Efficient and Highly Selective Copper(II) Transport across a Bulk Liquid Chloroform Membrane Mediated by Lipophilic Dipeptides

Article abstract of DOI:10.1021/jo9703257

Several structurally simple N-monoalkylated and -dialkylated dipeptides made of α-amino acids Gly, Phe, and Leu, 1-11, were synthesized and investigated as carriers for the transport of Cu(II), Zn(II), and Ni(II) from an aqueous pH = 5.6 buffer source to a 0.1 M HCl receiving phase across a bulk chloroform membrane. The proton-driven translocation was followed during the process by analyzing the metal ion concentrations in the three phases. The transport efficiency depends on the ease of formation of a neutral complex with Cu(II) (the peptide group and carboxylic acid being deprotonated) at the source-chloroform interface and on that of its disruption by protonation at the receiving phase: the carrier's lipophilicity favors the metal ion uptake and not the release. By modulating the length of the N-alkyl chains and the hydrophobicity of the dipeptide moiety, a quite remarkable transport efficiency was observed for Cu(II), in most cases superior to that of the industrial extractant Kelex 100. Moreover, using L,L- and L,D-N-octyl-PheLeu as carriers, remarkable diastereomeric effects were observed in the rate of uptake and release of Cu(II) ion although the differences mutually compensate in the overall transport rate. Under the conditions used the carriers are much less effective in the translocation of Zn(II) and Ni(II) and their transport efficiency drops dramatically in the presence of Cu(II), the latter being favored by factors of 1.2 × 10³ and > 10⁴, respectively. Such very high selectivities depend on the fact that only Cu(II) among other transition metal ions can form neutral complexes at the pH value of the source phase.

Full text of DOI:10.1021/jo9703257

5592
J . Org. Chem. 1997, 62, 5592-5599
Efficien t a n d High ly Selective Cop p er (II) Tr a n sp or t a cr oss a Bu lk
Liqu id Ch lor ofor m Mem br a n e Med ia ted by Lip op h ilic Dip ep tid es
Marco C. Cleij,†,§ Paolo Scrimin,‡ Paolo Tecilla,† and Umberto Tonellato*,†
Department of Organic Chemistry and Centro CNR Meccanismi di Reazioni Organiche,
University of Padova, Via Marzolo 1, 35131 Padova, Italy, and Department of Chemical Sciences,
University of Trieste, Via Giorgieri 1, 34127 Trieste, Italy
Received February 19, 1997X
Several structurally simple N-monoalkylated and -dialkylated dipeptides made of R-amino acids
Gly, Phe, and Leu, 1-11, were synthesized and investigated as carriers for the transport of Cu(II),
Zn(II), and Ni(II) from an aqueous pH ) 5.6 buffer source to a 0.1 M HCl receiving phase across
a bulk chloroform membrane. The proton-driven translocation was followed during the process by
analyzing the metal ion concentrations in the three phases. The transport efficiency depends on
the ease of formation of a neutral complex with Cu(II) (the peptide group and carboxylic acid being
deprotonated) at the source-chloroform interface and on that of its disruption by protonation at
the receiving phase: the carrier’s lipophilicity favors the metal ion uptake and not the release. By
modulating the length of the N-alkyl chains and the hydrophobicity of the dipeptide moiety, a quite
remarkable transport efficiency was observed for Cu(II), in most cases superior to that of the
industrial extractant Kelex 100. Moreover, using L,L- and L,D-N-octyl-PheLeu as carriers,
remarkable diastereomeric effects were observed in the rate of uptake and release of Cu(II) ion
although the differences mutually compensate in the overall transport rate. Under the conditions
used the carriers are much less effective in the translocation of Zn(II) and Ni(II) and their transport
efficiency drops dramatically in the presence of Cu(II), the latter being favored by factors of 1.2 ×
103 and >104, respectively. Such very high selectivities depend on the fact that only Cu(II) among
other transition metal ions can form neutral complexes at the pH value of the source phase.
The selective transport of metal ions across a mem-
selective extractions4 from an aqueous into an organic
phase, fewer studies dealing with their transport across
a membrane have been reported5-13 in spite of a growing
demand for these systems in waste water treatment,
medicine, and metallurgy. Investigations of the perme-
ation of biological membranes or of closely related
vesicular double layers are not easily accessible,9 and
most studies in the field have been addressed to liquid
membranes, made usually of chloroform. As pointed out
by Menger,10 liquid bulk organic solvents bear very little
similarity with biological membranes; however, they may
be useful for evaluating the factors playing a role in the
translocation from and to aqueous pools through an
brane is known to play an essential role in many
biological processes.1 Over the years, a large number of
molecules featuring proper binding sites, particularly
crown ethers, have been synthesized and demonstrated
to act as selective carriers of alkali or ammonium ions
across a membrane.2 Theoretical models aimed at cor-
relating the efficacy and selectivity of the transport with
the structure of the carrier molecules have been also
proposed and experimentally tested.3 In the case of
transition metal ions, although a large number of effec-
tive ligands have been characterized and utilized in
*Corresponding author. Phone: +39-49-8275269. Fax: +39-49-
8275239. E-mail: tonum@chor02.chor.unipd.it.
(4) Reviews: (a) Freiser, H. Anal. Sci. 1995, 11, 191. (b) Freiser, H.
