Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254

Contents lists available at ScienceDirect

Chemical Engineering Research and Design

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

A comprehensive study on equilibrium and

kinetics of morpholine extraction from aqueous
stream with CA in toluene: Experimental
evaluation, extraction model and parametric
optimization employing desirability function

Alka Kumari a,b , B. Karuna a , B. Satyavathi a,b,∗

a Process Development and Design Group, Chemical Engineering Division, CSIR – Indian Institute of Chemical
Technology, Hyderabad, Telangana 500007, India
b Academy of Scientific and Innovative Research (AcSIR), CSIR – Indian Institute of Chemical Technology Campus,

Hyderabad, Telangana 500007, India

a r t i c l e i n f o a b s t r a c t

Article history: The present study explores the equilibrium and kinetics for the reactive extraction of mor-
Received 9 September 2017 pholine, an important industrial reagent from its aqueous stream using capric acid, CA
Received in revised form 12 March (extractant) in toluene (diluent). An equilibrium model that employs the mass action law
2018 was also developed to evaluate the optimum apparent equilibrium constants (KE ), stoichio-
Accepted 16 March 2018 metric ratio (m) and physical constant (S). Equilibrium model was valid in representing the
Available online 26 March 2018 mechanism of morpholine extraction. Moreover, for the first time the study availed the
simultaneous optimization of two significant response characteristics; extraction efficiency
Keywords: (%E) and loading ratio (Z) using Box–Behnken design (33 ) employing multivariate desirabil-
Reactive extraction ity function. The statistical models predicted %E of 80.2 and Z of 0.64 for the optimum
Loading ratio combination of process parameters as follows: CMO = 5%, CCAO = 5% and T = 303.15 K with the
Box–Behnken design desirability of 0.947. Further, the intrinsic kinetics of the extraction model demonstrated
Desirability function that the reactions between morpholine and CA fall in regime 3 (fast chemical reaction in
Extraction model diffusion film), follows first order kinetics with respect to morpholine and displays exother-
Kinetic study micity of the process as revealed from thermodynamic studies. The equilibrium and kinetic
data is useful for the development, design and reliable scale-up of the extraction process.
© 2018 Published by Elsevier B.V. on behalf of Institution of Chemical Engineers.

1. Introduction
4-Methyl-1,3-dioxolan-2-one or morpholine has emerged as a com-
mercially competent product having widespread applications in
petrochemical plants, rubber industries, dye and resin industries, cok-
ing as well as in many other organic chemical industries. It has often
been selected for temperature-dependent multi-component solvent
system applications such as in hydroformylation reaction and used
in fruit waxing (Behr et al., 2008; Lee et al., 2014; Kumari et al., 2017).
It is also used as separating agent and implemented as a building
block in the preparation of antibiotics and, anti-cancer agents (Lee
et al., 2014; McGuire and Dimitrios, 1999). In this context, the sep-
aration of morpholine from aqueous waste streams is a statutory
obligation. In the preparation of morpholine by the dehydration of
diethanolamine with sulphuric acid or hydrochloric acid a reaction
mixture is obtained which contains the morpholine as an acid salt

