Chiral separation of underivatized amino acids by reactive extraction with palladium-BINAP complexes

Article abstract of DOI:10.1021/jo901002d

(Figure Presented) In answer to the need for a more economic technology for the separation of racemates, a novel system for reactive enantioselective liquid-liquid extraction (ELLE) is introduced. Palladium (S)-BINAP complexes are employed as hosts in the separation of underivatized amino acids. The system shows the highest selectivity for the ELLE of tryptophan with metal complexes as hosts reported to date and shows a good selectivity toward a range of natural and unnatural amino acids. Furthermore, the host can be prepared in situ from commerically available compounds. Bulk-membrane transport in the form of U-tube experiments demonstrates the enantioselective and catalytic nature of the transport. The dependency of the system on parameters such as pH, organic solvent, and host-substrate ratio has been established. ³¹P NMR spectroscopy has been used to confirm the preferred enantiomer in the extraction experiments. The intrinsic selectivity was deduced by determination of the association constants of the palladium complex with the tryptophan enantiomers.

Full text of DOI:10.1021/jo901002d

pubs.acs.org/joc
Chiral Separation of Underivatized Amino Acids by Reactive Extraction
with Palladium-BINAP Complexes
Bastiaan J. V. Verkuijl,† Adriaan J. Minnaard,*,† Johannes G. de Vries,*,†,‡ and
Ben L. Feringa*,†
†Stratingh Institute for Chemistry, University of Groningen, Nijenborgh 4, 9747 AG Groningen, The
Netherlands, and ‡DSM Pharmaceutical Products-Innovative Synthesis & Catalysis, PO BOX 18, 6160 MD
Geleen, The Netherlands
a.j.minnaard@rug.nl; hans-jg.vries-de@dsm.com; b.l.feringa@rug.nl
Received May 13, 2009
In answer to the need for a more economic technology for the separation of racemates, a novel system
for reactive enantioselective liquid-liquid extraction (ELLE) is introduced. Palladium (S)-BINAP
complexes are employed as hosts in the separation of underivatized amino acids. The system shows
the highest selectivity for the ELLE of tryptophan with metal complexes as hosts reported to date and
shows a good selectivity toward a range of natural and unnatural amino acids. Furthermore, the host
can be prepared in situ from commerically available compounds. Bulk-membrane transport in the
form of U-tube experiments demonstrates the enantioselective and catalytic nature of the transport.
The dependency of the system on parameters such as pH, organic solvent, and host-substrate ratio
has been established. 31P NMR spectroscopy has been used to confirm the preferred enantiomer in
the extraction experiments. The intrinsic selectivity was deduced by determination of the association
constants of the palladium complex with the tryptophan enantiomers.
Introduction
resolution,3 kinetic resolution4 with enzymes or chiral cata-
lysts, or dynamic kinetic resolution.5 Sometimes the un-
wanted isomer can be racemized so that it can be recycled.6
Despite the theoretical maximum of 50% yield classic reso-
lution is still a widely applied method in the fine chemical and
pharmaceutical industries.7 However, this production meth-
od is laborious, sometimes suffers reproducibility problems,
and is relatively expensive.8 Thus new effective separa-
tion methods are highly desirable. Among the methods
being explored are resolution with in situ racemization,9
The production and availability of enantiomerically pure
compounds is of immediate importance to the pharmaceu-
tical industry.1 Methods available to provide enantiopure
compounds include resolution of racemates, isolation from
natural sources, fermentation, asymmetric catalysis, and
biocatalysis, with resolution still being used most often.2
Among the available resolution methods are resolution via
crystallization of diastereomeric salts, also called classical
€
*To whom correspondence should be addressed.
(1) Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kesseler, M.;
Sturmer, R.; Zelinski, T. Angew. Chem., Int. Ed. 2004, 43, 788–824.
(2) Carey, J. S.; Laffan, D.; Thomson, C.; Williams, M. T. Org. Biomol.
Chem. 2006, 4, 2337–2347.
(3) Liu, W. Handbook of Chiral Chemicals, 2 ed.; Ager, D. J., Ed.; CRC-
Taylor & Francis: Boca Raton, FL, 2006; pp 75-95.
(4) Patel, R. N. Stereoselective Biocatalysis; Marcel Dekker: New York,
2000.
(5) Pamies, O.; Backvall, J. E. Chem. Rev. 2003, 103, 3247–3261.