Chem. Rev. 1988, 88, 611. (c) Lobana, T. S.; Sandhu, S. S. Coord. Chem.
Rev. 1982, 47, 283. (d) Uhlig, E. Coord. Chem. Rev. 1982, 43, 299.
(5) (a) Hiratani, K.; Taguchi, K.; Ohhashi, K.; Nakayama, H. Chem.
Lett. 1989, 2073. (b) Hiratani, K.; Hirose, T.; Fujiwara, K.; Saito, K.
Chem. Lett. 1990, 1921. (c) Hiratani, K.; Taguchi, K. Chem. Lett. 1990,
725. (d) Hiratani, K.; Hirose, T.; Saito, K. J . Org. Chem. 1992, 57, 7083.
(e) Kasuga, K.; Hirose, T.; Takahashi, T.; Hiratani, K. Chem. Lett. 1993,
2183. (f) Hiratani, K.; Sugihara, H.; Kasuga, K.; Fujiwara, K.; Ha-
yashita, T.; Bartsch, R. A. J . Chem. Soc., Chem. Commun. 1994, 319.
(6) Di Casa, M.; Fabbrizzi, L.; Perotti, A.; Poggi, A.; Riscassi, R.
Inorg. Chem. 1986, 25, 3984.
(7) (a) Tsukube, H.; Adachi, H.; Morosawa, S. J . Chem. Soc., Perkin
Trans. I 1989, 89. (b) Maruyama, K.; Tsukube, H. J . Am. Chem. Soc.
1980, 102, 3246.
(8) (a) Kishii, N.; Araki, K.; Shiraishi, S. J . Chem. Soc. Dalton Trans.
1985, 373. (b) Kishii, N.; Araki, K.; Shiraishi, S. J . Chem. Soc., Chem.
Commun. 1984, 103.
(9) (a) Moss, R. A.; Byeong, D. P.; Scrimin, P.; Ghirlanda, G. J . Chem.
Soc., Chem. Commun. 1995, 1627. (b) Ghirlanda, G.; Scrimin, P.;
Tecilla, P.; Tonellato, U. J . Org. Chem. 1993, 58, 3025. (c) Scrimin, P.;
Tecilla, P.; Tonellato, U. J . Am. Chem. Soc. 1992, 114, 5086. (d)
Kragten, U. F.; Roks, M. F. M.; Nolte, R. J . M. J . Chem. Soc., Chem.
Commun. 1985, 1275.
(10) Menger, F. M.; Lee, J . J . J . Org. Chem. 1993, 58, 1909.
(11) Ameerunisha, S.; Zacharias, P. S. Polyhedron 1994, 13, 2327.
(12) (a) Scrimin, P.; Tonellato, U.; Zanta, N. Tetrahedron Lett. 1988,
29, 4967. (b) Scrimin, P.; Tecilla, P.; Tonellato, U. Tetrahedron 1995,
51, 217.
† University of Padova.
‡ University of Trieste.
§ European E.C. Research Fellow (1993-1994).
X Abstract published in Advance ACS Abstracts, J uly 1, 1997.
(1) (a) Westley, J . W.; Evans, R. H.; Williams, T.; Stempel, A. J .
Chem. Soc., Chem. Commun. 1970, 71. (b) Maier, C. A.; Paul, I. C. J .
Chem. Soc., Chem. Commun. 1971, 181. (c) Ring, K. Angew. Chem.,
Int. Ed. Engl. 1970, 9, 345. (d) Ovchinnikov, Y. A.; Ivanov, V. T.;
Shkrob, A. M. Membrane Active Complexons; Elsevier: Amsterdam,
1974. (e) Urry, D. W. Top. Curr. Chem. 1985, 128, 175. (f) Linder, M.
C. Biochemistry of Copper; Plenum: New York, 1991.
(2) Selected examples: (a) Chrisstoffels, L. A. J .; Struijk, W.; de
J ong, F.; Reinhoudt, D. N. J . Chem. Soc., Perkin Trans. 2 1996, 1617.
(b) Gokel, G. W.; Murillo, O. Acc. Chem. Res. 1996, 29, 425. (c) Xie, Q.;
Gokel, G. W.; Hernandez, J .; Echegoyen, L. J . Am. Chem. Soc. 1994,
116, 690. (d) Pregel, M. J .; J ullien, L.; Lehn, J . M. Angew. Chem., Int.
Ed. Engl. 1992, 31, 1637. (e) Izatt, R. M.; Gypzy, C. L.; Biernat, J . F.;
Bochenska, M.; Bruening, R. L.; Bradshaw, J . S.; Christensen, J . J . J .
Inclusion Phenom. Mol. Recognit. Chem. 1989, 7, 487. (f) Masushima,
K.; Kobayashi, H.; Nakatuji, Y.; Okahara, M. Chem. Lett. 1983, 792.
(g) Schchori, E.; Grodzinskira, J . J . J . Appl. Polym. Sci. 1976, 20, 773.