∗
Corresponding author at: Process Development and Design Group, Chemical Engineering Division, CSIR – Indian Institute of Chemical
Technology, Hyderabad, Telangana 500007, India.
E-mail addresses: alka@iict.res.in (A. Kumari), satya@iict.res.in (B. Satyavathi).
https://doi.org/10.1016/j.cherd.2018.03.027
0263-8762/© 2018 Published by Elsevier B.V. on behalf of Institution of Chemical Engineers.
244 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254
dissolved in excess aqueous acid. An amount of alkali added renders
the mixture strongly alkaline, which is generally separated by steam
distillation. The concentration of aqueous morpholine so produced
depends on the quantity of excess alkali present during steam distil-
lation (George and Laemmle, 1957). The recovery of morpholine from
the aqueous stream has not been in practice by distillation owing to its
economical constraint. The prevalence of strong bonding between the
amino functional group in morpholine and water yields a highly non-
ideal mixture, making the separation difficult and requires a large and
costly distillation column. Besides, morpholine boil at higher tempera-
tures than water, the entire water needs to be vaporized by distillation.
This introduces the use of vast amounts of heat, steam and cooling
water (Somekh and Hawkes, 1969). To preclude the difficulties encoun-
tered in the distillation operation, some of the prior art work resorted
the processes like utilization of concentrated NaOH solution to dewa-
ter an aqueous morpholine solution. To regenerate the concentrated
NaOH solution necessitates evaporating the large amounts of water by
distillation, as previously indicated, costly and undesirable. Another
method utilizes ethyl ether as an extracting agent, which is again
extremely inefficient and expensive due to high solvent losses (Somekh
and Hawkes, 1969). Amidst the various intensified techniques, reactive
extraction has emerged as a promising one that leads to the separation
of close volatility or non-ideal mixtures with a substantial reduction in
the overall cost as compared to other methods (Marti et al., 2011). In this
context, reactive extraction technique was explored to avoid the above
stated difficulties encountered during separation of morpholine from
its aqueous stream. The extraction parameters such as initial concen-
tration in the aqueous phase, extractant concentration, temperature,
types of complexes formed, properties of diluents and extractants play
a vital role in influencing the reactive extraction process (Kahya et al.,
2001). In this study, capric acid (CA) has been selected as a complex-
ion agent, which is immiscible in water and highly soluble in selected
diluents (toluene).
Additionally, an equilibrium model employing mass action law with
suitable assumptions was utilized to regress the stoichiometries in the
reactive extraction. For an amine based aqueous solution, literature
is found scarce given a model and correlation of apparent equilib-
rium constants. Further, Box–Behnken multivariate optimization based
on desirability approach was employed to obtain the global optimum
combination of design variables, where two dependant variables were
selected as the response characteristics. In general trend, only one
response characteristics i.e., distribution coefficient (KD ) or extraction
efficiency (%E) has been investigated in the optimization of the reactive
extraction process, although the loading ratio (Z) imparts a crucial role
in the optimization since it decides the stoichiometry of the amine–acid
complex formed. Therefore, in the present study two response charac-
teristics, %E and z were simultaneously optimized in the multivariate
process.
Fundamentally, overall extraction rate depends on two different
models; interfacial and homogeneous reaction models (Steensma et al.,
2007). In the kinetic study, a homogeneous reaction model was adopted
owing to the limited solubility of morpholine in the organic phase. The
Vantoff’s expression was fitted to the thermodynamic data to obtain
the enthalpy and entropy for extraction of morpholine.
2. Theoretical consideration
2.1. Equilibrium models
The equilibrium model is interpreted by mass action law pro-
posed by Guldberg and Waage in 1864 describes the physical
and chemical phenomena and also explains the interaction
mechanism between the components (morpholine and water)
of the aqueous phase and organic (CA and toluene) phase
(Datta and Kumar, 2013). It is assumed that particular concen-
tration of the components is proportional to the mass action
law activities of the aqueous and organic phase species and
the non-ideals coupled with the reactive system (Datta et al.,
Fig. 1 – Mechanism of reactive extraction of morpholine
with capric acid.
2014). Therefore, a mathematical model is developed using the
apparent equilibrium constant for the reaction equilibrium.
This model can be classified into two types (i) physical extrac-
tion and (ii) chemical extraction. The details of the physical
and chemical equilibrium, Box–Behnken design, desirability
functions and kinetic models (Table S1) are described in the
supporting information under theoretical consideration.
The interaction of amine-acid molecules in the chem-
ical extraction happened through (i) hydrogen bonding of
undissociated amine molecules (ii) by ion-pair formation. The
mechanism of interfacial reactive extraction between mor-
pholine with CA is displayed in Fig. 1.
The morpholine molecules protonated in the aqueous
phase and forms HM+ ions and moves towards organic phase.
The reaction occurs at the interface between aqueous and
organic phase and forms acid–amine complex, which stabi-
lizes in the organic phase. In the present study, the reaction
between morpholine and CA occurs mainly through H-bond
formation because most of the morpholine is in the undis-
sociated form in aqueous phase. The undissociated form of
morpholine in the aqueous phase transferred (diffuse) to the
interface, CA diffuses through the organic phase and moves
to the interface, where amine-acid associates through H-
bonding and the complex dissolves in the bulk organic phase.
3. Experimental section
3.1. Chemicals
All the chemicals used were of analytical grade. The morpho-
line (purity ≥ 99.0%) procured from Sigma–Aldrich was stored
in desiccators to prevent from moisture attack and dust con-
tamination. Toluene (purity ≥ 99.5%) was acquired from s–d
fine chemical Ltd., Mumbai, India and stored in a segregated
and approved area. CA (purity ≥ 99%) was supplied from Alfa-
Aesar, Heysham, England and stored away from extreme heat.
Sodium 1,2-naphthoquinone-4-sulfonate (purity ≥ 97%) was
provided by Alfa-Aesar, Heysham, England. Distilled water
was used for the experimentation to prepare required con-
centrations of aqueous solutions.
3.2. Instruments
The mixtures were prepared using a high precision Sartorius
CP224S electronic balance with an accuracy of ±0.0001 g. The
equilibrium studies were carried out using a shaking incubator
(Model No.-LSI4018R) provided by Daihan Labtech India Pvt.,
Ltd., capable of maintaining the temperature within ±0.1 K. A
Polyscience refrigerating and heating thermostat model 9702
with a digital temperature controller was used to circulate
hot water through the jacketed reactor for maintaining the
 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254
Table 1 – Levels of variables in Box–Behnken
experimental design.
Independent variables Symbols Levels
−1 0 +1
Temperature (◦ C) A 30 40 50
Feed concentration (%) B 0.5 2.75 5 2.25
Extractant concentration (%) C 5 10 15
constant temperature (±0.01 ◦ C accuracy). An overhead stirrer
supplied by Remi Motors Ltd. (Mumbai, India), equipped with
a digital display was used for stirring the reaction mixture. A
platinum resistance type thermocouple with an uncertainty of
±0.05 K was used to measure the temperature inside the reac-
tor. The pH of the solution was measured using a Sartorious
PB-11 pH meter. The morpholine concentration in aqueous
phase was analyzed using gas chromatography (Nucon, Model:
5765) equipped with a thermal conductivity detector (TCD).
Lab India 3000+ Model UV/vis spectrophotometer was used
for the aliquot sample analysis of morpholine concentration
in the aqueous phase.
3.3. Equilibrium studies
All physical and chemical extraction experiments were car-
ried out in 100 ml iodine flasks. The aqueous phase was
prepared by dissolving pre-required amounts of morpho-
line in distilled water to make different feed concentrations
(0.06–0.57) kmol m−3 . The organic phase (0.29–0.86) kmol m−3
was prepared by dissolving the different quantities of extrac-
tants (CA) into diluents (toluene). Equal volumes (20 ml) of
the organic and aqueous phases were added to the flask
without pH alteration. The flask containing mixtures were
equilibrated for 3 h at a shaking speed of 200 rpm at tempera-
tures 303.15–323.15 K and left for 12 h to obtain clear phases.
Thereafter the layers were separated, and the aqueous layer
was collected and measured for pH and morpholine concen-
tration. Some of the experiments were performed in duplicate
to test the consistency of the data, which was found within
±0.05.
3.4. Box–Behnken experimental design coupled to RSM
Box–Behnken design is an efficient technique for process
optimization and designing a model, wherein the extreme
combinations of factor are excluded (Islam et al., 2009;
Varala et al., 2016b). In the present research, a Box–Behnken