(6) Fujima, Y.; Ikunaka, M.; Inoue, T.; Matsumoto, J. Org. Proc. Res.
Dev. 2006, 10, 905–913.
(7) Collins, A. N.; Sheldrake, G. N.; Crosby, J. Chirality in industry II;
Wiley and Sons: New York, 1997.
(8) Houlton, S. Manuf. Chem. 2001, 72 (1), 16–17.
(9) Noorduin, W. L.; Izumi, T.; Millemaggi, A.; Leeman, M.; Meekes, H.;
Van Enckevort, W. J. P.; Kellogg, R. M.; Kaptein, B.; Vlieg, E.; Blackmond,
D. G. J. Am. Chem. Soc. 2008, 130, 1158–1159.
6526 J. Org. Chem. 2009, 74, 6526–6533
Published on Web 07/29/2009
DOI: 10.1021/jo901002d
r
2009 American Chemical Society

Verkuijl et al.
JOCArticle
TABLE 1. Tryptophan Extraction with Palladium and Platinum (S)-
BINAP Complexesa
[host] (mM) D(org/aq)b Rop
c
entry
complex
aq
1
2
3
4
5
6
7
8
9
[PdCl2((S)-BINAP)] H2Od
[PdCl2((S)-BINAP)] H2O
[PtCl2((S)-BINAP)] H2O
[PtCl2((S)-BINAP)] H2O
[PdCl2((S)-BINAP)] pH 7.0
[PdCl2((S)-BINAP)] pH 7.0
[PdCl2((S)-BINAP)] Trp-Nae
[PdCl2((S)-BINAP)] Trp-Na
[PtCl2((S)-BINAP)] Trp-Na
1
5
1
5
1
5
1
5
1
5
0.1
0.2
0.1
0.1
0.5
3.4
0.5
1.6
0.7
5.4
2.2
2.1
1.9
1.9
2.4
2.4
2.4
2.4
2.2
2.2
10 [PtCl2((S)-BINAP)] Trp-Na
aConditions: organic solvent=dichloromethane; T=6 °C; [subst-
rate]=2.0 mM, Vorg=Vaq=0.40 mL. bThe distribution D(org/aq) of
the substrate (AA) over the two phases is defined as the ratio of the
concentration of the substrate over the two phases (D=[AA]org/[AA]aq).
cThe operational selectivity (Rop) is defined as the ratio of distribution
of enantiomers (Rop=DD-AA/DL-AA). dDouble distilled water (dd). eTrp
with 1 equiv of NaHCO3 in dd.
FIGURE 1. Schematic representation of ELLE: (gray symbols)
(S)-substrate; (black symbols) (R)-substrate; and (black rectangles)
host.
chromatograph containing an MTBE solution of bis-
1,4-(dihydroquinidinyl)phthalazine as the stationary chiral
host solution and were able to fully separate the herbicide
2-(2,4-dichlorophenoxy)propionic acid (dichlorprop), which
was fed as a solution in aqueous buffer as the mobile phase.23
As these methods are not scalable we have developed the use
of centrifugal separators as a highly efficient method for
continuous extraction.24-26 Applying a number of these in
series allows for the full separation of a racemate.27
Over and above the technological aspects, the develop-
ment of improved chiral host compounds is still required. In
particular the enantioselective extraction of underivatized
amino acids is one of the great challenges within the field of
ELLE. Only a few studies have been reported in this field to
date. Cram et al. have carried out pioneering work showing
that high selectivities could be achieved for a range of amino
acid perchlorate salts by using crown ethers with a functio-
nalized BINOL backbone.28,29 Rebek et al. developed a
receptor that although not enantioselective, showed high
selectivity for underivatized aromatic amino acids.30 Subse-
quently, De Mendoza and co-workers combined a chiral
guanidinium group with a crown-ether to extract tryptophan
enantioselectively.31 Metal complexes have been employed
as hosts including lanthanide β-diketonate32 and copper-
proline complexes.20,33 Copper complexes with chiral diamino
membrane-assisted separations,10 diastereomer separation
by distillation,11 supercritical extraction,12-14 fractional en-
antioselective extraction,15,16 and chiral simulated moving
bed (SMB).17
In enantioselective liquid-liquid extraction (ELLE), an
enantiopure host is used as an extractant to react enantios-
pecifically and reversibly with a racemic substrate (Figure 1).