(h) Choy, E. M.; Evans, D. F.; Cussler, E. L. J . Am. Chem. Soc. 1974,
96, 7085.
(3) (a) Lamb, J . D.; Izatt, R. M.; Garrick, D. G.; Bradshaw, J . S.;
Christensen, J . J . J . Membr. Sci. 1981, 9, 83. (b) Izatt, R. M.; Clark,
G. A.; Bradshaw, J . S.; Lamb, J . D.; Christensen, J . J . Sep. Purif.
Methods 1986, 15, 21. (c) Behr, J . P.; Kirch, M.; Lehn, J . M. J . Am.
Chem. Soc. 1985, 107, 241. (d) Visser, H. C.; Reinhoudt, D. N.; de J ong,
F. Chem. Soc. Rev. 1994, 75 and references cited therein.
(13) Kimura, E.; Dalimunte, C. A.; Yamashita, A.; Machida, R. J .
Chem. Soc., Chem. Commun. 1985, 1041.
S0022-3263(97)00325-3 CCC: $14.00 © 1997 American Chemical Society

Efficient and Highly Selective Cu(II) Transport
J . Org. Chem., Vol. 62, No. 16, 1997 5593
Ch a r t 1
Sch em e 1
tures of compounds 6-9 are just variations of that of 4
aimed at increasing its lipophilicity: in 6 the N-terminal
amino acid is Phe, in 7 and 8 the C-terminal amino acids
are Leu and Phe, respectively, and in 9 both amino acids
of the dipeptide are Phe. A special case is that of 10 and
11: they are the L,L and L,D diastereomers of the N-octyl-
PheLeu and were conceived to verify the possible effect
of diastereoselectivity on the Cu(II) transport. Moreover,
to compare the effectiveness of these carriers to that of
extractants used in industry, we included in our study
the active component of the industrial extractant Kelex
100,17 i.e, the 7-(4-ethyl-1-methyloctyl)-8-hydroxyquino-
line.
organic phase. From the published studies, also from this
laboratory,12 the transport through liquid membranes is
a more complicated process than it may appear, involving
a subtle interplay of many factors that are not easy to
estimate: thus it has been shown that a ligand that is
very efficient and selective in extracting transition metal
ions from an aqueous phase into an organic phase may
be extremely poor in its transport ability5a,b and vice
versa.5b,c
In this paper, we report our efforts to produce simple
ligands that are highly effective and selective carriers
in the transport of Cu(II) across a bulk chloroform
membrane. We focused our attention on simple dipep-
tides, like glycylglycine, which are known to strongly bind
Cu(II) in aqueous solution to give a neutral complex in
which the amido group is deprotonated14-16 also under
slightly acidic conditions. The dipeptide coordinates with
its amino and amide nitrogens and with a carboxylate
oxygen to three of the four square-planar coordination
sites of the Cu(II) ion as schematically represented in
Chart 1. Deprotonation of the amide group under
slightly acidic or neutral pH only occurs upon coordina-
tion with Cu(II) and not with Ni(II), Zn(II), or Co(II).16
Therefore, dipeptide derivatives are potential candidates
for the selective transport of Cu(II) across a membrane,
and their selectivity may be modulated on changing the
pH of the solution.
Resu lts a n d Discu ssion
Syn th esis of th e Ca r r ier s. Scheme 1 shows the
synthetic routes followed to obtain the ligands. Route
a, involving as a first step the alkylation of the proper
symmetrically substituted amine with the ethyl ester of
bromoacetic acid, was convenient for the synthesis of
1-5, and 7, and 8. Compound 6 was obtained following
route b through the dialkylation of the ethyl ester of
L-phenylalanine with octyl bromide. Compounds 9-11
were prepared via route c, by reacting the Boc-protected
phenylalanine with the phenylalanine or leucine ethyl
ester and, after deprotection, by alkylating the amino
group of the dipeptide with octyl bromide. The yields of
the latter step were very low (<25%) due to the competi-
tive formation of diketopiperazines; however, since the
optical purity of the diastereomeric ligand was relevant,
we preferred this route because of the absence of any
significant racemization (as ensured by the 1H-NMR
analysis of the products).
Tr a n sp or t a n d Rela ted Exp er im en ts. The experi-
mental setup for the transport experiments is shown in
Figure 1. A U-shaped glass apparatus, the inner diam-
eter of the arms being 16.5 mm, was placed in a
thermostatted bath kept at 25 °C. The transport experi-
ments were carried out using the following procedure.
The bulk liquid membrane made of a 1.0 mM solution of
the ligand in 20 mL of CHCl3 separated the two aqueous
phases I and II. Aqueous phase I was a 5.0 mM solution
On these premises, we synthesized and investigated
the dipeptides listed in Chart 1. Compounds 1-5 are
GlyGly derivatives with the amino group alkylated with
a pair of paraffinic chains of increasing (from ethyl to
n-dodecyl) length and, hence, lipophilicity. The struc-
(14) Dobbie, H.; Kermack, W. O. Biochem. J . 1955, 59, 246.
(15) Datta, S. P.; Rabin, B. R. Trans. Faraday Soc. 1956, 52, 1117,
1123.
