatha,
Kota, et al. 2017).
8.53ℎ 2 4 = V (3.28)
3.4. Results and Discussion
3.4.1 Equilibrium Studies
3.4.1.1 Physical equilibrium
The physical equilibrium studies were carried out with levulinic acid and MIBK at various temperatures (293-333 K).The carboxylic acid interaction in both aqueous and organic phases in the physical extraction is explained above, and the mechanism is illustrated in Figure. 3.1. The physical extraction is studied in terms of partition ( ) and dimerization coefficients ( ) at different temperatures and various acid concentrations in the aqueous phase. The correlation for distribution coefficient
( ) values determined using equation (3.14) at different temperatures for various aqueous phase acid concentrations are shown in Table 3.4. The , and values determined at different temperatures are shown in Table 3.5. For a given
temperature, the is found to decrease with an increase in initial acid concentration. It varies from 0.16 to 0.56 as initial acid concentration varies from 1.0 to 0.1 kmol m-3 at 293 K. Similar results are observed for other acid concentrations too. These results are relevant to fermentation broths because carboxylic acid concentrations in them are generally low. Table 3.5 shows that and decrease
with increasing temperature which explains the decrease in value with
61 | Page increasing temperature. The decrease in and values with increasing temperature could be attributed to increased acid solubility in the aqueous phase as temperature increases.
Figure 3.1 Physical extraction mechanism
Table 3.4 Effect of temperature on for the extraction of levulinic acid using
MIBK
,0
kmol m-3 293 303 313 323 333 K
0.1 0.56 0.51 0.48 0.41 0.36
0.2 0.53 0.48 0.42 0.36 0.31
0.3 0.51 0.43 0.37 0.30 0.25
0.4 0.46 0.38 0.33 0.28 0.22
0.5 0.42 0.35 0.29 0.24 0.18
0.6 0.34 0.28 0.23 0.19 0.15
0.7 0.31 0.24 0.20 0.16 0.13
0.8 0.28 0.19 0.17 0.14 0.11
62 | Page 0.9 0.21 0.16 0.13 0.10 0.07
1.0 0.16 0.11 0.08 0.05 0.03
Table 3.5 Partition ( ) and dimerisation coefficient ( ) values for the extraction of levulinic acid using MIBK at different temperatures
Temp (K)
293 0.28 0.051
303 0.23 0.043
313 0.18 0.031
323 0.14 0.013
333 0.11 0.010
3.4.1.2 Chemical extraction
Physical extraction of levulinic acid with MIBK was found to be not suitable to recover the acid in this work. The lower acid strength (pKa = 4.78) and dipole moment (3.148 Debye) of levulinic acid may be the reasons for the lower distribution
coefficient ( ) values observed in this process. The lower values of distribution coefficients in physical extraction indicate that extractants need to be used to improve the extraction efficiency. In this work, tri-n-octylamine (TOA) was used as the extractant at different concentrations (0.229, 0.458 and 0.678 kmol m-3) in MIBK
(diluent). The KD values for different initial acid concentrations and temperatures are given in Table 3.7. The chemical extraction involving TOA and MIBK leads to distribution coefficient ( ) values that are higher than those obtained for physical extraction (Table 3.7). In this process, there are significant improvements in and
63 | Page % values for different concentrations of TOA. The and % values vary from
4.69 to 0.82 and 82.44 to 45.11%, respectively as levulinic acid concentration in the aqueous phase varies from 0.1 to 1.0 kmol m-3 at 293 K for TOA concentration of
-3 0.229 kmol m . For a constant levulinic acid concentration, both and % values increase with increasing TOA concentration. The and % values increase from
2.27 to 21.72 and from 69.4 to 95.60 %, respectively as TOA concentration
-3 increases from 0.229 to 0.678 kmol m at 293 K. The highest values of and % are observed for TOA concentration of 0.678 kmol m-3 at 293 K. The effect of
-3 temperature on and % for TOA concentration of 0.229 kmol m in MIBK is shown in Figure 3.3. As the temperature increases, both and % values decrease due to the disruption of the acid-amine interaction with increasing thermal energy.
Figure 3.2 Chemical extraction mechanisms at equilibrium
64 | Page 80 3 %E 70 KD 2.5 60 2 50
40 1.5 KD % E
30 1 20 0.5 10
0 0 293 303 313 323 333 Temp K
Figure 3.3 Influence of temperature on % and for TOA concentration of 0.229 kmol m-3 at 293 K, MIBK used as the diluent.
3.4.1.3 Loading Ratio ( )
The loading ratio ( ) is defined as the ratio of acid concentration in the organic phase to amine concentration in the organic phase. It depends on the strength of acid-base interaction and the concentration of acid in the aqueous phase. The overall extraction process stoichiometry depends on the loading ratio. The equilibrium concentrations of levulinic acid in the aqueous phase for different concentrations of
TOA in MIBK are shown in Table 3.6. It can be noted that the loading ratio decreases with increasing concentrations of TOA. The change in loading ratio with
TOA concentration happens only when complexes involving more acid molecules are formed at lower concentrations of TOA. Since there is a decrease in loading with increasing TOA concentration, it can be suggested that only complexes involving single amines are formed. The 2:1 complex is formed if the organic phase is highly
65 | Page loaded usually at low TOA concentration and high acid concentrations and when
equation (3.23) is satisfied. To confirm the formation of 2:1 complex, the values of
∗ 2 -3 (2 ‒ ) are plotted against , in Figure 3.4 for TOA concentration of 0.229 kmol m
in MIBK for different temperatures. The equilibrium complexation constant values
can be obtained as slopes of straight lines. The slopes ( 21 ) of the straight lines are
found to be 89.0, 61.58, 43.0, 23.0 and 15.54 at temperatures 293, 303, 313, 323,
and 333 K, respectively. Based on these results, it can be concluded that levulinic
acid forms a 2:1 complex with TOA at high concentrations. It can be also suggested
that chemical extraction is better than the physical extraction because it leads to
higher KD and % values. Similar results have been reported by previous studies
(Uslu & Kırbaslar 2008; Uslu & Kirbaşlar 2008).
Table 3.6 Variation of loading ratio ( ) with changing TOA concentration 293 K
TOA concentration (kmol m-3)
0.229 0.458 0.678
∗ - ∗ ∗ ∗ , 0(kmol m , , ,
3)
0.1 0.018 0.36 0.004 0.21 0.002 0.145
0.2 0.058 0.62 0.01 0.415 0.007 0.285
0.3 0.092 0.91 0.021 0.61 0.013 0.423
0.4 0.130 1.18 0.034 0.80 0.020 0.560
0.5 0.170 1.44 0.051 0.98 0.039 0.68
0.6 0.245 1.55 0.078 1.14 0.058 0.80
0.7 0.283 1.82 0.095 1.32 0.083 0.91
0.8 0.367 1.89 0.113 1.50 0.108 1.02
66 | Page 0.9 0.453 1.95 0.135 1.67 0.134 1.13
1.0 0.549 1.97 0.162 1.83 0.153 1.25
3 293 K 303 2.5 K 313 K 2 323 K
1.5 z/(2-z)
1
0.5
0 0 0.02 0.04 0.06 0.08 C*2LAH, aq (kmol m-3)2
∗ 2 Figure 3.4 (2 ‒ ) versus , plots for the reactive extraction of levulinic acid using
0.229 kmol m-3 TOA in MIBK at different temperatures
3.4.1.3 Thermodynamic Studies
The three factors that affect % in the chemical extraction process are: (1) the strength of acid-amine association (2) the hydrophobicity of the solute, and (3) the steric effect between the solute and the extractant (Li, Qin & Dai 2002). The effect of temperature on the reactive extraction of levulinic acid can be seen from the results shown in Table 3.7. The reactive extraction of carboxylic acids by extractants occurs by intermolecular hydrogen bonding or ion exchange between the extractant group and the acid. The extraction of levulinic acid by acid-amine complexation is expected
67 | Page to be exothermic which would make the system higher ordered. This increase in order would therefore lead to a decrease in the entropy of the system. Furthermore, the system would become higher ordered if the interaction between the acid and the extractant is stronger.
The enthalpy and entropy of the reaction are assumed to be constant over the temperature range used in this work. The equilibrium complexation constant KE21 is related to temperature by equation (3.29) (Tamada, Kertes & King 1990).
‒∆ ∆ 21 = + (3.29) where is the enthalpy of reaction (kcal mol-1) and ΔS is the entropy of reaction
(cal mol-1 K-1). Based on equation 3.29, a graph of complexation coefficient
1 (ln(KE21)) vs 1/T was drawn. Equation (3.29) indicates that a plot of 21 versus should give a straight line as shown in Figure 3.5. The slope of the straight line is proportional to the enthalpy of reaction ( ), and its intercept is proportional to the entropy( ) , Gibbs free energy was then calculated using these values in equation
3.30 (Keshav, Wasewar & Chand 2008a). The more exothermic the reaction is, the more sensitive the equilibrium will be to the change in temperature. Table 3.7 shows the values of and obtained by fitting equation (3.29) to the results obtained in this work. Levulinic acid-amine complexation was found to be an exothermic reaction that leads to an increase in the order (decrease in entropy) of the system. The changes in Gibbs free energy (∆ ) calculated using equation are shown in Table 3.8 for different temperatures.
∆ =‒ 21 (3.30)
Table 3.8 Gibbs free energy at different temperatures.
Temp (K)∆ (kcal K-1 mol-1)
68 | Page 293 -2.60
303 -2.47
313 -2.33
323 -2.0
333 -1.80
69 | Page Table 3.7 Effects of temperature (293-333 K) on reactive extraction of levulinic acid using TOA in MIBK.
Temp LAH TOA (kmol m-3) 21 ∆ ∆
(K) (kmol m-3) 0.229 0.458 0.678 (kcal mol-1) (cal mol-1 K-1)
% z % Z % z
0.1 4.69 82.44 0.36 25.18 96.18 0.21 58.17 98.31 0.145 -8.61 -20.36
0.2 2.45 70.99 0.62 19.14 95.04 0.415 28.54 96.62 0.285
0.3 2.27 69.46 0.91 13.55 93.13 0.61 21.72 95.60 0.423
0.4 2.08 67.56 1.18 10.90 91.60 0.80 18.69 94.92 0.56
0.5 1.94 65.95 1.44 8.77 89.77 0.98 11.83 92.21 0.68
293 0.6 1.45 59.16 1.55 6.70 87.02 1.14 9.42 90.40 0.80 89.02
0.7 1.47 59.54 1.82 6.33 86.37 1.32 7.43 88.14 0.91
0.8 1.18 54.10 1.89 6.08 85.88 1.50 6.38 86.45 1.02
0.9 0.98 49.62 1.95 5.66 84.98 1.67 5.72 85.13 1.13
1.0 0.82 45.11 1.97 5.18 83.81 1.83 5.56 84.75 1.25
70 | Page 0.1 3.52 77.86 0.34 10.90 91.60 0.20 18.69 94.92 0.14
0.2 2.08 67.56 0.59 8.35 89.31 0.39 10.81 91.53 0.27
0.3 1.98 66.41 0.87 6.70 87.02 0.57 9.42 90.40 0.40
0.4 1.79 64.12 1.12 5.55 84.73 0.74 8.84 89.84 0.53
0.5 1.65 62.29 1.36 5.01 83.36 0.91 7.43 88.14 0.65
303 0.6 1.36 57.63 1.51 4.46 81.68 1.07 6.08 85.88 0.76 61.58
0.7 1.30 56.60 1.73 3.80 79.17 1.21 4.99 83.30 0.86
0.8 1.10 52.38 1.83 3.40 77.29 1.35 4.13 80.51 0.95
0.9 0.92 47.84 1.88 3.05 75.32 1.48 3.62 78.35 1.04
1.0 0.80 44.43 1.94 2.88 74.20 1.62 3.28 76.61 1.13
0.1 1.98 66.41 0.29 6.70 87.02 0.19 7.43 88.14 0.13
0.2 1.70 62.98 0.55 5.55 84.73 0.37 5.56 84.75 0.25
0.3 1.62 61.83 0.81 4.69 82.44 0.54 4.36 81.36 0.36
0.4 1.51 60.11 1.05 4.04 80.15 0.70 3.92 79.67 0.47
0.5 1.42 58.62 1.28 3.71 78.78 0.86 3.40 77.29 0.57
313 0.6 1.20 54.58 1.43 3.37 77.10 1.01 2.93 74.58 0.66 43.04
71 | Page 0.7 1.14 53.32 1.63 3.15 75.90 1.16 2.53 71.67 0.74
0.8 0.98 49.52 1.73 3.00 75.00 1.31 2.28 69.50 0.82
0.9 0.83 45.29 1.78 2.54 71.75 1.41 2.11 67.80 0.90
1.0 0.71 41.45 1.81 2.39 70.53 1.54 1.92 65.77 0.97
0.1 1.34 57.25 0.25 4.69 82.44 0.18 5.56 84.75 0.125
0.2 1.22 54.96 0.48 4.04 80.15 0.35 4.36 81.36 0.24
0.3 1.15 53.43 0.7 3.52 77.86 0.51 3.78 79.10 0.35
0.4 1.09 52.10 0.91 3.09 75.57 0.66 3.21 76.28 0.45
0.5 1.07 51.75 1.13 2.74 73.28 0.80 2.93 74.58 0.55
323 0.6 0.98 49.62 1.3 2.54 71.75 0.94 2.77 73.45 0.65 23.00
0.7 0.90 47.44 1.45 2.26 69.35 1.06 2.41 70.71 0.73
0.8 0.80 44.37 1.55 2.08 67.56 1.18 2.19 68.65 0.81
0.9 0.69 40.71 1.60 1.87 65.14 1.28 1.78 64.03 0.85
1.0 0.63 38.47 1.68 1.72 63.20 1.38 1.52 60.34 0.89
0.1 1.02 50.38 0.22 3.52 77.86 0.17 4.36 81.36 0.12
0.2 0.97 49.24 0.43 3.09 75.57 0.33 3.54 77.97 0.23
72 | Page 0.3 0.96 48.85 0.64 2.74 73.28 0.48 3.32 76.84 0.34
0.4 0.91 47.52 0.83 2.59 72.14 0.63 2.93 74.58 0.44
0.5 0.88 46.72 1.02 2.39 70.53 0.77 2.73 73.22 0.54
333 0.6 0.85 45.80 1.20 2.19 68.70 0.90 2.47 71.19 0.63 15.54
0.7 0.80 44.49 1.36 1.95 66.08 1.01 2.20 68.77 0.71
0.8 0.72 41.79 1.46 1.79 64.12 1.12 1.95 66.11 0.78
0.9 0.62 38.17 1.50 1.64 62.08 1.22 1.78 64.03 0.85
1.0 0.56 35.72 1.56 1.53 60.46 1.32 1.61 61.70 0.91
73 | Page Furthermore, the increase in ∆ values with increasing temperature indicates that lower temperatures are favourable for the reactive extraction process. The overall effect of temperature depends on the parameters such as the acid pKa, the acid- amine interaction, the solubility of the acid in both phases, the extractant basicity, and water co-extraction. The parameters such as pKa, the acid-amine interaction, the solubility of the acid were studied also by Riki and Aharon (2004), who concluded that the pKa values of common carboxylic acids decrease only slightly with increasing temperature.
5.2
4.7
4.2
3.7 ln (KE21) 3.2
2.7
2.2 0.0025 0.003 0.0035 0.004 1/T (K-1)
Figure 3.5 Determination of apparent enthalpy and entropy of reaction for the extraction of levulinic acid with 0.458 kmol m-3 TOA in MIBK.
3.4.1.4 Optimization of reactive extraction process
In this study, Taguchi ( 18 ) mixed array design consisting mixed levels of factors was used (Table 3.2). Figures 3.6a and 3.6b show the effects of variables on response parameters. The experimental data were converted into / ratio (by adopting larger is better criterion) and analysed using Minitab 17 software. It can be seen that and % values decrease with increasing levulinic acid concentration in the aqueous
74 | Page phase as well as with increasing TOA concentration. There is a noticeable difference in the change of and % values as the TOA concentration changes from 0.229 to
0.458 kmol m-3 compared to those when concentration changes from 0.458 to 0.678
-3 kmol m . Also, both and % values decrease significantly as temperature increases from 293 to 333 K. From the ranks obtained for each factor, it can be inferred that the extractant concentration ( 2 ) has the highest effect followed by temperature ( 3 ) and levulinic acid concentration ( 1 ). The optimum conditions obtained for both the responses from the response table are 1st factor at level 1, 2nd factor at level 3, and 3rd factor at level 1.
Analysis of variance was studied to find the individual significance of each factor on response parameters. It can be observed from Table 3.10 that the extractant concentration ( 2 ) has higher sum of squares value followed by temperature ( 3 ) and levulinic acid concentration ( 1 ). It implies that extractant concentration is the factor that has the greatest effect on and % compared to the other two. The p- values in the ANOVA table indicate the significance of each factor. These values depict that, except acid concentration ( 1 ), the other two factors has high significance (p<0.05) for . In the case of extraction efficiency (% ), all factors are highly significant (p<0.05). Figures 3.7a and 3.7b exhibit the percentage contribution of individual factor to the response parameters. The percentage contribution for both the responses is high for extractant concentration ( 2 ) (40.09 % for and 50.50 % for (% ) followed by temperature (3 ) and levulinic acid concentration ( 1 ).
The desirability (di) of individual response was calculated for each response parameter to determine the overall desirability (Table 3.3). The and values indicate the maximum and minimum values of the response parameters chosen from
75 | Page th 18 experimental data. L16 experiment (16 experiment-6, 1 and 3) gives low values for the response parameters (0.56 of and 35.72 of (% ). The values of response parameters are those found in experiment 9 (11.83 of and 92.21 of % ).
Desirability index (r) is taken as 1 and it is assumed to approach the target value.
The highest value of desirability appeared in 9 experiment at different levels (3, 3 and 1) of variables. The lowest value (i.e zero) of desirability appeared in experiment
16 in the combination of variables at 6, 1, 3 levels indicating that / ratio cannot be calculated for this experiment. Therefore, the overall desirability, which is calculated by the means desirability, is chosen to find out the optimum conditions (Akhtar,
Akhtar & Khalid 2008; Eda, Sumalatha, Kumari, et al. 2017). From Tables 3.8 and
-3 Table 3.3, the optimum response parameters are predicted as 1 = 0.3 kmol m , 2 =
-3 0.678 kmol m and 3 = 293 K and the response parameters are determined as
12.78 for and 94 % for % . To confirm the validity of predicted values, experiments were conducted using predicted values of 1 , 2 and 3 . The response parameters for these experiments are found to be 11.83 for and 92.2 % for % as shown in Table 3.11 These values are closer to the predicted values indicating that the model reliably represents the investigated system. From statistical analysis, the maximum distribution coefficient ( ) and extraction efficiency (% ) of levulinic acid are found for TOA concentration of 0.678 kmol m-3 at 293 K.
76 | Page Main Effects Plot for SN ratios Main Effects Plot for SN ratios Data Means Data Means
A B C A B C 16 39
14 38 s s o o
i 12 i t t a a r r N 10 N S S 37 f f o o n
8 n a a e e M M 36 6
4 35 2 0.1 0.3 0.5 0.7 0.9 1.0 0.229 0.458 0.678 293 303 333 0.1 0.3 0.5 0.7 0.9 1.0 0.229 0.458 0.678 293 303 333
Signal-to-noise: Larger is better Signal-to-noise: Larger is better
(a) (b)
Figure 3.6 Main effect plots ( / ratios) for (a) distribution coefficient, (b) extraction efficiency, %
Table 3.9 Response table for signal-to-noise ratios ( / ; larger is better)
Factors
1 2 3
Levels % % %
1 15.65 38.59 2.86 34.78 14.57 38.37
2 11.35 37.78 13.24 38.01 10.86 37.27
3 11.12 37.38 14.017 38.32 4.684 35.46
4 9.35 36.70 ------
5 6.24 36.02 ------
6 6.52 35.74 ------
∆max-min 9.414 2.85 11.15 3.54 9.89 2.91
Rank ( ) 31 2
Rank (% ) 312
77 | Page Table 3.10 ANOVA for distribution coefficient ( ) and extraction efficiency (% ) in
18 array design
Factors SSa DOFb MSc Fd p-valuee PCf
For distribution coefficient ( )
1 40.56 5 8.09 1.57 0.299 23.16
2 70.2 2 35.09 6.73 0.025 40.09
3 57.41 2 28369 5.5 0.041 32.81
Error 6.92 8 5.5
Total 175.09 17
For extraction efficiency (% )
1 839.89 5 178.78 25.97 0 17.74
2 2391.04 2 1195.52 173.67 0 50.50
3 1393.95 2 696.98 101.25 0 29.45
Error 55.07 8 6.88
Total 4733.95 17 asum of squares; bdegrees of freedom; cmean square; dFischer’s ratio; eprobability of significance; fpercentage contribution
78 | Page 3.94 % X1 2.27 % X1 X2 X2 X3 23.16 % 17.74 % X3 other s 29.45 % others 32.81 %
40.09 % 50.54 %
(a) (b)
Figure 3.7 Percentage contributions of factors for (a) distribution coefficient ( ) (b) extraction efficiency (%E)
Table 3.11 Multivariate optimization of and % and confirmation of the
optimum condition
Confirmation run Predicted
Multiple optimization (Experimental values) values
-3 LAH concentration ( 1 ) (kmol m ) 0.3 0.3
-3 TOA concentration ( 2 )(kmol m ) 0.678 0.678
Temperature ( 3)(K) 293 293
Distribution concentration ( ) 11.83 12.78
Extraction Efficiency (% ) 92.2 94
Standard deviation 0.67 ( ) 1.22 (% )
79 | Page 3.4.2 Kinetic studies
In kinetic studies, experiments were conducted at optimum conditions obtained from
Taguchi analysis. The specific rate of extraction ( ) was obtained (equation 3.2) by plotting the concentration of acid ( ) in the aqueous phase as a function of time as shown in Figure 3.8. Since the equilibrium complexation reaction is reversible, only initial rates determined using the inclining section of the curve (Figure 3.8) was considered in the kinetics study to avoid errors due to reversibility.
0.4
0.35
0.3
0.25
0.2
0.15
CLAH, aq (kmol m-3) 0.1
0.05
0 0 20 40 60 80 Time (min)
Figure 3.8 Variation of acid concentration in the aqueous phase with time
3.4.2.1 Determination of individual mass transfer coefficients for water-levulinic acid-
MIBK-TOA system
To determine the reaction regime in the reactive extraction of levulinic acid from the aqueous phase by TOA-MIBK mixture, the interfacial mass transfer coefficient value is required. Therefore, an indirect approach was used to estimate the mass transfer
80 | Page coefficient values. In this approach, the mass transfer coefficient for a reference system (acetic acid-water-kerosene) was measured and used to estimate the mass transfer coefficient for the system used in this work (Juang & Lo 1994; Kiani, Bhave
& Sirkar 1984; Komasawa, Otake & YamadaA 1980; Kumar, TP et al. 2010). Since acetic acid has a sufficiently small value of distribution coefficient in a water/kerosene system, the resistance of aqueous diffusion film can be neglected.
The overall mass transfer coefficient for the acetic acid-water-kerosene system can be determined by measuring the initial and transient concentrations of acetic acid in the organic phase and using equation (3.30).
0 = kaat (3.30) ( )
0 where and are the concentrations of initial acetic acid and concentration of acetic acid in organic phase, respectively, is the interfacial area and is time.
0 A linear plot of the group in L.H.S ( ) of equation (3.30) against (a.t) will yield ( ) the value of the mass transfer coefficient as the slope of the straight line (Figure
3.9). Thus, the mass transfer coefficient, , for acetic acid-water-kerosene system is
-5 found as 1.08 × 10 . But the mass transfer co-efficient, can also be determined using the following relationship (Doraiswamy, & Sharma, 1984; Hanna & Noble
1985; Juang & Huang 1995; Jun, YS et al. 2005; Kiani, Bhave & Sirkar 1984; Prasad et al. 1986).
2 1 3 ‒ 6 α ɣ (3.31)
81 | Page where and ɣ are diffusivity of the solute and kinematic viscosity of solvent, respectively.
In the present study, the value for water-levulinic acid-MIBK-TOA system is estimated using equation (3.31) in the following form:
2 1 = 2 1 (3.32) ‒ ‒ 3 6 3 6 ( ɣ ) ( ɣ )
The diffusivities of levulinic acid in MIBK and TOA in MIBK were estimated using the
Minhas-Hayduk method (Akhtar, Akhtar & Khalid 2008; Juang & Huang 1995) in which diffusivity is proportional to the ratio of Parachor values of solvent to solute and inversely proportional to molar volume and viscosity of the solvent. The kinematic viscosities of kerosene and MIBK at 298 K are 1.52 x 10-6 and 7.23 x10-7 m2 s-1, respectively. The molar volumes of kerosene, MIBK, levulinic acid, TOA, and
2:1 acid-amine complex are 259.9, 190.82, 135, 555.9 and 825.9 cm3 mol-1, respectively. The Parachor values (which depends on the structure of the component) for MIBK, levulinic acid, TOA, 1:1 acid-amine complex and 2:1 acid- amine complex are 275.37, 257.9, 1024, 2143.22, and 2800.5 cm3 g1/4 s-1/2 , respectively (Kumar, TP et al. 2010).
The diffusion and mass transfer coefficient values estimated using Minhas-Hayduk equation and equation (3.32), respectively are shown in Table 3.12 for both acetic acid-water-kerosene and levulinic acid-TOA-MIBK systems.