If the host is confined to one phase in a biphasic system, an
enantiomeric separation of the substrate can occur between
the two phases in a single step. If the separation is imperfect,
a fractional extraction scheme is needed.18,19 A minimal
selectivity of 1.5 is generally viewed as being necessary to
avoid the requirement for an excessive number of fractional
extraction steps.20 An important feature of this system is its
potential versatility. With a versatile host, for instance, the
separation of racemates of an entire class of compounds is
potentially achievable.
Typically U-tubes21 or membranes are employed in se-
paration schemes where the chiral host is applied in a
catalytic fashion. The use of a number of membranes in
series has even allowed for complete separation.22 Maier and
Lindner have reported the use of a centrifugal partition
(10) van der Ent, E. M.; Thielen, T. P. H.; Stuart, M. A. C.; van der Padt,
A.; Keurentjes, J. T. F. Ind. Eng. Chem. Res. 2001, 40, 6021–6027.
(11) Kozma, D.; Fogassy, E. Chirality 2001, 13, 428–430.
(23) Gavioli, E.; Maier, N. M.; Minguillon, C.; Lindner, W. Anal. Chem.
2004, 76, 5837–5848.
(24) Schuur, B.; Floure, J.; Hallett, A. J.; Winkelman, J. G. M.; de Vries,
J. G.; Heeres, H. J. Org. Proc. Res. Dev. 2008, 12, 950–955.
(25) Schuur, B.; Jansma, W. J.; Winkelman, J. G. M.; Heeres, H. J. Chem.
Eng. Proc. 2008, 47, 1484–1491.
(26) Hallett, A. J.; Kwant, G. J.; de Vries, J. G. Chem.;Eur. J. 2008,
2111–2120.
(27) (a) Schuur, B. Enantioselective liquid-liquid extraction in centrifugal
contactor separators. Thesis, University of Groningen, 2008. (b) Schuur, B.;
Hallett, A. J.; Winkelman, J. G. M.; de Vries, J. G.; Heeres, H. J. Org. Proc. Res.
Dev. Submitted for publicaiton.
ꢀ
ꢀ
(12) Keszei, S.; Simandi, B.; Szekely, E.; Fogassy, E.; Sawinsky, J.;
ꢀ
Kemeny, S. Tetrahedron: Asymmetry 1999, 10, 1275–1281.
ꢀ
ꢀ
(13) Simandi, B.; Keszei, S.; Fogassy, E.; Kemeny, S.; Sawinsky, J.
J. Supercrit. Fluids 1998, 13 (1-3), 331–336.
ꢀ
(14) Simandi, B.; Keszei, S.; Fogassy, E.; Sawinsky, J. J. Org. Chem. 1997,
62, 4390–4394.
(15) Steensma, M. Chiral Separation of amino-alcohols and amines by
fractional reactive extraction; Ph. D. Thesis, University of Twente, 2005.
(16) Eliel, E.; Wilen, S.; Mander, L. Miscellaneous Separation Methods.
In Stereochemistry of organic compounds; Wiley: New York,, 1994; pp 416-
424.
(17) Quallich, G. J. Chirality 2005, 17, S120–S126.
(18) Robbins, L. A. Liquid-liquid extraction. In Handbook of separation
technology for chemical engineers, 3rd ed.; Schweitzer, P. A., Ed.; McGraw-
Hill: New York, 1997; pp 1419-1447.
(19) Steensma, M.; Kuipers, N. J. M.; de Haan, A. B.; Kwant, G. Chem.
Eng. Proc. 2007, 46, 996–1005.
(28) Peacock, S. S.; Walba, D. M.; Gaeta, F. C. A.; Helgeson, R. C.;
Cram, D. J. J. Am. Chem. Soc. 1980, 102, 2043–2052.
(29) Lingenfelter, D. S.; Helgeson, R. C.; Cram, D. J. J. Org. Chem. 1981,
46, 393–406.
(30) Rebek, J.; Askew, B.; Nemeth, D.; Parris, K. J. Am. Chem. Soc. 1987,
109, 2432–2434.
ꢀ
(20) Steensma, M.; Kuipers, N. J. M.; de Haan, A. B.; Kwant, G.
Chirality 2006, 18, 314–328.
(31) Galan, A.; Andreu, D.; Echavarren, A.; Prados, P.; De Mendoza, J.