(16) Sigel, H.; Martin, R. B. Chem. Rev. 1982, 82, 385.
(17) Boumezioud, M.; Tondre, C. Polyhedron 1988, 7, 513.

5594 J . Org. Chem., Vol. 62, No. 16, 1997
Cleij et al.
Ta ble 1. Resu lts of Cu (II) Tr a n sp or t, Extr a ction , a n d
Decom p lexa tion Exp er im en ts Obta in ed w ith Liga n d s
1-11 a t 25 °Ca
Ee
b
c
d
carrier
T24 (%/24 h)
k1 (%/h)
k2 (%/h)
1
2
3
4
5
6
7
8
0
9
0
1
3
8
8
11
12
19
4
11
24
6
38
38
23
18
17
22
17
7
ndf
ndf
24
47
49
53
54
23
9
67
66
39
0.67
0.69
0.70
0.67
0.69
0.73
0.47
0.91
0.94
ndf
9
10
11
7
28
19
ndf
F igu r e 1. Experimental setup used in the transport experi-
Kelex 100
ments. See also the text.
a
b
For conditions, see text. Percent of initial Cu(II) that is
transported after 24 h. c Initial rate of Cu(II) extraction into the
CHCl3 phase expressed as the percentage of ligand that has formed
a complex with Cu(II) in the CHCl3 phase per hour. See also the
d
text. Initial rate of release of Cu(II) from CHCl3 into phase II
in the decomplexation experiments. The numbers give the per-
centage of the total amount of Cu(II) that is released per hour.
See also the text. e The fraction of ligand in the CHCl3 phase that
has formed a complex with Cu(II) when equilibrated against phase
I. See also the text. f Not determined.
ion was measured using preformed solutions of their
copper complex in chloroform. In each case, the metal
ion concentrations were determined by means of atomic
absorption spectroscopic analyses.
Tr a n sp or t of Cu (II) Ion s. Preliminary extraction
tests, using the standard aqueous phase I and the
chloroform solution, indicated that ligands 1 and 2 and
their Cu(II) complexes partition quite favorably in water,
3 is an intermediate case (interestingly the free ligand
is more water soluble than its complex which almost
entirely resides in the CHCl3 phase), and all the other
ligands 4-11 and their complexes are virtually absent
in the aqueous phase after equilibration.
F igu r e 2. Percent of initial Cu(II) in aqueous phase I (curve
a), aqueous phase II (curve b), and CHCl3 phase (curve c)
during a transport experiment with 4. See also the text and
for conditions Table 1.
(6.0 mL) of M(NO3)2 (M being Cu2+, Zn2+, or Ni2+) in 0.2
M MES buffer at pH ) 5.6,18 and aqueous phase II was
a 0.1 M HCl solution (6.0 mL). All three phases were
stirred at such a rate as to ensure efficacy and smooth-
ness at the interfaces. Under these conditions, the
carrier transports the M2+ ions from aqueous phase I to
II (where it is protonated) and protons from aqueous
phase II to I. Thus, the process is driven not only by a
difference in M2+ concentration but also by a difference
in proton concentration: as a consequence, more than
50% of the M2+ ions can be transported and, as long as
the pH difference of the aqueous phases is maintained,
all the ions can be translocated (see infra and Figure 2).
Then we carried out a systematic analysis of the
transport of Cu(II) using carriers 1-11 and Kelex 100.
The main results are shown in Table 1 where the data
are reported in a rather straightforward fashion in the
following terms. (a) T24 is the percentage of the total
amount of Cu ions transported in 24 h from aqueous
phase I to phase II. The T24 values range from 0 for 1
up to 67 for 10, and it is here emphasized that many of
the new ligands are considerably more efficient in the
transport of Cu(II) than Kelex 100 (T24 ) 39). (b) k1, here
referred to as the initial rate of uptake, indicates the
percentage of ligand that forms a complex and transfers
the metal ion from the donor phase (I) into the CHCl3
phase per hour at the beginning of the transport experi-
ment. These rates range from 0 for 1 to 24 for 11. (c)
k2, here referred to as the initial rate of release and
measured in the associated experiments described above
as (ii), is the fraction of metal ion initially contained in
the CHCl3 which is released per hour into the acidic
aqueous phase II. The rates of release change from 7
for ligands 8 and 9 to 38 for 1 and 2. (d) E, the extraction
efficiency value, is the ratio of the ligand concentration
and that of the Cu(II) that is extracted from aqueous
phase I (pH ) 5.6) into the CHCl3 phase and measured
in the associated experiment described above as (i). The
E values range from 0.47 for 9 to 0.94 for 11.
Associated to the transport experiments we carried out
simple extraction experiments and in particular two
sequential runs, referred to as (i) and (ii), using simple
glass test tubes of inner diameter 16.5 mm kept at 25
°C. (i) A 1 mM solution of ligand in CHCl3 (20 mL) was
equilibrated against aqueous donor phase I (6 mL), and
the amount of Cu(II) extracted in CHCl3 was determined.