3.4.2.2 Effect of speed of agitation on specific rate of extraction RA
The speed of agitation is one of the factors that can influence the reaction regime, and in the present study the criterion proposed by Doraiswamy, and Sharma (1984)
82 | Page was adopted to find the reaction regimes for levulinic acid and TOA. Mass transfer with chemical reaction in systems like stirred vessels is controlled either by diffusion or reaction kinetics. The extraction rate will increase with increasing stirrer speed if
0.16 0.14 0.12 0.1 0.08
ln(Ca0/Ca) 0.06 0.04 0.02 0 0 5000 10000 15000 a × t (m-1 sec)
Figure 3.9 Determination of mass transfer coefficient for acetic acid in kerosene/water
Table 3.12 Diffusion and mass transfer coefficients at 293 K
Diffusion coefficient Mass transfer coefficient System (Solute/medium) (m2 s-1) (m s-1)
Acetic acid/kerosene 7.68 x 10-10 1x 10-5
Levulinic acid/MIBK 2.10 x 10-9 2.2x 10-5
TOA/MIBK 1.18 x 10-9 1.48x 10-5
2:1 levulinic acid-TOA complex 5.13 x 10-10 0.85x10-5
mass transfer is diffusion-controlled but will remain unchanged if it is kinetics- controlled (Hanna & Noble 1985). In kinetics-controlled systems, the film adjacent to the interface will become thinner as the reaction progresses and therefore, the
83 | Page influence of diffusion will be lower compared to that of the chemical reaction. The speed of agitation in the present system was varied between 0.833 to 1.33 rev s-1.
The specific rate of extraction ( ) profile was plotted at levulinic acid concentration of 0.3 kmol m-3 and TOA concentration of 0.678 kmol m-3 in MIBK with different speeds of agitation (0.833, 1, 1.16, 1.33 rev s-1) as shown in Figure 3.10. The plot of the specific rate of extraction ( ) and the speed of agitation depicted in Figure 3.10 shows that there is no change in with increasing stirrer speed. It can be said that the reactive extraction of levulinic acid is occurring in regimes 1 or 3 among the 4 reaction regimes defined by Doraiswamy and Sharma (Doraiswamy, & Sharma
1984), which are very slow (regime 1), slow (regime 2), fast (regime 3), and instantaneous reaction regimes (regime 4).
0.000003
0.0000025
0.000002
0.0000015 RA (kmol m-2 s-1)
0.000001 0.8 1 1.2 1.4 Stirring speed (rev s-1)
Figure 3.10 Effect of speed of stirring on specific rate of extraction
3.4.2.3 Order of reaction with respect to initial levulinic acid concentration
The effect of initial levulinic acid concentration on the rate of extraction was studied to determine the order of reaction with respect to initial levulinic acid concentration.
For initial levulinic acid concentrations of 0.4 to 0.8 kmol m-3, the concentration
84 | Page profiles of levulinic acid in the aqueous phase for TOA concentration of 0.678 kmol m-3 in MIBK and 1 rev s-1 of agitation were obtained experimentally. The specific rate of extraction, , which were calculated from initial rates, are then plotted against
∗ as shown in Figure 3.11 and the order of reaction (m) with respect to levulinic acid is found to be 1 based on the straight line and rate constant is found from the slope of the straight line as 8.69 x10-6 s-1(Kumar, TP et al. 2010). Similarly the effect of TOA on was studied to determine the order of reaction with respect to TOA concentration in the organic phase. The concentration of TOA was varied from 0.229 kmol m-3 to 0.678 kmol m-3 while keeping the levulinic acid concentration as 0.3 kmol m-3 and speed of agitation as 1.16 rev s-1 and the order of reaction with respect to
TOA, n, is found to be one from the straight line in the plot of specific rate of
-6 extraction, versus TOA (Figure 3.12) and the rate constant is found as 6.74x10 from the slope intercept (Kumar, TP et al. 2010).
0.0000045
0.000004
0.0000035
0.000003 RA (kmol m-2 s-1)
0.0000025
0.000002 0.25 0.3 0.35 0.4 0.45 0.5 C*LAH (kmol m-3)
Figure 3.11 Effect of initial levulinic acid concentration on the specific rate of extraction at 0.678 kmol m-3 TOA, at 1.16rev s-1
85 | Page 5.50E-06 5.00E-06 4.50E-06 4.00E-06 3.50E-06 3.00E-06 2.50E-06
RA (kmol m-2 s-1) 2.00E-06 1.50E-06 1.00E-06 0.2 0.3 0.4 0.5 0.6 0.7 0.8 [TOA] (kmol m-3)
Figure 3.12 Effect of TOA concentrations on the specific rate of extraction at 0.3 kmol m-3 initial levulinic acid, 1.16 revs s-1 speed of agitation
3.4.2.4 Determination of rate constant, k2
The overall order of the reaction with respect to levulinic acid and TOA is second order. The second order rate constant 2 , was obtained by plotting the concentrations of levulinic acid and TOA versus rate of extraction ( ) in Figure
3 -1 -1 3.13 and the value of the rate constant 2 is found to be 0.01 m kmol s (Kumar, TP et al. 2010).
3.4.2.5 Criteria for reaction regime
To determine the reaction regime, the value of Hatta number MH, was evaluated. For
1/2 2 ∗ ‒1 ( +1 [ ] [ ] ) m=1 and n=1, the Hatta number, MH, becomes
1/2 ( 2[ ]) , which is found to vary from 10 to 17.34 for TOA concentrations
86 | Page 1.50E-03 1.40E-03 1.30E-03 1.20E-03 1.10E-03 1.00E-03
RAa (Kmol m-3 s-1) 9.00E-04 8.00E-04 7.00E-04 0.01 0.03 0.06 0.08 CLAH × [TOA]
Figure 3.13 Determination of second order reaction rate constant, 2 .
1/2 2 ∗ ‒1 ( +1 [ ] [ 0] ) range used in this work at 1.16 rev s-1 i.e. >>1. This shows that the reaction that is taking place in the film is fast or instantaneous
(Doraiswamy & Sharma 1984). To determine whether the reaction that is taking
[ 0] place in the organic film is instantaneous or fast, the value of ∗ calculated [ ] for 2:1 acid-amine complex using = 0.5 is found to vary from 2.98 to 10.44 for the range of concentrations of levulinic acid and TOA used in this work and this value is less than the Hatta number. This confirms that the reaction is an instantaneous reaction and is taking place in the organic film (Levenspiel 1999).
3.5. Conclusions
The present chapter shows that chemical extraction is more suitable than physical extraction to recover levulinic acid from aqueous solution. Low temperatures are found to be suitable for the reactive extraction process because the extraction efficiency decreases with increasing temperature. The highest extraction efficiency
87 | Page (% ) and distribution coefficient ( ) are achieved for a TOA concentration of 0.678 kmol m-3 at 293 K. The chemical equilibrium is achieved due to the formation of 2:1 complex as described by the mass action law model. The equilibrium complexation constants are estimated to be 89.0, 61.58, 43, 23 and 15.54 at 293, 303, 313, 323 and 333 K respectively. The enthalpy ( ) value for the extraction of levulinic acid is determined to be -8.61 kcal mol-1 and the corresponding entropy ( S) value is found to be -20.36 cal (mol-1 K-1). The reaction for levulinic acid-amine complex is found to be exothermic in nature. The values, estimated to find the feasibility of the system, increase with increasing temperature indicating that low temperatures are suitable for the recovery of levulinic acid through the reactive extraction process. The process variables are optimising using Taguchi mixed design multivariate approach,
1 2 -3 18 (6 3 ) array. It reveals that the optimum variables are 1 = 0.3 kmol m of
-3 levulinic acid, 2 = 0.678 kmol m of TOA and 3 = 293 K. Kinetic studies were performed at optimum conditions. To find the effect of stirrer speed on specific rate of extraction, the stirrer speed was varied from 0.8 to 1.33 rev s-1. There is no influence of stirrer speed on specific rate of extraction. The overall order of reaction is found to be 2 with respect to levulinic acid and TOA concentrations. The second order rate constant is determined to be 0.01 m3 kmol-1 s-1. Based on the Hatta number, the reaction regime is found to be instantaneously occurring in the organic film.
88 | Page Chapter 4
Recovery of Succinic acid by Reactive Extraction using Tri-n-octylamine in 1-Decanaol: Optimization of Equilibrium Studies using Response Surface Method and Kinetic Studies
89 | Page 4.1 Introduction
Succinic acid and its derivatives are widely used as polymers, foods, pharmaceuticals, and cosmetics. At present, commercial succinic acid is commonly produced from maleic anhydride derived from petroleum by a chemical process; however, due to the high conversion costs, succinic acid supply is limited for a wide range of applications. Production of succinic acid by fermentation includes product recovery, concentration, acidification, and purification. Anaerobiospirillum succiniciproducens (Jabalquinto et al. 2004; Lee, PC et al. 2003; Nghiem et al. 1997;
Podkovyrov & Zeikus 1993; Samuelov et al. 1999), Actinobacillus succinogenes
(Guettler, Rumler & Jain 1999; Kim et al. 2004; Park, D & Zeikus 1999; Urbance et al. 2004) and Mannheimia succiniciproducens (Lee, P et al. 2002) (isolated from bovine rumen) are facultative anaerobic bacteria currently considered as effective by succinic acid manufacturers due to the high demand for succinic acid.
Existing conventional processes for the recovery of organic acids from fermentation broths include stripping, adsorption, electrodialysis, direct distillation, solvent extraction, evaporation, chromatography, ultra filtration, reverse osmosis and drying, each process having its own advantages and disadvantages. Research is therefore targeted towards the selection of a unit process that is safer, cheaper, and more compact, consumes less energy and produces less waste. An intensified process such as reactive extraction aims in this direction. The advantage of adopting this approach is the significant improvement in both the reaction and separation.
Researchers have used reactive extraction to recover carboxylic acids from industrial waste streams in the chemical manufacture of cellulose and vinyl acetate. The process, a combination of both chemical and physical phenomena, leads to a higher
90 | Page distribution coefficient than other processes. Thus reactive extraction is a promising method for carboxylic acids recovery (Wardell & King 1978; Wennersten 1983).
Recovery of carboxylic acids from their aqueous solutions was first suggested by
Baniel et al.(Baniel, Blumberg & Hajdu 1981) via an extraction process with an immiscible water extractant of at least one tertiary or secondary amine dissolved in an organic solvent. Literature also cites tartaric acid extraction with tri-iso- octylamaine, lactic acid extraction using Alamine 336 in 1-decanol, succinic acid using long chain tertiary amines in 1-octanol, citric acid and levulinic acid using tri- propylamine (TPA) and levulinic acid with TOA dissolved in two ketones, five esters, and five alcohols (Hong, Y & Hong, W 2000; Jeong & Park 2009; Marti, Gurkan &
Doraiswamy 2011; Nikhade & Pangarkar 2005; Poposka et al. 2000; Rani et al.
2010; Uslu 2009; Wasewar, KL et al. 2002a).
Optimisation of the process parameters can be studied by statistical experimental design in order to avoid the limitations of the classical method (Demirkol et al. 2006).
Response surface methodology (RSM) is a statistical method for building models and designing experiments, investigating complex processes for the optimisation of target yield or value (Rashid et al. 2009). In this study, an RSM method containing a central composite design (CCD) was used to estimate the interactive effect and optimise the process parameters of the equilibrium studies of succinic acid recovery through reactive extraction. The initial succinic acid concentration (kmol m-3), amine composition (% v/v), and temperature (K) were selected as the critical independent variables, and the efficiency of extraction (% ) was selected as the dependent variable. An empirical model was developed to define the relationships and the interaction effects between the process parameters to obtain the maximum
91 | Page extraction efficiency. Kinetic studies were carried out using a stirred cell to find the reaction regime.
Minitab 17, statistical software was used for graphical analysis and regression of the experimental data. The quality of the model fit was evaluated using the coefficients of determination (R2) and analysis of variance (ANOVA). Response surfaces and surface plots were plotted by fitting a polynomial quadratic equation obtained from the regression analysis, keeping one of the independent variables at a constant value corresponding to the stationary point and changing the other two variables.
The stepwise procedure adopted is given below Figure 4.1.
Figure 4.1 Procedure chart for response surface methodology
92 | Page 4.2 Materials and Methods
4.2.1 Materials
TOA (98%) [CH3 (CH2)7]3N (density, 0.809 g/ml), a tertiary amine, supplied by
Sigma-Aldrich chemicals was used as an extractant, 1-decanol (99%) (density- 0.83 g ml-1) supplied by S.D.Fine-Chem Ltd. (Mumbai, India) was used as a diluent.
Phenolphthalein indicator (pH 8.2-9.8) was used in titration supplied by S.D. Fine-
Chem Ltd. (Mumbai, India). Succinic acid (98%) supplied by Sigma-Aldrich was of technical grade and, sodium hydroxide was used as a base in the titration of acid samples, was of reagent grade. All the chemicals were used in this study without further purification. Double distilled water was used for all experiments.
Equipment
Figure 4.2 Shaking incubator
Equilibrium studies were conducted in a shaking incubator (Model no: 4018R) as shown in Figure 4.2 equipped with a universal shaking platform, independent alarms and microprocessor for temperature and speed adjustment. The shaker provides
93 | Page orbital shaking motion, adjustable between 25 to 400 rpm. The shaking speed was set via the digital LED control panel. The shaking incubator has a versatile digital timer which can be set from 1 second to 9 days. The reaction temperature can be set from 288 to 353 K.
4.2.2 Methods
4.2.2.1 Equilibrium studies of reactive extraction
Equilibrium studies were carried out as follows: A 25 ml aqueous solution of 0.1 kmol m-3 succinic acid was prepared, and to this 25 ml solution of 10 % TOA (v/v) (0.229 kmol m-3) in 1-decanol (diluent) was added. The immiscible mixture was kept at a constant temperature (298-333K) in the shaking incubator for 6 h at 150 rpm; afterwards, the two phases were allowed to settle for 2h at the same temperature.
The aqueous phase was separated and then titrated with fresh 0.1 N NaOH solution using phenolphthalein as an indicator to evaluate the concentration of succinic acid still present in the aqueous phase. The same procedure was repeated for other TOA compositions.
The extraction efficiency (% ) was determined as a ratio of succinic acid concentration ( ) in the organic phase (TOA+1-decanol) at equilibrium to initial succinic acid concentration in the aqueous phase ( , , ) as given in equation
(4.1).
, % = ×100 (4.1) , ,
94 | Page 4.2.2.2 Response surface methodology (RSM) approach and design of experiments
RSM consists of three stages (i) selecting the parameters and designing the experiments, (ii) response surface modeling through regression analysis and (iii) optimisation of the response variable (Benyounis, Olabi & Hashmi 2005; Jaouachi,
Hassen & Sakli 2007). In order to optimise the process variables, there are different techniques available in RSM such as central composite design (CCD), Box Behnken statistical experimental design and full factorial designs. For statistical calculation, the following relationship has been established between coded value ( ) and real value ( ) as
‒ 0 = = , = 1, 2, 3,…. (4.2) ∆ where ∆ is the step change and 0 is the real value at the center point.
As the input parameters interact in a linear-to-linear and a linear-to-quadratic manner, taking all interactions into consideration, the system behavior can be described as
∑ ∑ 2∑ ∑ (4.3) % = 0 + =1 + =1 < + where i and j are linear, quadratic coefficients; n is the number of parameters (n=3),
β0 constant coefficient, linear effect or slope of input factor , is the effect of quadratic on input factor xi and βij is the linear interaction effect between input factors xj and xi , ε is the residual error (Wasewar, KL 2012).
4.2.3.3 Kinetic studies
The kinetic studies were conducted in a 500 ml glass jacketed reactor, fitted with a directly controlled digital overhead stirrer with a motor rating of 83/75 W, a max torque of 85 N cm. The diameter and height of the cylindrical stirred cell are 0.1m
95 | Page and 0.13m, respectively, with a round bottom and a paddle impeller with four blades.
The reactor has four necks with glass lids which can be used for periodic withdrawal of liquid samples during the course of the reaction. The agitation speed of the reactor stirrer was maintained using a controlled variable speed drive with a digital display. A
Poly-Science refrigerating and heating circulator (model 9702) with a digital temperature controller was used for maintaining the temperature (±0.01°C accuracy) in the reactor. A schematic diagram of the reactor assembly is shown in Figure 4.3
Figure 4.3 Schematic diagram of experimental setup: R-Jacketed reactor, F-Feed inlet, S-Stirrer, M-Motor, WS-Water source, WI-Water inlet, WO-Water outlet, M-
Motor, SP-Sampling point, TI-Temperature indicator
4.2.2.4 Specific rate of extraction ( ) with different initial acid concentrations and TOA concentration
The specific rate of extraction for different initial succinic acid concentrations was determined by adding 150 ml of aqueous solution containing succinic acid to 150 ml of the organic phase (1-decanol with 0.229 kmol m-3 TOA) in the stirred cell. The
96 | Page stirrer was operated at a fixed speed of 1.16 rev s-1. Samples were collected at various intervals of time. The samples were allowed to settle into separate aqueous and organic phases based on their density differences. The acid concentration in the aqueous phase was determined by titrating it against NaOH solution with phenolphthalein as indicator. The concentration of acid in the organic phase was determined by mass balance. As the reaction between TOA and succinic acid is reversible at high concentrations of TOA in the organic phase, the initial specific rate of extraction measurement was considered for determining the kinetics to avoid complications due to reversibility. The initial specific rate was obtained from the slope of the , (concentration of succinic acid in the aqueous phase) versus t
(time) plot at time =0 according to the following equation.
1 RA= ‒ a │t=0 (4.4)
-3 RA values for different initial succinic acid concentrations of 0.1 to 0.7 kmol m were determined using 150 ml of the organic phase containing 1-decanol with 0.229 kmol m-3 TOA. Similarly, the specific rate of extraction for experiments with different initial
TOA concentrations was determined using the above process using TOA concentrations of 0.229, 0.458, and 0.687 kmol m-3 with 0.5 kmol m-3 of succinic acid.
4.3 Results and Discussion
4.3.1 Effect of process variables on extraction efficiency (% )
The effect of initial succinic acid concentration ( , , ), amine composition ( ), temperature ( ) on the extraction efficiency (% ) were determined by plotting
97 | Page response surface plots on the two dimensional planes for known variables as shown in Figures 4.4-4.6.
4.3.2 Effect of succinic acid and TOA concentrations on extraction efficiency
(% ) at =318 K
The influences of the initial concentration of succinic acid and TOA composition on the efficiency of extraction (% ) at the zero level of temperature ( =318 K) can be seen in Figure 4.4, which depicts the interaction between two parameters ( , , and ). The efficiency of extraction (% ) decreases with an increase in succinic acid concentration at a fixed composition of TOA. The attachment ability of acid molecules to the TOA molecules becomes greater at higher concentrations of succinic acid; hence only fewer succinic acid molecules can be extracted by the TOA molecules, decreasing the efficiency of extraction (% ). The extraction efficiency reaches the maximum value of 92.5 % as TOA concentration increases from 3.36 to
36.38% (v/v), at a lower succinic acid concentration (0.0417 kmol m-3). It indicates that sufficient numbers of TOA molecules are available in the organic phase to form the complex with acid molecules at this acid concentration. Therefore, higher efficiency of extraction is achieved at higher levels of amine concentration. It is also observed that an increase in TOA concentration will affect the efficiency of extraction significantly up to a certain limit and its influence is lower after that.
98 | Page 100
90
80
70
60
50
40
30 3.67 % TOA
Extraction efficiency (%E) 10 % TOA 20 20 % TOA 10 30 % TOA 36.38 % TOA 0 0 0.2 0.4 0.6 0.8 1 Succinic acid concentration (kmol m-3)
Figure 4.4 Effect of succinic acid and TOA concentrations on extraction efficiency (%
)
4.3.3 Effect of temperature (T) on extraction efficiency (% )
The change in the extraction efficiency at various temperatures as a function of succinic acid concentration at a constant TOA concentration of (20 % v/v), and as a function of TOA concentration at a constant succinic acid concentration of
(0.45 kmol m-3) are shown in Figures 4.5 and 4.6, respectively. The study of temperature effects is important from the point of view of the back extraction/regeneration step. Generally industrial scale fermenters operate in the temperature range of 303 to 333 K for the production of carboxylic acids. Thus, an extractant can be considered efficient if it operates in this temperature range.
Reactive extraction is an exothermic process and a decrease in extraction efficiency is expected with an increase in temperature. However, the decrease in extraction efficiency is also the function of the extractant and diluent chosen i.e. the extracting medium. The various feed and reactor conditions under which industrial fermenters
99 | Page operate make the study of temperature effect more important (Kumar, KV &
Sivanesan 2005). The results from Figure 4.5 and Figure 4.6 show that the distribution of succinic acid in the organic phase decreases with an increase in the temperature because the back extraction of succinic acid molecules from the organic phase to the aqueous phase takes place at higher temperatures. The equilibrium complexation reaction occurring between succinic acid and TOA at the aqueous– organic interface is anticipated to be exothermic in nature. The acid-amine complex formation also makes the system high ordered which decreases the system’s entropy and randomness. Moreover, the thermal and kinetic energies of molecules increase with an increase in temperature and it interrupts the interactions and probable combinations of acid and amine molecules to form stable acid–amine complexes at the interface.
100 90 80 70 60 50 40 293.5 K 303 K 30 318 K Extraction efficiency (%E) 20 333 K 10 342.5 K 0 0 10 20 30 40 TOA composition (% v/v)
Figure 4.5 Effect of TOA concentration and temperature (K) on extraction efficiency (
% )
100 | Page 100 90 80 70 60 50
40 0.0475 kmol m-3 30 0.2 kmol m-3 20 0.45 kmol m-3 Extraction efficiency (% E) 0.7 kmol m-3 10 0.858 kmol m-3 0 290 310 330 350 370 Temperature (K)
Figure 4.6 Effect of succinic acid concentration and temperature (K) on extraction efficiency (% )
4.3.4 Experimental Design with Response Surface Methodology
In the reactive extraction process, the parameters such as initial acid concentration (
1), TOA composition ( 2) and temperature ( 3) affect the extraction process and therefore in this study these factors are considered as design parameters. The CCD
(central composite design) is applied to the design the experiments. A five level (-α, -
1, 0, +1, +α) three parameter CCD was chosen in this study to find the applicability of the reactive extraction process and the process variables that affect the reactive extraction of succinic acid. The levels +1 and -1 are higher and lower levels, 0 is center point while + ‒ and represent star points ( ) . The percentage of succinic acid extraction has been taken as the response of the system. Initial acid
-3 concentration ( , : 0.2 to 0.7 kmol m ), amine composition ( : 10-30 % v/v) and temperature (303-333 K) were chosen as the input parameters. The parameters and levels of the central composite design are given in Table 4.1. The total number
101 | Page of experiments to be conducted for the optimisation of parameters were determined using the formula
= 2 +2 + (4.5) where n is the number of parameters to be optimised and Nc is no of central points.
For four parameters, the total number of runs (20) will consist of three parts:
(a) 23 = 8 runs corresponding to factorial points
(b) 2×3=6 runs corresponding to axial points, fixed according to rotatability condition at a distance
(c) Nc = 6 runs replicating the center points used to determine the experimental error and check the reproducibility of the data.
Table 4.1 Selected process parameters and their levels for succinic acid extraction
Coded Coded variable levels
Variables Variables -α -1 0 1 α
Initial acid concentration (kmol m-3) 1 0.041 0.2 0.45 0.7 0.85
TOA composition % (v/v) 2 3.67 10 20 30 36.38
Temperature (K) 3 293.65 303 318 333 342.4
4.3.4.1 CCD Analysis
To evaluate the combined effects of initial acid concentration, TOA composition, and temperature, experiments were conducted according to the design matrix and the results obtained are reported in Table 4.3. The regression equation was developed by fitting the experimental data to linear, interactive, and quadratic models. To select the suitable model, that explains the reactive extraction of succinic acid using TOA in
1-decanol, the tests applied are lack of fit, the sequential model sum of squares and model summary statistics (Kumar, TP et al. 2010). The sequential model sum of
102 | Page squares and model summary statistics tests are sufficient to check the adequacy of the model and only these methods are applied in this study. Results obtained are listed in Table 4.2 and Table 4.3. From Table 4.3, it can be seen that, by comparing the values of probability factor ‘ ’ for different models, the value is less than 0.001 for the quadratic model. In the model summary statistics table, the values of the
“adjusted square” and “ square” were closer to unity for the quadratic model. By considering the value, “adjusted square” and “ square” values, it can be concluded that the quadratic model is the best fit for the extraction of succinic acid equilibrium studies.
Table 4.3 Different values for reactive extraction of succinic acid
Model summary
Source Predicted Adjusted Remarks
Linear 0.57 0.702 0.764
Linear + square 0.855 0.975 0.941
Linear + interaction 0.406 0.710 0.817
Full quadratic 0.966 0.987 0.994 Suggested
103 | Page Table 4.2Experimental Design Values with the Responses
Coded Variables Response
-3 Exp. No (kmol m ) 1 (%,v/v) 2 Temp (K) 3 % (Experimental) % (Predicted)
1 0.2 -1 10 -1 303 -1 50.22 50.88
2 0.7 1 10 -1 303 -1 48.10 49.48
3 0.2 -1 30 1 303 -1 90.41 92.89
4 0.7 1 30 1 303 -1 66.10 67.94
5 0.2 -1 10 -1 333 1 34.25 36.65
6 0.7 1 10 -1 333 1 37.10 38.86
7 0.2 -1 30 1 333 1 77.88 80.74
8 0.7 1 30 1 333 1 55.81 59.40
9 0.45 0 20 0 318 0 71.98 72.60
10 0.45 0 20 0 318 0 71.98 72.60
11 0.45 0 20 0 318 0 71.98 72.60
12 0.45 0 20 0 318 0 71.98 72.60
13 0.0417 -1.633 20 0 318 0 80.14 77.81 104 | Page 14 0.858 1.633 20 0 318 0 61.67 59.24
15 0.45 0 3.67 -1.633 318 0 26.70 25.71
16 0.45 0 36.38 1.633 318 0 80.56 76.78
17 0.45 0 20 0 293.5 -1.633 73.76 72.68
18 0.45 0 20 0 342.49 1.633 57.77 54.09
19 0.45 0 20 0 318 0 71.98 72.60
20 0.45 0 20 0 318 0 71.98 72.60
105 | Page 4.3.4.1.1 Fitting of second order polynomial equation and statistical analysis
The quadratic expression obtained by fitting the experimental data on the basis of
CCD experimental design and input variables for succinic acid extraction is given in the following equation.