design coupled to response surface methodology (RSM) was
employed wherein; initial morpholine concentration, temper-
ature and extractant concentration were identified as the most
significant process variables in the extraction studies. The pro-
cess variables were designated with A, B and C in the design
matrix and the levels of the variables considered (low −1, cen-
ter 0 and high +1) were shown in Table 1.
Statistical design software (Design expert: version 10) was
utilized to generate the (33 ) BBD design matrix and is pre-
sented in Table S2. According to a randomized experimental
design, a total of 17 experiments with 5 center points need
to be carried out to draw a complete picture of the effect of
factors (A, B and C) and reported along with experimental and
predicted responses for %E and Z.
245
3.5. Determination of extraction kinetics
For kinetic studies, a glass stirred reactor was used with a
∇x
constant interface area of 500 ml capacity (D: 0.062 and H:
0.092 m). The reactor consists of a flat bottom and a stirrer with
10
two paddle impellers of four blades each. The temperature
inside the reactor was maintained at constant temperature
5 303.15 K by circulating water through the jacket of the reactor.
The reactor was provided with an arrangement of withdraw-
ing aqueous samples with uninterrupted stirring. The organic
phase (100 ml) was first introduced to this reactor and the posi-
tion of the four-blade paddle impellers were adjusted to 5 mm
below and 5 mm above the interface. Then, the aqueous solu-
tion was added cautiously without disturbing the interface,
subsequently stirring of the mixtures started. Samples were
withdrawn from the aqueous phase at a precise time interval
and analyzed for morphine concentration. The specific rate of
extraction RA (kmol m−2 s−1 ) were calculated as described by
Tang et al. (2012) and Datta and Kumar (2013).
3.6. Analytical method
The morpholine concentration in aqueous layer was analyzed
employing a gas chromatography technique equipped with a
TCD detector. A Tenax column was used for the separation of
components and Helium was used as carrier gas with a flow
rate of 40 ml min−1 , injector and detector temperatures were
maintained at 150 and 220 ◦ C, respectively. The oven temper-
ature was maintained at 220 ◦ C. The lower concentrations of
morpholine in aqueous phase exceeding the detection limit of
GC were analyzed by spectrophotometric technique. The sam-
ple aliquots after dilution (dilution factor) were mixed with
three drops of 1N sodium hydroxide solution and 0.2 ml of
sodium 1,2-naphthoquinone-4-sulfonate. Then, the mixture
was allowed to stand for 20 min, but not over 30 min and the
absorbance was measured at 480 nm with 1 cm cell quartz
cuvettes against reagent blank (Stevens and Skov, 1965).
Each sample was analyzed twice and the error was noticed
within ±0.05. The morpholine concentration in the organic
phase was calculated by material balance. The error in pH
measurement was ±0.02.
4. Results and discussion
The experiments have been carried out to investigate the
physical (pure toluene) and chemical (CA in toluene) equilib-
rium studies for extraction of morpholine.
4.1. Physical extraction
The physical equilibrium studies were carried out for
the extraction of morpholine with toluene alone for 1:1
volume ratio of phases varying the morpholine concentra-
tion (0.06–0.57) kmol m−3 at different operating temperatures
(303.15–323.15) K. The physical equilibrium isotherms were
illustrated in Fig. 2 at different temperatures described a cor-
relation between morpholine concentrations in aqueous and
organic phase. The isotherm followed Henry’s law type model
at 303.15 K whereas at higher temperature, it displayed non-
ideal behavior due to deviation from Henry’s law (Wasewar
et al., 2002). It was observed that as initial morpholine concen-
tration increases from (0.06 to 0.57) kmol m−3 , the distribution
coefficients (Kdiluent ) decreases from (0.19 to 0.10) at a constant
D
temperature 303.15 K. And also the distribution coefficients
246 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254
Fig. 2 – Physical equilibrium isotherms for extraction of
morpholine in toluene at temperatures 303.15 K , 313.15 K
䊉, 323.15 K .
(Kdiluent
D ) decreased from (0.19 to 0.12) with an increase in tem-
perature at a constant feed concentration of 0.06 kmol m−3 .
Higher temperature facilitates the more back extraction of
solute molecules from organic to an aqueous phase. The
Kdiluent
D < 1, which is not much significant for extraction of
morpholine, accounted the morpholine affinity towards water
than organic solvent, renders the difficulty of separation of
physical extraction (Wasewar et al., 2011).
Also, it was concluded that the morpholine concentra-
tion in the organic phase increases with that of aqueous
phase at equilibrium. A statistical analysis was performed to
obtain the parameters signifying partition (P) and dimeriza-
tion coefficient (D) in regression equation Eq. (6) (supporting).
It was observed that P varies from 0.16 to 0.12 in the tem-
perature range of (303.15–323.15) K and negligible contribution
to Kdiluent
D by the dimerization of solute–solute interaction in
the organic phase. The P and Kdiluent
D were almost identical for
the morpholine displaying the existence of only one chemical
form of solute in physical extraction (Rewatkar et al., 2016).
As toluene is aprotic solvent, the solubilisation of mor-
pholine is dominated by the formation of ␲-bonded complex
attributed to the morpholine imino hydrogen bonding with
the ␲-electrons of toluene. Aroney et al. (1964) have reported
in their study that molecules of morpholine interact in the
inert diluents, nevertheless weak, come into view to increase
in the order as N-H < N-lone-pair. They have also established
the dominance of ␲-hydrogen bonding between solute-
solvent rather than solute-solute association, evidenced by
the infrared and nuclear magnetic resonance spectra. The
partition coefficient, P and distribution coefficient Kdiluent D
decreases with temperature, which is an indicative of the
exothermicity of solvation process of morpholine in toluene
(Rewatkar et al., 2016). The increase in temperature increases
excess molar volumes manifested by weakening of interac-
tions between molecules of toluene and morpholine through
breaking off ␲-hydrogen bonding.
4.2. Chemical equilibrium
The chemical (reactive extraction) equilibrium data was
obtained for extraction of morpholine from aqueous
solution (0.06–0.57) kmol m−3 using CA as extractant
(0.29–0.87) kmol m−3 dissolved in toluene at 303.15 K. The
concentration of extractant was limited upto 15% due to the
Fig. 3 – Chemical equilibrium isotherms for extraction of
morpholine with various concentrations of CA in toluene at
303.15 K. pure toluene , 5% CA 䊉, 10% CA , 15% CA .
high viscosity of CA to avoid mass transfer resistance as well
to restrict the three-phase formation in layer separation.
A plot of chemical equilibrium isotherms was shown in
Fig. 3, which shows aqueous and organic phase equilibrium
concentrations at 303.15 K.
The effect of extractant (CA) is evident on extraction of
morpholine. It can be observed that at a fixed concentration
of morpholine, the distribution coefficient increases with an
increase in CA concentrations, whereas upon varying the acid
concentration for a fixed extractant concentration, KD value
decreases. It was concluded that chemical extraction is far
better than physical extraction. As ascribed in the above the-
oretical section, the reactive extraction of morpholine was
mainly dominated by the hydrogen bond formation between
solute-extractant because most of the morpholine is in the
undissociated form in aqueous phase. This fact was certified
by the pH measurement of the initial aqueous solution. The
pH value plays a vital role in the existing form of solute in the
aqueous phase, when the equilibrium pH value in the aque-
ous solution is larger than the pKa , dominates the existing
form of morpholine M which undermine the ion-pair com-
plex mechanism whereas, smaller pH than pKa dominates
the MH+ form of morpholine (Chai et al., 2016). In the present
study the higher pH of aqueous solution (∼10) > pKa (8.63) of
morpholine, quantify the negligible morpholine dissociation.
Thereby, hydrogen bonding dominates the mechanism. Mor-
pholine being a secondary amine have a lone pair of electrons
on the nitrogen atom attached to the heterocyclic ring co-
ordinated with the CA in toluene, the extractants bind the
amine molecules with a dipole–dipole or dipole-ion type of
intermolecular hydrogen bonding. The bond strength depends
on the stability of the partially charged proton acceptor (mor-
pholine) and donor (CA) atoms. The electronegativity of the
atoms decides the stability of its charges which is a measure
of the tendency of an atom to pull a shared pair of electrons
toward it. The higher electronegativity difference between
hydrogen (2.1) and nitrogen (3.0) creates stronger hydrogen
bonding (Rewatkar et al., 2016).
This study includes the utilization of diluents in reactive
extraction to improve the physical properties (viscosity, den-
sity, surface tension at the interface) of the extractants and
solvation of the acid–extractant complexes formed. The dilu-
ent (toluene) chosen play a decisive role with regard to the
bonding with solute–extractant complex formed in the reac-
 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254 247