(ii) The thus obtained CHCl3 solutions of metal complex
were then exposed to the receiving acid aqueous phase
II (6 mL, in the absence of phase I), both phases were
stirred, and the rate (see below) of release of metal ion
into the aqueous phase was measured. Since ligands 1
and 2 are too hydrophilic, the rate of release of metal
(18) A reviewer suggested that the actual transported species could
be Cu(OH)+. However at pH ) 5.6, which is the highest pH value used
in the present work, copper ion is present approximately 94% as Cu2+
and only 6% as Cu(OH)+ (see Basolo, F.; Pearson, R. G. Mechanism of
Inorganic Reaction; J . Wiley and Sons: New York, 1967). In view of
that, we omitted the presence of the latter species when drawing Figure
1. Moreover, studies with similar ligands16 have shown that the most
stable complex formed is neutral and involves Cu2+ and not Cu(OH)+.
Figure 2 shows the changes of the amount of Cu(II) in
the three phases in a transport experiment (note that
the ion is fully translocated at the end of the experiment,
after ca. 50 h) using 4 as a carrier, and this diagram can

Efficient and Highly Selective Cu(II) Transport
J . Org. Chem., Vol. 62, No. 16, 1997 5595
protonation of the complex in the receiving phase is the
rate-limiting step in the transport process. On the other
hand, when the pH of the receiving phase was varied
between 1 and 2, no relevant changes in the k2 values
were observed thus indicating that, in this pH range, the
transfer of the chelating part of the complex to the
receiving phase (or hydration) becomes rate determining
for the release of Cu(II). On the whole, in the standard
conditions, both semiprocesses of the transport, uptake
of Cu(II) at the donating phase and release at the
receiving phase, appear to be controlled by the transfer
of the complex between the interfaces of the different
phases, the uptake of the metal ion being slightly slower
than its release. The finding that the transport is rate-
limited by diffusion across the membrane interfaces is
in agreement with what was already reported in the case
of alkali3 and also Cu(II)10 ions.
F igu r e 3. Extraction efficiency of Cu(II) into the CHCl3 phase
by 4 at different pH values in the absence (O) and presence
(b) of 0.2 M MES buffer. See also the text.
Interestingly, in another set of experiments employing
ligand 4, the rate of extraction of Cu(II) from aqueous
phase I to CHCl3 was found to be first order in ligand
concentration at least up to a concentration of 10 mM,
10 times larger than that used under the standard
conditions (see Table 1). Also the rate of release of Cu(II)
from CHCl3 into aqueous phase II is first order in
complex concentration in the CHCl3 phase at least up to
10 mM. These observations indicate that both the free
ligand and the complex are not yet saturated at the
CHCl3-water interfaces also with a carrier concentration
at least 10 times larger than those used so that, by
increasing the ligand concentrations, a very high Cu(II)
flux can be achieved. The transport efficiency depends
also on the nature of the bulk membrane. Using CCl4
the T24 value for 4 was reduced to 23 as compared to 47
with CHCl3. The presence of 10% (v/v) 1-octanol in
CHCl3 also resulted in a less efficient transport for 4 (T24
) 35).
The above results and in particular the analysis of the
data of Table 1 allow the following general considerations
concerning the transport of Cu(II) ions using 1-11. In
the case of the dialkylated carriers 1-7, the rate of
extraction k1 increases as the lipophilicity increases, and
vice versa, the increase of lipophilicity results in a
decrease in the rate of release k2. As a consequence of
the balance of the lipophilicity/hydrophilicity effects, the
most effective carriers are ligands 6 and 7 with a T24
value of ca. 54. Considering the diglycine ligands, the
effect of the increase in size of the alkyl chains from two
ethyl (1) up to two dodecyl (5) groups indicates that a
plateau value for T24, k1 and k2 is reached when two octyl
chains are attached to the N-terminus as in 4 and further
elongation to two dodecyl residues is not relevant. On
the other hand, as pointed out above, when the chains
are shorter than two octyl residues, the carrier and its
complex are soluble in water so that the formation and
the decomposition of the complex occur not only at the
interface but also in solution thus impairing the extrac-
tion on the one hand (k1 ) 0 for 1) and highly favoring
the release on the other hand (k2 ) 38 for 1 and 2 and
can be taken as the upper limit for this type of carrier).
In the case of ligands 4 and 5 the processes of the
extraction and release of Cu(II) can be assumed to be
virtually confined to the CHCl3-water interfaces, and the
k1 and k2 values seem to stabilize around 8 and 17,
respectively. The results obtained using the N-dialkyl-
ated ligands 6-9, when compared to their structurally
related parent compound 4, indicate that the increase in
apparent hydrophobicity brings about unexpected ef-
be taken as typical also for carriers 5-11. In the first
few hours, the amount of Cu(II) ions in phase I drops
rapidly and likewise increases in the CHCl3 phase so that
little transport occurs. Then, a steady state situation is
reached in which the amount of Cu(II) in the CHCl3
phase changes very little whereas the decrease in the
donating phase and the increase in the receiving phase
are constant and equal. Finally, when all the Cu(II) is
extracted from aqueous phase I, the concentration of the
metal ions in the CHCl3 phase drops and the rate of
release into aqueous phase II slows down. This plot
shows that the rate of extraction from the donor aqueous
phase I is zeroth order in Cu(II) concentration in phase
I. Even at a very low Cu(II) concentration the rate of
extraction remains constant. This implies that at these
Cu(II) concentrations the rate of extraction of Cu(II) does
not depend on the rate of complex formation at the
aqueous phase I-CHCl3 interface and that the amount
of complex at this interface is constant. Thus by all
evidence, rate determining for the process of metal ion
extraction from aqueous phase I to chloroform is the
transfer (or dehydration) of the hydrophilic part of the
amphiphilic complex through the interface of the two
phases.