% =-12.67+3.56 1 + 5.671 2 + 0.826 3 - 0.2447 1 1 - 0.08008 2 2 - 0.01536 3 3
- 0.2355 1 2 + 0.0240 1 3 - 0.00346 2 3 (4.6)
The analysis of variance (ANOVA) of data was carried out to check the significance of the model, and the results of the analysis are listed in Table 4.4. Large Fisher F- test values and corresponding lower p values ( <0.0001) suggest that the regression is statistically highly significant. The correlation coefficient ( 2 ) and adjusted 2 values used to check the fitness of model are shown in Figure 4.7. Fairly moderate values of 2 =0.994 and adjusted 2 =0.987 support the good correlation between observed and predicted values. The significances of initial acid concentration, amine composition and temperature on the reactive extraction process were determined by sum of squares values, F-values and p values (Table
4.4). Model terms are significant for < 0.05 while for > 0.1 terms are considered as insignificant.
106 | Page 100 90 80 70 60 50 40 30
Predicted efficiecy (% E) 20 10 0 0 20 40 60 80 100 Experimental efficiency (%E)
Figure 4.7RSM Model predicted v/s experimental values of degree of extraction
Table 4.4 Analysis of variance (ANOVA) for response surface quadratic model.
Source SS DF Adj F-value P-value
Model 542.50 10 5425.05 149.63 < 0.0011
Blocks 50.83 1 50.83 14.02 < 0.001
Linear 1374.34 3 4123.02 379.07 < 0.001
X1 431.05 1 431.05 118.89 < 0.001
X2 3259.89 1 3259.89 899.14 < 0.001
X3 432.08 1 432.08 119.18 < 0.001
Square 321.71 3 965.13 88.73 < 0.001
X1X1 25.90 1 25.90 8.52 0.017
X2X2 846.83 1 846.83 233.57 < 0.001
X3X3 157.83 1 157.83 43.53 < 0.001
2-Way 286.07 3 3 95.36 < 0.001
X1X2 277.42 1 277.42 76.52 < 0.001
107 | Page X1X3 6.50 1 6.50 1.79 0.213
X2X3 2.15 1 2.15 0.59 0.461
Error 3.63 9 32.63
Lack-of Fit 6.53 5 32.63
Pure 0 0 4
Total 19 19 5425.05
R2 = 0.994Adj R2 = 0.987Predicted R2 = 0.969
SS-Sum of squares, DOF-Degrees of freedom
4.3.4.2 Process Optimisation of process variables
The optimum values for the selected process variables were obtained by solving the
regression equation (4.10) using Minitab statistical software. The model was
employed to find the values of the process variables that gave maximum degree of
extraction. The predicted optimal values obtained from the model equation for
-3 maximum extraction efficiency are =0.2 kmol m , = 33 (% v/v), and =
305K (Table 4.5). The model predicts that the highest extraction efficiency that can
be obtained at the optimum conditions of the process variables is 93.75%. Extraction
efficiency from an experiment at these optimum conditions was 91%, which is very
close to the predicted value.
Table 4.5 Verification of experimental results at optimum process variables.
Optimum conditions Degree of extraction
-3 (kmol m ) (% v/v) Temperature (K) Experimental results Model predicted
0.2 33 305 91% 93.75%
108 | Page 4.3.5 Kinetic Studies
4.3.5.1 Determination of individual mass transfer coefficients
To determine the reaction regime in the reactive extraction of succinic acid from the aqueous phase by TOA-1-decanol mixture, the interfacial mass transfer coefficient value is required. Therefore, an indirect approach was used to estimate the mass transfer coefficient values. In this approach, the mass transfer coefficient for a reference system (acetic acid-water-kerosene) was measured and used to estimate the mass transfer coefficient for the system used in this work (Juang & Lo 1994;
Kiani, Bhave & Sirkar 1984; Komasawa, Otake & Yamada 1980; Kumar, TP et al.
2010). Since acetic acid has a sufficiently small distribution coefficient in a water/kerosene system, the resistance of aqueous diffusion film can be neglected.
The overall mass transfer coefficient for the acetic acid-water-kerosene system can be determined by measuring the initial and transient concentrations of acetic acid in the organic phase and using equation (4.7).
0 = kaat (4.7) ( )
0 where and are the concentrations of initial acetic acid and concentration of acetic acid in organic phase, respectively, is the interfacial area and is time.
0 A linear plot of the group in L.H.S ( ) of equation (4.7) against (a.t) will yield the ( ) value of the mass transfer coefficient as the slope of the straight line (Figure 4.8).
Thus, the mass transfer coefficient, , for acetic acid-water-kerosene system was
-5 -1 found to be 1.08 × 10 m s . But the mass transfer co-efficient, can also be
109 | Page determined using the following relationship (Juang & Huang 1995; Kiani, Bhave &
Sirkar 1984; Prasad et al. 1986).
2 1 3 ‒ 6 α ɣ (4.8)
where and ɣ are diffusivity of the solute and kinematic viscosity of solvent, respectively.
In this study, the mass transfer coefficient value for the water-succinic acid-1- decanol-TOA system is estimated using equation (4.9) in the following form:
2 1 = 2 1 (4.9) ‒ ‒ 3 6 3 6 ( ɣ ) ( ɣ )
The diffusivities of succinic acid in 1-decanol and TOA in 1-decanol were estimated using the Minhas-Hayduk method (Akhtar et al., 2008; Eda et al., 2015; Hanna &
Noble 1985; Juang & Huang 1995) in which diffusivity is proportional to the ratio of the Parachor values of the solvent and solute, and it is inversely proportional to the molar volume and viscosity of solvent. The kinematic viscosities of kerosene and 1- decanol at 298 K are 1.52 x 10-6 and 142.08 x10-6 m2 s-1, respectively. The molar volumes of kerosene, 1-decanol, succinic acid and TOA are 259.9, 192, 83 and
555.9 cm3 mol-1, respectively. The Parachor values (which depend on the structure of the component) for 1-decanol, succinic acid, and TOA are 442.68, 217.29 and
1024, cm3 g1/4 s-1/2, respectively (Kumar et al. 2010).
The diffusion and mass transfer coefficient values estimated using Minhas-Hayduk equation and equation (4.9), respectively are shown in Table 4.6 for both acetic acid- water-kerosene and succinic acid-TOA-1-decanol systems.
110 | Page 0.16
0.14
0.12
0.1
0.08
ln(C0a/Ca) 0.06
0.04
0.02
0 0 5000 10000 15000 a×t
Figure 4.8 Determination of mass transfer coefficient for acetic acid in kerosene/water.
Table 4.6 Diffusion and mass transfer coefficients at 303 K
Diffusion coefficient Mass transfer coefficient System (Solute/medium) (m2 s-1) (m s-1)
Acetic acid/kerosene 7.68 x 10-10 1.08x 10-5
Succinic acid/1-decanol 1.63 x 10-10 2.65x 10-6
TOA/1-decanol 8.71 x 10-11 1.75x 10-6
4.3.5.2 Effect of agitation speed on specific rate of extraction RA
According to standard reaction modeling, the reaction rate can be controlled either by the kinetics of the chemical reaction or diffusion of the process taking place in the system. In a diffusion-controlled process, the rate of extraction increases with increasing stirrer speed, but in a kinetics-controlled process, stirrer speed has no effect (Levenspiel 1999).
111 | Page The effect of agitation speed on the specific rate of extraction is shown in Figure 4.9.
It depicts that the specific rate of extraction does not depend on the stirrer speed above 60 rpm. From this observation, it can be concluded that the extraction occurs in the kinetics-controlled regime. As shown in the figure, the specific rate of extraction is not exactly constant beyond80 rpm. Thus, an agitation speed of 70 rpm was selected to conduct the experiments. There is no effect of stirrer speed on the specific rate of extraction from 60 to 80 rpm. From this observation, the reaction can be concluded to occur in either in regime 1 or 3.
0.0000054
0.0000049
0.0000044
0.0000039
0.0000034
0.0000029RA kmol m-2 s-1
0.0000024
0.0000019 40 60 80 100 RPM
Figure 4.9 Effect of speed of stirring on specific rate of extraction
4.3.5.3 Order of reaction with respect to initial succinic acid and TOA concentration
To determine the order of reaction with respect to succinic acid, its initial concentration in the aqueous phase was varied from 0.3 to 0.7 kmol m-3 to determine the specific rate of extraction under these conditions. Figure 4.10(a) shows the effect of acid concentration on the specific rate of extraction. It is found that the extraction rate is increasing linearly with increasing acid concentration. The order of the reaction with respect to succinic acid, m, is determined to be 1 and the rate constant
112 | Page is determined to be 3x10-5 s-1 from the slope of the straight line shown on the graph.
Similarly, the effect of TOA concentration on the specific rate of extraction was studied to determine the order of reaction with respect to TOA concentration in the organic phase. The system was studied using TOA concentration in the range of
0.114 to 0.6875 kmol m-3 in 1-decanol. The values of specific rate of extraction are found to increase with increasing TOA concentration and the order with respect to
TOA, n, is determined to be 1, and the second order rate constant is found from the slope of the straight line on the graph as 1x10-5 s-1 (Figure 4.10(b)).
0.000006
0.000005
0.000004
0.000003
0.000002 RA kmol m-2 s-1
0.000001
0 0 0.05 0.1 0.15 0.2 C*A org kmol m-3
0.00001 0.000009 0.000008 0.000007 0.000006 0.000005 0.000004
RA kmol m-2 s-1 0.000003 0.000002 0.000001 0 0 0.2 0.4 0.6 0.8 [TOA] kmol m-3
113 | Page Figure 4.10 (a) Effect of initial succinic acid concentration on the rate of extraction ,
0.2291 kmol m-3 TOA, at 1 rev s-1, (b) effect of TOA concentration on the rate of extraction for 0.5 kmol m-3 initial succinic acid concentration, 1.16 revs s-1 is the speed of agitation
4.3.5.4 Rate constant, k2
∗ The order of the reaction with respect to succinic acid concentration [ ] and
[TOA]0 is second order. The second order rate constant k2 is obtained from Figure
4.11 as 0.002 m3 kmol-1 s-1.
0.0006 0.00055 0.0005 0.00045 0.0004 0.00035 0.0003 RAa kmol m-2 s-1 0.00025 0.0002 0.00015 0 0.05 0.1 0.15 0.2 C*SAH[TOA]0
Figure 4.11 Determination of second order reaction rate constant, k2
4.3.5.5 Criteria for reaction regime
To confirm the reaction regime, the value of Hatta number MH, was evaluated. For m
1/2 2 ∗ ‒1 ( +1 [ ] [ ] ) = 1 and n = 1, the Hatta number, MH, becomes
114 | Page 1/2 ( 2[ ] ]) , and is found to vary from 7.2 to 21.5 for different concentrations of
1/2 2 ∗ ‒1 ( +1 [ ] [ 0] ) TOA at an agitation speed of 1.16 rev s-1, i.e. >>1. This shows that the reaction taking place in the film phase is fast or instantaneous
[ 0] (Doraiswamy & Sharma1984). To confirm the reaction regime, the value of ∗ [ ] as mentioned by Doraiswamy and Sharma (1984) was determined for the 1:1 acid- amine complex using the z value (loading ratio) of 0.5 for the organic phase. The value of the group was found to vary from 2.15 to 12.9 for the range of succinic and
TOA concentrations used in this work. These values are lower than the Hatta number mentioned above. Based on these findings, it can be confirmed that the reaction is instantaneous and occurring in the organic film.
4.4 Conclusions
In this chapter, a five level central composite design was presented to study the interaction effects of parameters on the reactive extraction of succinic acid. A second order polynomial equation was used to predict the functional relationship between the efficiency of extraction and the process variables. According to the analysis of variance, R2 was found to be 0.994, and adj R2 was found to be 0.987. The optimum
-3 conditions are found to be: , , = 0.2 kmol m , =33 (% v/v), and T= 305K with % of 93.75% by the RSM model. The estimated % E of 93.75% compared well with the experimental value of % E= 91% obtained using optimum conditions. These results are useful in the design and optimisation of the process variables for the reactive extraction of succinic acid from aqueous solution. The concept of extraction accompanied by the chemical reaction has been used to obtain the extraction kinetics. The kinetic studies were carried out based on the initial extraction rates to
115 | Page determine the specific rate of extraction, RA. It was shown that there was not much variation in the value of specific rate of extraction with different speeds of agitation in the range of 0.8 to 1.4 rev s-1. The reactions have been found to be first order with respect to succinic acid and TOA. Based on the Hatta number, the reaction between the succinic acid and TOA was found to be instantaneously occurring in the film.
116 | Page Chapter 5
Regeneration of Levulinic Acid from Loaded-Organic Phase: Equilibrium, Kinetic Studies and Process Economics
117 | Page 5.1 Introduction
Recovery of volatile fatty acids from aqueous solutions is a growing requirement of fermentation-based industries. Volatile fatty acids like acetic, propionic, and butyric acids can be recovered using conventional solvent extraction, adsorption, distillation, and extractive distillation (Kertes & King 1986; Kuo et al. 1987). For higher boiling point acids (non-volatile acids), however, energy intensive operations such as distillation are not suitable. Several methods are available to separate carboxylic acids from aqueous streams. One of the conventional methods includes the addition of calcium hydroxide to form calcium salt of carboxylic acids, to which sulphuric acid is added to liberate the free acid. This method of recovery is expensive as the separation, and final downstream operations involve 50% of the total production cost
(Chaudhuri & Phyle 1992; Eyal & Bressler 1993). Moreover, this method is environmentally unfriendly because it consumes large quantities of lime and sulphuric acid, and generates calcium sulphate sludge as a solid waste (Shreve &
Brink Jr 1977). The accumulation of sludge, which contains the valuable carboxylic acids, in fermentation broth also affects the product quality. These problems can be resolved if the acids are recovered from fermentation broth using suitable alternative processes.
Recovery of carboxylic acids through reactive extraction is an emerging technology.
In Chapter 3, an intensified approach involving reactive extraction was employed because it does not consume any additional reagent and does not produce large waste streams like those in conventional processes (Keshav & Wasewar 2010;
Wasewar, KL, Heesink, et al. 2004; Wasewar, KL, Yawalkar, et al. 2004). Reactive extraction with suitable extractants giving higher distribution coefficients has been
118 | Page proposed as a promising technique for the recovery of carboxylic acids (Wardell &
King 1978; Wennersten 1983), but the efficiency of liquid-liquid extraction process for recovering carboxylic acids from aqueous streams is greatly influenced by the nature of the solvent used.
Reactive extractants like tertiary amines, phosphates, and phosphine oxides can be used to improve the physical properties of solvents especially in increasing the capacity and selectivity of solvents. Preferred extractants for reactive extraction of carboxylic acids from aqueous phase include long chain aliphatic primary, secondary, and tertiary amines. Primary amines are characterised by a high solubility in the aqueous phase. On the other hand, secondary amines have the highest distribution coefficient but tend to form amides in the downstream regeneration operation like distillation. Tertiary amines offer advantages over the other extractants on the grounds of lower cost and higher equilibrium distribution coefficients KD (Ricker, Pittman & King 1980). Regardless of the type of amine used, the stability of the acid-amine complex is affected by the basicity of the amine, but this can be altered by the addition of various diluents.
Recovery of citric acid from fermentation broth was studied by Baniel et al., using a tertiary amine as the extractant with an alcohol modifier in a hydrocarbon as the diluent (Baniel, Blumberg & Hajdu 1981). They regenerated the acid by back extraction into the aqueous phase using the temperature swing method. Tamada and King (1990) established a process for the back-extraction of citric acid and deduced that back extraction can be achieved by distillation which leads to a temperature swing and consequent change in diluent composition. The total concentration of acid that can be achieved by this method is restricted by the extent to which the distribution equilibrium can be changed in two steps. The first step
119 | Page involves the extraction of the solute as a solute-extractant complex leading to the generation of a relatively solute-free aqueous raffinate. The second step is the regeneration step.
Figure 5.1 shows a flow sheet for the total process of reactive extraction (forward and back extraction/regeneration). The regeneration step is necessary for stripping the solute from the loaded-organic phase to obtain extractant-free aqueous solute as a product (regeneration) (Wasewar, KL 2012).
Figure 5.1 Process flow sheet for the recovery of carboxylic acids
Various methods that are available for the recovery of carboxylic acids from a loaded-organic phase in the regeneration process can be summarised as follows:
1. NaOH method: In this method, the loaded-organic phase is mixed with NaOH
solution to achieve acid regeneration. Yabannavar and Wang (Yabannavar &
120 | Page Wang 1991a) suggested the NaOH method is an effective method for the
regeneration of lactic acid.
2. HCl Method: In this method, which was also suggested by Yabannavar and
Wang (Yabannavar & Wang 1991a), concentrated HCl solution was used to
separate the acid from loaded-organic phase (lactic acid+Alamine336+oleyl
alcohol). Yabannavar and Wang concluded that it is possible to regenerate
the solvent by distilling diluent in the organic phase.
3. Temperature-swing method: Tamada and King (1990) suggested that low
temperatures are favourable for the forward extraction process. For the
regeneration, however, a fresh aqueous phase was contacted with the
loaded-organic phase at higher temperatures to facilitate the extraction of acid
from the acid-extractant complex.
4. Tri-methylamine (TMA) method: Poole and King (1991) suggested that a
suitable method for the regeneration of carboxylic acids from the loaded-
organic phase was to use a stronger volatile tertiary amine (TMA), which has
a very low boiling point of 3°C, in the aqueous phase. Ammonia is the most
water soluble volatile base, but the ammonia and both primary, secondary
amines will form strong ammonium carboxylates when they are heated with
carboxylic acids (Mitchell & Reid 1931; Poole & King 1991; Streitwieser et al.
1992), so it is difficult to regenerate and recover the acid.
5. Diluent swing method: The diluent swing is based on a shift in the equilibrium
distribution of the acid from the aqueous phase to the organic phase between
forward- and back extraction processes due to the change in the composition
of the diluent with which the extractant is mixed. The diluent composition
swing facilitates the back extraction of the acid into an aqueous phase.
121 | Page (Baniel, Blumberg & Hajdu 1981; Hong, Hong & Hong 1999; Tamada & King
1990).
This chapter aims to investigate the regeneration of levulinic acid from the loaded- organic phase, which involves the back-extraction of the acid into the aqueous phase. The aim is to determine the best back-extraction technique among the following four methods using extraction efficiency as the key criterion:
1. NaOH method
2. Temperature swing method
3. Diluent swing method
4. Tri-methylamine (TMA) method.
This chapter also contains a kinetic study of the best back-extraction method (TMA method) chosen. A detailed economic evaluation is carried out to determine the financial viability of TMA method at a larger scale.
5.2 Theory
5.2.1 Determination of liquid-liquid interfacial area
The liquid-liquid interfacial area is dependent on several factors such as interfacial tension, impeller speed, reactor and impeller diameter, and liquid density and viscosity. The interfacial area can be obtained as follows:
The average radius of droplets in dispersion present in a cylindrical reactor is estimated using the following equation (Starks 1999).
1 (2 3 ‒ 1)( ) = (5.1) 2 2
122 | Page where n is the number of droplets formed during splitting, is the density of the liquid
(kg m-3), is the rotational speed of the stirrer s-1, is the distance of the shell of liquid from the center of the reactor and Υ is the interfacial tension.
For a cylindrical reactor with radius and height ℎ , operating at a stirrer speed of rpm, the volume of dispersed phase in a cylindrical shell can be estimated using equation (5.2),
=2 (ℎ) . (5.2)
3 where is the volume of the dispersed liquid (m ), is the distance of the shell of liquid from the center of the reactor, is the thickness of the shell, and ∅ is the volume fraction of the dispersed phase in the continuous phase.
The total number of dispersed droplets in the cylindrical shell is calculated by dividing by the average droplet volume. When this number is multiplied by the average interfacial area per droplet, we obtain equation (5.3) for the total interfacial area in the thin shell.
3 2 (Area) = = 4 4 = (5.3) 3 3
3 where is the average droplet volume (m ) and is the average interfacial area per droplet (m2).
By substituting equations (5.1) and (5.2) for and , respectively in equation (5.3) and integrating the resulting equation between the limits = 0 and = , we get an equation for the interfacial area as,
Interfacial area =8.53ℎ 2 4 (5.4)
123 | Page By dividing the equation for the interfacial area (equation 5.4) by the liquid volume (
), we can get an equation for the specific interfacial area ( ) as,
8.53ℎ 2 4 = . (5.5)
5.3 Experimental
5.3.1 Chemicals
TOA (98% purity), a tertiary amine, supplied by Sigma Aldrich chemicals was used as an extractant. MIBK (99% purity) supplied by S. D. Fine-Chemicals Ltd., Mumbai,
India was used as a diluent. Levulinic acid (98%) supplied by Sigma-Aldrich was of technical grade, and Millipore water was used to prepare the aqueous solutions of levulinic acid. Reagent grade sodium hydroxide supplied by S. D. Fine-Chemicals
Ltd. Mumbai, India was used for the preparation of NaOH solution, which was used in back-extraction experiments and the titration of acid. TMA (assay 30% in water) was supplied by Loba Chemie, Mumbai, India, and heptane and toluene (inert diluents) were supplied by S. D. Fine Chemicals Ltd. Mumbai, India. Ammonium dihydrogen phosphate supplied by Alfa-Aesar was used as the mobile phase for
HPLC. Phosphoric acid supplied by Sigma-Aldrich was used in the adjustment of mobile phase pH. Phenolphthalein indicator (pH 8-10) was supplied by Thermo
Fisher Scientific India Pvt Ltd. The properties of all chemicals are shown in Table
5.1.
124 | Page Table 5.1 Properties of chemicals
Name M.W.(g.mol-1) Density (g.cc-1) B.P. (0C) Molecular structure
Levulinic acid 116 1.14 245-246
Tri-methylamine 59.11 0.67 2.9
Tri-n-octylamine 353.67 0.809 365-367
Methyl isobutyl ketone 100.16 0.802 117-118
Toluene 92.14 0.87 111
Heptane 100.21 0.679 98.1-98.7
5.3.2 Equilibrium Studies
5.3.2.1 Equipment
The equilibrium study was carried out in a shaking incubator (Model no: 4018R)
equipped with a universal shaking platform, independent alarms, and
microprocessors for temperature and speed adjustments. The shaker provided
orbital shaking motion, which can be adjusted between 25 to 400 rpm. The shaking
speed was set using a digital LED control panel. The shaking incubator has a
versatile digital timer, which can be set from 1 second to 9 days. The incubator
temperature can be varied from 288 to 353 K.
5.3.2.2 Experimental methods
NaOH method: For the regeneration of acid by NaOH, the loaded-organic phase was
mixed with a required amount of NaOH solution, and the conical flask containing the
125 | Page mixture was kept in the shaking incubator operating at 220 rpm and 298 K for 6 h, and then allowed to settle for 2 h at 298 K. The resulting two phases were separated using a separating funnel, and their volumes were measured. NaOH solution was titrated with oxalic acid to find the concentration of NaOH, which was then used as blank to find the concentration of acid in the aqueous phase. The levulinic acid concentration in the aqueous phase was measured by titrating it with 0.5 N oxalic acid using phenolphthalein as the indicator, and the final concentration of levulinic acid was obtained by subtracting this concentration from the initial oxalic acid concentration (blank).
Temperature swing method: For the regeneration of acid by the temperature swing method, equal volumes (20 ml each) of the loaded-organic phase and the aqueous phase were mixed, and the mixture was allowed to equilibrate in a shaking incubator operating at 220 rpm at 298 K over a period of 6 h, and then allowed to settle for 2 h at 298 K to obtain two separate phases. The acid concentration of the aqueous phase was determined by titrating it against 0.5 N NaOH solution using phenolphthalein as the indicator. The acid concentration in the organic phase was then calculated by mass balance. Similar procedure was used at four different temperatures in the range of 313-343 K.
Diluent swing method: In the diluent swing method, the loaded-organic phase was mixed with chosen diluent (heptane and toluene) and water, and allowed to equilibrate in a shaking incubator at 298 K over a period of 6 h. The equilibrated mixture was then allowed to settle over a period of 2 h at 298 K and separate into two separate phases. The aqueous phase was titrated with 0.5 N NaOH solution to determine the acid transferred using phenolphthalein as the indicator. The
126 | Page corresponding acid concentration in the organic phase was then calculated by mass balance.
TMA method: In the TMA method, known volumes of aqueous phase with TMA and levulinic acid loaded-organic phase of known concentration were mixed and allowed to equilibrate in a shaking incubator for 6 h at 220 rpm and 298 K. The mixture was then allowed to settle for 2 h to ensure the separation of the two phases in a separating vessel. The concentration of the acid in the aqueous phase was determined using high-performance liquid chromatography (HPLC) system consisting of binary pumps, PDA detector, and a column oven. Acetonitrile was used as the stationary phase and aqueous ammonium dihydrogen phosphate solution, adjusted to a pH value of 2.2 by the addition of phosphoric acid, was used as the reverse phase in the C-18 column. Levulinic acid was detected at 210 nm. The acid content of the organic phase was determined by mass balance. The recovery of levulinic acid was calculated using equation (5.6).
Recovery efficiency (% )= ×100 (5.6) , where and , are the concentrations of acid in the aqueous phase and its initial concentration in the organic phase, respectively.
5.3.3 Kinetic studies
The kinetic experiments were carried out in a 500 ml jacketed glass reactor equipped with an overhead stirrer. The reactor was a cylindrical glass vessel of 0.1m diameter and 0.13m height with round bottom. The stirrer was a paddle impeller with four blades whose speed was controlled by a variable speed drive. The reactor assembly consisted of the reactor with four necks with glass stoppers to facilitate the periodic withdrawal of liquid samples during the reaction. A refrigerating and heating circulator water bath (Julabo) with a digital temperature controller was used for
127 | Page maintaining the temperature of the reactor within the range of ± 0.010C of the set temperature by circulating water through the jacket. A schematic diagram of the reactor assembly is shown in Figure 5.2.