tive extraction. Generally, the bonding can be categorized into Substituting the values of C̄m1 and CM from Eqs. (2) and (5),
specific interaction and the general solvation (Wasewar et al., respectively in Eq. (8) is derived as:
2011). Toluene being aromatic and aprotic diluent provides a
better distribution because of the solvation occurred through qKE CM,total m−1 CCA s
KD = + (9)
the interaction of the aromatic ␲-electrons of toluene with the CH+ CH+
1+ Ka 1+ Ka
complex. The increase in extraction efficiency is due to the
increase in acidic extractant concentration. As the concentra-
The free extractant concentration, C̄CA in the organic phase
tion of acidic extractant CA increases in the organic phase,
at equilibrium is represented as
enhancement in %E was observed, as the higher concentration
fulfils the stoichiometric obligation of the extractant.
C̄CA = C̄0CA − C̄m1 (10)

4.3. Modeling and optimization of stoichiometric Using Eq. (10), the equation is deduced as follows:
parameters (KE , m and S)
C̄0CA
An equilibrium model was developed to evaluate the optimum C̄CA = (11)
(1 + KE CM m )
apparent equilibrium constants (KE ), stoichiometric ratio of
morpholine to CA in toluene (m) and physical constant (S).
The value of C̄CA from Eq. (11) is substituted in Eq. (9), which
The mechanism of the reactive extraction process may be pro-
results in Eq. (12):
posed based on the following assumptions: (i) morpholine is
extracted into the organic phase through the physical solubil- mKE C̄0CA Cm−1 s
ity of the solute in the diluent of the organic phase (physical KD =  M,total
m +   (12)
Ka
constant, S) (ii) poor solubility of extractant and diluents in 1+ Ka
+ KE Cm 1+ [H+ ]
[H+ ] M
water, (iii) For simplicity, it was assumed that the solute (mor-
pholine) carries negligible amount of water in the organic To estimate the optimized values of KE , m and S for the reac-
phase, and (iv) negligible change in the volume of each phase tive extraction of morpholine, an objective function based on
at equilibrium (Kumar et al., 2011; Chai et al., 2016; Thakre least-squares method (know as root mean-square deviation,
et al., 2016). The extraction of morpholine (undissociated part, rmsd) was minimized for predicting
 the stoichiometry of the
n
M) with extractant (CA) is an interfacial reaction described by (K −KD,pred )
2
1 D,exp
the following equation. complexation reaction. rmsd = n , where n is
the number of data points. By fitting the experimental data to
KE Eq. (12), the obtained optimized parameters such as KE , m and S
mM + CA↔Mm (CA) (1)
were reported in Table 2 and subsequently used for predicting
the KD values (Uslu et al., 2014).
With an expression of equilibrium constant as follows: The kE were found to be almost constant with increase
in concentration of morpholine (0.06–0.57) kmol m−3 due to
C̄m1 an appropriate concentration of CA (0.29–0.87) kmol m−3 . The
KE = (2)
CM m C̄CA kE values decreased with increasing CA concentration due to
availability of free extractant at higher concentration of CA
Morpholine dissociates at equilibrium in the aqueous (Kumar et al., 2011). The kE show dependency on the acid-
phase as mentioned in Eq. (1) and the dissociation constant ity of the CA and the specific basicity of morpholine (Zhou
(Ka ) is given by Eq. (3): et al., 2013). The stoichiometry coefficient m depends on the
dielectric constant of the diluent, which might be considered
CM CH+ for solute-solvent interaction considering the restrictions of
Ka = (3)
CMH+ solute solubilisation either by solvent or extractant attributed
to the existence of ionisable group in the solute chemical
Therefore, the total morpholine concentration in the aque- structure (Kumar et al., 2011). The values of S reveal that
ous phase, CM can be expressed in terms of undissociated physical extraction can be considered for extraction of mor-
solute concentration CM , Ka, and proton concentration CH+ . pholine in toluene and decreases with the increase in initial
morpholine concentration, support the outcome observed in
CM,total = CM + CMH+ (4) physical extraction. As shown in Fig. 4, the calculated val-
ues of KD , are in very good resemblance to the experimental
 Ka
 values. Standard deviation (SD) was calculated to test the
CM,total = CM 1 + (5)
[H+ ] consistency of the model with the experimental values. The
computed R2 values of fit of the model were obtained ≥0.99 for
Physical extraction is expressed as: reactive extraction using extractant in toluene. Therefore, the
developed model is adequate in representing the equilibrium
S mechanism of morpholine extraction.
Maq ↔Morg (6)
The stoichiometry of the overall extraction reaction gen-
erally depends on the loading ratio in the organic phase, Z
CMorg
S= (7) given in supporting Eq. (9). In the present study, in most of the
CMaq
cases the organic phase is not highly concentrated i.e. loading
ratios (Z < 0.5), it is expected that the 1:1 complex is formed.
CMorg qCm1 sCM It is apparent that 1:1 amine–acid complex is formed in the
KD = = + (8)
CM + CMH+ CM + CMH+ CM + CMH+ organic phase, as the loading ratio decreases with increas-
248 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254

Table 2 – Values of equilibrium complexation constants (kE ), reacting molecules (m) and physical extraction constant (S)
with CA as extractant in toluene with different concentrations of morpholine.
 
CA (kmol m−3 ) kE m S SD CMO (kmol m−3 ) kE m S SD
o

0.29 7.96 1.13 0.13 0.26 0.057 3.64 1.00 0.19 0.14
0.41 3.04 0.90 0.13 0.25 0.143 3.96 1.11 0.14 0.27
0.58 2.24 1.02 0.13 0.27 0.316 3.92 1.18 0.12 0.25
0.75 2.07 0.96 0.13 0.24 0.459 4.11 0.94 0.10 0.26
0.87 2.06 0.97 0.13 0.18 0.574 4.17 1.03 0.10 0.29

CMO : initial concentration of morpholine in aqueous phase, SD: standard deviation.

It can be noticed from KE1 value that it is approximately

equal to the model parameters investigated from the above
studies validate the formation of 1:1 amine:acid complex.