Further experiments were then carried out in order to
complement the results shown in Table 1 and to gain
insight into the system investigated. We measured the
E values in the case of 4 using buffered (MES, 0.2 M)
and unbuffered donating aqueous solutions of Cu(NO3)2
(5 mM) of different pHs. The results are illustrated in
Figure 3 and indicate a quite substantial effect of the
buffer. Assuming that the effective chelating agent is
the ligand in its deprotonated form, the curves of Figure
3 appear as titration curves, and one may estimate that
in the absence of buffer the pKa of the amido group is
4.5 and that in the presence of the buffer the apparent
pKa increases to 5.4. Whatever the reasons for such an
effect, this means that in the presence of 0.2 M MES, as
in the standard transport experiments (Table 1), the
extraction efficiency (E ) 0.69) is not as large as in the
absence of buffer. We also noticed that the T24 and k1
values increased when the buffer concentration was
reduced to 0.1 M. However, we preferred to use the
higher buffer concentration since this guarantees a stable
pH and a steady efficiency of the system for at least 3
days. As a matter of fact, the pH gradient is of para-
mount importance: we observed that if the receiving
phase was a 0.2 M aqueous solution of MES buffer of pH
) 5.6 (just as the donating phase) the k2 value for 4 drops
from 18 to 0.2, indicating that at this pH value the

5596 J . Org. Chem., Vol. 62, No. 16, 1997
Cleij et al.
fects: in particular it appears that the presence of a
phenyl group as the R4 substituent induces a sharp
decrease in k2 and hence in T24. Such an adverse effect
is magnified by the presence of two phenyl groups in 9:
in this case both k1 and k2 are low indicating a small
extraction affinity of the complex (the E value is the
lowest in the series) for the CHCl3 phase. We are at odds
in explaining such an effect. Although it may be the
result of a slightly higher pKa of the peptide group due
to substituent effects, the small E value may be related
to the fact that in the complex with 9 the two very bulky
phenyl groups of the dipeptide face each other, being
located on the same side of the coordination plane. We
argue that the interaction between the aromatic group
may lead to steric repulsion in the CHCl3 phase and
retard the extraction into the bulk membrane. That
stereochemical factors are at play in the extraction and
transport of Cu(II) is indicated by the results obtained
in the case of single-tailed diastereomeric ligands 10 and
11. These appear the most effective carriers of the series
in terms of the T24 values (66 and 67) due to the optimal
balance between the rate of extraction and that of release.
Their case deserves a further look.
a
b
Cu (II) Tr a n sp or t by Dia ster eom er ic Ca r r ier s.
The extraction efficiency (E) at pH ) 5.6 of the diaster-
eomeric carriers is remarkably larger than that of the
N-dialkylated dipeptides, and this may be due to the
lower pKa of the amido group. Although the T24 values
are identical, the k1 and k2 values are different: the L,D
stereoisomer 11 is much faster in extracting Cu(II) from
aqueous phase I than the L,L stereoisomer 10; on the
other hand, 11 is much slower in the release of Cu(II)
into aqueous phase II than 10. Thus the differences in
k1 and k2 almost completely compensate, at least after
24 h. A picture of the transport course is shown in Figure
4a that reports for both diastereomers the amount of
Cu(II) present in aqueous phases I and II as a function
of time. In the case of 11, due to the high extraction rate,
the amount of metal ions in the donating phase rapidly
decreases and is always lower than in the case of 10. On
the other side, in the receiving phase II, the amount of
Cu(II) is almost similar for both carriers and the slope
in the steady state regime is slightly greater for 11 than
for 10. The diastereoselectivity effect is clearly il-
lustrated in Figure 4b where the amount of Cu(II) in the
CHCl3 phase is plotted as a function of time. Using 11,
up to 33% of the ion resides in the CHCl3 phase, whereas
no more than 13% is present with 10.
The diastereomeric effect may be explained as follows.
It has been reported that in water the Cu(II) complexes
of lipophilic L,L dipeptides are more stable than the
complexes of the diastereomeric L,D dipeptides.16,19 In the
first case, assuming the formation of square-planar
complexes, the lipophilic side chains of the amino acid
residues are located on the same side of the coordination
plane and close to each other; in the second case, with
the L,D diastereomer, these side chains are located on
opposite sides of the coordination plane. The larger
stability in water of the L,L complexes is attributed to
the close proximity and hence to the positive (micellar-
like) hydrophobic interaction. Thus apparently, the
hydration mantle of the L,D complex is less stable, and
its complex is relatively more lipophilic than the L,L
complex. This may account for the higher k1 values
observed for the L,D stereoisomer 11 since the extraction
F igu r e 4. (a) Percent of initial Cu(II) in aqueous phases I
and II during the transport experiments with carriers 10 (O)
and 11 (b). (b) Percent of initial Cu(II) in the CHCl3 phase
(symbols as in panel a).
of Cu(II) from phase I into the CHCl3 phase is driven by
lipophilicity and for the smaller k2 rate since the release
of Cu(II) into aqueous phase II is retarded by a higher
lipophilicity of the complex.