J. Am. Chem. Soc. 1992, 114, 1511–1512.
(21) Mohler, L. K.; Czarnik, A. W. J. Am. Chem. Soc. 1993, 115, 2998–
2999.
(22) Maximini, A.; Chmiel, H.; Holdik, H.; Maier, N. W. J. Membr. Sci.
2006, 276, 221–231.
(32) Tsukube, H.; Shinoda, S.; Uenishi, J.; Kanatani, T.; Itoh, H.; Shiode,
M.; Iwachido, T.; Yonemitsu, O. Inorg. Chem. 1998, 37, 1585–1591.
(33) Takeuchi, T.; Horikawa, R.; Tanimura, T. Anal. Chem. 1984, 56,
1152–1155.
J. Org. Chem. Vol. 74, No. 17, 2009 6527

Verkuijl et al.
JOCArticle
FIGURE 2. The dependency of the distribution and the operational selectivity (Rop) on pH. Conditions: organic solvent=dichloromethane;
T=6 °C; metal precursor=[PdCl2(CH3CN)2]; ligand=(S)-BINAP; [metal]=1.0 mM; [ligand]=1.0 mM; [substrate]=2.0 mM.
ethane derivatives34 or hydroxyproline derivatives35 have
been employed to transport amino acids through a liquid
membrane enantioselectively. However, the challenge of
developing a chiral metal complex that is facile to synthesize,
which binds with high operational selectivity (Rop) to a
variety of underivatized amino acids and can be applied in
fractional ELLE, remains.
It has been reported that metal complexes can show good
performance in enantioselective extraction.36,37 Palladium
phosphine complexes in general have shown the ability to
bind underivatized amino acids.38,39 On the other hand, the
chiral bisphosphine BINAP has proven to be a highly
versatile ligand in asymmetric catalysis.40 Since it is possible
to bind other ligands by exchange of the counterion it see-
med logical to access palladium and platinum complexes of
(S)-BINAP for their ability to extract amino acids in general
and tryptophan (Trp) in particular in an enantioselective
manner.
In this paper, we disclose our recent finding that the use of
the metal complex [PdCl2((S)-BINAP)] as a host results in
the highest operational selectivity for underivatized trypto-
phan in an ELLE system using metal complexes reported to
date. This system also shows good selectivity toward a range
of natural and unnatural amino acids. Furthermore, the host
can be prepared in situ from commercially available com-
pounds.
increased to 0.2 by increasing the host concentration (entry
2). In the case of a buffered aqueous solution at pH 7.0 (entry
5) the distribution is increased to 0.5 and the Rop increases to
2.4. A further increase of the distribution was obtained by
increasing the host concentration (entry 6). When Trp-Na is
used as a substrate (entry 7), the distribution and selectivity
compare to that using a pH 7.0 buffered solution (entry 5).
PtCl2((S)-BINAP) shows a higher distribution (entries 9 and
10) albeit with a lower Rop (2.2).
The pH of the aqueous phase is in most cases a key issue in
ELLE. Normally, the pH should affect the distribution of the
substrate, but not the Rop.
In Figure 2, the relationship between the pH of the
aqueous phase and the distribution and Rop is depicted. In
the measured pH range, significant physical partition (i.e.,
the host-absent distribution of the substrate) was not ob-
served (P(org/aq) = 0.0). Below a pH of 4, Trp does not
show a significant distribution. As the pH increases from
5.0 to 7.0, the distribution rises from 0.1 to 0.7. The
corresponding Rop remains constant at 2.1. When the
pH is >8.0, the Rop drops, whereas at pH 10.0, D = 2.3
and Rop =1.2 are observed. The increased distribution in
response to increasing pH suggests that the Trp anion
complexes to the palladium complex.
The observation that the Rop stays constant in the pH
range below 7.0 and the distribution below 1 shows that in
this pH range, Trp binds enantioselectively to palladium,
most probably in a 1:1 stoichiometry. At pH values above
8.0, it is possible that a second tryptophan binds to the
complex (D > 1 and c(host) = 1.0 mM), providing an
explanation for the decrease in Rop.
Another important parameter in ELLE is the host con-
centration. Increasing the host concentration normally
causes the distribution to increase. The Rop should be
independent of the host concentration. In Figure 3 it is
shown that the distribution increases up to 0.7 when the host
concentration is increased to 2.5 mM. The Rop remains
constant at 2.35. The linear fit depicted in the graph on the
right shows an Rop between 2.32 and 2.38. These results show
that excellent control of the distribution can be achieved by
varying the host concentration while the Rop remains con-
stant.