4.4. Box–Behnken multi-response design

In this study, experiments were conducted according to the

Fig. 4 – Comparison of experimental and predicted KD for
different concentration of morpholine (0.057 , 0.143 ,
0.316 , 0.459 䊉, 0.574×) kmol m−3 .
Fig. 5 – Plot of z/1 − z versus C∗aq for estimation of the 1:1
equilibrium complexation constant KE1 of morpholine with
CA in toluene at 303.15 K.
ing CA concentration, which was satisfied using supporting
Eq. (10). For confirmation z/1 − z vs. C∗aq was fitted in Fig. 5
which yields the equilibrium complexation constant (KE1 ) val-
ues in the range of (7.94–2.05) m3 kmol−1 for the 1:1 amine:acid
complex for the extraction of morpholine with different con-
centration of CA (0.29–0.87) kmol m−3 dissolved in toluene. In
this case, a ratio of moles of solute to the mole of extraction in
the organic phase is less than unity (Thella et al., 2011). For the
higher concentration of morpholine (0.75 and 0.87) kmol m−3 ,
the extractant is more loaded (z > 0.5) with the morpholine
molecules compared to lower morpholine concentrations.
Box–Behnken design and the results obtained were employed
for the statistical significance. In regard to describe the sta-
tistical significance of combined parameters on responses, it
was identified from supporting information provided in Table
S3 that amid all, the quadratic model was the most suitable
one describing the effect of process parameters on responses.
The goodness of the model in the present study is accred-
ited through R2 and adj R2 values and they acquired 0.993 and
0.984 for %E, 0.981 and 0.956 for Z, respectively. For extraction
model, the R2 obtained a value of 98.1, i.e., 98.1% of total vari-
ation revealed a good adjustment of the experimental data
for morpholine extraction and only 1.9% is left with resid-
ual variability. The variation in adj R2 and pred R2 should
be approximate to 0.2 to be in reasonable agreement. In this
study, the pred R2 is 0.88, and adj R2 is 0.98 for %E and, which
implies that the variation in adj R2 and pred R2 is in reason-
able agreement that exhibits the adequacy of the %E response
model. In case of loading ratio, the pred R2 = 0.70 is not close to
adj R2 = 0.956 i.e. the difference is >0.2, owing to the difference
might exhibit block effect (Varala et al., 2016b). The adequacy
of the model is evident from the fact that the data points in
Fig. S1 sprawl alongside straight line as the residuals of the
prediction of the responses are less than 2%.
Additionally, the analysis of variance (ANOVA) was
performed to test the statistical significance of process param-
eters towards response (Varala et al., 2016a). The p-value
Prob > F ≤ 0.05 indicates that the model terms are statistically
significant. The p-value Prob > F ≤ 0.1 indicates marginally sig-
nificant and >0.1 indicates statistically insignificant (at 95%
confidence limit) (Varala et al., 2016b).
ANOVA results presented in Table 3 indicate the F-values
of the model for %E was 110.78, the high values of F imply
that the model is statistically significant (at 95% confidence
limit). There is only a 0.01% chance for a larger F-value which
could occur due to noise. The analysis signifies that the linear
effects of A, BC and square effects of A2 , B2 and interaction
of AC, BC were statistically significant for %E, whereas a high
probability value of B2 (P = 0.286 > 0.05) and interaction of AB
(P = 0.89 > 0.05) for a sequential and the quadratic models indi-
cate non-significant effect towards extraction of morpholine.
Hence, the model can be used to navigate the design space
(Ozer et al., 2009).
A pie chart is shown in Fig. 6 with a series of bars besides
where length reveals the frequency or impact of parameters.
The factors represented by the big bars are relatively more
 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254 249

Table 3 – Analysis of variance of the response surface quadratic models, %E.

Source SS DOF MS F-Value p-value Prob > F Remarks

Model 2022.44 9 224.72 110.78 <0.0001 Significant

A-Temperature 590.72 1 590.72 291.20 <0.0001
B-Feed 787.54 1 787.54 388.23 <0.0001
C-Extractant 519.20 1 519.20 255.95 <0.0001
AB 0.04 1 0.04 0.02 0.8904
AC 38.08 1 38.08 18.77 0.0034
BC 13.95 1 13.95 6.88 0.0343
A2 30.02 1 30.02 14.80 0.0063
B2 2.71 1 2.71 1.34 0.2856
C2 34.11 1 34.11 16.81 0.0046
Residual 14.20 7 2.03
Lack of fit 14.20 3 4.73
Pure error 0.00 4 0.00
Cor total 2036.64 16

R2 = 0.993, Pred R2 = 0.888, Adj R2 = 0.984, adequate precision = 35.37, C.V% = 3.64, std. dev = 1.42.

Fig. 6 – Schematic representation of contribution of factors to the extraction efficiency, %E.

significant. It was observed that the first order terms TPCi
manifested the highest significance with a total contribution
of 94.1%, followed by quadratic terms TPCii with a total con-
tribution of 3.31% and then the interaction terms indicative
of the least level of significance with a total contribution of
2.58. % to the extraction efficiency. It is evident from bar graph
that the initial feed (B) showed the highest level of significance
with a contribution of 39.05%, making temperature being the
second factor with 29.29% followed by extractant concentra-
tion with 25.74% in first order terms as compared to other
components for %E.
complex at the interface of aqueous and organic phases. The
disturbed interaction of CA and morpholine exhibits lesser %E
(Kumar et al., 2011). The extraction decreases with increase
in feed concentration at a constant extractant concentration
(CA) of 0.29 kmol m−3 . It was illustrated in Fig. 7b that at a
fixed temperature 303.15 K, with an increase in extractant
concentration (CA) %E increases, however it decreases with
increase in feed concentration. For a constant feed concentra-
tion 0.06 kmol m−3 , %E decreases with increase in temperature
and increases with increase in extractant concentration (CA)
as shown in Fig. 7c.

4.4.1. Interaction effect of process parameters: response
surface plots
The 3-D plots illustrated in Fig. 7 facilitate an examination
of interactive effects of parameters on response, which pro-
vides valuable information about the behavior of the system
within the experimental design. From surface plots, it can be
observed that the interactions of AC, BC have a significant
effect towards response (%E), whereas the interaction AB is
undermined concerning the statistical significance. The max-
imum separation of morpholine is indicated by the surface
confined in the smallest curve of the plot with the other vari-
ables maintained at zero levels. From Fig. 7a it can be found
that as temperature increases, the %E decreased from 56 to
38.9%. The increase in temperature results in an increase in
the thermal energy which interrupts the probable interac-
tions between morpholine and CA to form a stable acid–amine
4.5. Multi-response optimization using desirability
function
The lower and upper limit values concerning all the
responses in the multivariate optimization employing desir-
ability approach were considered and given in supporting
Table S4. The larger-the-better (LTB) funtions of the stan-
dard desirability was used for the optimization of extraction
of morpholine. To achieve this, the importance of each goal
was adjusted with respect to other goals (Islam et al., 2009).
As the initial feed concentration contributes significantly to
the responses, the goal was assigned as “maximum” with
the respective importance of 5, as the higher concentration
is mainly favored in the extraction system. Under these set-
tings and boundary condition, multivariate optimization was
conducted.
250 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254

Fig. 7 – 3 D response surface plots (a) AB with %E (b) BC with %E (c) ACwith %E.

Fig. 8 – Desirability ramp for numerical optimization of five goals.