Selective Tr a n sp or t of Cu (II). We investigated the
transport efficiency of Zn(II) and Ni(II) using carrier 4
in the absence and presence of Cu(II). Figure 5a shows
that in the absence of Cu(II), the translocation of Zn(II)
occurs at a moderate extent, the T24 value being 2.5 as
compared with the T24 for Cu(II) of 47 (see Table 1), and
the rate remains constant for 3 days. Thus, under these
conditions the selectivity in terms of the ratio T24(Cu)/
T24(Zn) is 19. We did not observe any accumulation of
Zn(II) in the CHCl3 phase which means that k2 is much
larger than k1. Under which form Zn(II) is present in
the CHCl3 membrane is only a matter of speculation:
perhaps, the carboxylate groups of two ligand molecules
form an ion-pair with the metal ion. In the presence of
Cu(II), the transport of Zn(II) is almost totally suppressed
whereas the transport of Cu(II) is essentially unaffected
(cf. Figures 5a and 3). The amount of Zn(II) that is
transported after 24 h amounts to only 0.04%, and this
gives a selectivity factor T24(Cu)/T24(Zn) ) 1200. The
transport of Zn(II) remains inhibited as long as even very
low amounts of Cu(II) are present in the donating
phase: after 40 h, when there is only 1% Cu(II) left in
aqueous phase I, 0.08% Zn(II) has been transported as
compared to 88% for Cu(II). Only when the donating
phase is almost depleted of the latter ion does the rate
of transport of Zn(II) start to increase.
A very similar behavior was observed for the transport
of Ni(II) in the absence and presence of Cu(II) as shown
in Figure 5b, but the selectivities are much higher. In
the absence of Cu(II), T24(Ni) amounts to 0.10% and the
ratio T24(Cu)/T24(Ni) is 460. In the presence of Cu(II),
T24(Ni) is <0.01% (we could not detect any Ni(II) in the
(19) Nakon, R.; Angelici, R. J . J . Am. Chem. Soc. 1974, 96, 4178.

Efficient and Highly Selective Cu(II) Transport
J . Org. Chem., Vol. 62, No. 16, 1997 5597
uptake and disfavors the release, this being kinetically
rather than thermodynamically controlled. Such a point
is illustrated by the limiting case of the two diastereo-
meric carriers 10 and 11 for which there are no argu-
ments to indicate any difference in lipophilicity and hence
in the partitioning between the aqueous and chloroform
phases (as a matter of fact, the E values of Table 1 are
the same); yet, their complexes are taken up and released
at different rates due likely to a subtle difference in the
lipophilicity of the complexes and the consequent ease
of uptake in and release from the organic phase. What
we learned also from this study is that the chemical
changes at the interface play a fundamental role in the
transport of the metal ion. In the present case, the
carrier facing the donor phase with its hydrophilic portion
may easily form a neutral lipophilic complex with Cu(II)
that is transferred to the bulk chloroform and from this
to the donor phase where it undergoes protonation thus
releasing the metal ion and a neutral free carrier which
turns over. In the case of Zn(II) or Ni(II) complexation
with involvement of the amide moiety at the pH used is
unlikely, and the ion can be taken out into the organic
phase only as an ionic system at a rate which cannot
compete with that of the Cu(II) complex. This point
deserves further attention, and studies are under way
in this laboratory to further investigate the selectivity
also toward other metal ions and to exploit the effect of
the pH of the donating phase to control the transport of
different ions by modulating the acidity of the source and
receiving phases.
a
b
F igu r e 5. (a) Percent of initial Zn(II) transported by 4 in the
absence (O) and presence (b) of Cu(II). (b) Percent of initial
Ni(II) transported by 4 in the absence (O) and presence (b) of
Cu(II). Conditions are as described for the transport experi-
ments in Table 1; the initial concentration of Cu(NO3)2,
Zn(NO3)2, and Ni(NO3)2 in phase I was 5.0 × 10-3 M.
Exp er im en ta l Section
receiving phase II) which gives a selectivity T24(Cu)/
T24(Ni) > 5000. Only after 48 h when 97% Cu(II) was
translocated could we detect the presence of 0.01% Ni(II)
in phase II, and this allows to estimate a selectivity ratio
T48(Cu)/T48(Ni) ) 104. After 48 h, when all the Cu(II)
has disappeared from aqueous phase I, the transport rate
of Ni(II) increases as in the absence of the other ion.
Thus carrier 4 under the conditions used is extremely
selective in the transport of Cu(II) from a mixture of other
transition metal ions such as Zn(II) or Ni(II) also when
it is present in relatively small amounts. Quite likely,
the observed selectivity is due to a substantial change
in the mode of transport: through a neutral, relatively
stable complex in the case of Cu(II) and through saltlike
ionic species in the case of the other transition metal ions,
as long as the aqueous donating phase is a neutral or
slightly acidic solution.
