Equilibrium of chiral extraction of 4-nitro-d,l-phenylalanine with BINAP metal complexes

Article abstract of DOI:10.2478/s11696-013-0419-4

The enantioselective extraction of 4-nitro-phenylalanine (Nphy) was studied with metal-2,2′-bis(diphenylphosphino)-1,1′-binaphthyl (BINAP) complexes as the chiral selector. The complex with palladium (BINAP-Pd) exhibits the highest selectivity out of the selectors studied, which is solubilised in the organic phase and preferentially extracts d-Nphy from the aqueous phase. Efficiency of extraction depends, often substantially, on a number of process variables, including types of organic solvents and metal precursors, concentration of ligand, pH, and temperature. A reactive extraction model was established to interpret the experimental data. The equilibrium formation constants and other important parameters required by the model were determined experimentally. The equilibrium formation constants were 6.73 and 1.93 for d-Nphy and l-Nphy. By way of modelling and experiment, an optimal extraction condition with pH of 7 and BINAP-Pd concentration of 1 mmol L^-1 was obtained with enantioselectivity (α) of 3.37, which was close to the maximum of 3.48, and a performance factor (pf) of 0.195. The model was verified experimentally with excellent results.

Full text of DOI:10.2478/s11696-013-0419-4

Chemical Papers
DOI: 10.2478/s11696-013-0419-4
ORIGINAL PAPER
Equilibrium of chiral extraction of 4-nitro-D,L-phenylalanine
with BINAP metal complexes
a
b
a
b
aJia-Jia Liu, Guo-Hui Wu, Ke-Wen Tang*, Xiong Liu, Pan-Liang Zhang
aCollege of Chemistry and Chemical Engineering, Central South University, Changsha 410083, Hunan, China
bDepartment of Chemistry and Chemical Engineering, Hunan Institute of Science and Technology,
Yueyang 414006, Hunan, China
Received 30 December 2012; Revised 13 March 2013; Accepted 7 April 2013
The enantioselective extraction of 4-nitro-phenylalanine (Nphy) was studied with metal-2,2ꢀ-
bis(diphenylphosphino)-1,1ꢀ-binaphthyl (BINAP) complexes as the chiral selector. The complex
with palladium (BINAP-Pd) exhibits the highest selectivity out of the selectors studied, which is
solubilised in the organic phase and preferentially extracts D-Nphy from the aqueous phase. Effi-
ciency of extraction depends, often substantially, on a number of process variables, including types
of organic solvents and metal precursors, concentration of ligand, pH, and temperature. A reactive
extraction model was established to interpret the experimental data. The equilibrium formation
constants and other important parameters required by the model were determined experimentally.
The equilibrium formation constants were 6.73 and 1.93 for D-Nphy and L-Nphy. By way of mod-
elling and experiment, an optimal extraction condition with pH of 7 and BINAP-Pd concentration
of 1 mmol L−1 was obtained with enantioselectivity (α) of 3.37, which was close to the maximum of
3.48, and a performance factor (pf) of 0.195. The model was verified experimentally with excellent
results.
c
ꢀ 2013 Institute of Chemistry, Slovak Academy of Sciences
Keywords: enantioselective extraction, 4-nitro-phenylalanine, equilibrium formation, BINAP,
modelling
Introduction
developed, such as enzymatic resolution or biocatal-
ysis (Okudomi et al., 2007), inclusion complexation
(Matsumoto et al., 2004; Miyamoto et al., 2000; Tsun-
oda et al., 1997), chromatography (Davankov, 1994),
simulated moving bed chromatography (SMB) (Qual-
lich, 2005), ionic liquids (Gutierrez et al., 2011), cap-
illary electrophoresis (Jira et al., 1997; Kodama et al.,
2001), and membrane-based approaches (Keurentjes
et al., 1996; Maximini et al., 2006). However, these
chiral separation methods entail drawbacks such as
low versatility or high cost. Consequently, a more ef-
fective chiral separation method is urgently needed.
Chiral resolution by chiral liquid–liquid extraction has
attracted a great deal of interest (Koska & Haynes,
2001; Schuur et al., 2011). A partial separation can
be obtained in one-stage extraction, whereas staging
The availability of enantiopure compounds is ex-
tremely important for the pharmaceutical and also to
some extent for the agrochemical, flavour, and fra-
grances industries (Tombo & Belluš, 1991; Garcia-
Chavez et al., 2012). Two racemic enantiomers pos-
sess only a slight structural difference; however, this
often exhibits a different biological and pharmaco-
logical activity when ingested by living organisms.
In most cases, one has the desired effect, whereas
the other enantiomer can have negative side-effects
or even toxic effects. For this reason, chiral sepa-
ration methods are of primary importance (Jiao et
al., 2007; Zhou et al., 2013). Many methods follow-
ing a chiral separation principle have recently been
*Corresponding author, e-mail: tangkewen@sina.com

ii
J. J. Liu et al./Chemical Papers
in a fractional extraction scheme can be used to ob-
tain both enantiomers at the desired level of purity
(Steensma et al., 2007a, 2007b; Schuur et al., 2010).