Results and Discussion
Extraction Experiments. In Table 1 the results of the
extraction of tryptophan with palladium and platinum (S)-
BINAP complexes are summarized. In entries 1-4 it is
shown that the complexes are able to extract tryptophan
enantioselectively with a selectivity of 2.4 in the case of
the palladium complex under completely neutral conditi-
ons. The distribution (D(org/aq)) is quite low, but can be
(34) Scrimin, P.; Tecilla, P.; Tonellato, U. Tetrahedron 1995, 51, 217–230.
(35) Pickering, P. J.; Chaudhuri, J. B. Chirality 1997, 9, 261–267.
(36) Reeve, T. B.; Cros, J. P.; Gennari, C.; Piarulli, U.; de Vries, J. G.
Angew. Chem., Int. Ed. 2006, 45, 2449–2453.
(37) Konishi, K.; Yahara, K.; Toshishige, H.; Aida, T.; Inoue, S. J. Am.
Chem. Soc. 1994, 116, 1337–1344.
(38) Bravo, J.; Cativiela, C.; Chaves, J. E.; Navarro, R.; Urriolabeitia,
E. P. Inorg. Chem. 2003, 42, 1006–1013.
(39) Dialer, H.; Mayer, P.; Polborn, K.; Beck, W. Eur. J. Inorg. Chem.
2001, 1051–1055.
(40) Berthod, M.; Mignani, G.; Woodward, G.; Lemaire, M. Chem. Rev.
2005, 105, 1801–1836.
The organic solvent can have a profound influence on both
distribution and the operational selectivity. In enantioselec-
tive host-guest complexations, the solvent can influence the
6528 J. Org. Chem. Vol. 74, No. 17, 2009

Verkuijl et al.
JOCArticle
FIGURE 3. The dependency of the distribution and the operational selectivity (Rop) on the host concentration. Conditions: organic solvent=
dichloromethane; T=6 °C; pH(aqueous phase)=7.0; metal precursor=[PdCl2(CH3CN)2]; ligand=(S)-BINAP; [substrate]=2.0 mM.
TABLE 2. Solvent Effect on Enantioselective Extraction of Tryptophan
TABLE 3. Substrate Scopea
with [PdCl2((S)-BINAP)]a
entry
substrate
D(org/aq)
Rop
entry
solvent
D(org/aq)a
Rop
1
2
3
4
5
6
7
8
Asp
His
0.7
1.2
0.6
0.7
2.8
3.4
0.7
0.0
1.1
2.0
2.0
1.9
2.3
2.4
2.1
1
2
3
4
5
dicholoromethane
1,2-dichloroethane
chloroform
chlorobenzene
toluene
0.5
0.4
0.6
0.3
1.0
2.4
2.8
1.9
1.2
1.2
Met
Pgeb
Phe
Trp
Val
aConditions: organic solvent = dichloromethane; T = 6 °C; pH-
(aqueous phase)=7.0; metal precursor=[PdCl2(CH3CN)2]; ligand=
(S)-BINAP; [metal]=1.0 mM; [ligand]=1.0 mM; [substrate]=2.0 mM.
Pglc
aConditions: solvent=dichloromethane; T=6 °C; pH (aqueous ph-
ase) = 7.0; metal precursor: [PdCl2(CH3CN)2]; ligand: (S)-BINAP;
[PdCl2(CH3CN)2]=5.0 mM; [ligand]=5.0 mM; [substrate]=2.0 mM;
bPge=Phenylglycine; cPgl=Phenylglycinol
noncovalent interactions and affect the conformation of the
chiral partners.41
In Table 2, the distribution and operational selectivity of
the ELLE of Trp by palladium BINAP complexes with five
different organic solvents is shown. The chlorinated solvents
show the same distribution trend, whereas the operational
selectivities differ considerably. The highest selectivity is
achieved with 1,2-dichloroethane (entry 2) inducing an Rop
of 2.8 with a D of 0.4. In chlorobenzene (entry 4) no
significant Rop was observed. The aromatic nature of the
solvent is likely to influence the enantioselective complexa-
tion negatively, suggesting that π-π interactions between
tryptophan and the palladium complex are important in
chiral recognition. Toluene (entry 5) stands out, being the
only nonchlorinated solvent to show both high distribution
and no significant operational selectivity. Visual inspection
of the system showed a suspension in the organic layer, which
confirms precipitation of the complex.