Figure 5.2 Schematic diagram of experimental setup: R-Jacketed reactor, F-Feed inlet, S-Stirrer, M-Motor, WS-water source, WI-Water inlet, WO-Water outlet, M -
Motor, SP-Sampling point, TI-Temperature indicator
5.3.3.1 Experimental method
To determine the specific rate of extraction, the reactor was initially charged with 150 ml of aqueous solution with TMA and 150 ml of loaded-organic phase. The agitator speed was maintained at 1.16 rps (70 rpm) during the reaction. Samples of the aqueous phase were collected at regular intervals to find the concentration of acid transferred from loaded-organic phase to the aqueous phase. The acid concentration in the aqueous phase was determined by HPLC whereas the acid in the organic phase was determined by mass balance. A selected few experiments were repeated twice to determine the reproducibility of the results. A similar procedure was followed
128 | Page for various acid and TMA concentrations at different impeller speeds. Only the initial rates of reaction were considered in the evaluation of reaction kinetics so to avoid the problems due to reaction reversibility. The initial rate was determined from the linear plot of levulinic acid concentration in the aqueous phase ( ) versus time at time =0 as per equation (5.7),
1 RA= ‒ a │t=0 (5.7)
where is the specific rate of extraction, is the specific interfacial area.
5.4. Results and discussion
5.4.1 Equilibrium studies
5.4.1.1 NaOH method
Regeneration of levulinic acid is done by reversible complexation, which allows the recovery of the acid in the pure water phase. However, reversing the reaction is a complicated step because chemical complex should be a strong bond to influence the forward extraction. Therefore, regeneration of levulinic acid from loaded-organic phase was carried out in this work using NaOH solutions. Different stoichiometric ratios of NaOH to levulinic acid concentrations ranging from 0.2 to 1.8 were used.
These experiments were also carried out at various concentrations of levulinic acid in loaded-organic phase ranging from 0.058 to 0.314 kmol m-3.
The recovery of levulinic acid is found to increase linearly with increasing NaOH to levulinic acid molar concentration ratio up to 1.0 and remain more or less constant after that (Figure 5.3). The recovery of levulinic acid by NaOH solution is above 75 % for all concentrations (0.05780 to 0.314 kmol m-3) of acid in organic phase when
NaOH concentration is such that the molar concentration ratio of NaOH to levulinic
129 | Page acid is greater than 1.0. This result can be attributed to increasing pH of the organic solution with increasing NaOH concentration. The formation of acid-amine complex, which occurs pre-dominantly in the forward extraction process, is not favoured at high pH values and therefore the acid is transferred back to the aqueous phase leading to the generation of TOA while NaOH leads to the formation of sodium levulate. By increasing the levulinic acid concentration in the loaded-organic phase, the recovery of acid is found to be lower because the pH of the solution becomes low with increased levulinic acid concentration thereby leading to lower recovery of acid.
115
105
95
85
75
65 1% acid
% Recovery 55 2% acid 45 3% acid 35 4% acid 25 5% acid 15 0 0.5 1 1.5 2 NaOH/[C]org
Figure 5.3 Recovery of levulinic acid from loaded-organic phase using NaOH method, 1 % = 0.0578, 2 % = 0.115, 3 % = 0.235, 4 % = 0.251 and 5 % = 0.313 kmol m-3 of levulinic acid in organic phase
5.4.1.2 Temperature swing
The main objective of this work is to determine whether temperature-swing regeneration can be employed for recovering the acid from the loaded-organic phase
130 | Page at a higher temperature, especially in the range of 313-343 K. From a thermodynamic view point, the acid molecules in the organic phase are more ordered because they exist as a complex. Thus, in the forward extraction, the transfer of acid from the aqueous phase to the organic phase increases the order and reduces entropy.
The recovery of acid from the loaded-organic phase with different acid concentrations is shown in Figure 5.4 as a function of temperature (313-343 K). The acid recovery ranges are 73-89, 68-84, 64-78%, 61-76%, and 60-75% for loaded phase acid concentrations of 0.057, 0.115, 0.235, 0.251 and 0.314 kmol m-3 respectively. These results show that the percentage of acid recovery increases with increasing temperature regardless of the acid concentration in the loaded phase indicating that temperature-swing is a successful technique to regenerate the acid.
However, it appears that percentage acid recovery is the highest at 343K especially at low acid concentrations, which could be attributed to the larger extent of interaction of TOA and acid in the organic phase leading to the back distribution of acid into the aqueous phase at higher temperatures. Similar work was carried out by
Baniel et.al (1981) to study the regeneration of citric acid using the temperature swing technique by changing the process temperature. They reported that temperature change led to changes in different parameters such as the acid value, acid-amine interaction, and acid solubility in both phases, extractant basicity, etc.
131 | Page 95 1 % acid 90 2 % acid
85 3 % acid 4 % acid 80 5 % acid
75
% Recovery 70
65
60
55 310 320 330 340 350 Temp (K)
Figure 5.4 Recovery of levulinic acid from loaded-organic phase using temperature swing, 1 % = 0.0578, 2 % = 0.115, 3 % = 0.235, 4 % = 0.251 and 5 % = 0.313 kmol m-3 of levulinic acid in organic phase
5.4.1.3 Diluent Swing
Using this method, the fresh aqueous solution was contacted with the loaded-organic phase in the presence of diluents, toluene or heptane, to produce an acid-laden aqueous solution. Two loaded-organic phase to diluent volume ratios, 1:0.5 and 1:1, were used in the experiments. Experiments for toluene and heptane were conducted separately. The percentage recovery of levulinic acid at 298 K is shown in Figure 5.5 for loaded-organic phase with different acid concentrations. For all runs, the percentage acid recovery values are higher for high loaded-organic phase to diluent volume ratio (1:1) compared to those for low volume ratio (1:0.5). Also, the percentage acid recovery is the highest for the loaded-organic phase with the lowest acid concentration (i.e., 0.058 kmol m-3). For all cases, the acid recovery is higher when heptane (aliphatic hydrocarbon) is used as the diluent compared to toluene
132 | Page (aromatic hydrocarbon). As mentioned above, the diluents aid the recovery of levulinic acid from the loaded-organic phase by lowering its distribution ratio between the organic and aqueous phases, and heptane performs better in this process than toluene. Thus, diluent swing is a flexible method for both the extraction and regeneration equilibria into appropriate distribution ranges. However, the recovery of acid is about 15% lower than those obtained with temperature swing method at 343
K (Figure 5.4).
80 1:0.5(organic:heptane) 1:0.5(organic:toulene)
70 1:1(organic:heptane) 1:1(organic:tolune)
60
50
40
Recovery 30 %
20
10
0 0.058 0.116 0.236 0.251 0.314 [C]org (kmol m-3)
Figure 5.5 Recovery of levulinic acid from loaded-organic phase using diluent swing
5.4.1.4 TMA method
TMA is a stronger volatile tertiary amine. TMA is used in the regeneration of acid from the loaded-organic phase because it is easily removable by heating the mixture
(Keshav & Wasewar 2010; Wasewar, KL, Heesink, et al. 2004; Wasewar, KL,
Yawalkar, et al. 2004). The TMA regeneration method does not involve any by-
133 | Page product formation. TMA can be recycled by thermal decomposition. The thermal decomposition provides acid as a liquid product and tri-methylamine as vapor, which can be condensed and recycled to regeneration step. The most water-soluble volatile base is ammonia. However, ammonia is not suitable for the regeneration of the acid from loaded-organic phase because ammonia, when heated with carboxylic acids, will form strong ammonium carboxylates from which it is difficult to regenerate the acid. A similar phenomenon is observed with primary and secondary amines too.
The transfer of acid in reactive extraction using TOA and its regeneration with TMA are graphically represented in Figure 5.6.
Figure 5.6 Transport mechanism of acid in reactive extraction and regeneration for
TMA method Red: levulinic acid, Blue-TOA, Green-TMA.
Experimental results of percentage recovery of levulinic acid using TMA are shown in Figure 5.7 as a function of TMA concentration. The acid concentration in loaded-
-3 organic phase was varied from 0.058 to 0.314 kmol m , which corresponds to TMA to acid molar concentration ratios from 0.2 to 2. Figure 5.7 depicts that, around the
TMA to acid molar concentration ratio of 1, the acid recovery is 65 and 80% for acid
134 | Page concentrations of 0.314 and 0.058 kmol m-3, respectively. At TMA to acid molar concentration ratio of 1.5, the acid recovery is above 80% for all acid concentrations.
110 100 90 80 70 60 1 % acid 50 % Recovery 2 % acid 40 3 % acid 30 4 % acid 20 5 % acid 10 0 1 2 3 [TMA]/[C]org
Figure 5.7 Recovery of levulinic acid from loaded-organic phase using TMA method,
1 % = 0.0578, 2 % = 0.115, 3 % = 0.235, 4 % = 0.251 and 5 % = 0.313 kmol m-3 of levulinic acid in organic phase
Comparing all the regeneration methods used in this work, the recovery of acid is relatively high in TMA and NaOH methods at high TMA to acid and NaOH to acid molar concentration ratios (1 or > 1), respectively. Although the percentage recovery of levulinic acid in both methods is greater than 80 %, it is advantageous to use the
TMA method because the highly volatile base (TMA) and levulinic acid can be easily regenerated using evaporation. The recovery of levulinic acid using the temperature and diluent swing methods are low (75-85 %) compared to NaOH and TMA methods.
5.4.1.5 Mass action law model
135 | Page The effect of TMA on the regeneration of levulinic acid from loaded-organic phase can be explained by mass action law model, which was proposed by Kertes and
King (1986). In mass action law model, activity coefficients of the aqueous and organic phase species are assumed to be proportional to respective concentrations of the species, and the equilibrium constant takes care of the proportionality constant
( , and ) or the non-idealities associated with the reactive system.
Therefore, an apparent equilibrium constant is used in mass action law model for developing a mathematical model to predict the extraction efficiency at reaction equilibrium.
The dissociations of levulinic acid, TMA and TOA in the regeneration process are as follows.
⇔ ˉ+ ⁺
3 ⁺ ⇔ 3 + ⁺
⁺ ⇔ + +
+ + where is levulinic acid, 3 is trimethylamine, and is tri-n- octylamine.
The dissociation constants for levulinic acid ( ), TMA ( ) TOA ( ) are calculated using the equations shown below:
[ ‒ ][ + ] = [ ]
+ [ 3 ][ ] = + [ 3 ]
136 | Page [ ‒ ][ + ] = [ + ]
The interaction of the levulinic acid molecule with TOA is as shown below:
1 + ⇔ ‒
The interaction of levulinic acid with TMA is as shown below:
2 3 + ⇔ ‒ 3
Combining the above two equations, we get
3 ‒ + 3 ⇔ ‒ 3 +
The equilibrium complexation constants are calculated using the equations shown below:
[ ‒ ] K a1 = [ ] [[ ]
[ ‒ 3 ] K = a2 [ ][ ] 3
[ ‒ 3 ] [ ][ ] 3 = = × [ 3 ][ ‒ ] [ ‒ ]
where 1 , 2 , and 3 are equilibrium complexation constants. Using the definition of extraction efficiency (% ) of acid in the regeneration step, the following equation can be written:
[ ] % = ×100 [ ] ,
137 | Page The extraction efficiency (% ) in terms of equilibrium complexation constant is as follows:
[ ‒ 3 ] % E = ×100 [ ‒ ]
[ 3 ] = 3 [ ] ×100
[ 3 ] % = ×100 [ ‒ ]
Therefore, the percentage of acid recovery %E in the regeneration process is written as:
[ 3 ] % = 100 (5.8) [ ‒ ]
where , and are the dissociation constants of levulinic acid, TMA and TOA, respectively. Equation (5.8) represents the relationship between the efficiency of acid recovery in the regeneration process and TMA concentration. The recovery efficiency is directly proportional to . It implies that at high concentrations of TMA, the percentage recovery of acid will be high. In other words, the percentage acid recovery will increase with increasing TMA concentration, and this phenomenon is seen in Figure 5.7 to a certain extent.
5.4.2 Kinetic studies for TMA method
Among the above four methods used for the regeneration of levulinic acid from loaded organic phase the extraction efficiency was found to be the highest for the
TMA method. Therefore, kinetic studies were performed only using the TMA method.
In kinetic studies, the experimental data obtained in the TMA method was used in equation (5.2) to determine the specific rate of extraction ( ) by plotting the
138 | Page concentration of acid ( ) in the aqueous phase as a function of time as shown in
Figure 5.8. Since the equilibrium complexation reaction is reversible, only initial rates determined using the inclining section of the curve (Figure 5.8) were considered in the kinetics study to avoid errors due to reversibility.
0.35
0.3
0.25
0.2 CA aq kmol m-3
0.15
0.1 0 20 40 60 80 Time (min)
Figure 5.8 Variation of acid concentration in the aqueous phase with time.
5.4.2.1 Effect of stirrer speed on specific rate of extraction
The RA results shown in Figure 5.9 indicate that it is relatively constant over the stirred speed range used in this work (1.0 to 1.67 rps). Mass transfer with chemical reaction in systems like stirred vessels is controlled either by diffusion or reaction kinetics. The extraction rate will increase with increasing stirrer speed if mass transfer is diffusion-controlled system but will remain unchanged if it is kinetics- controlled (Hanna & Noble 1985). In kinetics-controlled systems, the film adjacent to the interface will become thinner as the reaction progresses and therefore, the influence of diffusion will be lower compared to that of the chemical reaction. Based on this theory, it can be concluded that the rate of extraction in the TMA method is
139 | Page controlled by the chemical reaction. It can be said that the regeneration is occurring in regimes 1 or 3 among the 4 reaction regimes defined by Doraiswamy and Sharma
(1984) which are very slow (regime 1), slow (regime 2), fast (regime 3), and instantaneous reaction regimes (regime 4). To choose between regimes 1 and 3, the
effect of phase volume ratio on ( ) is considered below. ( )
0.000025
0.00002
0.000015
0.00001 RA kmol m-2 s-1
0.000005
0 0.8 1 1.2 1.4 1.6 1.8 2 Stirrer speed (rev s-1)
Figure 5.9 Effect of stirrer speed rate on specific rate of extraction
5.4.2.2 Effect of phase volume ratio () on
The RA results shown in Figure 5.10are relatively constant over the volume ratio range used in this work (0.8 to 1.2) indicating that the regeneration of levulinic acid is occurring in a narrow region adjacent to the interface which confirms that the process is taking place in regime 3 (fast regime).
140 | Page 0.000025
0.00002
0.000015
0.00001 RA kmol m-2 s-1
0.000005
0 0.7 0.9 1.1 1.3 Vaq/Vorg
Figure 5.10 Effect of volume ratio on specific rate of extraction ( )
5.4.2.3 Order of reaction with respect to levulinic acid concentration and
TMA concentration
The order of the reaction involved in the TMA method was estimated by keeping the concentration of one component constant and varying the concentration of the other
at a constant stirrer speed for of 1.0. Experiments were carried out with ( ) loaded-organic phase with different levulinic acid concentrations ranging from 0.0578 to 0.313 kmol m-3. The specific rate of extraction is found to increase linearly with increasing acid concentrations in the aqueous phase as shown in Figure 5.11, which indicates that the order of the reaction for levulinic acid is one. Similarly, the value of specific rate constant was measured as a function TMA concentration (Figure 5.12).
The concentration of TMA was varied from 0.046 to 0.229 kmol m-3 while keeping the acid concentration in loaded-organic phase at 0.313 kmol m-3 at a stirrer speed
141 | Page of 1.33 rps. The order of the reaction for this case is found to be 0.5 by regression analysis as can be seen from Figure 5.12.
2.00E-05 1.80E-05 1.60E-05 1.40E-05 1.20E-05 1.00E-05 8.00E-06 RA kmol m-2 s-1 6.00E-06 4.00E-06 2.00E-06 0.00E+00 0 0.05 0.1 0.15 CA* aq kmol m-3
Figure 5.11Effect of levulinic acid concentration in aqueous phase on specific rate of extraction
2.50E-05
2.00E-05
1.50E-05
1.00E-05
RA kmol m-2 s-1 5.00E-06
0.00E+00 0 0.1 0.2 0.3 [TMA] kmol m-3
Figure 5.12 Effect of TMA concentration in aqueous phase on the specific rate of extraction
Thus, the overall order of the reaction with respect to levulinic acid and TMA is found to be 1.5 from Figures 5.12 and 5.13. The rate constant k1.5, was obtained by plotting
142 | Page the concentration of levulinic acid and TMA against the rate of extraction ( a) and fitting a liner relationship for the data shown in Figure 5.13. The value of the reaction rate constant is found to be 0.0317(m3) -0.5 kmol-0.5 s-1.
0.0025
0.002
0.0015
0.001 RA a kmol m-3 s-1
0.0005
0 0 0.02 0.04 0.06 0.08 C*A aq × [TMA]00.5
Figure 5.13 Determination of reaction rate constant, k1.5
5.4.3 Evaporation of TMA
Evaporation experiments were carried out using the aqueous phase, which contains acid and TMA solution in a rotary evaporator. The aim of the process was to generate the concentrated levulinic acid solution by removing all TMA and water.
The aqueous phase containing acid and TMA was heated at 104-140 0C under atmospheric pressure for 30 minutes. The residue remaining in the evaporator was analysed using HPLC to determine the percentage recovery of levulinic acid which was found to be 80%. The evaporated TMA solution was collected in a condensate collection tank. The condensate can be reused to regenerate levulinic acid.
5.4.4 Economic evaluation of the reactive extraction process scale-up
143 | Page An economic analysis was carried out to determine the financial viability of the reactive extraction process, including both forward extraction and regeneration. The analysis was carried out based on the results of our lab-scale experiments for the design and operation of a facility with a capacity of 2 m3 h-1. The process flow diagram used in the analysis is shown in Figure 5.14. It is assumed that the plant is to be built to recover value added carboxylic acids that are present in low concentrations (<10 %) in the fermentation broth.
Figure 5.14 Schematic diagram for scaled-up reactive extraction process to recover levulinic acid from the aqueous stream.
5.4.4.1 Capital investment
The major equipment required for the reactive extraction unit and their costs are listed in Table 5.2. The capital cost of the unit is estimated to be approximate $US
9,817 based on 2001 and 2016 plant cost indices of 397 and 556.8, respectively
144 | Page (Turton et al. 2008). Storage tanks are not included in the capital cost estimation because they are normally expected to be available in the plant.
Table 5.2List of equipment and their capital costs in the reactive extraction plant with a capacity of 2 m3 h-1to recover levulinic acid
Item Capacity Quantity Total cost (USD)
Feed pumps (lpm) 35 4 673
Aqueous phase pump (lpm) 35 1 168
Vacuum pump (lpm) 35 1 4,207
Condenser pump 1 140
Vacuum gauge 1 140
Stirred vessel 1 1,683
Evaporator 1 2,805
Total capital cost 9,817
5.4.4.2 Operating cost and maintenance cost
The operating cost including raw material and power costs are shown in Table 5.3.
The assumptions made in the operating cost estimation include an acid concentration of 7 wt % in the loaded-organic phase and an acid recovery of 80% in the regeneration. The plant is assumed to operate 20 hours day-1. The interest cost is assumed to be 5% of the total capital investment (TCI), and the annual depreciation is taken as 10% TCI.
Table 5.3 Operating cost of the reactive extraction plant with a capacity of 2 m3 h-1 for the recovery of levulinic acid
145 | Page Operating cost
Loaded-organic phase feed capacity (m3 h-1)2.0
Acid in the feed (7%) (m3 h-1)0.14
Recovery (80%) (m3 h-1) 0.112
Power cost
Feed pumps (4) (kW) 1.48
Aqueous phase pump (kW) 0.74
Condenser pump (kW) 0.37
Vacuum brandt (kW) 0.4
Stirrer motor (kW) 1.5
Evaporator (kW) 20
Total power consumption (kW) 24.29
Price per unit 0.1
Total power cost (USD) 2.43
Raw material cost
TMA per liter (USD) 2.64
TMA required per hour (L) 100
TMA loss per hour (L) 100
No. of working hours per day 20
Total TMA required per day (L) 500
TMA required per year (L) 1,80,000
MIBK required (L) 4,200
MIBK cost (USD) 55,588
TOA required 600
146 | Page TOA cost (USD) 6,93,529
Labour cost 1.0
No. of working hours per day 20
Total labour cost per day 20
Total labour cost per year 7,200
Total operating cost per year (USD) 12,49,395
Depreciation cost (10% of capital cost) (USD) 552
Interest on capital (5% of capital cost) (USD) 276
Utility cost (5% of capital cost) (USD) 276
Total cost per year (USD) 12,50,499
The total operating costs, annual revenue, and profit details are shown in Table
5.4.The production (recovery) cost is estimated to be $US1.55 per liter of levulinic acid recovered. The annual profit is estimated to be US$ 2,780,641 by determining the difference between the annual revenue due to sales of acid and the annual operating cost. The payback period is determined to be 0.49 years by dividing the annual profit by TCI.
Table 5.4 Evaluation of profitability of the reactive extraction plant with a capacity of
2 m3 h-1 for the recovery of levulinic acid
Recovery
Quantity of levulinic acid recovery per hour (L) 112
Operation time per day (h) 20.0
Quantity of levulinic acid generated per day (L) 2,240
Quantity of levulinic acid generated per year (L) 8,06,400
147 | Page Cost of levulinic acid recovered per liter (USD) 1.55
For a selling price of 5 USD per liter
Annual revenue (USD) 40,32,000
Annual operating cost (USD) 12,51,359
Annual profit (USD) 27,80,641
Pay out time (year) 0.49
5.5 Conclusions
Regeneration of levulinic acid was studied using different methods including NaOH, temperature swing, diluent swing, and TMA methods. In NaOH method, the equilibrium data show that the percentage acid recovery was high for all concentrations of acid in the loaded-organic phase. In temperature swing method, the recovery of acid was high at 343 K compared to other temperatures at all concentrations of acid in the loaded-organic phase. Recovery of acid was 89% at
343 K for 0.057 kmol m-3 of acid in loaded-organic phase. In diluent swing method, which involved the use of heptane and toluene as diluents, percentage acid recovery was higher for volume ratio 1:1 compared to volume ratio (1:0.5). Between the two diluents, the acid recovery for heptane was 73% higher than that for toluene.
Among all the regeneration methods, percentage recovery of acid was the highest for TMA method. For TMA to acid molar concentration ratio of 1 or above, the acid recovery from loaded-organic phase was greater than 80% for all concentrations of acid in the loaded-organic phase. A mass action model was developed for TMA method to relate the recovery of acid and TMA concentration. Kinetic studies performed for TMA method in a stirred cell indicated that the reaction regime is ‘fast
148 | Page regime.’ Changes in stirrer speed and phase volume ratio did not have any significant effects on the specific rate of extraction. It was concluded that the order of the reaction with respect to acid and TMA was 1.5. The rate constant was found to be 0.0317 (m3) -0.5 kmol-0.5 s-1. It was found that 80% of the acid in the TMA solution can be recovered from the aqueous phase by evaporation at 104-140 0C under atmosphere pressure. An economic evaluation revealed that 80% of levulinic acid can be recovered at the cost of 1.55 USD per liter with a payback period of 0.49 years for a loaded-organic phase rate of 2 m3 h-1.
149 | Page Chapter 6
Recovery of Multi-acids (volatile fatty acids) by Reactive Extraction using Tri-n-Octylamine and Tri- Butyl Phosphate in Different Solvents: Equilibrium Studies, pH and Temperature Effect, and Optimisation using Multivariate Taguchi Approach
150 | Page 6.1 Introduction
Liquid-liquid reactive extraction is an economical and feasible technique to separate the carboxylic acids and it has been used in pharmaceutical, food and fermentation industries to reduce the environmental contamination. There is a need to separate value added volatile fatty acids (VFAs) from aqueous effluent stream or fermentation broth because the use of VFAs is widespread in the pharmaceutical, food and biotechnological industries. In this work, the separation of volatile fatty acids from fermentation broth was carried out using reactive extraction i.e. process intensification method, using TBP and TOA as extractants and 1-decanol and MIBK as diluents to facilitate the desired recovery. VFAs such as formic acid (FA), acetic acid (AA), propionic acid (PA) and butyric acid (BA) are the short-chain fatty acids produced either synthetically from fossil resources or obtained as metabolic intermediates during fermentation. Recovery of VFAs from the broth is important as they are used in the production of important chemicals including alcohols, aldehydes, ketones, esters and olefins. Acetic acid and propionic acid, in particular, have widespread applications as taste-enhancing additives and preservatives in the food industry, as buffer solutions in the pharmaceutical industry and skin lightening or anti-acne agents in the cosmetic industry. Butyric acid is more often used as an animal feed supplement, fishing bait additives and food additives and flavouring
(Solichien et al. 1995; Zacharof & Lovitt 2013). Although the reactive extraction process is effective, it is used cautiously in the fermentation and food industries. It is possible that the extractants or diluents may be carried over in to the final food product thereby adulterating it. However, the concerns can be overcome by using effective techniques like evaporation and distillation to remove the traces of extractants using specialized process equipment or by effective process control.
151 | Page The major challenge in the production of VFAs by fermentation is their recovery from the resultant broths. Existing conventional processes for the recovery of organic acids from fermentation broths include stripping, adsorption, electrodialysis, direct distillation, solvent extraction, evaporation, chromatographic methods, ultrafiltration, reverse osmosis and drying. Each of these processes has its advantages and disadvantages. Current research is therefore aimed towards the selection of an intensified approach such as reactive extraction that would overcome the weaknesses of the existing processes. The literature cites a few examples for the recovery of carboxylic acids by reactive extraction from aqueous waste streams in cellulose and vinyl acetate manufacturing units. As the process is a combination of both chemical and physical phenomena, a higher distribution coefficient is achieved making the process a promising technique for the recovery of carboxylic acids
(Wardell & King 1978; Wennersten 1983). The efficiency of liquid-liquid extraction process for recovering carboxylic acids from aqueous streams is strongly influenced by the nature of the solvent used. Classical extractants (alcohols, ketones, aliphatic, aromatic, etc.) give low distribution coefficients, resulting in poor extraction efficiency of carboxylic acids. On the other hand, the use of aminic or organ-phosphoric compounds as extractants has yielded higher efficiency with better distribution coefficients compared to classical extractants.