In reactive extraction, enantioselectivity (α) is
dominated by the application of an enantioselective
extractant (Steensma et al., 2006; Tang & Zang,
2011a). With high selectivity of a chiral selector,
the stages required for a certain enantiomeric pu-
rity can be considerably reduced. Examples of chi-
ral selectors are cyclodextrins, macrocyclic antibi-
otics, carbohydrates, chiral crown ethers, and pro-
teins. However, the low enantioselectivities of these
selectors restrict their wider application. Metal com-
plexes have shown good performance in enantioselec-
tive extraction. H¨odl et al. (2008) reported the use
of copper(II), cobalt(II), nickel(II), and zinc(II) com-
plexes of sugar acids as chiral selectors for the enan-
tioseparation of amino acids and glycyl dipeptides.
2,2ꢀ-bis(diphenylphosphino)-1,1ꢀ-binaphthyl (BINAP)
is well known as a highly versatile ligand in asymmet-
ric catalysis. More recently, Verkuijl et al. (2009) intro-
duced the palladium complex of (s)-BINAP to obtain
a high performance in chiral liquid–liquid extraction.
Although ample literature is available for enantioselec-
tive extraction, only a few studies provide fundamen-
tal insights into the reaction engineering mechanisms
(Schlichting et al., 1987; Bora et al., 1997; Koska et
al., 2001; Schuur et al., 2008; Tang et al., 2011b, 2012).
Amino acids are extensively used in the synthesis of
several products used in the chemical, pharmaceutical,
food, and health industries. 4-Nitro-L-phenylalanine
(L-Nphy) is an important intermediate in many syn-
theses (Sassatelli et al., 2006; Mishra et al., 2003). This
paper reports on the enantioselective reactive extrac-
tion of Nphy enantiomers using the complex of BINAP
with palladium (BINAP-Pd) as a selector. The effects
of process variables such as types of organic solvents
and metal precursors, concentration of selector, pH,
and temperature on extraction efficiency were inves-
tigated. The mechanism of the extraction is discussed
and a reactive extraction model with an interfacial
reaction of D- and L-Nphy with BINAP-Pd was estab-
lished to optimise the extraction process.
Fig. 1. Diagram of mechanism of reaction extraction; M –
metal.
aqueous phase, while Nphy enantiomers are highly hy-
drophilic. Hence, the interfacial reaction mechanism
was applied here (Fig. 1), in which the main reactions
are restricted to the interface between the phases. The
mass transfer and chemical reaction are described as
serial processes: the reactants are transported to the
interface where they are adsorbed, the reaction takes
place at the interface, the products are adsorbed and
transported into the bulk (Steensma et al., 2007b).
In this mechanism (Fig. 1), the equilibrium for-
mation constants of D- and L-Nphy with the complex
BINAP-Pd at the interface can be defined as:
[PdBD]org[Cl−]w
[PdB]org[D−]w
KD
=
(1)
(2)
[PdBL]org[Cl−]w
[PdB]org[L−]w
KL =
where [PdB]org is the concentration of BINAP-Pd
in the organic phase in equilibrium, [PdBD]org and
[PdBL]org are the concentrations of the complexes of
PdCl2-(s)-BINAP with D- and L-Nphy in the organic
phase, [D−]w and [L−]w are the concentrations of ionic
D- and L-Nphy in the aqueous phase. [Cl−]w is the con-
centration of chloride ion in the aqueous phase.
Theoretical
BINAP-Pd and its complex with Nphy enan-
tiomers cannot be dissolved in the aqueous phase,
while the Nphy enantiomers are highly hydrophilic.
Therefore, Eqs. (1) and (2) are appropriate for describ-
ing the equilibrium constants for these processes. By
linear regression of the experimental data according to
Eqs. (1) and (2), the equilibrium formation constants
are evaluated from the slope of the fitting lines.
Knowledge of the mechanism can improve our un-
derstanding of the behaviour of an extraction system.
In most cases, either a homogeneous reaction model or
an interfacial reaction model is applied to understand
a reactive extraction system. The main difference be-
tween the two models is the locus of the main reac-
tions for reactive extraction. In the reactive extraction
of D- and L-Nphy with BINAP-Pd as selector, the re-
actions may take place in either the organic phase, the
aqueous phase or at the phase interface. Solution ex-
periments demonstrated that BINAP-Pd and its com-
plex with Nphy enantiomers cannot be dissolved in the
Mathematical modelling
On the basis of understanding the extraction mech-
anism, the equilibrium of the extraction system can
be modelled by combining the following equilibrium

J. J. Liu et al./Chemical Papers
iii
equations and mass balance equations. The phase
compositions of the two-phase system in equilibrium
involving the formation of multiple chemical com-
plexes can be calculated by simultaneous solution of
the equations.
With a further treatment of Eq. (10), the following
equation concerning [PdB]org can be deduced:
(KDKL − KDA − KLA + A2)[PdB]o3rg
+ (2KDACPtodtB + 2KLACPtodtB − 3A2C
+
tot
PdB
+
The dissociation constant of D- and L-Nphy in the
aqueous phase can be described as follows:
+ KDKLCDtot − KDCDtotA + KDKLCLtot
−
− KLCtotA − KDKLCPtodtB)[PdB]2
+
L
org
[D−]w[H+]w
[DH]w
[L−]w[H+]w
[LH]w
+ (KDCDtotAC
+ KLCLtotAC
−
tot
tot
ka =
=
(3)
PdB
PdB
2KDA − Ctot 2KLA +
tot
− C
PdB
PdB
Since volume of water phase and organic phase is
equal, the following equations represent the mass bal-
ance for D- and L-Nphy:
3
+ 3A2C
2)[PdB]org − A2Ctot = 0
(11)
tot
PdB
PdB
Therefore, [PdB]org can be obtained by solving Eq.