A feature of a successful ELLE system is the substrate
scope. Several natural and one unnatural amino acid have
been extracted to test their distribution and operational
selectivity. From the data in Table 3 it is evident that aspartic
acid shows good distribution of 0.7 but, unfortunately, the
Rop is not significant. The second carboxylic acid group of
Asp does not seem to have a positive effect on D and Rop
compared to Trp. The distributions summarized in entries
2-7 show a wide range of amino acid enantiomers that can
be successfully separated by extraction. All of the amino
acids other then Asp tested could be extracted with an Rop of
at least 1.9. The amino acids with a benzyl, phenyl, or an
FIGURE 4. Scheme of the U-tube used for bulk-membrane
transport.
isopropyl side group could be extracted with high opera-
tional selectivities. These results demonstrate the versatility
of the [PdCl2((S)-BINAP)] complex as an enantioselective
extractant. The nonsignificant distribution of phenylglycinol
(entry 8) illustrates that the carboxylic acid group is required
for extraction.
After separation of the two liquid phases the amino acid
attached to the metal complex can be re-extracted into a
second aqueous phase at a different pH, i.e., back extraction.
Back-extraction of Trp is possible by using a 1 M HCl (aq)
solution. Trp-Na was extracted with 1 equiv of [PdCl2((S)-
BINAP)]; subsequent back-extraction yielded the expected
(41) Schug, K. A.; Lindner, W. Chem. Rev. 2005, 105, 67–113.
J. Org. Chem. Vol. 74, No. 17, 2009 6529

Verkuijl et al.
JOCArticle
FIGURE 5. Transport rate and ee values of the bulk-membrane transport. Conditions: organic solvent=dichloromethane; T=6 °C. Lag time
of the transport through the transport phase (8 h) is omitted.
FIGURE 6. 31P NMR spectra of the titration of the palladium complex upon addition of D-Trp (spectra 1-7: [D-Trp]=0.0, 0.5, 1.0, 1.5, 2.0,
5.0, 10.0 mM). Extraction conditions: organic solvent=CDCl3; T=6 °C.
values of c(aq)=0.20 mM with the predicted ee of 36%. The
remaining complex was successfully re-employed in a sub-
sequent extraction, demonstrating the full reversibility of the
system.
with respect to the carrier are transported in this experiment,
albeit that the ee decreases from 30% to 20%. These results
demonstrate the enantioselective and reversible nature of the
system. The low rate is due to the poor phase mixing in the
U-tube and should not be seen as indicative of the rate of a
preparative separation.
Bulk-Membrane Transport. Reversibility of the amino
acid binding to the [PdCl2((S)-BINAP)] complex was also
proven by bulk-membrane transport experiments. The U-
tube depicted in Figure 4 was used to transport Trp from the
aqueous feeding phase to the aqueous receiving phase
through the organic (DCM) transport phase, using the
[PdCl2((S)-BINAP)] complex as an enantioselective carrier.
Figure 5 shows the transport rate and the ee of the receiving
phase. After 100 h, the amount of transported Trp exceeds
the amount of carrier present in the transport phase. The
transport rate is constant during the transport, which sug-
gests a constant concentration of substrate in the transport
phase. The pH of the feeding phase starts at 8.1 and declines
during the transport process up to 7.4. Up to 4 equiv of Trp
31P NMR Spectroscopy. The 31P NMR titration spectra in
Figures 6 and 7 show the 31P NMR shifts of BINAP in the
two diastereomeric complexes formed from [PdCl2((S)-
BINAP)] and both enantiomers of Trp. (S)-BINAP is used
as an internal standard (c=0.2 mM), since an excess showed
no change in extraction. The [PdCl2((S)-BINAP)] complex
appears as a singlet absorption at 29.7 ppm.
At higher concentrations of D-Trp (Figure 6), the ab-
sorption at 29.7 ppm decreases in intensity, whereas a
singlet absorption grows at 27.4 ppm. At c = 10.0 mM,
the initial absorption is below detection limits, leaving only
the new singlet absorption. Together with ES-MS data
6530 J. Org. Chem. Vol. 74, No. 17, 2009

Verkuijl et al.