Preferred extractants for reactive extraction of carboxylic acids from aqueous phase include long chain aliphatic primary, secondary and tertiary amines. Primary amines are characterised by a high solubility in the aqueous phase, while secondary amines have the highest distribution coefficient but tend to form amides in the downstream regeneration by distillation. Tertiary amines offer advantages over the other extractants on the grounds of its lower cost and generally higher equilibrium
152 | Page distribution coefficient KD (Juang, Huang & Wu 1997). Acid-amine complex stability is affected by the basicity of the amine, which can be altered using various diluents.
The majority of the literature is focused on the reactive extraction of single acid in the aqueous phase. The literature is scarce for the extraction of multi-acids from aqueous streams. Only very few reports are available on the extraction of multi- acids. Among them Juang et al(1997) studied the recovery of two acids in the aqueous phase using TOA dissolved in xylene and the recovery of multi-acids i.e.
VFAs (AA, BA, PA and VA) using TOPO dissolved in kerosene. Alkayya et al.
(Alkaya et al. 2009) have studied the recovery of VFAs using tri-octylphosphine oxide (TOPO) in kerosene and their results indicate the partition coefficient of VFAs is relatively low.
In this chapter, the potential of the reactive extraction technique is explored for the effective recovery of VFAs (AA, PA and BA) from an aqueous stream using TOA and
TBP dissolved in 1-decanol and MIBK. The aim of the extraction experiment is to assess the applicability of reactive extraction process to influence the distribution coefficient and extraction efficiency by varying the pH and the composition of extractants (TOA and TBP in 1-decanol and MIBK) on the recovery of VFAs from fermentation broth at different temperatures. Taguchi 36 mixed design was used to determine the optimum combination of factors for multivariate response characteristics based on the desirability approach. In this work, the influences of diluents type, 1 (1-decanol, MIBK), extractant type, 2 (TOA, TBP), extractant concentration, 3 (10, 20 and 30%), temperature, 4 (293.15, 303.15, 313.15 and
323.15 K) and pH, 5 (2.5, 3.5 and 4.5) on the reactive extraction of VFAs were studied.
153 | Page 6.2 Materials and Methods
6.2.1 Chemicals and Reagents
Glacial acetic acid (AA) (mass fraction purity> 0.995) was supplied by Molychem,
Mumbai, India. Propionic acid (PA) and butyric acid (BA) of mass fraction purity >
0.99 were procured from Sigma Aldrich. TOA (mass fraction purity > 0.98) and TBP
(mass fraction purity > 0.97) supplied by Sigma Aldrich were used as extractants. 1-
Decanol (mass fraction purity > 0.99, density-0.819 g ml-1) and MIBK (mass fraction purity > 0.99, density- 0.802 g ml-1) supplied by S. D. Fine-Chem Ltd., Mumbai, India were used as diluents. Phenolphthalein indicator (pH 8.2-9.8) was supplied by S. D.
Fine-Chem. Ltd., Mumbai, India. Sodium hydroxide used in the experiments was of reagent grade. Hydrochloric acid (HCl) (mass fraction purity> 36.5 supplied by
Molychem, Mumbai, India) of 0.1N was prepared by dissolving an appropriate amount in de-mineralized water. All the chemicals were used without further purification. De-mineralized (DM) water was used for all the experiments.
6.2.2 Preparation of VFA Solution
The synthetic mixture of acids was prepared based on the composition of fermentation broth, which was collected from Bio-chemical and environmental lab
CSIR-IICT, Hyderabad, India a facility that uses bio production to produce volatile fatty acids. Fermentation broth samples were collected at regular intervals over a period of 20 days and the collected samples were analysed many times to check the reproducibility of composition. The average composition was then used to prepare the synthetic mixture. The synthetic mixture was prepared by dissolving 8.26 g. L-1 of
AA, 3.08 g. L-1 of PA and 2.66 g. L-1 of BA in de-mineralized (DM) water to mimic the
154 | Page real composition of fermentation broth. The extraction of acids from the aqueous stream into the organic stream may not be affected by the by-products of fermentation and biomass because the solubility of acids in the organic phase
(solvent + extractant) and affinity of acids towards the bases (extractants) are relatively higher than those of by-products.
6.2.3 Instrumentation
The acid composition was measured using high-performance liquid chromatography
(HPLC) (Model: Shimadzu SPD-M20A) equipped with UV detector. The equilibrium studies were carried out using shaking incubator (Model No.-LSI4018R) provided by
Daihan Labtech India Pvt. Ltd., capable of maintaining the temperature within ±0.1 K.
The pH of the solution was measured using Sartorius PB-11 pH meter. A Radleys-
RR98072 magnetic stirrer was used for stirring the VFA mixture. A high precision
Citizen balance (accuracy of ±0.0001g) was used for weighing the accurate amount of acids.
6.2.4 Reactive extraction experiments
All reactive extraction experiments were conducted in batch mode. The experimental procedure used is detailed in the following steps; (1) Preparation of aqueous solutions by diluting acids to the total concentration of 14 g L-1 (containing AA, PA and BA in the concentrations of 59, 22 and 19vol%, respectively) with DM water. (2)
The total concentration of the acids in the aqueous solution was evaluated by APHA method. (3) The individual acid concentration was determined by HPLC method. (4)
A fixed quantity of organic phase (containing extractants TOA or TBP mixed with diluents 1-decanol or MIBK) was added to the aqueous solution. (5) Aqueous and organic phases were mixed in a conical flask and kept under shaking motion at 200
155 | Page rpm for 5 hours, which was determined to be a suitable equilibrium time from preliminary evaluations. (6) The solution was then allowed to settle for 3 hours during which samples were withdrawn periodically to analyze and ensure the attainment of equilibrium (7) After the extraction of acids into the organic layer, the total concentration of acid in the aqueous layer was determined using the APHA method.
(8) Recovery of individual acids was determined by measuring their concentrations in the aqueous layer using HPLC method.
In reactive extraction experiments, the main focus was to determine the recovery of acids by the extractants, whereas the main focus of the equilibrium study was to determine the distribution of acids between the aqueous and the organic phases.
Experimental results were used to determine distribution coefficient ( ) and extraction efficiency (% E). The distribution coefficient is expressed by equation
(6.1)
∗ CA,org KD = ∗ (6.1) CA,aq
The degree of extraction (% ) is expressed by equation (6.2)
KD % E = × 100 (6.2) (1 + KD)
6.2.4.1 Equilibrium studies
This section on equilibrium studies is a subsection of reactive extraction studies. The equilibrium studies were carried out by mixing 20 ml aqueous solution containing 14 g.L-1 of VFAs with equal volume organic phase (containing TOA or TBP (v/v) in 1- decanol or MIBK). The mixture was kept at a constant temperature (chosen in the range of 293.15-333.15K) in a shaking incubator for 5 hours at 200 rpm, and then
156 | Page allowed to settle for 3 hours before separating the two separate layers. The total acid
∗ concentration ( , ) in the aqueous phase was determined by titration. The
∗ corresponding total VFA concentration in the organic phase ( , ) was determined by mass balance. Then, equation (6.1) was used to determine the value. Figures
6.1-6.6 show the experimental data. These values are also reported in Table 6.3. All of these experiments were repeated twice and the standard deviation in experimental results was observed to be ±1.
6.2.5 Optimization of Reactive Extraction Process: Taguchi L36 Mixed Design
A Taguchi design was employed to determine the combination of optimum parameters, which influence the response characteristics the most. In the present study, a Taguchi ( 36) mixed array design consisting mixed levels of factors as listed in Table 6.1 was used for the optimisation of process parameters in the reactive extraction of volatile fatty acids (VFAs) from aqueous solution. The process parameters that would influence the multi-response variables were identified to be the extractant type ( 1) , diluents type ( 2) , extractant concentration ( 3 ), temperature ( 4 ) and pH of the aqueous solution ( 5 ). The distribution coefficient (
) and extraction efficiency (% ) were selected as the response variables. In Taguchi method, signal (S) denotes the desired characteristics and noise (N) represents the undesired value. S/N ratio is the log transformation of mean square deviation of desired performance, which measures the response characteristics deviating from the desired value (Daneshvar et al. 2007; Radhakumari et al. 2014). An appropriate
S/N ratio: (larger-the-better) was employed as the criterion. Thus, Taguchi ( 36) mixed design and larger-the-better approach was adopted in this chapter so as to maximise and % . The selected experimental factors with their respective levels
157 | Page suggested that 36 runs in total need to be carried out to study the influence of these factors on response variables. The details of design matrix along with their experimental results are presented in Table 6.2. For the criteria opted, the equation representing the response variable ( ) for observations is shown below:
1 1 = ‒ 10 [ ] (6.3) 10 ∑ 2 =1
Table 6.1 36 Factors along with their levels by Taguchi mixed design
Levels
Factors 123
Diluent type ( 1) 1-decanol MIBK --
Extractant type ( 2) TOA TBP --
Extractant concentration 10% 20% 30% ( 3)
Temperature (K) ( 4 ) 293.15 303.15 323.15
pH ( 5 ) 2.5 3.5 4.5
Table 6.2 36 mixed array (levels of five different factors and obtained results).
Exp. 1 2 3 4 5 ratio (db) % ratio (db) No
158 | Page 1 1 1 1 1 1 4.28 12.63 81.06 38.18 0.62
2 1 1 2 2 2 7.33 17.78 87.99 38.95 1.00
3 1 1 3 3 3 2.35 7.42 70.14 36.92 0.36
4 1 1 1 1 1 4.28 12.63 81.06 38.18 0.62
5 1 1 2 2 2 7.33 17.78 87.99 38.95 1.00
6 1 1 3 3 3 2.35 7.42 70.14 36.92 0.36
7 1 1 1 1 2 6.06 15.65 85.84 38.67 0.81
8 1 1 2 2 3 2.44 7.75 70.90 37.01 0.37
9 1 1 3 3 1 5.06 14.08 83.50 38.43 0.71
10 1 2 1 1 3 1.00 -0.03 49.90 33.96 0.07
11 1 2 2 2 1 1.21 1.66 54.75 34.77 0.13
12 1 2 3 3 2 1.17 1.36 53.92 34.63 0.12
13 1 2 1 2 3 0.90 -0.92 47.37 33.51 0.05
14 1 2 2 3 1 1.04 0.34 50.98 34.15 0.09
15 1 2 3 1 2 1.82 5.20 64.54 36.20 0.26
16 1 2 1 2 3 0.90 -0.92 47.37 33.51 0.05
17 1 2 2 3 1 1.04 0.34 51.05 34.15 0.09
18 1 2 3 1 2 1.82 5.20 64.54 36.20 0.26
19 2 1 1 2 1 2.14 6.61 68.15 36.67 0.32
159 | Page 20 2 1 2 3 2 1.92 5.67 65.75 36.36 0.28
21 2 1 3 1 3 1.85 5.34 64.95 36.25 0.27
22 2 1 1 2 2 2.16 6.69 68.35 36.70 0.32
23 2 1 2 3 3 1.25 1.94 55.56 34.89 0.14
24 2 1 3 1 1 2.61 8.33 72.30 37.18 0.40
25 2 1 1 3 2 1.86 5.39 65.03 36.26 0.27
26 2 1 2 1 3 1.76 4.91 63.77 36.09 0.25
27 2 1 3 2 1 2.42 7.68 70.76 37.00 0.37
28 2 2 1 3 2 1.34 2.54 57.26 35.16 0.16
29 2 2 2 1 3 1.11 0.91 52.61 34.42 0.10
30 2 2 3 2 1 1.80 5.11 64.32 36.17 0.26
31 2 2 1 3 3 0.85 -1.41 45.95 33.24 0.00
32 2 2 2 1 1 1.57 3.92 61.12 35.72 0.21
33 2 2 3 2 2 1.82 5.20 64.60 36.20 0.26
34 2 2 1 3 1 1.32 2.41 56.93 35.11 0.15
35 2 2 2 1 2 1.63 4.24 62.00 35.85 0.22
36 2 2 3 2 3 1.14 1.14 53.35 34.54 0.11
1: Diluent type, 2 :Extractant type, 2 : Extractant concentration, 3 : Temperature, 3 : pH, : Overall desirability. where n is the number of replicate runs and is the response variable. The unit of
S/N ratio is decibel (dB), often used in communication engineering. The S/N ratio
160 | Page values calculated using equation (6.3) for individual responses is reported in Table
6.4.
Table 6.4 Response Table for signal-to-noise ratios ( ) - Larger is better.
Factors
Levels 1 2 3 X4 X5
% % % KD %E KD %E
1 6.83 36.28 8.52 37.04 4.86 35.70 6.11 36.25 6.27 36.34
2 4.25 35.77 2.04 34.88 4.91 35.82 5.77 36.10 6.97 36.50
3 6.09 36.35 3.97 35.52 2.61 35.04
∆ ‒ 2.57 0.47 6.47 2.16 1.22 0.66 2.13 0.74 4.36 1.46
Rank( ) 315 4 2
Rank(% ) 514 3 2
∆ ‒ : Range is the difference between maximum and minimum levels of factors.
6.2.5.1 Desirability approach
Desirability approach is an established and useful technique for the simultaneous optimisation of factors, which verifies the optimum performance of multiple responses. In this approach, each response is converted into desirability function that varies from 0 to 1 (lowest to highest). The numerical multivariate optimisation has been carried out to identify the combination of factors that maximises the overall desirability of the process. In the present study, the desirability function for two response characteristics ( and % ) was calculated for satisfying the larger-the- better (LTB) criterion according to the following equation:
161 | Page If ≤ =0
‒ If ≤ ≤ = (6.4) ( ‒ )
If ≥ = 1
where indicates the response characteristic, and are the lower and upper
ℎ tolerance limits of response characteristics, respectively, is the desirability of response and is the desirability index. In a multi-response condition, the ideal case is achieved if the desirability of all responses equals to one, and the case is unacceptable if the desirability of response reaches 0. Therefore, the overall desirability is computed as the geometrical mean of all the individual desirability ( ) according to equation (6.5)
1 = ( 1 2……. ) (6.5) where D is the overall desirability and is the total number of responses. MINITAB
17 was used for the design of experiments, analysis of variance analysis (ANOVA) and optimisation of the extraction process. ANOVA was performed to study systematically the relative importance of factors, which indicates the contribution of each factor on responses relative to the total contribution.
6.2.6 Analytical Method
Individual carboxylic acid (AA, PA and BA) concentration was analyzed using HPLC
(Model: Shimadzu SPD-M20A) method equipped with C18 reverse phase column with dimensions of 250 x 4.6 mm and specifications including particle size of 5 μm, flow rate of 0.8 ml min-1 and wave length of 210 nm. The mobile phase consisted of
162 | Page 75/25 acetonitrile/water and 0.1 N H2SO4 at a pH of 2.5. A sample injection volume of 20 μl was used. The overall acid concentration in the aqueous phase was analyzed using APHA method (Federation & Association 2005). Aqueous sample (5 ml) was diluted to 100 ml by adding DM water and the mixture was titrated with 0.1 N
HCl up to the end point with pH of 3.0 and then titrated with 0.1 N NaOH up to the end point with pH of 6.5.
6.3 Results and Discussion
The reactive extraction data for a system involving aqueous solution of VFAs (14 g.L-
1) and organic phase involving extractants (TOA or TBP) and diluents (1-decanol or
MIBK) are presented in Table 6.3for pH ranging from2.5 to 4.5 and temperature ranging from 293.15 to 323.15 K. The results are presented in terms of distribution coefficient ( ) and extraction efficiency (% ). In the following sections, the influences of different parameters are discussed and the optimisation of the reactive extraction process is explained.
6.3.1 Effects of Diluent and Extractant
The present study was carried out with two types of diluents and extractants in the extraction of VFAs (AA, PA and BA). The diluents, 1-decanol and MIBK, belong to alcohol and ketone groups, respectively. The extractants investigated in the present study are TOA and TBP. Distribution coefficient ( ) and extraction efficiency (% ) for both the diluents without extractant and with different concentrations of extractants are shown in Figure 6.1(a).
163 | Page 35 0.6 (a) T=303.15 K, pH=3.5 34 % E 0.5 33 KD 0.4 32
31 0.3 KD % E
30 0.2 29 0.1 28
27 0 1-decanol MIBK
100 9 (b) T=303.15 K, pH=3.5 % E 90 8 KD 80 7 70 6 60 5 % E
50 KD 4 40 3 30 2 20
10 1
0 0 30 % TOA in 30 % TBP in 30 % TOA in 30 % TBP in 1-decanol 1-decanol MIBK MIBK
164 | Page Table 6.3 Individual distribution coefficients of AA, PA and BA, overall distribution coefficients and extraction efficiency % E at different pH and temperatures.
pH=2.5
293.15 K 303.15 K 313.15 K 323.15 K
10 20 30 10 20 30 10 20 30 10 20 30
AA-2.99 AA-5.55 AA-6.16 AA-2.33 AA-4.02 AA-4.53 AA-1.94 AA-3.12 AA-3.52 AA-1.74 AA-2.92 AA-3.31 TOA in individual PA-6.68 PA-11.96 PA-13.10 PA-4.05 PA-9.48 PA-11.29 PA-4.62 PA-6.95 PA-7.81 PA-4.82 PA-6.12 PA-8.49 1- KD BA-24.34 BA-46.25 BA-53.33 BA-24.18 BA-35.06 BA-45.46 BA-18.36 BA-31.04 BA-37.93 BA-18.09 BA-33.85 BA-37.72 decanol Overall KD 4.27 8.01 8.86 3.38 5.96 6.77 3.01 4.67 5.24 2.80 4.32 5.06
Efficiency 81.06 88.91 89.85 77.19 85.63 87.13 75.07 82.37 83.98 73.70 81.34 83.50
AA-0.73 AA-0.79 AA-0.98 AA-0.73 AA-0.78 AA-0.915 AA-0.47 AA-0.56 AA-0.61 AA-0.45 AA-0.50 AA-0.57
TBP in individual PA-2.74 PA-2.75 PA-3.07 PA-0.92 PA-1.69 PA-2.65 PA-1.52 PA-1.72 PA-2.03 PA-1.66 PA-1.86 PA-2.16
1- KD BA-10.38 BA-11.11 BA-17.83 BA-3.41 BA-3.84 BA-5.71 BA-7.10 BA-8.26 BA-9.5 BA-6.85 BA-7.65 BA-9.06 decanol Overall KD 1.40 1.53 1.76 1.00 1.21 1.52 0.96 1.08 1.18 0.94 1.04 1.16
Efficiency 58.36 60.61 63.80 50.17 54.90 60.34 49.01 52.12 54.32 48.66 51.04 53.72
AA-1.28 AA-1.47 AA-1.51 AA-1.24 AA-1.34 AA-1.38 AA-1.05 AA-1.12 AA-1.15 AA-1.03 AA-1.09 AA-1.12
individual PA-3.68 PA-4.20 PA-4.62 PA-3.44 PA-3.78 PA-4.21 PA-3.07 PA-3.11 PA-3.12 PA-3.34 PA-3.45 PA-3.37 TOA in KD BA-50.62 BA-54.20 BA-57.08 BA-28.04 BA-31.11 BA-57.08 BA-14.75 BA-13.52 BA-13.79 BA-14.31 BA-13.01 BA-12.8- MIBK Overall KD 2.23 2.51 2.60 2.13 2.29 2.41 1.83 1.90 1.94 1.82 1.89 1.92
Efficiency 69.07 71.58 72.29 68.07 69.60 70.70 64.66 65.59 66.0 64.64 65.48 65.82
165 | Page AA-0.76 AA-0.86 AA-1.08 AA-0.76 AA-0.858 AA-1.01 AA-0.73 AA-0.85 AA-1.01 AA-0.69 AA-0.81 AA-0.96
individual PA-2.70 PA-2.81 PA-3.47 PA-2.49 PA-2.82 PA-3.33 PA-2.20 PA-2.54 PA-2.81 PA-2.37 PA-2.71 PA-3.23 TBP in KD BA-10.17 BA-11.23 BA-14.06 BA-10.35 BA-11.28 BA-12.90 BA-10.38 BA-11.28 BA-13.08 BA-9.8 BA-10.06 BA-12.6 MIBK Overall KD 1.44 1.57 1.90 1.41 1.56 1.80 1.34 1.52 1.74 1.32 1.48 1.73
Efficiency 59.02 61.11 65.56 58.54 60.96 64.32 57.35 60.44 63.59 56.93 59.77 63.45
pH=3.5
293.15 K 303.15 K 313.15 K 323.15 K
10 20 30 10 20 30 10 20 30 10 20 30
AA-4.38 AA-6.01 AA-7.19 AA-3.28 AA-5.01 AA-5.60 AA-2.43 AA-3.71 AA-4.15 AA-1.838 AA-2.83 AA-3.14 TOA in individual PA-7.57 PA-13.12 PA-16.01 PA-7.72 PA-11.35 PA-13.19 PA-5.68 PA-8.77 PA-10.44 PA-4.58 PA-6.49 PA-8.47 1- KD BA-28.55 BA-47.36 BA-60.86 BA-33.10 BA-43.3 BA-55.59 BA-2.29 BA-38.11 BA-45.66 BA-21.73 BA-30.66 BA-37.55 decanol Overall KD 6.06 8.65 10.36 4.93 7.32 8.24 3.72 5.55 6.26 2.91 4.28 4.862
Efficiency 85.83 89.64 91.2 83.16 87.99 89.17 78.84 84.74 86.2 74.44 81.08 82.95
AA-0.82 AA-0.88 AA-0.91 AA-0.80 AA-0.92 AA-0.94 AA-0.506 AA-0.53 AA-0.58 AA-0.49 AA-0.53 AA-0.585
TBP in individual PA-2.93 PA-3.26 PA-5.28 PA-3.18 PA-3.19 PA-3.52 PA-1.98 PA-2.04 PA-2.40 PA-1.78 PA-1.86 PA-2.083
1- KD BA-5.63 BA-8.81 BA-15.62 BA-4.28 BA-5.22 BA-12.03 BA-7.61 BA-8.01 BA-9 BA-7.44 BA-8.81 BA-9.35 decanol Overall KD 1.45 1.60 1.82 1.41 1.56 1.72 1.05 1.08 1.19 1.02 1.07 1.16
Efficiency 59.25 61.6 64.57 58.50 60.96 63.33 51.2 52.1 54.35 50.50 51.87 53.81
AA-1.54 AA-1.60 AA-1.61 AA-1.22 AA-1.30 AA-1.40 AA-1.06 AA-1.12 AA-1.15 AA-1.03 AA-1.18 AA-1.22 TOA in individual PA-4.31 PA-4.61 PA-4.76 PA-4.62 PA-5.474 PA-.23 PA-3.24 PA-3.51 PA-3.59 PA-3.93 PA-2.81 PA-2.87 MIBK KD BA-16.97 BA-18.13 BA-20.11 BA-15.73 BA-17.22 BA-18.41 BA-13.37 BA-13.86 BA-14.2 BA-11.66 BA-11.78 BA-1.91
166 | Page Overall KD 2.52 2.62 2.66 2.16 2.32 2.50 1.84 1.95 1.99 1.86 1.92 1.96
Efficiency 71.6 72.42 72.73 68.36 69.92 71.44 64.87 66.10 66.62 65.03 65.79 66.32
AA-0.79 AA-0.89 AA-1.09 AA-0.79 AA-0.83 AA-1.01 AA-0.71 AA-0.83 AA-0.95 AA-0.70 AA-0.81 AA-0.95
individual PA-2.68 PA-3.13 PA-3.66 PA-2.54 PA-3.26 PA-3.55 PA-2.38 PA-2.80 PA-3.49 PA-2.38 PA-2.80 PA-3.49 TBP in KD BA-10.71 BA-11.97 BA-14.11 BA-8.5 BA-11.03 BA-14.11 BA-10.56 BA-11.66 BA-3.1 BA-10.56 BA-11.66 BA-13.14 MIBK Overall KD 1.46 1.63 1.93 1.43 1.57 1.82 1.34 1.53 1.75 1.34 1.51 1.74
Efficiency 59.50 62 65.91 58.86 61.1 64.60 57.37 60.46 63.63 57.28 60.14 63.57
pH=4.5
293.15 K 303.15 K 313.15 K 323.15 K
10 20 30 10 20 30 10 20 30 10 20 30
AA-1.55 AA-2.18 AA-2.28 AA-1.32 AA-1.52 AA-1.67 AA-1.32 AA-1.52 AA-1.58 AA-1.03 AA-1.40 AA-1.43 TOA in individual PA-4.01 PA-4.43 PA-4.61 PA-3.49 PA-3.99 PA-4.11 PA-2.95 PA-3.57 PA-4.03 PA-2.69 PA-3.76 PA-3.91 1- KD BA-16 BA-20.53 BA-23.8 BA-14.10 BA-15.15 BA-16.80 BA-14.92 BA-17.38 BA-20.53 BA-11.35 BA-14.38 BA-17.08 decanol Overall KD 2.48 3.24 3.39 2.19 2.43 2.62 2.10 2.41 2.54 1.72 2.28 2.34
Efficiency 71.3 76.46 77.26 68.70 70.89 72.41 67.75 70.68 71.78 63.34 69.51 70.13
AA-0.575 AA-0.59 AA-0.71 AA-0.36 AA-0.37 AA-0.40 AA-0.414 AA-0.43 AA-0.45 AA-0.32 AA-0.35 AA-0.359
TBP in individual PA-1.43 PA-1.45 PA-1.54 PA-2.56 PA-2.81 PA-2.90 PA-1.05 PA-1.25 PA-1.30 PA-1.20 PA-1.21 PA-1.41
1- KD BA-4.50 BA-5.09 BA-6.27 BA-5.33 BA-6.27 BA-7.53 BA-5.85 BA-6.56 BA-7.04 BA-5.91 BA-6.06 BA-6.69 decanol Overall KD 0.99 1.03 1.18 0.90 0.94 0.99 0.808 0.86 0.89 0.739 0.77 0.81
Efficiency 49.9 50.79 54.16 47.37 48.46 49.79 44.69 46.34 47.32 42.51 43.5 44.81
TOA in individual AA-1.02 AA-1.06 AA-1.16 AA-0.703 AA-0.74 AA-0.76 AA-0.68 AA-0.70 AA-0.72 AA-0.68 AA-0.70 AA-0.72
167 | Page MIBK KD PA-2.50 PA-2.57 PA-2.37 PA-2.37 PA-2.43 PA-2.65 PA-1.99 PA-2.01 PA-2.04 PA-1.79 PA-1.88 PA-1.92
BA-10.36 BA-12.2 BA-14.59 BA-8.0 BA-9.37 BA-9.87 BA-8.42 BA-8.96 BA-9.18 BA-8.42 BA-8.70 BA-8.83
Overall KD 1.68 1.75 1.85 1.30 1.37 1.41 1.24 1.27 1.29 1.22 1.255 1.27
Efficiency 62.7 63.74 64.95 56.53 57.80 58.6 55.35 56.11 56.51 54.99 55.67 56.08
AA-0.52 AA-0.59 AA-0.62 AA-0.50 AA-0.55 AA-0.59 AA-0.46 AA-0.49 AA-0.54 AA-0.47 AA-0.52 AA-0.55
individual PA-1.49 PA-1.70 PA-1.87 PA-1.37 PA-1.73 PA-1.90 PA-1.22 PA-1.86 PA-1.90 PA-0.99 PA-1.09 PA-1.57 TBP in KD BA-6.98 BA-7.83 BA-11.42 BA-6.22 BA-7.79 BA-8.70 BA-5.28 BA-8.19 BA-8.62 BA-5.2 BA-5.63 BA-8.70 MIBK Overall KD 1.0 1.11 1.18 0.95 1.06 1.14 0.88 1.02 1.08 0.85 0.91 1.45
Efficiency 50.02 52.64 54.28 48.76 51.62 53.35 46.77 50.60 52.03 45.98 47.90 51.11
168 | Page 100 8 (c) T=303.15 K, pH=3.5 % E 90 KD 7 80 6 70 5 60
50 4 KD % E
40 3 30 2 20 1 10
0 0 20 % TOA in 20 % TBP in 20 % TOA in 20 % TBP in 1-decanol 1-decanol MIBK MIBK
90 6 (d) T=303.15 K, pH=3.5 % E 80 KD 5 70
60 4
50 3 KD % E 40
30 2
20 1 10
0 0 10 % TOA in 10 % TBP in 10 % TOA 10 % TBP in 1-decanol 1-decanol in MIBK MIBK
Figure 6.1 A plot of distribution coefficient , extraction efficiency (% ) (a) for 1- decanol and MIBK, (b) 10, (c) 20, and (d) 30 % composition of extractants in 1- decanol and MIBK diluents. Temperature = 303.15 K, pH = 3.5.