(11).
CDtot = [D]w + [D−]w + [PdBD]org
CLtot = [L]w + [L−]w + [PdBL]org
(4)
The distribution coefficient of each species is a
complex function of the concentrations of the free
species present in the two phases, serving to indi-
cate that the overall phase equilibria are linked to the
chemical equilibria in each phase (Koska et al., 2001).
In accordance with Eqs. (6) and (7), the distribution
coefficients (kD and kL) can be calculated from the
following equations:
(5)
where, CDtot and CLtot are the total concentrations of D-
and L-Nphy, respectively, [D]w and [L]w are the con-
centrations of molecular D- and L-Nphy in the aqueous
phase, respectively.
Combining Eqs. (1)–(5), it may be deduced that:
KD[PdB]org
ꢀ
ꢁ
ꢁ
kD
=
(12)
[H+]
[PdB]org[D−]wKD
[D−] [H+]w
[Cl−]w 1 +
CDtot
=
+[D−]w+
+ [L−]w +
(6)
w
ka
[Cl−]w
ka
KL[PdB]org
[PdB]org[L−]wKL
ꢀ
kL =
(13)
[L−] [H+]w
[H+]
ka
CLtot
=
(7)
w
[Cl−]w 1 +
[Cl−]w
ka
Enantioselectivity (operational selectivity, aop) is de-
fined by the ratio of the distribution coefficient aop
= kL/kD, assuming kL > kD. Its upper limit is the
intrinsic selectivity (aint), which is the ratio of the
complexation constants, indicating that the separa-
tion power is determined solely by the nature of the
selector-enantiomer interaction aint = KL/KD.
A common alternative measure of the system per-
formance is provided by the enantiomeric excess (ee).
For the system described here, the reaction takes place
in the organic phase. Therefore, the ee in the organic
phase is defined as:
The following equation concerns the mass balance of
BINAP-Pd:
tot
PdB
C
= [PdB]org+[PdBD]org+[PdBL]org
(8)
Combining Eqs. (1) and (2), Eq. (9) is derived as:
[PdB]org[D−]wKD
tot
PdB
C
= [PdB]
+
+
org
[Cl−]w
[PdB]org[L−]wKL
+
(10)
[Cl−]w
CLtot
CDtot
ꢀ
ꢁ
−
[H+]
ka
1
kD
1
kL
tot
PdB
1 +
1 +
Let A = 1 +
, then C
can be defined as:
ee =
(14)
CLtot
CDtot
+
1
kD
1
kL
CDtot[Cl−]w
[PdB]orgKD + A[Cl−]w
1 +
1 +
[PdB]orgKD
tot
PdB
C
= [PdB]
+
org
[Cl−]w
The fraction of the solute i extracted (f ) into the aque-
CL[Cl−]w
ous phase is given by:
[PdB]orgKL
[PdB]orgKL + A[Cl−]w
Ci,org
+
(10)
fi =
(15)
[Cl−]w
Ci

iv
J. J. Liu et al./Chemical Papers
where, Ci,org represents the total concentration of the
solute i in the organic phase in equilibrium, and Ci
represents the initial total concentration of the solute
i.
The extraction performance factor (pf) is useful for
optimising enantioselective extraction and is defined
as:
mmol L−1, and the concentration of BINAP-Pd was
varied from 0.25 mmol L−1 to 4 mmol L−1. After ex-
traction, phase compositions were analysed by HPLC.
The amount of negatively charged racemic Nphy was
calculated using the Henderson–Hasselbalch equation
and the other concentrations through mass balances.
The apparent equilibrium formation constants were
obtained by regression analysis of the experimental
data according to Eqs. (1) and (2).
pfi = fiee
(16)
A performance factor close to unity indicates a high
enantiomeric purity in both phases.
The quantification of Nphy enantiomers in the
aqueous phase was performed by HPLC using an Ag-
ilent LC 1200 series apparatus (Agilent Technologies,
USA). A UV detector operated at 230 nm was applied.
The column was Inertsill/Wondasil C18, 5 µm parti-
cle size of the packing material, 250 mm × 4.6 mm
i.d. (GL Sciences, Japan). The mobile phase was a
mixture of methanol and an aqueous solution (ϕr =
1 : 4) containing 0.75 mmol L−1 of cupric sulphate
and 1.5 mmol L−1 of L-proline with pH = 5.0 (pH was
adjusted with ammonium acetate). The flow-rate was
Experimental
Materials and methods
Bi(acetonitrile)dichloropalladium
(PdCl2(CH3-
CN)2, mass fraction > 99 %) was purchased from
the Metallurgy Institute of Zhejiang (China). Bis(tri-
phenylphosphine)nickel(II) ([(C6H5)3P]2NiCl2, mass
fraction > 99 %), tetrakis(acetonitrile)copper(I) hexa-
fluorophosphate ([(CH3CN)4Cu]PF6, mass fraction
> 99 %) were purchased from Hewei Chemical
(China). BINAP (mass fraction > 99 %) was pur-
chased from Shengjia Chemical (China). Nphy (mass
fraction > 99 %) were purchased from Saen Chemi-
cal (China). The solvent for chromatography was of
HPLC grade. All other chemicals were of analytical-
reagent grade and from different suppliers.