By seeking from 54 starting points in the response surface
changes, the best optimum condition with overall maximum
desirability was shown in Fig. 8. The best optimum conditions
were achieved at a temperature (A) of 313.15 K, initial feed con-
centration (B) of 5%, and an extractant concentration (C) 5%
with desirability of 0.947.
The experiment was perfomed for the confirmatory run at
optimized conditions and the results obtained were %E: 35.5,
Z: 0.80 which were in reasonable agreement with the predicted
ones %E: 38.15 Z: 0.65.
4.6. Kinetics investigation
Reactive extraction of morpholine using CA involves the prin-
ciple of mass transfer accompanied by a chemical reaction.
Since the solubility of extractant (the reactive component in
organic phase) is negligible in the aqueous phase, whereas
morpholine (the reactive component in the aqueous phase)
has a significant solubility in the organic phase (in diluent),
resistance to mass transfer is determined by interfacial chem-
ical kinetics and molecular diffusion of components towards
and away from the interface.
4.6.1. Physical mass transfer coefficient (kL )
Physical mass transfer coefficient, kL is essential for con-
firming the extraction regime. Because of high distribution
coefficient between morpholine and water system, the aque-
ous diffusion film resistance was neglected, and only the
resistance of organic diffusion film was considered. The trans-
fer of morpholine molecules from the aqueous phase to the
 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254
   Fig. 10 – Effect of speed of agitation, N on the initial rate of
Fig. 9 – A Plot between ln C∗org / C∗org − Corg & time to
estimate kL at N = 1 rev.s−1 , T = 303.15 K.
organic phase, the mass transfer rate (molar flux) was given
by:
Vorg dCorg 
= kL C∗org − Corg
A dt
where A is the interface area (m2 ), Vorg is the organic phase
volume (m3 ) and C∗org is the concentration of morpholine at
the interface. The kL was obtained by conducting a physical
extraction experiments for a morpholine concentration range
of 0.06–0.57 kmol m−3 in pure diluent only (toluene).
The kL was calculated by fitting the experimental data to a
straight line, as shown in Fig. 9 for equal volumes (1:1 aq/org)
and the value obtained to be 4.4 × 10−6 .
4.6.2. Effect of process parameters on the initial rate of
extraction, RA
As explained by Doraiswamy and Sharma (1984), the effects
of the volume ratio of the phases (Vorg /Vaq ), the speed of
agitation, initial feed concentration and extractant concentra-
tion (CA) on the specific rate of extraction must be examined
to determine the reaction regime. Therefore, the experi-
mental data were collected to calculate the specific rate
of extraction in a stirred reactor with varying parameters
viz. speed of agitation for 1.0–1.5 rev.s−1 , initial feed con-
centration of 0.06–0.57 kmol m−3 , initial CA concentration of
0.29–0.87 kmol m−3 in toluene and volume ratio of the phases
at Vorg /Vaq of 1:0.8–1:2. The reaction regime was established by
discerning mechanism proposed by Doraiswamy and Sharma
(1984) as described in Table S1.
4.6.2.1. Effect of agitation speed (N). Effect of stirring speed
was verified to identify the contribution of mass transfer
to the overall extraction regime. If the rate enhances with
stirring speed, diffusion governs the extraction process and
the boundary layer thickness on each side of the interface
decreases. Whilst, if the mass transfer rate is independent
of the stirring speed, the extraction process is under kinetic
control regime. To examine the hydrodynamic effects, the agi-
tation speed varied from 1.0 to 1.5 rev.s−1 , while keeping in
view of accomplishing the absolute renewal of surface (Tang
et al., 2012). While, the agitation speed less than 1.0 rev.s−1 ,
RA values were very low, which is probably due to the partial
renewal of the surface, resulting in a mass transfer resistance-
controlled regime (Marti et al., 2011).
251
extraction, RA . CMO = 0.06 kmol m−3 , CCAO = 0.290 kmol m−3 ,
Vorg /Vaq = 1:1, T = 303.15 K.
(13)
Fig. 11 – Effect of phase volume ratio (Vorg /Vaq ) on initial
rate of extraction, RA . N = 1 rev.s−1 , CMO = 0.06 kmol·m−3 ,
CCAO = 0.29 kmol m−3 , T = 303.15 K.
As shown in Fig. 10, there is no effect of agitation speed on
RA concluded that the extraction of morpholine is governed by
the kinetic control regime in the stirring speed range investi-
gated. And the chemical reaction between morphine and CA
in toluene in stirred cell reactor falls under either regime 1
(very slow) or regime 3 (fast). In this range of agitation speed
the liquid–liquid interface was flat and the interfacial area is
equal to cross sectional area of the stirred cell (Rani et al.,
2010).
4.6.2.2. Effect of phase volume ratio (Vorg /Vaq ). The effect of
phase volume ratio on RA for reactive extraction of morpholine
with CA in toluene was evaluated to differentiate between the
regime 1 and 3. The volume ratio of organic to aqueous phase
was varied in the range of 1:0.8–1:2 at a constant stirring of
1 rev.s−1 .
The plot illustrated in Fig. 11 showed the independency of
the phase volume ratio on the specific rate of extraction RA ,
since the bulk of the volume does not involve in the rate of
extraction as the reaction mostly happens in a narrow region
close to the interface (Marti et al., 2011). This demonstrates
that the reaction between morpholine and CA falls in regime 3
i.e., extraction accompanied by a fast chemical reaction occur-
ring in the diffusion film. Very slow reaction regime, regime 1
is mass transfer independent and thus regime 1 is improbable
in the present case (Gadgil et al., 2014).
252 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254

Fig. 12 – Effect of initial feed concentration on initial rate of

Fig. 13 – Effect of initial CA concentration on initial rate of
extraction, RA . N = 1 rev.s−1 , CCAO = 0.29 kmol m−3 ,
extraction, RA . N = 1 rev.s−1 , CMO = 0.06 kmol m−3 ,
Vorg /Vaq = 1:1, T = 303.15 K.
Vorg /Vaq = 1:1, T = 303.15 K.
4.6.2.3. Order of reaction w.r.to morpholine. The order of the
reaction was determined by changing the initial concen-
tration of morpholine in the aqueous phase from (0.06 to
0.57) kmol m−3 , while keeping the other variables constant.
A plot between RA vs. C∗org illustrated in Fig. 12 provides the
order of reaction concerning initial morpholine concentra-
tion. It was found that RA increases with increase in initial
morpholine concentration. By fitting the experimental data
yields a linear trend gives the order of the reaction m = 1, for
morpholine. The increase in morpholine concentration in the
aqueous phase indicates that the more quantity of morpholine
is offered for reaction by diffusion into the organic phase.

4.6.2.4. Order of reaction w.r.to capric acid. The order of reac-

tion was estimated with respect to capric acid by varying
its concentration from (0.29 to 0.87) kmol m−3 in the organic
phase by keeping other variables constant. The plot between
RA and initial CA concentrations in Fig. 13 signifies that there
was no effect of extractant concentration on RA , indicating the
order of reaction w.r.to extractant is zero (n = 0).
Fig. 14 – Dependency of lnKE on T1 for extraction of
morpholine, CCAO = 0.87 kmol m−3 .
and CA (extractant) concentration. This circumstance further
supports the true intrinsic kinetics observed in the stirred cell
experiments (Wasewar et al., 2002).