169 | Page Tertiary amines (TOA) and organo-phosphorous (TBP) compounds are effective extractants and provide higher distribution coefficient and extraction efficiency than carbon-bonded oxygen-bearing extractants. The chemical stability of the chosen extractants plays an important role in determining the efficiency of the extracting solution enabling good separation with solutions containing chemically similar elements. TOA and TBP compounds contain amine and phosphoryl groups respectively, which are stronger Lewis bases than other extractants like carbonyl group, which enable TOA and TBP lead to higher extraction efficiency. TOA has a higher viscosity (8.39 mPa.s) than TBP (3.39 mPa.s), but when used with diluents, it leads to good separation. TOA in 1-decanol shows the highest value of distribution coefficient ( = 8.24) with an extraction efficiency (% ) of 89.17%. The and % values for TBP in 1-decanol are found to be 1.72 and 63.25%, respectively. From
Figures 6.1b, 6.1c and 6.1d, it can be observed that the capability of TBP in extracting volatile fatty acids is lower than that of TOA. It is also clear that 1-decanol has a greater influence on the extraction efficiency than MIBK. Among the four organic systems used in this work to extract VFAs, the combination of TOA in 1- decanol has the highest extraction efficiency followed by TOA, MIBK, TBP in MIBK, and then TBP in 1-decanol (Figures 6.1b, 6.1c and 6.1d). This trend can be attributed to the higher solubilities of VFAs in 1-decanol compared to those in MIBK.
Although, 1-decanol-TOA mixture has a higher viscosity compared to other systems, the effect of viscosity on mass transfer would have been overcome by the turbulent flow conditions produced during the operation of shaking incubator at 200 rpm which could lead to an increase in the convective mass transfer coefficient and the generation of fine droplets which increase the interfacial area for mass transfer
170 | Page 6.3.2 Effect of Extractant Concentration
The reactive extraction experiments were carried out with 10, 20, and 30 % extractants (TOA and TBP) dissolved in 1-decanol and MIBK. The concentrations of extractants in the extractant-diluent mixture were restricted to 30% due to high viscosities of TOA and TBP. Higher concentration of extractant in the organic phase was found to result in the formation of three phase mixture during the extraction process. The and % values obtained for the extraction of total acids and individual acid using the four extractant-diluent systems are shown in Figure 6.2.For all four systems, the and % values are higher for butyric acid followed by propionic acid and then acetic acid. Also, the and % values increase with increasing TOA and TBP concentrations in the organic phase regardless of the diluent used. Among the four systems studied, TOA+1-decanol has the highest and % values followed by TOA+ MIBK, TBP + MIBK and TBP + 1-decanol.
The extraction efficiency usually depends on the following factors:
1. The association ability between the acid and extractant, (pKb – pKa.)
2. The hydrophobicity of the solute, (logP)
3. The steric effect between the solute and the extractant (Li, Qin & Dai 2002).
The association ability and hydrophobicity values for the three acids are as follows: pKa = 4.75 and log P = -0.33for acetic acid, pKa= 4.67 and log P = 0.33for propionic acid, and pKa= 4.82 and log P = 0.79for butyric acid. On analysing these values along with the data shown in Table 6.3, it can be stated that the higher and % observed for butyric acid are due to its hydrophobic nature.
171 | Page 120.00 70 (a) TOA in 1-decanol at T= 293.15 K, pH=3.5
% E 60 100.00 KD 50 80.00 40
60.00 30 KD % E
20 40.00 10 20.00 0
0.00 -10 AA PA BA AA PA BA AA PA BA 10 % 20 % 30 % TOA TOA TOA
100.00 18 (b) TBP in 1-decanol at T= 293.15 K, pH=3.5 90.00 % E 16 80.00 KD 14 70.00 12 60.00 10 50.00 % E 8 KD 40.00 6 30.00
20.00 4
10.00 2
0.00 0 AA PA BA AA PA BA AA PA BA 10 % 20 % 30 % TBP TBP TBP
172 | Page (c) 120.00 25 TOA in MIBK at T= 293.15 K, pH=3.5
100.00 % E 20 KD
80.00 15
60.00 KD
% E 10 40.00
5 20.00
0.00 0 AA PA BA AA PA BA AA PA BA 10 % 20 % 30 % TOA TOA TOA
(d) 100.00 TBPin MIBK at T= 293.15 K, pH=3.5 16 90.00 % E 14 80.00 KD 12 70.00 10 60.00
50.00 8 KD % E 40.00 6 30.00 4 20.00 2 10.00
0.00 0 AA PA BA AA PA BA AA PA BA 10 % 20 % 30 % TBP TBP TBP
Figure 6.2 Distribution coefficient ( ) and extraction efficiency (% ) for different organic phases. (a) TOA+1-decanol, (b) TBP+1-decanol, (c) TOA+MIBK, and (d)
TBP+MIBK
173 | Page 6.3.3 Effect of Temperature
In the large-scale production of carboxylic acids, fermenters usually operate in the temperature range of 303.15 to 313.15 K. Therefore, an extractant will be suitable only if it is effective in this temperature range. Reactive extraction is an exothermic process and therefore a decrease in the extractability is expected with increasing temperature because the extractability is a function of the properties of extractants and diluents, which are expected to change with changing temperature. Therefore, it is important to study the effect of temperature on the reactive extraction. Many researchers have studied the influence of temperature on % and reported that the extractability decreases with increasing temperature (Keshav, Wasewar & Chand
2008a, 2009; Wasewar, K & Keshav 2010).
In the present study, the effect of temperature on the extraction of VFAs was studied by varying the temperature from 293.15 to 323.15 K. Figure 6.3 shows the experimental results of and % for individual acids (AA, PA and BA) as a function of temperature for the extractant concentration of 30 %. Figure 6.4 presents
and % values for the mixture of VFAs as a function of temperature for the extrcatant concentration of 30 %. It can be observed that the distribution of both total and individual acids between the organic and aqueous phases decreases with increasing temperature. The values for all three acids exhibit nearly a flat trend with increasing temperature for TBP in MIBK system and non-linear trends for other systems. Effect of temperature on the extraction of VFAs was studied also for 10 and
20% of extractants (TOA and TBP) in the solvents (1-decanol and MIBK). The trends in the results of these experiments are similar to those observed for 30 % extractant concentration. The changes in and % E with changing temperature can be attributed to the effect of temperatures on different parameters such as pKa, the
174 | Page amine-acid interaction, the solubility of the acid in the extractant, the extractant basicity, and water co-extraction. The solubility of acids in both the aqueous and extractant phases is affected by temperature. It is well known that the solubility of acids in water increase as the temperature increases. Similarly, the solubility and, therefore, the distribution coefficient ( ) of acetic and propionic acids is particularly sensitive to temperature during the reactive extraction. On the other hand, only slight variation in solubility (or ) was observed for butyric acid as temperature changes from 303.15 to 323.15K for all the systems investigated. Therefore, it can be concluded that the extraction of VFAs can be carried out over a range of temperatures. The results on the overall extraction of acid mixtures at different temperatures shown in Figure 6.4a to 6.4d indicate that both and % decrease with increasing temperature.
100 120 AA(KD) (a) TOA in 1-Decanol 90 PA(KD) 100 BA(KD) 80 AA(%E) 70 PA(%E) 80 60 BA(%E)
KD 50 60 %E 40 40 30
20 20 10
0 0 293.15 303.15 313.15 323.15 Temp (K)
175 | Page 50 100 (b) TBP in 1-Decanol AA(KD) 45 90 PA(KD)
40 80 BA(KD) AA(%E) 35 70 PA(%E) 30 60 BA(%E)
KD 25 50 %E 20 40
15 30
10 20
5 10
0 0 293.15 303.15 313.15 323.15 Temp (K)
TOA in MIBK 40 100 (c) AA(KD) 90 35 PA(KD) 80 BA(KD) 30 AA(%E) 70 PA(%E) 25 60 BA(%E) 20 50 KD
40 %E 15 30 10 20 5 10
0 0 293.15 303.15 313.15 323.15 Temp (K)
176 | Page TBP in MIBK
(d) 20 100 AA(KD) PA(KD) 18 90 BA(KD) 16 80 AA(%E) 14 70 PA(%E) 12 60 BA(%E) 10 50 KD
8 40 %E 6 30 4 20 2 10 0 0 293.15 303.15 313.15 323.15 Temp (K)
Figure 6.3 Distribution coefficient ( ) and extraction efficiency (% ) for individual acids (AA, PA, and BA) at different temperatures (a) TOA+1-decanol, (b)TBP+1- decanol, (c) TOA+MIBK and (d) TBP+MIBK.
TOA in 1-Decanol TBP in 1-Decanol 100 2 100 12 90 1.8 10 80 80 1.6 70 1.4 8 60 60 1.2 50 1
6 KD % E KD 40 40 0.8 % E 4 30 0.6 %E 20 20 0.4 2 %E KD 10 0.2 KD 0 0 0 0 293.15 303.15 313.15 323.15 293.15 303.15 313.15 323.15 Temp (K) Temp (K)
(a) (b)
177 | Page TOA in MIBK TBP in MIBK 80 5 100 5 70 4.5 4 60 80 4 3.5 50 3 60 3 40 2.5 KD % E %E KD 30 2 40 2 1.5 20 %E 1 20 %E 1 10 KD 0.5 KD 0 0 0 0 293.15 303.15 313.15 323.15 293.15 303.15 313.15 323.15 Temp (K) Temp (K) (c) (d)
Figure 6.4 Overall extraction efficiency (%E) and overall distribution coefficient (KD) at different temperatures (a) TOA+1-decanol, (b) TBP+1-decanol, (c) TOA+MIBK, and (d) TBP+MIBK
6.3.4 Effect of pH
In the acid producing sludge fermentation process, the accumulation of free VFAs inhibits the action of acid-producing bacteria. pH adjustment is an important step to control the equilibrium concentration between the aqueous and organic phases. The extraction of carboxylic acids depends on the pH value in the aqueous phase.
Carboxylic acids are extracted more effectively at low pH values, i.e. lower than the pKa value of the respective acid where the acids are available in their un-dissociated forms. In the mixtures used in the present study, amines and phosphates extract carboxylic acids from an aqueous solution forming an acid-extractant complex with un-dissociated acid. The effect of pH on the extraction of organic acids was demonstrated by Chen et.al (2007), who studied the production of organic acid
178 | Page during sludge digestion under different pH conditions through the addition of methanogen inhibitor. They found that there was a good ecological niche for acid- producing bacteria when the pH is 6 and the accumulation of fermentation products improves the inhibition of acidogenic bacteria.
The effect of pH variation on the extractability of acetic, propionic and butyric acids in mixtures containing 30% extractant is shown in Figure 6.5. For all four systems used in this work, the and % values decrease as pH increases. Also, the variation in
and % for acetic and propionic acids is slight as pH increases from 2.5 to 4.5.
On the other hand, values of butyric acid decrease dramatically especially when pH varies from 3.5 to 4.5. Figure 6.6 illustrates the effect of pH on the overall extractability of VFAs. The maximum extraction is achieved at a pH of 3.5 for all four systems studied. Furthermore, it is clear that and % values increase with decreasing pH except at extremely high or low pHs, where they do not change significantly. A decrease in pH means an increase in the concentration of un- dissociated acids. Therefore, it can be concluded that the extractability is influenced by the concentration of un-dissociated acids.
179 | Page TOA in 1-Decanol at 293.15K (a) 90 120 AA(KD) 80 PA(KD) 100 BA(KD) 70 AA(%E) 60 80 PA(%E) BA(%E) 50 60 KD %E 40
30 40
20 20 10
0 0 2.5 3.5 4.5 pH
TBP in 1-Decanol at 293.15K 90 100 AA(KD) (b) PA(KD) 80 90 BA(KD) 80 70 AA(%E) 70 PA(%E) 60 BA(%E) 60 50 KD 50 %E 40 40 30 30 20 20
10 10
0 0 2.5 3.5 4.5 pH
180 | Page TOA in MIBK at 293.15K (c) 100 100 AA(KD) 90 90 PA(KD) BA(KD) 80 80 AA(%E) 70 70 PA(%E) BA(%E) 60 60
50 50 KD
40 40 % E
30 30
20 20
10 10
0 0 2.5 3.5 4.5 pH
TBP in MIBK at 293.15K (d) 25 120 AA(KD) PA(KD) 100 BA(KD) 20 AA(%E)
80 PA(%E) 15 BA(%E)
KD 60
10 %E 40
5 20
0 0 2.5 3.5 4.5 pH
Figure 6.5 Extraction efficiency (% E) and distribution coefficient (KD) for individual acids (AA, PA, and BA) at different pH(a) TOA+1-decanol, (b) TBP+1-decanol, (c)
TOA+MIBK, and (d) TBP+MIBK
181 | Page TOA in 1-Decanol TBP in 1-Decanol 100 20 100 5
80 80 4 15
60 60 3 KD
10 KD % E % E 40 40 2
%E 5 %E 20 20 1 KD KD
0 0 0 0 2.5 3.5 4.5 2.5 3.5 4.5 pH pH
(a) (b)
TBP in MIBK TOA in MIBK 100 5 100 5 90 4.5 80 4 80 4 70 3.5 60 3 60 3 KD 50 2.5 KD % E % E 40 2 40 2 30 1.5 20 1 20 %E 1 %E 10 KD 0.5 KD 0 0 0 0 2.5 3.5 4.5 2.5 3.5 4.5 pH pH
(c) (d)
Figure 6.6 Overall extraction efficiency and overall distribution coefficient at different pH (a) TOA+1-decanol, (b) TBP+1-decanol, (c) TOA+MIBK, and (d) TBP+MIBK
6.3.5 Optimization of Reactive Extraction Process: Taguchi L36 Mixed Design
182 | Page In Taguchi multivariate optimization method, and % were chosen as the two response variables. Figures 6.7a and 6.7b present the effects of experimental variables on response variables. Experimental data were transformed into S/N ratio
(using larger is better criterion) and analysed. It can be inferred that 1-decanol and
TOA has the largest influence on both and % . Also, there is negligible effect on the response characteristics as the extractant concentration increases from 10 to
20%. However, both and % increase when the extractant concentration increases from 20 to 30%. Both and % decrease as the temperature increases from 293.15 to 323.15 K. The largest changes in response are achieved for changes in the pH value especially at pH = 3.5. From the ranks obtained for each factor, it is evident that the extractant type is the factor, which has the most influence on both
and % whereas extractant composition and diluents type have the least influence (Table 6.4). For each response, optimum conditions were obtained as 1st factor at level 1, 2nd factor at level 1, 3rd factor at level 3, 4th factor at level 1 and 5th factor at level 2.
Analysis of variance test was performed to detect the relative significance of each factor to response. Fischer ratio (F-test) values were calculated from the experimental data and compared with the critical values shown in Table 6.5. On the basis of the calculated F-values, it can be concluded that the extractant type and pH have statistical significance (i.e. significant effect) on both and %E at 95% level of confidence. It can be observed from Table 6.5 that extractant type ( 2) has the largest sum of squares indicating that it is the most effective factor for followed by pH ( 5 ), diluent type ( 1 ), temperature ( 4 ) and extractant concentration ( 3 ).
However, in the case of %E, the most effective factor is the extractant type ( 2)
183 | Page followed by pH ( 5 ), temperature ( 4 ), extractant concentration ( 3 ) and diluents type
(). 1
Figure 6.7 Main effect plots ( / ratios) for (a) distribution coefficient ( ) (b) extraction efficiency (% )
The p-values in the ANOVA table were used to check the significance of each
coefficient. They indicate that all coefficients ( 1, 2, 4 and 5 ), except the
184 | Page extractant concentration ( 3), were highly significant for and % with very low p-values (p< 0.05). It can also be observed from the ANOVA table that there are possible interactions between the factors. The interaction of diluents type on extractant type is highly statistically significant with very low p-values.
However, no interaction was observed between the extraction concentration and temperature, and between the temperature and pH. For distribution coefficients, all interactions are statistically significant except for 2 5 and 3
5. In the case of % , the interaction between 1 2 and 2 3 is statistically significant with a value of p< 0.05. F-critical values were also calculated and included in Table 6.5 for factors of statistical significance such that > . In the case of distribution coefficient, all the factors are found to be significant.
Figs. 8a and 8b show the percentage contribution of each factor to the responses. The percentage contribution on both and % is maximum from the extractant type (X2) (47.49 and 53.09 %) followed by pH (17.12 and 19.89
%) and the interaction of 1 2 (18.08 and 14.41 %), respectively. 3-D surface plots shown in Figure 6.9 depict the effects of combination of factors ( 3 , 4 ),
( 5 , 6 ) and ( 3 , 5 ) on the responses and % . The interactions of temperature and extractant composition (X4, X3) with responses KD and % E as depicted in Figures 6.9 (a) and (e) can be explained as follows: at a constant pH =3.5, for an initial composition of extractant (X3)of10% and 293 K
(X4), the values of KD and (%E) are 2.23 and 69.07 %, respectively. However, an increase in the extractant composition to 30 % at the same temperature,
KD and (% E) values increase to 2.60 and 72.29%, respectively. On the other hand, when the temperature increases to 323 K for the extractant composition of 10%, KD and (% E) values decrease to 1.82 to 64.64%, respectively.
185 | Page Similarly, the interactions of pH (X5) and temperature (X4) with responses KD and % E are illustrated in Figures 6.9 (b) and (f). At a constant extractant composition (X3) of 20%, for an initial pH (X5) of 2.5 and 293 K (X4), KD and
(% E) values are 2.51 and 71.5%, respectively. When pH increases from to
3.5 to 4.5 at a constant temperature, KD and (% E) values are found to be higher at pH3.5 compared to that at 4.5.It means that at pH4.5, the extraction of acid decreases because the stability or strength of the complex formed in the organic phase become slow. By increasing the temperature, the values of
KD and (% E) decrease. Likewise Figure 6.9 (c) and (g) represents the interaction of extractant composition (X3) and pH (X5) with response parameters KD and (% E), at a constant initial temperature 293 K. For pH=2.5 and an extractant composition of 20%,KD and (% E) values are found to be
2.51 and 71.58%, respectively. However, when pH increases to 4.5 for the same extractant composition of 20%, KD and (% E) values decrease. When pH increases from 2.5 to 3.5, KD and (% E) values are found to be 2.66 and
72.73 %, respectively. From the above observations, it can be established that among all the interactions, pH and extractant composition govern mainly the values of the response parameters KD and (% E).
186 | Page X1 X1 1% 2% X2 1% 3% X2 X3 8% X3 X4 14% 18% X4 X5 X5 X1X2 X1X2 X2X3 20% 17% 54% X2X4 48% X2X4 5%
4% 1% 3%
Figure 6.8 Percentage contribution of factors for (a) distribution coefficient ( ) and extraction efficiency (% )
Individual desirability ( ) of responses and % were calculated and combined to evaluate the overall desirability (Table 6.2). The and are the maximum and minimum values of % and obtained in 36 experiments; 31 combination has the values (0.85 for and 45.95 % for % ), and 1 and 5 have the values (7.33 for and 87.99 % for % ). The desirability index, is taken as 1 as it approaches the target value. From Table 6.2, it can be seen that the combination 2,
2, 1, 3 and 3 (Exp. No.31) shows the overall desirability value as zero, which means that the S/N ratio cannot be calculated for this case. Therefore, a mean of overall desirability has been determined to find out the optimum combination (Akhtar, Akhtar
& Khalid 2008; Nandi et al. 2010). It can also be noticed that the combination 1, 1, 2,
2 and 2 (Experiments 2 and 5) show an overall desirability value of 1, which signifies that the maximum target responses have been achieved. From Table 6.4 and Figure
6.6, the optimum combination of parameters is predicted as: X1= 1-decanol, X2 =
TOA, X3 = 20 %, X4 = 293.15 K and X5 = 3.5 and the responses for and % at the
187 | Page optimum combination are 10.36 and 91.26 %, respectively, which is different from the overall desirability value one combination. A confirmation run was carried out to verify the validity of the optimum condition and the values for and % in the confirmation run were found to be 8.65 and 89.64%, respectively. These values are close to the predicted one reported in Table 6.6 (i.e., 10.36 and 91.26 %, respectively), which suggests that the model adequately represents the investigated system.
Therefore, from the statistical analysis of experimental data, it is concluded that the maximum extraction efficiency of VFAs can be achieved using 20% of TOA as an extractant diluted in 1-decanol at a temperature of 293.15 K and a pH of 3.5.
Table 6.5 ANOVA for distribution coefficient ( ) and extraction efficiency (% ) in
36array design.
a b c d e f g Factors SS DOF MS F Fcr P-value PC
For Distribution coefficient ( )
1 49.207 1 36.451 207.64 161.4 0.000 7.65
2 305.414 1 138.115 786.78 18.51 0.000 47.49
3 9.284 2 1.536 8.75 5.14 0.035 1.44
4 26.173 2 6.651 37.89 4.46 0.003 4.07
5 110.058 2 37.309 212.54 4.10 0.000 17.12
1 2 116.265 1 80.58 459.03 18.51 0.000 18.08
1 3 0.341 2 1.585 9.03 5.14 0.033 0.05
188 | Page 1 4 0.877 2 3.053 17.39 4.46 0.011 0.14
1 5 8.119 2 6.984 39.78 4.10 0.002 1.26
2 3 6.588 2 4.797 27.33 3.89 0.005 1.02
2 4 10.424 2 3.684 20.99 3.63 0.008 1.62
2 5 0.081 2 0.055 0.31 3.49 0.747 0.01
3 5 0.217 4 0.054 0.31 2.53 0.859 0.03
Residual 0.702 4 0.176
Total 643.751 29
For extraction efficiency (% )
1 1.870 1 1.603 53.35 161.4 0.002 2.93
2 33.849 1 13.965 464.87 18.51 0.000 53.09
3 2.205 2 0.022 0.72 5.14 0.539 3.46
4 2.963 2 1.092 36.34 4.46 0.003 4.65
5 12.680 2 3.020 100.52 4.10 0.000 19.89
1 2 9.187 1 5.962 198.47 18.51 0.000 14.41
1 3 0.058 2 0.058 1.92 5.14 0.261 0.09
1 4 0.110 2 0.101 3.37 4.46 0.139 0.17
1 5 0.024 2 0.074 2.47 4.10 0.200 0.04
189 | Page 2 3 0.222 2 0.245 8.16 3.89 0.039 0.35
2 4 0.372 2 0.173 5.75 3.63 0.067 0.58
2 5 0.133 2 0.037 1.22 3.49 0.387 0.21
3 5 0.083 4 0.021 0.69 2.53 0.634 0.13
Residual Error 0.120 4 0.030
Total 63.878 29
a b c d e sum of squares; degrees of freedom; mean square; Fischer’s ratio; F0.05(ν1, ν2) where ν1 is the f degree of freedom and ν2 = (a*n –a), a is factor number and n is number of levels; probability of significance; gpercentage contribution.
Table 6.6 Multivariate optimization of and %E and confirmation of the optimum condition.