The aqueous phase was prepared by dissolving
racemic Nphy in 0.1 mol L−1 KH2PO4/H3PO4 buffer
solution and the organic phase was prepared by
dissolving the (s)-BINAP-metal complex which was
generated in situ by adding the metal precursor
(PdCl2(CH3CN)2, [(C6H5)3P]2NiCl2, [(CH3CN)4Cu]
PF6), and (s)-BINAP in equal mole amounts to the
appropriate organic solvent. The reaction mixture was
stirred overnight and diluted to the desired concentra-
tion and the resulting solution was applied directly in
the extraction experiments. Equal volumes (each of
2 mL) of the aqueous and the organic phases were
transferred into a 10 mL centrifuge tube and shaken
sufficiently (12 h) before being kept in a water bath
at a constant temperature to achieve equilibrium.
After phase separation, the concentrations of Nphy
enantiomers in the aqueous phase were analysed by
HPLC. The total amounts of the enantiomers in the
organic and aqueous phases after extraction were con-
sistent with the initial amounts included in the aque-
ous phase. The concentration of D- or L-Nphy in the
organic phase was calculated from a mass balance.
The experiment of the equilibrium formation con-
stants was carried out at pH = 7 in the aqueous phase
with dichloromethane as the solvent. Equal volumes
(each of 2 mL) of the aqueous and organic phases
were mixed together and shaken sufficiently (12 h)
before being kept in a water bath at a constant tem-
perature to achieve equilibrium. Racemic Nphy was 2
set at 1 mL min−1
.
Results and discussion
Complexation constants KD and KL
A series of extraction experiments were performed
with different concentrations of extractant in the or-
ganic phase for determination of the equilibrium for-
mation constants KD and KL. Figs. 2a and 2b show the
regression analysis of the experimental data according
to Eqs. (1) and (2). The linearity of the plot confirms
a 1 : 1 reaction between BINAP-Pd and racemic Nphy.
KD and KL calculated from the slopes of the straight
lines are 6.73 and 1.93, respectively.
Screening of extraction process variables
The effect of the structural variation of the metal
precursor was investigated. Different extractants were
prepared by the reaction of different metal precursors
with (s)-BINAP. Table 1 shows that the structure of
the metal precursor has an obvious influence on the ex-
traction efficiency. The highest enantioselectivity was
achieved with PdCl2(CH3CN)2 as the metal precur-
sor. The lowest distribution ratios and operational se-
lectivity were obtained with [(C6H5)3P]2NiCl2 as the
metal precursor. Although a high distribution coeffi-
cient can be achieved with [(CH3CN4)Cu]PF6 as the
metal precursor, the enantioselectivity obtained is rel-
atively low. Consequently, PdCl2(CH3CN)2 was used
as the most suitable metal precursor for the reactive
extraction of Nphy enantiomers.
The distribution coefficient and enantioselectivity
for Nphy enantiomers with four different organic sol-
vents were investigated to find out the influence of
the organic solvent type. The nature of the organic
solvent can influence the distribution behaviour of the

J. J. Liu et al./Chemical Papers
v
Table 2. Influence of organic solvent on extraction efficiencya
Organic solvent
kD
kL
α
1,2-Dichloroethane
Trichloromethane
Dichloromethane
Chlorobenzene
0.981
1.727
1.369
0.265
0.197
0.561
0.406
0.108
3.30
3.07
3.37
2.45
a) Conditions of extraction: aqueous phase – CNphy = 2.0
mmol L−1; organic phase – CBINAP−Pd = 1 mmol L−1; pH
= 7.0; temperature = 25◦C.
was used, a high distribution coefficient was obtained
but with relatively low enantioselectivity. By contrast,
low distribution coefficients and low enantioselectivity
were obtained with chlorobenzene as solvent. The rel-
atively higher enantioselectivity and distribution coef-
ficients were achieved with alkyl chloride solvents, out
of which, dichloromethane had the highest operational
selectivity when compared with 1,2-dichloroethane.
Consequently, dichloroethane was an ideal organic sol-
vent for the reactive extraction of Nphy enantiomers.
The pH of the aqueous phase is a significant factor
in a reactive extraction system, where pH influences
the distribution coefficient of the substrate, but not
the enantioselectivity; this was predicted by the model
(Eqs. (12) and (13)). Experiments at different pH were
carried out to demonstrate the model predictions. A
comparison of the experimental data with the model
predictions is given in Figs. 3a and 3b, from which it
may be concluded that the model predictions of dis-
tribution coefficients are in good agreement with the
experiment, as well as the enantioselectivities.
The relationship between pH and the distribution
coefficient and enantioselectivity is caused by the shift
in the association–dissociation equilibrium of Nphy.
Fig. 3a shows that the distribution coefficients (kD
and kL) increase with the increase in pH. Meanwhile,
no significant physical partition was observed in the
pH range measured. According to the species fraction
plot (Fig. 4), the amount of the anionic Nphy increases
with increasing pH while molecular Nphy decreases at
the same time. When pH is lower than 3, most of the
enantiomers are present in the cationic form; just a
small fraction of the anionic species is available in the
aqueous phase, while it is the Nphy anion that binds to
the BINAP-Pd complex, through which Nphy is dis-
tributed into the organic phase. Therefore, the distri-
bution coefficients are relatively low. With an increase
in pH, the amount of anions to be transported into the
organic phase is increased, which leads to an increase
in distribution coefficients. However, the enantioselec-
tivity remains constant (Fig. 3b), which is determined
only by the complexation equilibrium constants of the
two complexes of BINAP-Pd with Nphy enantiomers.