JOCArticle
FIGURE 7. 31P NMR spectra of the titration of the palladium complex upon addition of L-Trp (spectra 1-7: [L-Trp]=0.0, 0.5, 1.0, 1.5, 2.0, 5.0,
10.0 mM). Extraction conditions: organic solvent=CDCl3; T=6 °C.
FIGURE 9. Differential UV-vis spectra of the palladium complex
upon addition of D-Trp ([D-Trp] = 0.0-10.0 mM). Extraction
conditions: solvent=dichloromethane; T=6 °C.
FIGURE 8. UV-vis spectra of the palladium complex upon addi-
tion of D-Trp ([D-Trp] = 0.0-10.0 mM). Extraction conditions:
solvent=dichloromethane; T=6 °C.
NMR time scale. The broad, ill-defined absorption in the
case of L-Trp is indicative of fast exchange processes
suggesting a less stable complex. This agrees with the
observation that D-Trp is the preferred enantiomer in the
extraction experiments.
UV-Vis Spectrosopy and Association Constants. The re-
lationship between the Rop and the association constants
between both enantiomers of the substrate and the host was
investigated by titration, using UV-vis spectroscopy. The
association constants of the palladium complex with the
enantiomers of tryptophan in a biphasic system were deter-
mined accordingly.
The UV-vis spectra shown in Figures 8 and 9 show the
effect of increased complexation of D-Trp to the palladium
complex. An isosbestic point is maintained at 328 nm, which
slightly shifts, probably due to a small absorption of Trp at
this wavelength.42 Above 328 nm, titration with D-Trp results
in a decrease in the absorption of the palladium complex.
taken from the organic phase (MS (ES) m/z 931.5 (Mþ)),
which confirms the complexation of D-Trp to the palladium
BINAP complex, a [PdCl((S)-BINAP)(D-Trp)] complex is
proposed.
Figure 7 shows the spectra obtained upon addition of
L-Trp to the palladium complex. The observed spectra are
different compared to the spectra with D-Trp (Figure 6).
As in Figure 6, the initial absorption at 29.7 ppm de-
creases in intensity upon increasing the L-Trp concentra-
tion. The newly formed complex, however, shows
a broad, ill-defined absorption, in the range of 26.1-
28.1 ppm. The difference between the absorption at 27.4
ppm of the [PdCl((S)-BINAP)(D-Trp)] complex and the
broad absorption at 26.1-28.1 ppm of [PdCl((S)-
BINAP)(L-Trp)] is evidence for a difference in complexa-
tion between the two Trp enantiomers. The singlet
absorption in the case of D-Trp suggests a symmetrical
complex with respect to the phosphorus atoms on the 31P
(42) Nishijo, J.; Yonetani, I.; Iwamoto, E.; Tokura, S.; Tagahara, K.;
Sugiura, M. J. Pharm. Sci. 1990, 79, 14–18.
J. Org. Chem. Vol. 74, No. 17, 2009 6531

Verkuijl et al.
JOCArticle
Which gives an operational selectivity that can be expressed
as:
Ki,D
Ki,L
DD-AA
¼
DL-AA
Rop
¼
Observations on all of the studied parameters are in line
with this model. The pH has an influence on the distribu-
tion, but not on the Rop, as does the host concentration.
Furthermore, the Rop corresponds to the intrinsic selectivity
and thus to the association constants of the host and the
substrate.
Conclusions
In conclusion, we have developed a new method of
efficient enantioselective extraction of underivatized amino
acids using a readily available host that showed unprece-
dented selectivities for tryptophan. Bulk membrane trans-
port experiments proved that the system is fully reversible.
The intrinsic selectivity determined corresponds well with the
operational selectivity. 31P NMR spectroscopy suggests that
a more stable complex is formed in the case of the preferred
enantiomer. An extraction model featuring a counterion
exchange extraction mechanism at the liquid-liquid inter-
face is proposed. Further investigations are underway to
expand the scope, increase the selectivity, and elucidate the
further details of the extraction mechanism.
FIGURE 10. Proposed extraction model.
Nonlinear regression of the titration data at 368 nm yielded
the following association constants:
KaðD-TrpÞ ¼ 856 ( 46
KaðL-TrpÞ ¼ 539 ( 33
The association constant of D-Trp is higher than that of
L-Trp. This is in agreement with the preferred extraction of
D-Trp in a typical extraction experiment. If the intrin-
sic selectivity is defined as the ratio of the Ka values, then
R
int =1.6. This value is somewhat lower than the Rop obs-
Experimental Section
erved in the extraction experiments (vide supra). Neverthe-
less, the intrinsic selectivity corresponds reasonably well to
the operational selectivity.