Multiple Optimization Confirmation run Predicted values
(Experimental Values)
and % E simultaneously optimised
Diluent type (X1) 1-decanol 1-decanol
Extractant type(X2) TOA TOA
Extractant concentration (X3) 20% 20%
Temperature(X4) 293.15 K 293.15 K pH (X5)3.53.5
Distribution coefficient ( ) 8.65 10.36
Extraction efficiency ( % E) 89.64 91.26
190 | Page Figure 6.9 3-D surfaces showing effect of combination of factors {( 3 , 4 ), ( 4 , 5 ) and ( 3 , 5 )}on responses. (a-c) distribution coefficient ( ), (e-g) extraction efficiency (% )
191 | Page 6.4 Conclusions
This chapter revealed that TOA+1-decanol system was effective in the separation of volatile fatty acids (AA, PA and BA) compared to TBP+1-decanol, TOA+MIBK and
TBP+MIBK in the reactive extraction process. The effect of pH on the reactive extraction of VFAs was investigated to determine a suitable pH for the recovery of
VFAs from aqueous streams or fermentation broths. Highest extraction efficiency and distribution coefficient were achieved at a pH of 3.5. The effect of temperature is an important variable in the process design; therefore the effect of temperature was studied on the separation of VFAs. The maximum extraction efficiency and distribution coefficient values were obtained at 303.15 K, which is the operating temperature of fermentation broths. The extraction efficiency of organic phases
TBP+1-decanol, TOA+1-decanol and TBP+MIBK was tested in batch reactive extraction process using process parameters that were optimised using multivariate response Taguchi ( 36 ) mixed design adapting desirability approach. The optimum condition obtained were X1 = 1-decanol, X2 = TOA, X3 = 20 %, X4 = 293.15 K and X5
=3.5 with a maximum distribution coefficient of 8.65 and an extraction efficiency of
89.64 %.The reactive extraction process was found to be a promising technique for the effective removal of the VFAs (AA, PA & BA) from fermentation broth.
192 | Page Chapter 7
Conclusions
193 | Page 7.1 Overall Conclusions
Carboxylic acids are organic acids consisting of one or more carboxylic acid group.
The properties of carboxylic acids depend on the carbon–chain length, molecular structure and the presence of additional functional groups. The carboxylic acids can be derived from the petroleum feed stock or carbohydrates via a fermentation process. Traditionally, all carboxylic acids were produced by biochemical processes.
In spite of the large amount of literature available on the reactive extraction of carboxylic acids, there is a significant shortage on literature on the optimisation and regeneration of carboxylic acids from a loaded-organic phase. This knowledge gap has created the need to develop a highly efficient economically viable reactive extraction process.
In this study, the value added carboxylic acids such as levulinic and succinic acids are recovered through the reactive extraction process and process variables have been optimise using the statistical tools of Taguchi and response surface methodology to maximise the response parameters of distribution coefficient and extraction efficiency. The process variables are acid concentration, extractant concentration and temperature.
The first part of this work was concerned with the first research question to study the reactive extraction of levulinic acid using TOA in MIBK at different temperatures
(293-333 K). Physical and chemical equilibrium were studied at various temperatures. Distribution coefficients ( ) and extraction efficiency (% ) were determined in each study. The results obtained in physical extraction studies were very low compared to chemical extraction. Chemical extraction was studied with three different concentrations of TOA. The highest and % were obtained at
194 | Page 0.678 kmol m-3 TOA and 293 K. Chemical extraction studies were interpreted with the formation of 2:1 complex. The complexation constants were determined at different temperatures (293-333 K). The complexation constants decreased with increasing temperatures. Enthalpy ( )( and entropy S) of the system were estimated and found that the reaction is exothermic in nature. Gibbs free energy ( ) values were estimated to find the feasibility of the system. The values increased with temperature. The process variables were optimised using Taguchi mixed design
1 2 18 (6 3 ) array. Levulinic acid concentration, TOA concentration and temperature were selected as process variables. Distribution coefficient ( ) and extraction efficiency (% ) were selected as response parameters. It reveals that the optimum
-3 -3 variables are 1 = 0.3 kmol m levulinic acid, 2 = 0.678 kmol m TOA and 3 =
293 K. Kinetic studies were carried out at optimum conditions. The order of the reaction was found to be 2 with respect to levulinic acid and TOA concentrations.
The reaction regime was found to be instantaneous occurring in the organic film.
The second research question in the general introduction chapter is explored in chapter 4. In this chapter, reactive extraction of succinic acid was studied using TOA in 1-decanol. Reactive extraction of succinic acid was explained by equilibrium and kinetic studies. Equilibrium studies were carried out in a shaking incubator and optimised using RSM. A five level central composite design was conducted to study the interaction of parameters. The process variables were selected as succinic acid concentration, TOA concentration and temperature. The response parameter selected as extraction efficiency (% ). The optimum conditions were found to be succinic acid concentration 0.2 kmol m-3, TOA concentration 0.678 kmol m-3 (33 volume %) and temperature 305 K with extraction efficiency (% ) was found to be
93.75 % and 91 % by RSM model and experimentally, respectively. The kinetic
195 | Page studies were carried out in stirred cell. The order of the reaction of succinic acid and
TOA was found to be first order. The reaction regime was found to instantaneous occurring in film.
The third research question of regeneration of carboxylic acids from loaded-organic phase was addressed in chapter 5 through the study of regeneration of levulinic acid from the loaded-organic phase (TOA+MIBK). The regeneration process was studied using the four methods found in the literature NaOH, temperature swing, diluent swing and TMA methods. Among the four methods the extraction efficiency was the highest for the TMA method. A mass action law model was developed for the TMA method to relate the recovery of acid with TMA concentration. Kinetic studies carried out with TMA in a stirred cell indicated that the reaction was in the fast regime. The order of the reaction with respect to levulinic acid and TMA was 1.5. The rate constant was determined as 0.0317 (m3)-0.5 kmol-0.5 s-1. Finally 80% of the acid was recovered by evaporating the TMA solution at 104-140oC under atmospheric pressure from the aqueous phase. Process economics were evaluated for the total reactive extraction process (forward and regeneration/ back extraction) of levulinic acid recovery from aqueous solution. It revealed that 80% of levulinic acid can be recovered at the cost of 1.55 USD per liter with a payback period of 0.49 years for a loaded-organic phase rate of 2 m3 h-1.
The fourth research question regarding the recovery of multi-acids from aqueous solutions addressed in the chapter 6. In this chapter the selected multi-acids are volatile fatty acids (VFAs) (acetic, propionic and butyric acids). Total concentration of the mixture was prepared according to the ratio based on the composition of fermentation broth, which was collected from a facility that uses bio production to produce volatile fatty acids. The multi-acids were recovered using four organic phase
196 | Page systems, they are TOA+1-decanol, TOA+MIBK, TBP+1-decanol and TBP+MIBK.
The concentration of extractants was varied from 0.229 to 0.678 kmol m-3 in diluent.
The pH of aqueous solution varied (2.5, 3.5 and 4.5) to determine the most suitable pH for the recovery of the VFAs. The highest extraction efficiency and distribution coefficient was obtained at pH 3.5. The effect of temperature was studied on separation of VFAs from aqueous solutions. The highest extraction efficiency was obtained at 303.15 K which is the operating temperature of fermentation broths. The process variables of type of solvent, type of extractant, extractant concentration, temperature and pH were optimised using Taguchi mixed design ( 36 ) adapting desirability approach. The optimum parameters obtained were solvent-1-decanol, extractant-TOA, extractant concentration-20%, temperature-293.15 K and pH -3.5 with maximum distribution coefficient ( ) of 8.65 and an extraction efficiency (% ) of 89.64%.
7.2 Recommendations for future work
Regeneration: Limited literature exits on this process and this plays main role in reactive extraction to complete the total process. The regeneration of carboxylic acids is done by reversible complexation, which allows the recovery of acid into the water phase and evaporation of TMA for recycling process. The acid recovered in this process can be used for a wide range of applications. The regeneration process developed in this work makes it possible to study in the regeneration of multi-acids from a loaded-organic phase. The regenerated multi-acids (VFAs) can be used in many applications.
Reactive extraction: The reactive extraction process, which is an economically feasible technique to separate carboxylic acids from fermentation broth, that has
197 | Page been developed in this work makes it possible to study the continuous process in lab and pilot scale as well. A scale up study would be commercially beneficial in the food, pharmaceutical and fermentation industries to separate products from by- products.
198 | Page References
Akhtar, K, Akhtar, MW & Khalid, AM 2008, 'Removal and recovery of zirconium from
its aqueous solution by Candida tropicalis', Journal of Hazardous Materials, vol.
156, no. 1, pp. 108-117.
Aljundi, IH, Belovich, JM & Talu, O 2005, 'Adsorption of lactic acid from fermentation
broth and aqueous solutions on Zeolite molecular sieves', Chemical
Engineering Science, vol. 60, no. 18, pp. 5004-5009.
Alkaya, E, Kaptan, S, Ozkan, L, Uludag-Demirer, S & Demirer, GN 2009, 'Recovery
of acids from anaerobic acidification broth by liquid–liquid extraction',
Chemosphere, vol. 77, no. 8, pp. 1137-1142.
Alter, JE & Blumberg, R 1981, Extraction of citric acid, Google Patents.
AsciYS & İnci̇, I 2010, 'Extraction equilibria of acrylic acid from aqueous solutions by
amberlite LA-2 in various diluents', Journal of Chemical & Engineering Data,
vol. 55, no. 7, pp. 2385-2389.
Baniel, AM, Blumberg, R & Hajdu, K 1981, Recovery of acids from aqueous
solutions, Google Patents.
Barbari, TA & King, CJ 1982, 'Equilibrium distribution coefficients for extraction of
chlorinated hydrocarbons and aromatics from water into undecane',
Environmental science & technology, vol. 16, no. 9, pp. 624-627.
Bart, H-J 2001, Reactive Extraction, Springer, Berlin.
199 | Page Bayazit, SaS, Uslu, H & İnci, I 2009, 'Comparative equilibrium studies for citric acid
by amberlite LA-2 or tridodecylamine (TDA)', Journal of Chemical &
Engineering Data, vol. 54, no. 7, pp. 1991-1996.
Beardslee, T & Picataggio, S 2012, 'Bio‐based adipic acid from renewable oils', Lipid
Technology, vol. 24, no. 10, pp. 223-225.
Benyounis, K, Olabi, A & Hashmi, M 2005, 'Effect of laser welding parameters on the
heat input and weld-bead profile', Journal of materials processing technology,
vol. 164, pp. 978-985.
Berglund, KA, Yedur, S & Dunuwila, DD 1999, Succinic acid production and
purification, Google Patents.
Bishai, M, De, S, Adhikari, B & Banerjee, R 2015, 'A platform technology of recovery
of lactic acid from a fermentation broth of novel substrate Zizyphus oenophlia',
3 Biotech, vol. 5, no. 4, pp. 455-463.
Bízek, V, Horáček, J & Koušová, M 1993, 'Amine extraction of citric acid: effect of
diluent', Chemical Engineering Science, vol. 48, no. 8, pp. 1447-1457.
Bizek, V, Horáček, J, Koušová, M, Heyberger, A & Prochazka, J 1992, 'Mathematical
model of extraction of citric acid with amine', Chemical Engineering Science,
vol. 47, no. 6, pp. 1433-1440.
Bizek, V, Horacek, J, Rericha, R & Kousova, M 1992, 'Amine extraction of
hydroxycarboxylic acids. 1. Extraction of citric acid with 1-octanol/n-heptane
solutions of trialkylamine', Industrial & engineering chemistry research, vol. 31,
no. 6, pp. 1554-1562.
200 | Page Blahušiak, M, Schlosser, Š & Cvengroš, J 2012, 'Simulation of a new regeneration
process of solvents with ionic liquid by short-path distillation', Separation and
Purification Technology, vol. 97, pp. 186-194.
Boonkong, W, Sangvanich, P, Petsom, A & Thongchul, N 2009, 'Comparison of an
ion exchanger and an in‐house electrodialysis unit for recovery of l‐lactic acid
from fungal fermentation broth', Chemical engineering & technology, vol. 32,
no. 10, pp. 1542-1549.
Canari, R & Eyal, AM 2004, 'Temperature effect on the extraction of carboxylic acids
by amine-based extractants', Industrial & engineering chemistry research, vol.
43, no. 23, pp. 7608-7617.
Cascaval, D & Galaction, A-I 2004, 'New extraction techniques on bioseparations: 1.
Reactive extraction', Hemijska industrija, vol. 58, no. 9, pp. 375-386.
Cascaval, D, Kloetzer, La & Galaction, A-I 2011, 'Influence of organic phase polarity
on interfacial mechanism and efficiency of reactive extraction of acetic acid with
tri-n-octylamine', Journal of Chemical & Engineering Data, vol. 56, no. 5, pp.
2521-2526.
Chaudhuri, J & Phyle, D 1992, 'Emulsion liquid membrane extraction of organic
acids—I. A theoretical model for lactic acid extraction with emulsion swelling',
Chemical Engineering Science, vol. 47, no. 1, pp. 41-48.
Chen, L, Zeng, A, Dong, H, Li, Q & Niu, C 2012, 'A novel process for recovery and
refining of L-lactic acid from fermentation broth', Bioresource technology, vol.
112, pp. 280-284.
201 | Page Chen, Y 2007, The study on the acetate production by fermentation from municipal
sewage sludge [MA dissertation], thesis.
Cheng, K-K, Zhao, X-B, Zeng, J, Wu, R-C, Xu, Y-Z, Liu, D-H & Zhang, J-A 2012,
'Downstream processing of biotechnological produced succinic acid', Applied
microbiology and biotechnology, vol. 95, no. 4, pp. 841-850.
Daneshvar, N, Khataee, A, Rasoulifard, M & Pourhassan, M 2007, 'Biodegradation
of dye solution containing Malachite Green: Optimization of effective
parameters using Taguchi method', Journal of Hazardous Materials, vol. 143,
no. 1, pp. 214-219.
Datta, D, Marti, ME, Pal, D & Kumar, S 2016, 'Equilibrium study on the extraction of
levulinic acid from aqueous solution with aliquat 336 dissolved in different
diluents: Solvent’s polarity effect and column design', Journal of Chemical &
Engineering Data, vol. 62, no. 1, pp. 3-10.
Datta, R 1992, Process for the production of succinic acid by anaerobic fermentation,
Google Patents.
Datta, R, Glassner, DA, Jain, MK & Roy, JRV 1992, Fermentation and purification
process for succinic acid, Google Patents.
Davison, BH, Nghiem, NP & Richardson, GL 2004, 'Succinic acid adsorption from
fermentation broth and regeneration', Applied biochemistry and biotechnology,
vol. 114, no. 1-3, p. 653.
Della Pina, C, Falletta, E & Rossi, M 2011, 'A green approach to chemical building
blocks. The case of 3-hydroxypropanoic acid', Green chemistry, vol. 13, no. 7,
pp. 1624-1632.
202 | Page Demirkol, S, Aksoy, HA, Tüter, M, Ustun, G & Sasmaz, DA 2006, 'Optimization of
enzymatic methanolysis of soybean oil by response surface methodology',
Journal of the American Oil Chemists' Society, vol. 83, no. 11, pp. 929-932.
Domańska, U & Lukoshko, EV 2013, 'Measurements of activity coefficients at infinite
dilution for organic solutes and water in the ionic liquid 1-butyl-1-
methylpyrrolidinium tricyanomethanide', The Journal of Chemical
Thermodynamics, vol. 66, pp. 144-150.
Doraiswamy, LK & Sharma, MM 1984, Heterogeneous reactions: Analysis, examples
and reactor design, vol. 2 fluid-fluid-solid reactions, Wiley, New York.
Dwidar, M, Park, J-Y, Mitchell, RJ & Sang, B-I 2012, 'The future of butyric acid in
industry', The Scientific World Journal, vol. 2012.
Eda, S, Kota, BJ, Thella, PK, Bankupalli, S, Bhargava, SK & Parthasarathy, R 2017,
'Regeneration of levulinic acid from loaded-organic phase: equilibrium, kinetic
studies and process economics', Chemical Papers, pp. 1-13.
Doi:10.1007/s1169-017-0188-6.
Eda, S, Kumar, TP, Satyavathi, B, Sudhakar, P & Parthasarathy, R 2016, 'Recovery
of succinic acid by reactive extraction using tri-n-octylamine in 1-decanol:
Equilibrium optimization using response surface method and kinetic studies',
IJCST, vol. 1, p. 1-14.
Eda, S, Kumari, A, Thella, PK, Satyavathi, B & Rajarathinam, P 2017, 'Recovery of
volatile fatty acids by reactive extraction using tri‐n‐octylamine and tri‐butyl
phosphate in different solvents: Equilibrium studies, pH and temperature effect,
203 | Page and optimization using multivariate taguchi approach', The Canadian Journal of
Chemical Engineering, vol. 95, no. 7, pp. 1373-1387.
Eda, S, Parthasarathy, R & Kumar, TP 'Reactive extraction of succinic acid from
aqueous solutions using Tri-N-Octylamine (TOA) in 1-decanol: Equilibria and
effect of temperature', Engineers Australia, p. 496.
Elizalde-González, M & García-Díaz, L 2010, 'Application of a Taguchi L 16
orthogonal array for optimizing the removal of Acid Orange 8 using carbon with
a low specific surface area', Chemical Engineering Journal, vol. 163, no. 1, pp.
55-61.
Errico, M & Rong, B-G 2012, 'Synthesis of new separation processes for bioethanol
production by extractive distillation', Separation and Purification Technology,
vol. 96, pp. 58-67.
Eyal, AM & Bressler, E 1993, 'Industrial separation of carboxylic and amino acids by
liquid membranes: Applicability, process considerations, and potential
advantage', Biotechnology and Bioengineering, vol. 41, no. 3, pp. 287-295.
Federation, WE & Association, APH 2005, 'Standard methods for the examination of
water and wastewater', American Public Health Association (APHA):
Washington, DC, USA, vol.
Ferner, RE & Aronson, JK 2016, 'Cato Guldberg and Peter Waage, the history of the
Law of Mass Action, and its relevance to clinical pharmacology', British journal
of clinical pharmacology, vol. 81, no. 1, pp. 52-55.
204 | Page Girisuta, B, Janssen, L & Heeres, H 2007, 'Kinetic study on the acid-catalyzed
hydrolysis of cellulose to levulinic acid', Industrial & engineering chemistry
research, vol. 46, no. 6, pp. 1696-1708.
Goldberg, I, Rokem, JS & Pines, O 2006, 'Organic acids: old metabolites, new
themes', Journal of Chemical Technology and Biotechnology, vol. 81, no. 10,
pp. 1601-1611.
Gonzalez, MI, Alvarez, S, Riera, FA & Alvarez, R 2008, 'Lactic acid recovery from
whey ultrafiltrate fermentation broths and artificial solutions by nanofiltration',
Desalination, vol. 228, no. 1-3, pp. 84-96.
Gu, Z, Glatz, BA & Glatz, CE 1998, 'Propionic acid production by extractive
fermentation. I. Solvent considerations', Biotechnology and Bioengineering, vol.
57, no. 4, pp. 454-461.
Guettler, MV, Rumler, D & Jain, MK 1999, 'Actinobacillus succinogenes sp. nov., a
novel succinic-acid-producing strain from the bovine rumen', International
Journal of Systematic and Evolutionary Microbiology, vol. 49, no. 1, pp. 207-
216.
Guo, X, Chang, RK & Hussain, MA 2009, 'Ion‐exchange resins as drug delivery
carriers', Journal of pharmaceutical sciences, vol. 98, no. 11, pp. 3886-3902.
Hábová, V, Melzoch, K, Rychtera, M & Sekavová, B 2004, 'Electrodialysis as a
useful technique for lactic acid separation from a model solution and a
fermentation broth', Desalination, vol. 162, pp. 361-372.
205 | Page Han, DH & Hong, WH 1996, 'Reactive extraction of lactic acid with
trioctylamine/methylene chloride/n-hexane', Separation science and
technology, vol. 31, no. 8, pp. 1123-1135.
Hanna, GJ & Noble, RD 1985, 'Measurement of liquid-liquid interfacial kinetics',
Chemical reviews, vol. 85, no. 6, pp. 583-598.
Hartl, J & Marr, R 1993, 'Extraction processes for bioproduct separation', Separation
science and technology, vol. 28, no. 1-3, pp. 805-819.
Heding, LG & Gupta, J 1975, 'Improvement of conditions for precipitation of citric
acid from fermentation mash', Biotechnology and Bioengineering, vol. 17, no. 9,
pp. 1363-1364.
Heyberger, A, Procházka, J & Volaufova, E 1998, 'Extraction of citric acid with
tertiary amine—third-phase formation', Chemical Engineering Science, vol. 53,
no. 3, pp. 515-521.
Hong, Y-G, Hong, W-H & Hong, T-H 1999, 'Separation characteristics of lactic acid
by using mixed tertiary amine extractants', KSBB Journal, vol. 14, no. 5, pp.
566-571.
Hong, Y & Hong, W 2000, 'Extraction of succinic acid with 1-octanol/n-heptane
solutions of mixed tertiary amine', Bioprocess and Biosystems Engineering, vol.
23, no. 5, pp. 535-538.
Hong, YK & Hong, WH 2000, 'Equilibrium studies on reactive extraction of succinic
acid from aqueous solutions with tertiary amines', Bioprocess and Biosystems
Engineering, vol. 22, no. 6, pp. 477-481.
206 | Page Huang, H-J, Ramaswamy, S, Tschirner, U & Ramarao, B 2008, 'A review of
separation technologies in current and future biorefineries', Separation and
Purification Technology, vol. 62, no. 1, pp. 1-21.
Huang, H, Cheng, G, Chen, L, Zhu, X & Xu, H 2009, 'Lead (II) removal from aqueous
solution by spent Agaricus bisporus: determination of optimum process
condition using Taguchi method', Water, air, and soil pollution, vol. 203, no. 1-4,
pp. 53-63.
Jabalquinto, A, González-Nilo, F, Laivenieks, M, Cabezas, M, Zeikus, J & Cardemil,
E 2004, 'Anaerobiospirillum succiniciproducens phosphoenolpyruvate
carboxykinase. Mutagenesis at metal site 1', Biochimie, vol. 86, no. 1, pp. 47-
51.
Jaouachi, B, Hassen, MB & Sakli, F 2007, 'Strength of wet spliced denim yarns after
sizing using a central composite design', AUTEX Research Journal, vol. 7, no.
3, pp. 159-165.
Jeong, G-T & Park, D-H 2009, 'Optimization of biodiesel production from castor oil
using response surface methodology', Applied biochemistry and biotechnology,
vol. 156, no. 1-3, pp. 1-11.
Jipa, I, Dobre, T, Stroescu, M & Stoica, A 2009, 'Acetic acid extraction from
fermentation broth experimental and modelling studies', Revista De Chimie, vol.
60, no. 10, pp. 1084-1089.
Juang, R-S, Huang, R-H & Wu, R-T 1997, 'Separation of citric and lactic acids in
aqueous solutions by solvent extraction and liquid membrane processes',
Journal of membrane science, vol. 136, no. 1-2, pp. 89-99.
207 | Page Juang, R-S & Huang, W-T 1995, 'Kinetic studies on the extraction of citric acid from
aqueous solutions with tri-n-octylamine', Journal of chemical engineering of
Japan, vol. 28, no. 3, pp. 274-281.
Juang, R-S & Lo, R-H 1994, 'Mass-transfer characteristics of a membrane
permeation cell and its application to the kinetic studies of solvent extraction',
Industrial & engineering chemistry research, vol. 33, no. 4, pp. 1001-1010.
Jun, Y-S, Lee, EZ, Huh, YS, Hong, YK, Hong, WH & Lee, SY 2007, 'Kinetic study for
the extraction of succinic acid with TOA in fermentation broth; effects of pH, salt
and contaminated acid', Biochemical Engineering Journal, vol. 36, no. 1, pp. 8-
13.
Jun, YS, Huh, YS, Hong, WH & Hong, YK 2005, 'Kinetics of the Extraction of
Succinic Acid with Tri‐n‐octylamine in 1‐Octanol Solution', Biotechnology
progress, vol. 21, no. 6, pp. 1673-1679.
Kahya, E, Bayraktar, E & Mehmetoglu, U 2001, 'Optimization of process parameters
for reactive lactic acid extraction', Turkish Journal of Chemistry, vol. 25, no. 2,
pp. 223-230.
Kertes, A & King, CJ 1986, 'Extraction chemistry of fermentation product carboxylic
acids', Biotechnology and Bioengineering, vol. 28, no. 2, pp. 269-282.
Keshav, A, Chand, S & Wasewar, KL 2008, 'Equilibrium studies for extraction of
propionic acid using tri-n-butyl phosphate in different solvents', Journal of
Chemical & Engineering Data, vol. 53, no. 7, pp. 1424-1430.
Keshav, A, Norge, P & Wasewar, KL 2012, 'Reactive extraction of citric acid using
tri-n-octylamine in nontoxic natural diluents: part 1—equilibrium studies from
208 | Page aqueous solutions', Applied biochemistry and biotechnology, vol. 167, no. 2, pp.
197-213.
Keshav, A & Wasewar, KL 2010, 'Back extraction of propionic acid from loaded
organic phase', Chemical Engineering Science, vol. 65, no. 9, pp. 2751-2757.
Keshav, A, Wasewar, KL & Chand, S 2008a, 'Extraction of acrylic, propionic, and
butyric acid using Aliquat 336 in oleyl alcohol: Equilibria and effect of
temperature', Industrial & engineering chemistry research, vol. 48, no. 2, pp.
888-893.
Keshav, A, Wasewar, KL & Chand, S 2008b, 'Extraction of propionic acid using
different extractants (tri-n-butylphosphate, tri-n-octylamine, and Aliquat 336)',
Industrial & engineering chemistry research, vol. 47, no. 16, pp. 6192-6196.
Keshav, A, Wasewar, KL & Chand, S 2009, 'Recovery of propionic acid by reactive
extraction-1. Equilibrium, effect of pH and temperature, water coextraction',
Desalination and Water Treatment, vol. 3, no. 1-3, pp. 91-98.
Kiani, A, Bhave, R & Sirkar, K 1984, 'Solvent extraction with immobilized interfaces
in a microporous hydrophobic membrane', Journal of membrane science, vol.
20, no. 2, pp. 125-145.
Kim, P, Laivenieks, M, McKinlay, J, Vieille, C & Zeikus, JG 2004, 'Construction of a
shuttle vector for the overexpression of recombinant proteins in Actinobacillus
succinogenes', Plasmid, vol. 51, no. 2, pp. 108-115.
Klement, T & Büchs, J 2013, 'Itaconic acid–a biotechnological process in change',
Bioresource technology, vol. 135, pp. 422-431.