The influence of BINAP-Pd concentration on the
distribution behaviour of Nphy enantiomers was inves-
tigated by varying the concentration from 0 mmol L−1
Fig. 2. Linear regression of experimental data for determining
complexation constant of KD (a) and KL (b): Linear
regression (solid line) and experimental data ( ).
Table 1. Influence of metal precursor type on extraction
efficiencya
Metal precursor
kD
kL
α
PdCl2(CH3CN)2
[(CH3CN)4Cu]PF6
[(C6H5)3P]2NiCl2
1.369
1.234
0.020
0.406
1.156
0.017
3.37
1.03
1.19
a) Conditions of extraction: aqueous phase
–
CNphy
=
=
2
mmol L−1
; organic phase (dichloromethane) – CPd
1 mmol L−1, CCu = 1 mmol L−1, CNi = 1 mmol L−1, CBINAP
= 1 mmol L−1; pH = 7.0; temperature = 25◦C.
enantiomers due to the fact that organic solvent can
influence the non-covalent interactions and affect the
formation of the ternary complex during the enantios-
elective host-guest complexations (Atkins, 1994). The
solvent molecules can influence stabilities and ligand
enantioselectivities, hence separation performance, in
many ways. Table 2 shows that when trichloromethane

vi
J. J. Liu et al./Chemical Papers
Fig. 5. Influence of selector concentration on kD (ꢀ), kL
( )
Fig. 3. Influence of pH on kD (ꢀ), kL
( ) (a) and α (b); model
predictions (solid line) and experimental data (sym-
bols). Conditions: organic solvent (dichloromethane);
•
•
(a) and
α (b); model predictions (solid line) and
experimental data (symbols). Conditions: CNphy
2.0 mmol L−1; pH = 7.0; temperature = 25◦C.
=
CBINAP-Pd = 1.0 mmol L−1; CNphy = 2.0 mmol L−1
;
temperature = 25◦C.
predictions is given in Figs. 5a and 5b. It may be con-
cluded from Fig. 5 that the model predictions are in
good agreement with the experiment.
Fig. 5a shows that kD and kL increase rapidly with
the increase in the concentration of BINAP-Pd. Nphy
enantiomers can form complexes with BINAP-Pd in
the organic phase; with the increase in BINAP-Pd
concentration, more complexes are formed in the or-
ganic phase and the distribution coefficient increases
as a consequence. At the same time, the enantioselec-
tivity is independent of the host concentration, which
is in good accord with the model prediction (Fig. 5b).
The influence of temperature on the distribution
coefficient and enantioselectivity was investigated in
the range from 5◦C to 30◦C. Fig. 6a shows that with
the increase in temperature the distribution coefficient
increases slightly and the enantioselectivity decreases.
The good enantioselectivity is achieved at a lower tem-
perature (Fig. 6b). However, it will be difficult to per-
form extraction at below 5◦C because of an increase in
the viscosity of the solution and the decrease in host
Fig. 4. Species fraction δ as a function of pH; cationic Nphy
(dashed line), neutral Nphy (dashed–dotted line), and
anionic Nphy (solid line).
to 2 mmol L−1, at pH = 7 and temperature = 25◦C. A
comparison of the experimental data with the model

J. J. Liu et al./Chemical Papers
vii
Fig. 6. Influence of temperature T on kD ( ) and kL
(
) (a) and α (b). Conditions: CBINAP-Pd = 1.0 mmol L−1; CNphy
=
•
2.0 mmol L−1; pH = 7.0.
and substrate solubility. Hence, 5◦C is considered to
be the optimal temperature for enantioselective ex-
traction.
Model prediction in the extraction system
A comparison of the experimental data with the
experimental results indicates that the model estab-
lished provides a powerful tool for predicting the
enantiomer-partition over a range of experimental
conditions. Hence, this model was used to explore the
influence of various operating conditions on extraction
efficiency in a single stage.
Figs. 7a and 7b show the distribution coefficients
of D- and L-Nphy as a function of pH and BINAP-Pd
concentration, respectively. It can be observed from
Figs. 7a and 7b that kD and kL exhibit a similar ten-
dency with the change in pH and BINAP-Pd concen-
tration. The increase in pH and BINAP-Pd concen-
tration can give rise to the distribution coefficients of
the two enantiomers.
Fig. 8 shows the ee in the organic phase for Nphy
enantiomers as a function of pH and BINAP-Pd con-
centration. The ee is relatively high under the condi-
tions where pH and BINAP-Pd concentration are low.
Under these conditions, the amount of Nphy enan-
tiomers partitioning into the organic phase is minimal,
which is not conducive to increasing the yield.
However, in addition to a satisfactory ee value, a
good yield is also required. Fig. 9 shows the fraction
of D-Nphy extracted into the organic phase. Increas-
ing the BINAP-Pd concentration enhances the extrac-
tion of D-Nphy, while the extraction fraction achieves
a plateau where the pH value is above 7 and BINAP-
Fig. 7. Calculated distribution coefficients kD (a) and kL (b) as
a function of pH and BINAP-Pd concentration. Condi-
tions: CNphy = 2.0 mmol L−1; pH = 7.0; temperature
= 25◦C.
Pd concentration is more than 1.5 mmol L−1
.