4.7. Determination of rate constant, k1 and reaction

5. Thermodynamic studies
regime

The first order reaction rate constant was calculated from
Fig. 12, for m = 1 and n = 0 the governing rate equation for the
initial rate of extraction reduced to:
the pKa value, the amine-acid interaction, the solubility of
RA = [C]∗Morg DAB k1 (14)
The experimental data were fitted to the above equa-
tion to obtain the value of the first order rate constant, k1
and found to be 6.30 s−1 . Diffusion coefficient (DAB ) was esti-
mated using Hayduk and Minhas equation and found to be
DAB = 1.92 × 10−9 m2 s−1 (Reid et√al., 1987). For m = 1 and n = 0,
DAB k1
Hatta number reduces to MH = kL
for non-aqueous (non-
electrolyte) system, where physical mass transfer coefficient
−6 m s−1 .
kL value is obtained from Fig. 9 with a value of 4.4 × 10√
D k
In order to confirm the reaction regime 3, the value of kAB 1 >
L
3 as per the guidelines provided by Doraiswamy and Sharma
(1984). The condition was satisfied for regime 3 i.e. the extrac-
tion accompanied by a fast chemical reaction taking place
in the diffusion film. The morpholine solubility in toluene is
quite high, whereas the DA solubility in water is not much
significant. This confirms that the reaction occurred in the
organic diffusion film and RA is independent of staring speed
The effect of temperature is an important study in the view
of operating temperature in the regeneration step, which is
attributed to the consequence of different parameters like
the solute in phases, the extractant properties, and water co-
extraction (Eda et al., 2017). If it is assumed that the entropy
and enthalpy of the reaction between morpholine and CA are
constant, the KE is related to temperature using Vantoff’s equa-
tion (Thella et al., 2011).
H S
lnKE = − + (15)
RT R
A plot lnKE between and T−1 gave a straight line in
Fig. 14, the slope and intercept represent the enthalpy and
entropy of extraction of morpholine. The results indicate that
amine–acid complex formed at the interface of phases is
exothermic in nature since H found to be −58.21 kJ mol−1 .
This is in agreement with the fact that hydrogen bond for-
mation in complex is an exothermic reaction (Bokhove et al.,
2012).
The more exothermicity of the reaction, most sensitive of
the equilibrium to changes in the temperature. And, S was
 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254
found to be −187.8 J mol−1 K−1 . From a thermodynamic point
of view, the decrease in entropy with temperature pronounces
that the formation of the acid–amine complex makes the reac-
tive extraction system more ordered and thus reduces the
system’s entropy and randomness (Robin et al., 2016).
6. Conclusions
A comprehensive study of extraction equilibrium of morpho-
line in aqueous phase along with modeling and optimization
of the reactive extraction process was carried out. Further,
kinetic studies were explored in the present study. The extrac-
tions of morpholine (0.06–0.57) kmol m−3 from the aqueous
phase using capric acid, CA (0.29–0.87) kmol m−3 diluted in
toluene have been carried out. Physical extraction revealed
that the extraction mainly happened through the parti-
tion of undissociated morpholine to the organic phase and
decrease with increase in temperature due to weakening of
␲–H complex. While, the chemical extraction is dominated by
complex formation due to H-bonding of undissociated amine
molecules with CA attributed to the fact pH > pKa . A reac-
tive extraction model signifies that the physical extraction
contribution to an extent. Subsequently, the stoichiometry of
the complex described by the loading ratio (Z) indicates 1:1
amine–acid complex formation between morpholine and CA.
Additionally, the reactive extraction was studied for simul-
taneous optimization of multi-response using Box–Behnken
employing desirability approach. It showed good correla-
tion between experimental and predicted data using the
full quadratic model. The best optimum multivariable com-
bination was achieved at initial feed concentration 5%,
temperature 30 ◦ C and extractant concentration 5% with
38.15% (extraction efficiency) and 0.64 (loading ratio) with a
desirability value of 0.947. Kinetic studies revealed that reac-
tion between morpholine and capric acid in toluene falls in
regime 3 (fast chemical reaction occurring in the organic dif-
fusion film). Thermodynamic study has demonstrated that
amine–acid complex formed at the interface of phases is
exothermic and makes the extraction process more ordered.
From the above results, it can be concluded that extractant;
CA dissolved in toluene was efficient for the extraction of a
basic compound in aqueous waste stream based processes
and the equilibrium and kinetic data might be useful for the
development, design and reliable scale-up of the acid extrac-
tion process. Also, a novel optimization technique for design
variables in the reactive extraction process introduces the
new era of multi-response optimization based on desirability
approach.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at https://doi.org/10.1016/
j.cherd.2018.03.027.
References
Aroney, M.J., Chen, C.Y., Lefevre, R.J.W., Saxb, J.D., 1964. Molecular
polarisabdity. The molecular conformations of morpholine
and of N-methylmorpholine in non-polar media. J. Chem. Soc.
0, 4269–4274.
Behr, A., Henze, G., Johnen, L., Awungacha, C., 2008. Advances in
thermomorphic liquid/liquid recycling of homogeneous
transition metal catalysts. J. Mol. Catal. A: Chem. 285, 20–28.
253
Bokhove, J., Schuur, B., de Haan, A.B., 2012. Equilibrium study on
the reactive liquid–liquid extraction of 4-cyanopyridine with
4-nonylphenol. Chem. Eng. Sci. 82, 215–222.
Chai, W., Zhu, X., Liu, W., Zhang, W., Zhou, Z., Ren, Z., 2016.
Extraction of aniline from wastewater: equilibria, model, and
fitting of apparent extraction equilibrium constants. RSC Adv.
6, 6125–6132.
Eda, S., Kumari, A., Thella, P.K., 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, and
optimization using multivariate Taguchi approach. Can. J.