General procedure for extraction experiments: cis-[PdCl2-
(CH3CN)2]2 and (S)-BINAP were dissolved in dcm in equimolar
amounts and the mixture was stirred overnight. The solution
was diluted to the desired concentration and used in the extrac-
tion experiments. The palladium and platinum complexes were
synthesized and purified following literature procedures and
spectroscopic and analytical data correspond with those re-
ported.44 The isolated complexes were used as a host, giving
similar results compared to the aforementioned in situ generated
host in the case of the palladium complex. The racemic amino
acid was dissolved in double distilled water (adding 1 equiv of
sodium bicarbonate in the case of Trp-Na) or in the appropiate
sodium phosphate buffer (c=0.100 M) at a concentration of c=
2.0 mM. The two stock solutions were put together in a vial in
equivolumous amounts (0.40 mL) and stirred overnight at 6 °C.
All extractions were performed at least in duplo. A sample of the
aqueous phase was analyzed by using a RP-HPLC equipped
with a Crownpak(þ) column.
Proposed Extraction Model. As a preliminary model,
based on the results presented here, we propose the interface
model for extraction.34,43 If the pH of the aqueous phase is at
7.0, the amino acid is in equilibrium between the zwitterionic
form and the anionic form. No significant physical partition
has been observed under these conditions (P(org/aq)=0.0).
Hence, complexation of the amino acid with the metal
complex is expected to occur at the liquid-liquid interface
as a counterion exchange with a chloride, according to the
model shown in Figure 10.
According to the model shown in Figure 10, the distribu-
tion for D-AA can be defined as:
½D-AAꢀorg;total
½D-AAꢀaq;total
½PdClððSÞ-BINAPÞðD-AAÞꢀorg
½D-AAꢀaq
DD-AA
¼
¼
General procedure for titration experiments: Extractions were
carried out as indicated above with the aqueous phase at pH 7.0.
The amino acid was enantiomerically pure and its concentration
was varied between 0.20 and 10.0 mM. The equivolumous
amounts of the liquid phases of the extraction were increased
to V=0.6 mL. In the 31P NMR titrations, CDCl3 was used as
solvent. CDCl3 gave extraction results which compared well
with CHCl3.
The equilibria at the interface can be defined as:
½PdClððSÞ-BINAPÞðD-AAÞꢀorg½Cl-ꢀaq
½D-AAꢀaq½PdCl2ððSÞ-BINAPÞꢀorg
Ki,D
¼
Bulk-membrane transport experiments: Experiments were
done in duplicate. Feeding phase: double distilled water, V=
5.0 mL; c(Trp-Na)=20.0 mM. Transport phase: dichlorome-
thane, V=10.0 mL, c([PdCl2((S)-BINAP)])=0.5 mM. Receiving
phase: 0.100 M HCl (aq). The phases were placed into the
U-tube depicted in Figure 4. The tube was placed in a cooled
chamber at T=6 °C. The magnetic stirrer was set to 900 rpm.
In this way the distribution ratio for one enantiomer be-
comes:
½PdCl2ððSÞ-BINAPÞꢀorg
DD-AA ¼ Ki,D
½Cl-ꢀaq
ꢁ
(44) Celentano, G.; Benincori, T.; Radaelli, S.; Sada, M.; Sannicolo, F.
J. Organomet. Chem. 2002, 643, 424–430.
(43) Pickering, P. J.; Chaudhuri, J. B. Chem. Eng. Sci. 1997, 52, 377–386.
6532 J. Org. Chem. Vol. 74, No. 17, 2009

Verkuijl et al.
JOCArticle
When a sample of 0.2 mL was taken out of the receiving phase, it
was replaced by 0.2 mL of 0.100 M HCl (aq).
tific Research (NWO) through the Separation Technology
program in cooperation with DSM and Schering-Plough.
Acknowledgment. We thank Dr. W. R. Browne and Dr. T.
Jerphagnon for useful discussions. This research was finan-
cially supported by The Netherlands Organisation for Scien-
Supporting Information Available: Titration graphs and
nonlinear regression fits. This material is available free of charge
via the Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 74, No. 17, 2009 6533