It may be concluded from Figs. 8 and 9, that
achieving a good ee value and a good yield are some-
what contradictory. To facilitate optimisation of the
reactive extraction system, performance factor, is in-
troduced, which is defined as the product of the ee in
the organic phase and the fraction of enantiomer ex-
tracted into the organic phase. A good performance
indicates that the given enantiomer can be purified to

viii
J. J. Liu et al./Chemical Papers
a high level with maximum yield. Performance fac-
tors calculated from the ee and f data in Figs. 8
and 9 are given in Fig. 10. This shows that the per-
formance factor is strongly influenced by BINAP-Pd
concentration and pH. Performance factor achieves
its maximum where the pH value is approximately
7 and the BINAP-Pd concentration is no lower than
1 mmol L−1
.
Conclusions
The enantioselective reactive extraction of Nphy
enantiomers with BINAP-Pd as chiral selector was in-
vestigated. The experimental results show that the
efficiency is strongly influenced by process variables
such as types of organic solvents and metal precur-
sors, concentration of selector, pH, and temperature.
The reactive extraction process proceeds according to
an interfacial reaction mechanism. The model, taking
into account acid-base dissociation of solute, interfa-
cial reaction equilibrium and mass balance, quantita-
tively predicts solute partitioning and enantioselectiv-
ities under various system conditions and solute con-
centrations with excellent accuracy, hence it may be
used for the optimisation of enantiomer extractions.
The best conditions involving a pH value of ap-
proximately 7 and BINAP-Pd concentration of above
1 mmol L−1 are obtained by the model relying on a
high performance factor. The model predictions were
validated by experiments. The data presented indicate
that the model provided a powerful tool for modelling
the two-phase enantioselective reactive extraction sys-
tem. Full separation of the Nphy enantiomers cannot
be achieved in a single equilibrium step but can be
achieved by multi-stage extraction.
Fig. 8. Calculated ee as a function of pH and BINAP-Pd con-
centration. Conditions: CNphy = 2.0 mmol L−1; pH =
7.0; temperature = 25◦C.
Fig. 9. Calculated fraction of Nphy extracted into organic
phase (forg) as a function of pH and BINAP-Pd con-
centration. Conditions: CNphy = 2.0 mmol L−1; pH =
7.0; temperature = 25◦C.
Acknowledgements. This work was financially supported
by the National Natural Science Foundation of China (No.
21171054), Programme for New Century Excellent Talents in
University, Hunan Provincial Natural Science Foundation of
China (No. 10JJ1004), and the Aid Programme for Science
for Technology Innovative Research Team in Higher Educa-
tional Institutions of Hunan Province.
Symbols
aop
aint
operational selectivity
intrinsic selectivity
BINAP 2,2ꢀ-bis (diphenylphosphino)-1,1ꢀ-binapht-
halene
Ci
initial total concentration of
solute i
mmol L−1
Ci,org the total concentration of the solute i in
the organic phase in equilibrium mmol L−1
CBINAP concentration of 2,2ꢀ-bis (diphenyl-
Fig. 10. Calculated pf as a function of pH and BINAP-Pd con-
centration. Conditions: CNphy = 2.0 mmol L−1; pH
= 7.0; temperature = 25◦C.
phosphino)-1,1ꢀ-binaphthalene
CBINAP-Pd concentration of PdCl2-(s)-
BINAP
mmol L−1
mmol L−1

J. J. Liu et al./Chemical Papers
ix
CCu
CD
concentration of tetrakis(acetonitrile)
copper(I) hexa-fluorophosp
ϕ
fraction of solute i extracted into the or-
ganic phase
mmol L−1
total concentration of D-Nphy in the aque-
mmol L−1
mmol L−1
Subscripts
ous phase
total concentration of D-Nphy
CDtot
[Cl−]w concentration of chloride ion in the aque-
D
int
L
op
org
w
D-enantiomer
intrinsic value
L-enantiomer
operational value
organic phase
aqueous phase
ous phase
mmol L−1
CL
total concentration of L-Nphy in the aque-
ous phase
total concentration of L-Nphy
mmol L−1
mmol L−1
CLtot
CNi
concentration of bis(triphenylphosphine)
nickel(II)
mmol L−1
References
CNphy concentration of 4-nitro-phenyl-
alanine
mmol L−1
mmol L−1
mmol L−1
mmol L−1
Atkins, P. W. (1994). Physical chemistry (5th ed.). New York,
NY, USA: Oxford University Press.
CPd
concentration of Bi(acetonitrile)
dichloropalladium
Bora, M. M., Ghosh, A. C., Dutta, N. N.,
& Mathur,
CPdB total concentration of PdCl2-(s)-
R. K. (1997). Reactive extraction of 6-aminopenicillanic
acid with aliquat-336: Equilibrium and kinetics. Cana-
dian Journal of Chemical Engineering, 75, 520–526. DOI:
10.1002/cjce.5450750305.
Davankov, V. A. (1994). Chiral selectors with chelating proper-
ties in liquid chromatography: Fundamental reflections and
selective review of recent developments. Journal of Chro-
matography A, 666, 55–76. DOI: 10.1016/0021-9673(94)80
370-6.
BINAP in the organic phase
total concentration of PdCl2-(s)-
tot
PdB
C
BINAP
[D−]w concentration of molecular D-Nphy in the
aqueous phase
concentration of molecular D-Nphy in the
aqueous phase
mmol L−1
[D]w
mmol L−1
ee
f
enantiomeric excess
solute fraction extracted into the organic
phase
distribution coefficient, organic/aquatic
concentration
distribution coefficient in aqueous phase
distribution coefficient (D-enantiomer), or-
ganic/aquatic concentration
distribution coefficient (L-enantiomer), or-
ganic/aquatic concentration
equilibrium formation constant
(D-enantiomer)
Garcia-Chavez, L. Y., Shazad, M., Schuur, B., & de Haan, A.