Chem. Eng. 95, 1373–1387.
Datta, D., Kumar, S., 2013. Equilibrium and kinetic studies of the
reactive extraction of nicotinic acid with tri-n-octylamine
dissolved in MIBK. Ind. Eng. Chem. Res. 52, 14680–14686.
Datta, D., Kumar, S., Uslu, H., 2014. Status of the reactive
extraction as a method of separation. J. Chem. 2014, 1–16.
Doraiswamy, L.K., Sharma, M.M., 1984. Heterogeneous Reaction:
Analysis, Examples, and Reactor Design, vol. 2:
Fluid-Fluid-Solid-Reactions., 1st ed. John Wiley & Sons, New
York, pp. 17–41.
Gadgil, O.D., Dalvi, V.H., Shenoy, K.T., Rao, H., Ghosh, S.K., Joshi,
J.B., 2014. Kinetics of extraction of uranium from phosphoric
acid by D2EHPA–TBP and D2EHPA–TOPO systems using
constant interfacial area stirred cell. Chem. Eng. Sci. 110,
169–184.
George, Jr., Laemmle, G.J., 1957. Recovery of morpholine from
aqueous solutions thereof. US Pat. 2776972 A.
Islam, M.A., Sakkas, V., Albanis, T.A., 2009. Application of
statistical design of experiment with desirability function for
the removal of organo phosphorus pesticide from aqueous
solution by low-cost material. J. Hazard. Mater. 170, 230–238.
Kahya, E., Bayraktar, E., Mehmeto, U., 2001. Optimization of
process parameters for reactive lactic acid extraction. Turk. J.
Chem. 25, 223–230.
Kumar, S., Datta, D., Babu, B.V., 2011. Estimation of equilibrium
parameters using differential evolution in reactive extraction
of propionic acid by tri-n-butyl phosphate. Chem. Eng.
Process. 50, 614–622.
Thella, P.K., Vishwanadham, B., Rani, K.N.P., Mallikarjun, M., Rao,
V.V.B., 2011. Reactive extraction of levulinic acid from aqueous
solutions with tri-n-octylamine (TOA) in 1-octanol: equilibria,
kinetics, and model development. Chem. Eng. Commun. 198,
572–589.
Kumari, A., Aniya, V., Rane, N.V., Thella, P.K., Satyavathi, B., 2017.
Isobaric phase equilibrium of morpholine + 1-decanol,
volumetric properties and molar refractivity from 293.15 to
333.15 K of morpholine + 1-decanol and 1-octanol + toluene
system with applications of Prigogine–Flory–Patterson theory.
Thermochim. Acta 649, 41–53.
Lee, T.G., Lim, S.Y., Song, K.H., Choe, J., 2014. Liquid–liquid
equilibria for the ternary systems of
4-methyl-1,3-dioxolan-2-one + 1,4-dimethylbenzene + octane,
decane, or dodecane and the ternary systems of
acetonitrile + morpholine + octane, decane, or dodecane at
313.15 K or 298.15 K. J. Chem. Eng. Data 59, 890–895.
Marti, M.E., Gurkan, T., Doraiswamy, L.K., 2011. Equilibrium and
kinetic studies on reactive extraction of pyruvic acidwith
trioctylamine in 1-octanol. Ind. Eng. Chem. Res. 50,
13518–13525.
McGuire, R.G., Dimitrios, D.A., 1999. Evaluation of shellac and
sucrose ester fruit coating formulations that support
biological control of post-harvest grapefruit decay. Biocontrol
Sci. Technol. 9, 53–65.
Ozer, A., Gurbuz, G., Ayla, C., Alimli, B., Korbahti, B.K., 2009.
Biosorption of copper(II) ions on enteromorphaprolifera:
application of response surface methodology (RSM). Chem.
Eng. J. 146, 377–387.
Rani, K.N.P., Kumar, T.P., Murthy, J.S.N., Sankarshana, T.,
Vishwanadham, B., 2010. Equilibria, kinetics, and modeling of
extraction of citric acid from aqueous solutions with alamine
336 in 1-octanol. Sep. Sci. Technol. 45, 654–662.
254 Chemical Engineering Research and Design 1 3 3 ( 2 0 1 8 ) 243–254
Reid, R.C., Prausnitz, J.M., Poling, B.E., 1987. The Properties of
Gases and Liquids, 4th ed. McGraw-Hill, New York, pp. 11–26.
Rewatkar, K., Shende, D.Z., Wasewar, K.L., 2016. Effect of
temperature on reactive extraction of gallic acid using
tri-n-butyl phosphate, tri-n-octylamine and Aliquat 336. J.
Chem. Eng. Data 61, 3217–3224.
Robin, S., Van den, R.B., Jessika, I., Zeiner, T., 2016. Purification of
terpenyl amine by reactive extraction. Ind. Eng. Chem. Res. 55,
5763–5769.
Somekh, G.S., Hawkes, E.N., 1969. Process for recovery of amines
from aqueous solutions by solvent treatment and distillation.
US Pat. 3433788 A.
Steensma, M., Kuipers, N.J.M., de Haan, A.B., Kwant, G., 2007.
Modelling and experimental evaluation of reaction kinetics in
reactive extraction for chiral separation of amines, amino
acids and amino-alcohols. Chem. Eng. Sci. 62, 1395–1407.
Stevens, W.H., Skov, K., 1965. A rapid spectrophotornetric method
for determining parts per million of morpholine in boiler
water. Analyst 90, 182–183.
Tang, K., Cai, J., Yang, C., Liu, Y., Zhang, P., Liu, Y., 2012. Kinetic
study on reactive extraction for chiral separation of
oxybutynin enantiomers. Sep. Purif. Technol. 92, 30–35.
Thakre, N., Prajapati, A.K., Mahapatra, S.P., Kumar, A., Khapre, A.,
Pal, D., 2016. Modeling and optimization of reactive extraction
of citric acid. J. Chem. Eng. Data 61, 2614–2623.
Uslu, H., Datta, D., Kumar, S., 2014. Reactive extraction of
oxoethanoic acid(glyoxylic acid) using amberlite-LA2 in
different diluents. Chem. Eng. Data 59, 2623–2629.
Varala, S., Kumari, A., Dharanija, B., Bhargava, S.K., Parthasarathy,
R., Satyavathi, B., 2016a. Removal of thorium(IV) from aqueous
solutions by deoiled karanja seed cake: optimization using
Taguchi method, equilibrium, kinetic and thermodynamic
studies. JECE 4, 405–417.
Varala, S., Dharanija, B., Satyavathi, B., Rao, V.V.B., Parthasarathy,
R., 2016b. New biosorbent based on deoiled karanja seed cake
in biosorption studies of Zr(IV): optimization using
Box–Behnken method in response surface methodology with
desirability approach. Chem. Eng. J. 302, 786–800.
Wasewar, K.L., Heesink, A.B.M., Versteeg, G.F., Pangarkar, V.G.,
2002. Equilibria and kinetics for reactive extraction of lactic
acid using Alamine 336 in decanol. J. Chem. Tech. Biotechnol.
77, 1068–1075.
Wasewar, K.L., Shende, D., Keshav, A., 2011. Reactive extraction of
itaconic acid using quaternary amine Aliquat 336 in ethyl
acetate, toluene, hexane, and kerosene. Ind. Eng. Chem. Res.
50, 1003–1011.
Zhou, Z., Li, Z., Qin, W., 2013. Reactive extraction of saturated
aliphatic dicarboxylic acids with trioctylamine in 1-octanol:
equilibria, model, and correlation of apparent reactive
equilibrium constants. Ind. Eng. Chem. Res. 52, 10795–10801.