209 | Page Komasawa, I, Otake, T & Yamada, A 1980, 'Diffusional resistance in extraction rate
of copper with hydroxyoxime extractant', Journal of chemical engineering of
Japan, vol. 13, no. 3, pp. 209-213.
Kulkarni, PS, Branco, LC, Crespo, JG, Nunes, MC, Raymundo, A & Afonso, CA
2007, 'Comparison of physicochemical properties of new ionic liquids based
onimidazolium, quaternary ammonium, and guanidinium cations', Chemistry-A
European Journal, vol. 13, no. 30, pp. 8478-8488.
Kumar, KV & Sivanesan, S 2005, 'Comparison of linear and non-linear method in
estimating the sorption isotherm parameters for safranin onto activated carbon',
Journal of Hazardous Materials, vol. 123, no. 1, pp. 288-292.
Kumar, R, Nanavati, H, Noronha, SB & Mahajani, SM 2006, 'A continuous process
for the recovery of lactic acid by reactive distillation', Journal of Chemical
Technology and Biotechnology, vol. 81, no. 11, pp. 1767-1777.
Kumar, S & Babu, B 2008, 'Process intensification for separation of carboxylic acids
from fermentation broths using reactive extraction', i-Manager's Journal on
Future Engineering and Technology, vol. 3, no. 3, p. 21.
Kumar, S & Babu, B 2009, 'Extraction of pyridine-3-carboxylic acid using 1-
dioctylphosphoryloctane (TOPO) with different diluents: Equilibrium studies',
Journal of Chemical & Engineering Data, vol. 54, no. 9, pp. 2669-2677.
Kumar, S, Datta, D & Babu, B 2011, 'Differential evolution approach for reactive
extraction of propionic acid using tri-n-butyl phosphate (TBP) in kerosene and
1-decanol', Materials and Manufacturing Processes, vol. 26, no. 9, pp. 1222-
1228.
210 | Page Kumar, S, Uslu, H, Datta, D, Rarotra, S & Rajput, K 2015, 'Investigation of Extraction
of 4-Oxopentanoic Acid by N, N-Dioctyloctan-1-amine in Six Different Diluents:
Equilibrium Study', Journal of Chemical & Engineering Data, vol. 60, no. 5, pp.
1447-1453.
Kumar, S, Wasewar, KL & Babu, B 2008, 'Intensification of nicotinic acid separation
using organophosphorous solvating extractants by reactive extraction',
Chemical engineering & technology, vol. 31, no. 11, pp. 1584-1590.
Kumar, TP, Vishwanadham, B, Rani, KP, Mallikarjun, M & Rao, VB 2010, 'Reactive
extraction of levulinic acid from aqueous solutions with tri-n-octylamine (TOA) in
1-octanol: equilibria, kinetics, and model development', Chemical Engineering
Communications, vol. 198, no. 4, pp. 572-589.
Kuo, Y, Munson, CL, Rixey, WG, Garcia, AA, Frierman, M & King, CJ 1987, 'Use of
Adsorbents for Recovery of Acetic Acid from Aqueous Solutions Part I—Factors
Governing Capacity', Separation and Purification Methods, vol. 16, no. 1, pp.
31-64.
Kurzrock, T & Weuster-Botz, D 2010, 'Recovery of succinic acid from fermentation
broth', Biotechnology letters, vol. 32, no. 3, pp. 331-339.
Labbaci, A, Kyuchoukov, G, Albet, J & Molinier, J 2009, 'Detailed investigation of
lactic acid extraction with tributylphosphate dissolved in dodecane', Journal of
Chemical & Engineering Data, vol. 55, no. 1, pp. 228-233.
Lateef, H, Gooding, A & Grimes, S 2012, 'Use of 1‐hexyl‐3‐methylimidazolium
bromide ionic liquid in the recovery of lactic acid from wine', Journal of
Chemical Technology and Biotechnology, vol. 87, no. 8, pp. 1066-1073.
211 | Page Lee, P, Lee, S, Hong, S & Chang, H 2002, 'Isolation and characterization of a new
succinic acid-producing bacterium, Mannheimia succiniciproducens MBEL55E,
from bovine rumen', Applied microbiology and biotechnology, vol. 58, no. 5, p.
663.
Lee, P, Lee, S, Hong, S, Chang, H & Park, S 2003, 'Biological conversion of wood
hydrolysate to succinic acid by Anaerobiospirillum succiniciproducens',
Biotechnology letters, vol. 25, no. 2, pp. 111-114.
Lee, SC & Kim, HC 2011, 'Batch and continuous separation of acetic acid from
succinic acid in a feed solution with high concentrations of carboxylic acids by
emulsion liquid membranes', Journal of membrane science, vol. 367, no. 1, pp.
190-196.
Levenspiel, O 1999, Chemical reaction engineering, 3rd edn, Wiley, New York.
Li, Z, Qin, W & Dai, Y 2002, 'Liquid− liquid equilibria of acetic, propionic, butyric, and
valeric acids with trioctylamine as extractant', Journal of Chemical &
Engineering Data, vol. 47, no. 4, pp. 843-848.
Liu, L, Zhu, Y, Li, J, Wang, M, Lee, P, Du, G & Chen, J 2012, 'Microbial production of
propionic acid from propionibacteria: current state, challenges and
perspectives', Critical reviews in biotechnology, vol. 32, no. 4, pp. 374-381.
Ma, C, Li, J, Qiu, J, Wang, M & Xu, P 2006, 'Recovery of pyruvic acid from
biotransformation solutions', Applied microbiology and biotechnology, vol. 70,
no. 3, pp. 308-314.
212 | Page Mackenzie, PD & King, CJ 1985, 'Combined solvent extraction and stripping for
removal and isolation of ammonia from sour waters', Industrial & Engineering
Chemistry Process Design and Development, vol. 24, no. 4, pp. 1192-1200.
Malmary, G, Albet, J, Putranto, A, Hanine, H & Molinier, J 2000, 'Recovery of
aconitic and lactic acids from simulated aqueous effluents of the sugar‐cane
industry through liquid–liquid extraction', Journal of Chemical Technology and
Biotechnology, vol. 75, no. 12, pp. 1169-1173.
Mao, F, Zhang, G, Tong, J, Xu, T & Wu, Y 2014, 'Anion exchange membranes used
in diffusion dialysis for acid recovery from erosive and organic solutions',
Separation and Purification Technology, vol. 122, pp. 376-383.
Marszałek, J & Kamiński, W 2012, 'Efficiency of acetone-butanol-ethanol-water
system separation by pervaporation', Chemical and Process Engineering, vol.
33, no. 1, pp. 131-140.
Marták, J & Schlosser, Š 2007, 'Extraction of lactic acid by phosphonium ionic
liquids', Separation and Purification Technology, vol. 57, no. 3, pp. 483-494.
Marti, ME, Gurkan, T & Doraiswamy, L 2011, 'Equilibrium and kinetic studies on
reactive extraction of pyruvic acid with trioctylamine in 1-octanol', Industrial &
engineering chemistry research, vol. 50, no. 23, pp. 13518-13525.
Matsumoto, M, Otono, T & Kondo, K 2001, 'Synergistic extraction of organic acids
with tri-n-octylamine and tri-n-butylphosphate', Separation and Purification
Technology, vol. 24, no. 1, pp. 337-342.
McCoy, M 2009, 'Big plans for succinic acid', Chemical & Engineering News, vol. 87,
no. 50, pp. 23-25.
213 | Page McKinlay, JB, Vieille, C & Zeikus, JG 2007, 'Prospects for a bio-based succinate
industry', Applied microbiology and biotechnology, vol. 76, no. 4, pp. 727-740.
Mikkola, J-P, Virtanen, P & Sjöholm, R 2006, 'Aliquat 336®—a versatile and
affordable cation source for an entirely new family of hydrophobic ionic liquids',
Green chemistry, vol. 8, no. 3, pp. 250-255.
Miller, C, Fosmer, A, Rush, B, McMullin, T, Beacom, D & Suominen, P 2011, 3.17 -
Industrial production of lactic acid A2 - Moo-Young, Murray, 2nd edn,
Comprehensive biotechnology, Academic Press, Burlington,
Mitchell, JA & Reid, EE 1931, 'The preparation of aliphatic amides', Journal of the
American Chemical Society, vol. 53, no. 5, pp. 1879-1883.
Nandi, G, Datta, S, Bandyopadhyay, A & Pal, PK 2010, 'Analyses of hybrid Taguchi
methods for optimization of submerged arc weld', paper presented to
Conference Proceedings, National Institute of Technology, Agartala,
Nghiem, NP, Davison, BH, Suttle, BE & Richardson, GR 1997, 'Production of
succinic acid by Anaerobiospirillum succiniciproducens', Applied biochemistry
and biotechnology, vol. 63, no. 1, pp. 565-576.
Nikhade, BP & Pangarkar, VG 2005, 'Equilibria and kinetics of extraction of citric acid
from aqueous solutions in Alamine 336–cyclohexanone system', Separation
science and technology, vol. 40, no. 12, pp. 2539-2554.
Oliveira, FS, Araújo, JM, Ferreira, R, Rebelo, LPN & Marrucho, IM 2012, 'Extraction
of L-lactic, L-malic, and succinic acids using phosphonium-based ionic liquids',
Separation and Purification Technology, vol. 85, pp. 137-146.
214 | Page Otto, C, Yovkova, V & Barth, G 2011, 'Overproduction and secretion of α-ketoglutaric
acid by microorganisms', Applied microbiology and biotechnology, vol. 92, no.
4, p. 689.
Pal, D & Keshav, A 2014, 'Extraction equilibria of pyruvic acid using tri-n-butyl
phosphate: Influence of diluents', Journal of Chemical & Engineering Data, vol.
59, no. 9, pp. 2709-2716.
Pal, D & Keshav, A 2016, 'Recovery of pyruvic acid using tri-n-butylamine dissolved
in non-toxic diluent (rice bran oil)', Journal of The Institution of Engineers
(India): Series E, vol. 97, no. 1, pp. 81-87.
Pal, D, Tripathi, A, Shukla, A, Gupta, KR & Keshav, A 2015, 'Reactive extraction of
pyruvic acid using tri-n-octylamine diluted in decanol/kerosene: equilibrium and
effect of temperature', Journal of Chemical & Engineering Data, vol. 60, no. 3,
pp. 860-869.
Park, C, Nam, H-G, Lee, KB & Mun, S 2014, 'Optimal design and experimental
validation of a simulated moving bed chromatography for continuous recovery
of formic acid in a model mixture of three organic acids from Actinobacillus
bacteria fermentation', Journal of Chromatography A, vol. 1365, pp. 106-114.
Park, D & Zeikus, J 1999, 'Utilization of electrically reduced neutral Red
byActinobacillus succinogenes: physiological function of neutral Red in
membrane-driven fumarate reduction and energy conservation', Journal of
bacteriology, vol. 181, no. 8, pp. 2403-2410.
Pazouki, M & Panda, T 1998, 'Recovery of citric acid-a review', Bioprocess and
Biosystems Engineering, vol. 19, no. 6, pp. 435-439.
215 | Page Podkovyrov, SM & Zeikus, JG 1993, 'Purification and characterization of
phosphoenolpyruvate carboxykinase, a catabolic CO2-fixing enzyme, from
Anaerobiospirillum succiniciproducens', Microbiology, vol. 139, no. 2, pp. 223-
228.
Polen, T, Spelberg, M & Bott, M 2013, 'Toward biotechnological production of adipic
acid and precursors from biorenewables', Journal of Biotechnology, vol. 167,
no. 2, pp. 75-84.
Poole, LJ & King, CJ 1991, 'Regeneration of carboxylic acid-amine extracts by back-
extraction with an aqueous solution of a volatile amine', Industrial & engineering
chemistry research, vol. 30, no. 5, pp. 923-929.
Poposka, FA, Nikolovski, K & Tomovska, R 1997, 'Equilibria and Mathematical
Models of Extraction of Citric Acid with Isodecanol/n-Parrafins Solutions of
Trioctylamine', Journal of chemical engineering of Japan, vol. 30, no. 5, pp.
777-785.
Poposka, FA, Nikolovski, K & Tomovska, R 1998, 'Kinetics, mechanism and
mathematical modelling of extraction of citric acid with isodecanol/n-paraffins
solutions of trioctylamine', Chemical Engineering Science, vol. 53, no. 18, pp.
3227-3237.
Poposka, FA, Prochazka, J, Tomovska, R, Nikolovski, K & Grizo, A 2000, 'Extraction
of tartaric acid from aqueous solutions with tri-iso-octylamine (HOSTAREX A
324). Equilibrium and kinetics', Chemical Engineering Science, vol. 55, no. 9,
pp. 1591-1604.
216 | Page Prasad, R, Kiani, A, Bhave, R & Sirkar, K 1986, 'Further studies on solvent extraction
with immobilized interfaces in a microporous hydrophobic membrane', Journal
of membrane science, vol. 26, no. 1, pp. 79-97.
Prochazka, J, Heyberger, A, Bizek, V, Kousova, M & Volaufova, E 1994, 'Amine
extraction of hydroxycarboxylic acids. 2. Comparison of equilibria for lactic,
malic, and citric acids', Industrial & engineering chemistry research, vol. 33, no.
6, pp. 1565-1573.
Radhakumari, M, Ball, A, Bhargava, SK & Satyavathi, B 2014, 'Optimization of
glucose formation in karanja biomass hydrolysis using Taguchi robust method',
Bioresource technology, vol. 166, pp. 534-540.
Rani, KP, Kumar, TP, Murthy, J, Sankarshana, T & Vishwanadham, B 2010,
'Equilibria, kinetics, and modeling of extraction of citric acid from aqueous
solutions with Alamine 336 in 1-octanol', Separation science and technology,
vol. 45, no. 5, pp. 654-662.
Rashid, U, Anwar, F, Ansari, TM, Arif, M & Ahmad, M 2009, 'Optimization of alkaline
transesterification of rice bran oil for biodiesel production using response
surface methodology', Journal of Chemical Technology and Biotechnology, vol.
84, no. 9, pp. 1364-1370.
Ren, Y-P, Wang, J-J, Li, X-F & Wang, X-H 2011, 'Reactive extraction of short-chain
fatty acids from synthetic acidic fermentation broth of organic solid wastes and
their stripping', Journal of Chemical & Engineering Data, vol. 57, no. 1, pp. 46-
51.
217 | Page Ricker, N, Pittman, E & King, C 1980, 'Solvent extraction with amines for recovery of
acetic acid from dilute aqueous industrial streams', J. Separ. Proc. Technol, vol.
1, no. 2, pp. 23-30.
Şahin, S, Bayazit, SaS, Bilgin, M & İnci, Is 2009, 'Investigation of formic acid
separation from aqueous solution by reactive extraction: effects of extractant
and diluent', Journal of Chemical & Engineering Data, vol. 55, no. 4, pp. 1519-
1522.
Samuelov, NS, Datta, R, Jain, MK & Zeikus, JG 1999, 'Whey fermentation by
Anaerobiospirillum succiniciproducens for production of a succinate-based
animal feed additive', Applied and Environmental Microbiology, vol. 65, no. 5,
pp. 2260-2263.
Sauer, M, Porro, D, Mattanovich, D & Branduardi, P 2008, 'Microbial production of
organic acids: expanding the markets', Trends in biotechnology, vol. 26, no. 2,
pp. 100-108.
Serrano-Ruiz, JC, West, RM & Dumesic, JA 2010, 'Catalytic conversion of
renewable biomass resources to fuels and chemicals', Annual review of
chemical and biomolecular engineering, vol. 1, pp. 79-100.
Shreve, RN & Brink Jr, JA 1977, Chemical Process Industries, McGraw-Hill Book
Co.
Siebold, M, Pv, F, Joppien, R, Rindfleisch, D, Schügerl, K & Röper, H 1995,
'Comparison of the production of lactic acid by three different lactobacilli and its
recovery by extraction and electrodialysis', Process biochemistry, vol. 30, no. 1,
pp. 81-95.
218 | Page Soccol, CR, Vandenberghe, LP, Rodrigues, C & Pandey, A 2006, 'New perspectives
for citric acid production and application', Food Technology & Biotechnology,
vol. 44, no. 2.
Solichien, M, O'Brien, D, Hammond, E & Glatz, C 1995, 'Membrane-based extractive
fermentation to produce propionic and acetic acids: Toxicity and mass transfer
considerations', Enzyme and Microbial Technology, vol. 17, no. 1, pp. 23-31.
Starks, CM 1999, 'Interfacial area generation in two-phase systems and its effect on
kinetics of phase transfer catalyzed reactions', Tetrahedron, vol. 55, no. 20, pp.
6261-6274.
Straathof, A 2011, 'The proportion of downstream costs in fermentative production
processes', Comprehensive biotechnology, vol. 2, pp. 811-814.
Straathof, AJ 2013, 'Transformation of biomass into commodity chemicals using
enzymes or cells', Chemical reviews, vol. 114, no. 3, pp. 1871-1908.
Straathof, AJ, Sie, S, Franco, TT & Van der Wielen, LA 2005, 'Feasibility of acrylic
acid production by fermentation', Applied microbiology and biotechnology, vol.
67, no. 6, pp. 727-734.
Straathof, AJJ & van Gulik, WM 2012, 'Production of fumaric acid by fermentation', in
X Wang, J Chen & P Quinn (eds), Reprogramming Microbial Metabolic
Pathways, Springer Netherlands, Dordrecht, pp. 225-240, <
Streitwieser, A, Heathcock, CH, Kosower, EM & Corfield, PJ 1992, Introduction to
organic chemistry, Macmillan New York.
219 | Page Sun, X, Luo, H & Dai, S 2011, 'Ionic liquids-based extraction: a promising strategy for
the advanced nuclear fuel cycle', Chemical reviews, vol. 112, no. 4, pp. 2100-
2128.
Tamada, JA, Kertes, AS & King, CJ 1990, 'Extraction of carboxylic acids with amine
extractants. 1. Equilibria and law of mass action modeling', Industrial &
engineering chemistry research, vol. 29, no. 7, pp. 1319-1326.
Tamada, JA & King, CJ 1990, 'Extraction of carboxylic acids with amine extractants.
3. Effect of temperature, water coextraction, and process considerations',
Industrial & engineering chemistry research, vol. 29, no. 7, pp. 1333-1338.
Tong, W-Y, Fu, X-Y, Lee, S-M, Yu, J, Liu, J-W, Wei, D-Z & Koo, Y-M 2004,
'Purification of L (+)-lactic acid from fermentation broth with paper sludge as a
cellulosic feedstock using weak anion exchanger Amberlite IRA-92',
Biochemical Engineering Journal, vol. 18, no. 2, pp. 89-96.
Tong, Y, Hirata, M, Takanashi, H, Hano, T, Matsumoto, M & Miura, S 1998, 'Solvent
screening for production of lactic acid by extractive fermentation', vol.
Turton, R, Bailie, RC, Whiting, WB & Shaeiwitz, JA 2008, Analysis, synthesis and
design of chemical processes, Pearson Education.
Urbance, SE, Pometto, AL, DiSpirito, AA & Denli, Y 2004, 'Evaluation of succinic
acid continuous and repeat-batch biofilm fermentation by Actinobacillus
succinogenes using plastic composite support bioreactors', Applied
microbiology and biotechnology, vol. 65, no. 6, pp. 664-670.
Uslu, H 2009, 'Reactive extraction of formic acid by using tri octyl amine (TOA)',
Separation science and technology, vol. 44, no. 8, pp. 1784-1798.
220 | Page Uslu, H, İsmail Kırbaşlar, S & Wasewar, KL 2009, 'Reactive extraction of levulinic
acid by amberlite LA-2 extractant', Journal of Chemical & Engineering Data,
vol. 54, no. 3, pp. 712-718.
Uslu, H & Kırbaslar, SI 2008, 'Equilibrium studies of extraction of levulinic acid by
(trioctylamine (TOA)+ ester) solvents', Journal of Chemical & Engineering Data,
vol. 53, no. 7, pp. 1557-1563.
Uslu, H & Kirbaşlar, SIs 2008, 'Investigation of levulinic acid distribution from
aqueous phase to organic phase with TOA extractant', Industrial & engineering
chemistry research, vol. 47, no. 14, pp. 4598-4606.
Van Maris, AJ, Geertman, J-MA, Vermeulen, A, Groothuizen, MK, Winkler, AA,
Piper, MD, van Dijken, JP & Pronk, JT 2004, 'Directed evolution of pyruvate
decarboxylase-negative Saccharomyces cerevisiae, yielding a C2-independent,
glucose-tolerant, and pyruvate-hyperproducing yeast', Applied and
Environmental Microbiology, vol. 70, no. 1, pp. 159-166.
Vane, LM 2005, 'A review of pervaporation for product recovery from biomass
fermentation processes', Journal of Chemical Technology and Biotechnology,
vol. 80, no. 6, pp. 603-629.
Waghmare, MD, Wasewar, KL, Sonawane, SS & Shende, DZ 2011, 'Natural
nontoxic solvents for recovery of picolinic acid by reactive extraction', Industrial
& engineering chemistry research, vol. 50, no. 23, pp. 13526-13537.
Wang, J, Pei, Y, Zhao, Y & Hu, Z 2005, 'Recovery of amino acids by imidazolium
based ionic liquids from aqueous media', Green chemistry, vol. 7, no. 4, pp.
196-202.
221 | Page Wardell, JM & King, CJ 1978, 'Solvent equilibriums for extraction of carboxylic acids
from water', Journal of Chemical and Engineering Data, vol. 23, no. 2, pp. 144-
148.
Wasewar, K & Keshav, A 2010, 'Seema. Physical extraction of propionic acid', Int. J.
Res. Rev. App. Sci, vol. 3, no. 3, pp. 290-302.
Wasewar, KL 2005, 'SeparationofLacticAcid: RecentAdvances', Chemical and
biochemical engineering quarterly, vol., pp. 159-172.
Wasewar, KL 2012, 'Reactive extraction: an intensifying approach for carboxylic acid
separation', International Journal of Chemical Engineering and Applications,
vol. 3, no. 4, p. 249.
Wasewar, KL, Heesink, ABM, Versteeg, GF & Pangarkar, VG 2002a, 'Equilibria and
kinetics for reactive extraction of lactic acid using Alamine 336 in decanol',
Journal of Chemical Technology and Biotechnology, vol. 77, no. 9, pp. 1068-
1075.
Wasewar, KL, Heesink, ABM, Versteeg, GF & Pangarkar, VG 2002b, 'Reactive
extraction of lactic acid using alamine 336 in MIBK: equilibria and kinetics',
Journal of Biotechnology, vol. 97, no. 1, pp. 59-68.
Wasewar, KL, Heesink, ABM, Versteeg, GF & Pangarkar, VG 2004, 'Intensification
of conversion of glucose to lactic acid: equilibria and kinetics for back extraction
of lactic acid using trimethylamine', Chemical Engineering Science, vol. 59, no.
11, pp. 2315-2320.
Wasewar, KL, Shende, D & Keshav, A 2011, 'Reactive extraction of itaconic acid
using tri‐n‐butyl phosphate and aliquat 336 in sunflower oil as a non‐toxic
222 | Page diluent', Journal of Chemical Technology and Biotechnology, vol. 86, no. 2, pp.
319-323.
Wasewar, KL & Shende, DZ 2011, 'Equilibrium study for reactive extraction of
caproic acid in MIBK and Xylene', Engineering, vol. 3, no. 08, p. 829.
Wasewar, KL, Yawalkar, AA, Moulijn, JA & Pangarkar, VG 2004, 'Fermentation of
glucose to lactic acid coupled with reactive extraction: a review', Industrial &
engineering chemistry research, vol. 43, no. 19, pp. 5969-5982.
Wennersten, R 1983, 'The extraction of citric acid from fermentation broth using a
solution of a tertiary amine', Journal of Chemical Technology and
Biotechnology, vol. 33, no. 2, pp. 85-94.
Xu, Z, Shi, Z & Jiang, L 2011, '3.18 - Acetic and Propionic Acids A2 - Moo-Young,
Murray', in Comprehensive Biotechnology (Second Edition), Academic Press,
Burlington,pp.189-199.
Yabannavar, V & Wang, D 1991a, 'Analysis of mass transfer for immobilized cells in
an extractive lactic acid fermentation', Biotechnology and Bioengineering, vol.
37, no. 6, pp. 544-550.
Yabannavar, V & Wang, D 1991b, 'Strategies for reducing solvent toxicity in
extractive fermentations', Biotechnology and Bioengineering, vol. 37, no. 8, pp.
716-722.
Yang, ST, White, SA & Hsu, ST 1991, 'Extraction of carboxylic acids with tertiary and
quaternary amines: effect of pH', Industrial & engineering chemistry research,
vol. 30, no. 6, pp. 1335-1342.
223 | Page Zacharof, M-P & Lovitt, RW 2013, 'Complex effluent streams as a potential source of
volatile fatty acids', Waste and biomass valorization, vol. 4, no. 3, pp. 557-581.
Zelle, RM, de Hulster, E, van Winden, WA, de Waard, P, Dijkema, C, Winkler, AA,
Geertman, J-MA, van Dijken, JP, Pronk, JT & van Maris, AJ 2008, 'Malic acid
production by Saccharomyces cerevisiae: engineering of pyruvate
carboxylation,oxaloacetate reduction, and malate export', Applied and
Environmental Microbiology, vol. 74, no. 9, pp. 2766-2777.
Zhang, L, Yu, F, Chang, Z, Guo, Y & Li, D 2012, 'Extraction equilibria of picolinic acid
with trialkylamine/n-octanol', Journal of Chemical & Engineering Data, vol. 57,
no. 2, pp. 577-581.
Zhou, J, Bi, W & Row, KH 2011, 'Purification of lactic acid from fermentation broth by
spherical anion exchange polymer', Journal of applied polymer science, vol.
120, no. 5, pp. 2673-2677.
Zolgharnein, J, Asanjarani, N & Shariatmanesh, T 2013, 'Taguchi L 16 orthogonal
array optimization for Cd (II) removal using Carpinus betulus tree leaves:
adsorption characterization', International Biodeterioration & Biodegradation,
vol. 85, pp. 66-77.
224 | Page