B. (2012). (Liquid + liquid) equilibrium data for the system
(propyleneglycol + water + tetraoctyl ammonium 2-methyl-
1-naphthoate). Journal of Chemical Thermodynamics, 55,
238–244. DOI: 10.1016/j.jct.2012.04.002.
Gutierrez, J. P., Meindersma, W., & de Haan, A. B. (2011).
Binary and ternary (liquid + liquid) equilibrium for{methyl-
cycloexane (1) + toluene (2) + 1-hexyl-3-methylimidazolium
tetracyanoborate (3)/1-butyl-3-methylimidazolium tetra-
cyanoborate (3)}. Journal of Chemical Thermodynamics, 43,
1672–1677. DOI: 10.1016/j.jct.2011.05.029.
Ho¨dl, H., Schmid, M. G., & Gu¨bitz, G. (2008). Chiral sep-
aration of amino acids and glycyl dipeptides by chiral
ligand-exchange capillary electrophoresis comparing Cu(II),
Co(II), Ni(II) and Zn(II) complexes of three different sugar
acids. Journal of Chromatography A, 1204, 210–218. DOI:
10.1016/j.chroma.2008.05.071.
Jiao, F. P., Chen, X. Q., Hu, W. G., Ning, F. R., & Huang,
K. L. (2007). Enantioselective extraction of mandelic acid
enantiomers by L-dipentyl tartrate and β-cyclodextrin as bi-
nary chiral selectors. Chemical Papers, 61, 326–328. DOI:
10.2478/s11696-007-0041-4.
k
ka
kD
kL
KD
KL
equilibrium formation constant
(L-enantiomer)
[L−]w concentration of molecular L-Nphy in the
aqueous phase
concentration of molecular L-Nphy in the
aqueous phase
mmol L−1
[L]w
mmol L−1
Nphy 4-nitro-L-phenylalanine
[PdB]org concentration of PdCl2-(s)-BINAP in
the organic phase at equilibrium mmol L−1
[PdBD]orgconcentration of the complexe of PdCl2-
(s)-BINAP with D-Nphy in the organic
Jira, T., Bunke, A., Schmid, M. G., & Gu¨bitz, G. (1997). Chiral
resolution of diols by capillary electrophores using borate-
cyclodextrin complexation. Journal of Chromatography A,
761, 269–275. DOI: 10.1016/s0021-9673(96)00836-9.
Keurentjes, J. T. F., Nabuurs, L. J. W. M., & Vegter, E. A.
(1996). Liquid membrane technology for the sepapation of
racemic mixtures. Journal of Membrane Science, 113, 351–
360. DOI: 10.1016/0376-7388(95)00176-x.
Kodama, S., Yamamoto, A., Matsunaga, A., & Hayakawa,
K. (2001). Direct chiral resolution of tartaric acid in food
products by ligand exchange capillary electrophoresis us-
ing copper(II)-D-quinic acid as a chiral selector. Journal
of Chromatography A, 932, 139–143. DOI: 10.1016/s0021-
9673(01)01228-6.
phase
mmol L−1
[PdBL]org concentration of the complexe of PdCl2-
(s)-BINAP with L-Nphy in the organic
phase
mmol L−1
pf
performance factor
SMB simulated moving bed chromatography
Greek Letters
Koska, J., & Haynes, C. A. (2001). Modelling multiple chemical
in chiral partition systems. Chemical Engineering Science,
56, 5853–5864. DOI: 10.1016/s0009-2509(00)00419-x.
α
δ
enantioselectivity
species fraction

x
J. J. Liu et al./Chemical Papers
Koska, J., Mui, C., & Haynes, C. A. (2001). Solvent effects in
chiral ligand exchange systems. Chemical Engineering Sci-
ence, 56, 29–41. DOI: 10.1016/s0009-2509(00)00418-8.
Matsumoto, K., Okamoto, T., & Otsuka, K. (2004). Optical res-
olution of acyclic α-hydroxy ketone derivatives by inclusion
complexation. Bulletin of the Chemical Society of Japan, 77,
2051–2056. DOI: 10.1246/bcsj.77.2051.
Steensma, M., Kuipers, N. J. M., de Haan, A. B., & Kwant, G.
(2006). Influence of process parameters on extraction equi-
libria for the chiral separation of amines and amino-alcohols
with a chiral crown ether. Journal of Chemical Technology
and Biotechnology, 81, 588–597. DOI: 10.1002/jctb.1434.
Steensma, M., Kuipers, N. J. M., de Haan, A. B., & Kwant,
G. (2007a). Analysis and optimization of enantioselec-
Maximini, A., Chmiel, H., Holdik, H., & Maier, N. W. (2006).
Development of a supported liquid membrane process of
separating enantiomers of N-protected amino acid deriva-
tives. Journal of Membrane Science, 276, 221–231. DOI:
10.1016/j.memsci.2005.09.050.
Mishra, A. K., Panwar, P., Chopra, M., Sharma, R. K., Chatal,
J. F. (2003). Synthesis of novel bifunctional Schiff-base lig-
ands derived from condensation of 1-(p-nitrobenzyl)ethylene-
diamine and 2-(p-nitrobenzyl)-3-monooxo-1,4,7-triazahepta-
ne with salicylaldehyde. New Journal of Chemistry, 27, 1054–
1058. DOI: 10.1039/b300621m.
Miyamoto, H., Sakamoto, M., Yoshioka, K., Takaoka, R.,
& Toda, F. (2000). Resolution of hydrocarbons by inclu-
sion complexation with a chiral host compound. Tetra-
hedron: Asymmetry, 11, 3045–3048. DOI: 10.1016/s0957-
4166(00)00269-x.
Okudomi, M., Shimojo, M., Nogawa, M., Hamanaka, A.,
Taketa, N., & Matsumoto, K. (2007). Easy separation of
optically active products by enzymatic hydrolysis of solu-
ble polymer-supported substrates. Tetrahedron Letters, 48,
8540–8543. DOI: 10.1016/j.tetlet.2007.09.141.
tive extraction in a multi-product environment with a
multistage equilibrium model. Chemical Engineering and
Processing: Process Intensification, 46, 996–1005. DOI:
10.1016/j.cep.2007.05.008.
Steensma, M., Kuipers, N. J. M., de Haan, A. B., & Kwant, G.
(2007b). Modelling and experimental evalution of reaction ki-
netics in reactive extraction for chiral separation of amines,
amino acids and amino-alcohols. Chemical Engineering Sci-
ence, 62, 1395–1407. DOI: 10.1016/j.ces.2006.11.043.
Tang, K. W., & Zhang, P. L. (2011). Enantioselective extraction
of terbutaline enantiomers with β-cyclodextrin derivaties as
hydrophilic selectors. Chemical Papers, 65, 273–279. DOI:
10.2478/s11696-011-0011-8.
Tang, K. W., Zhang, P. L., Pan, C. Y., & Li, H. J. (2011).
Equilibrium studies on enantioselective extraction of oxy-
buynin enantiomers by hydrophilic β-cyclodextrin deriva-
tives. AIChE Journal, 57, 3027–3036. DOI: 10.1002/aic.
12513.
Tang, K. W., Wu, G. H., Zhang, P. L., Zhou, C. S., & Liu,
J. J. (2012). Experimental and model study on enantiosele-
tive extraction of phenylglycine enantiomers with BINAP-
metal complexes. Industrial & Engineering Chemistry Re-
search, 51, 15233–15241. DOI: 10.1021/ie301976h.
Quallich, G. J. (2005). Development of the commercial pro-
ꢀ
cess for Zoloft /sertraline. Chirality, 17, S120–S126. DOI:
10.1002/chir.20113.
Tombo, G. M. R., & Belluš, D. (1991). Chirality and crop
protection. Angewandte Chemie International Edition, 30,
1193–1251. DOI: 10.1002/anie.199111933.
Tsunoda, T., Kaku, H., Nagaku, M., & Okuyama, E. (1997).
Deracemization of 2-alkylcyclohexanones utilizing host-gust
molecular association with optically active host compounds
in basic suspension media. Tetrahedron Letters, 38, 7759–
7760. DOI: 10.1016/s0040-4039(97)10071-5.
Verkuijl, B. J. V., Minnaard, A. J., de Vries, J. G., & Feringa,
B. L. (2009). Chiral separation of underivatized amino
acids by reactive extraction with palladium-BINAP com-
plexes. Journal of Organic Chemistry, 74, 6526–6533. DOI:
10.1021/jo901002d.
Zhou, C. S., Xu, P., Tang, K. W., Jiang, X. Y., Yang, T.,
& Zhang, P. L. (2013). Enantioselective extraction of hy-
drophilic 2-chloromandelic acid enantiomers by hydroxypro-
pyl-β-cyclodextrin: experiments and modeling. Chemical Pa-
pers, 67, 155–163. DOI: 10.2478/s11696-012-1268-6.
Sassatelli, M., Debition, É., Aboab, B., Prudhomme, M., &
Moreau, P. (2006). Synthesis and antiproliferaties activies
of indolin-2-one derivatives bearing amino acid moieties. Eu-
ropean Journal of Medicinal Chemisty, 41, 709–716. DOI:
10.1016/j.ejmech.2006.03.021.
Schlichting, E., Halwachs, W., & Schu¨gerl, K. (1987). Reac-
tive extraction of salicylic acid and D,L-phenylalanine in a
stirred cell. Chemical Engineering Communication, 51, 193–
205. DOI: 10.1080/00986448708911843.
Schuur, B., Winkelmam, J. G. M., & Heeres, H. J. (2008). Equi-
librium studies on enantioselective liquid–liquid amino acid
extraction using a cinchona alkaloid extractant. Industrial
& Engineering Chemistry Research, 47, 10027–10033. DOI:
10.1021/ie800668e.
Schuur, B., Winkelman, J. G. M., de Vries, J. G., & Heeres, H. J.
(2010). Experimental and modeling studies on the enantio-
separation of 3,5-dinitrobenzoyl-(R),(S)-leucine by continu-
ous liquid–liquid extraction in a cascade of centrifugal con-
tactor separators. Chemical Engineeering Science, 65, 4682–
4690. DOI: 10.1016/j.ces.2010.05.015.
Schuur, B., Verkuijl, B. J. V., Minnaard, A. J., de Vries, J.
G., Heeres, H. J., & Feringa, B. L. (2011). Chiral separa-
tion by enantioselective liquid–liquid extraction. Organic &
Biomolecular Chemistry, 9, 36–51. DOI: 10.1039/c0ob00610f.