Gen er a l Meth od s a n d Ma ter ia ls. Melting points are
uncorrected. 1H-NMR spectra were recorded on a 200 MHz
spectrometer, and chemical shifts in ppm are reported relative
to internal Me4Si. Microanalyses were performed by the
Laboratorio di Microanalisi of our department. Cu(NO3)2,
Zn(NO3)2, and Ni(NO3)2 were analytical-grade products. Metal
ion stocks solutions were titrated against EDTA following
standard procedures.20 The 2-morpholinoethanesulfonic acid
buffer (MES) was used as supplied by Fluka. Kelex 100 was
a gift of Dr. C. Tondre, Laboratoire d'Etude des Solutions
Organiques et Colloidales (LESOC), UACNRS 406, Universite
de Nancy I, B.P. 239, 54506 Vandoeuvre les Nancy, France.
The active component 7-(4-ethyl-1-methyloctyl)-8-hydroxyquin-
oline was purified according to a method described in the
literature.17
Gen er a l P r oced u r e for th e Syn th esis of Ca r r ier s 1-8.
Rou te a : Sch em e 1, Used for Ca r r ier s 1- 5, 7, a n d 8. The
proper dialkylamine (16.6 mmol) (diethylamine, dibutylamine,
dihexylamine, dioctylamine, or didodecylamine) and 3.5 g (21
mmol) of bromoacetic acid ethyl ester were dissolved in 10 mL
of ethanol and heated for 8 h at 60 °C. The reaction mixture
was then concentrated in vacuo and the residue dissolved in
CHCl3 and washed with a Na2CO3 solution. The CHCl3 layer
was dried over Na2SO4 and concentrated in vacuo. The residue
was purified by column chromatography (silica gel, CHCl3).
The yield amounted to ca. 65%. The thus obtained dialkyl-
glycine ethyl ester was hydrolyzed with NaOH in ethanol/
water (5:1, v/v). Subsequently, 25 mL of water was added,
and the pH of the solution was adjusted to pH ) 6.0 with HCl.
The solution was concentrated in vacuo, and the dry residue
was extracted with ethanol/CHCl3 ) 1:1 (v/v). A solid material
was removed by filtration, and the organic solution was
concentrated in vacuo giving the different dialkylglycine in an
almost quantitative yield. The products were used without
further purification.
Con clu sion s
The long-chained dipeptides 4-11 are very effective and
highly selective carriers of Cu(II) through a chloroform
membrane. They are an interesting breed of compounds,
structurally simple and relatively easy to modify. Al-
though a quantitative analysis similar to that developed
for crown ethers is not yet possible since it requires the
determination of not easily accessible parameters such
as the formation constants of the lipophilic carrier-metal
ion complexes, the present study brought to light some
interesting features which may allow to better under-
stand their functions and the design of proper carrier
molecules. The most delicate factor is the proper balance
of the lipophilicity/hydrophilicity features of the carrier
and its complexes: the lipophilicity favors the metal ion
(20) Holzbecher, Z. Handbook of Organic Reagents in Inorganic
Analysis; Wiley: Chichester, 1976.

5598 J . Org. Chem., Vol. 62, No. 16, 1997
Cleij et al.
N,N-Diet h ylglycin e: 1H-NMR (CD3OD) δ 1.25 (t, 6 H),
for C27H46N2O3: C, 72.60; H, 10.38; N, 6.27. Found: C, 72.51;
H, 10.42; N, 6.22.
3.18 (q, 4 H), 3.55 (s, 2 H). This is a hygroscopic compound.
N,N-Dioctylglycyl-L-leu cin e (7): 1H-NMR (CD3OD) δ 0.9
(m, 12 H), 1.3 (m, 20 H), 1.6 (m, 7 H), 2.9 (t, 4 H), 3.65 (AB
N,N-Dibu tylglycin e: 1H-NMR (CD3OD) δ 0.85 (t, 6 H), 1.3
(m, 4 H), 1.65 (m, 4 H), 3.05 (t, 4 H), 3.55 (s, 2 H); mp ) 118
quartet), 4.4 (m, 1 H); [R]20 ) +6.3 (c ) 1.0, MeOH). Anal.
°C.
D
Calcd for C24H48N2O3: C, 69.86; H, 11.72; N, 6.79. Found: C,
69.54; H, 11.81; N, 6.72.
N,N-Dih exylglycin e, N,N-Dioctylglycin e, a n d N,N-Di-
d od ecylglycin e: The 1H-NMR spectra closely resemble the
spectrum of N,N-dibutylglycine with the exception of the signal
at 1.3 ppm which has an intensity of 12 H, 20 H, and 36 H,
respectively. The melting points are 81, 74, and 101 °C
respectively.
N,N-Dioctylglycyl-L-ph en ylalan in e (8): 1H-NMR (CD3OD)
δ 0.9 (t, 6 H), 1.3 (m, 20 H), 1.55 (m, 4 H), 2.85 (t, 4 H), 3.2 (m,
2 H), 3.6 (AB quartet, 2 H), 4.6 (m, 1 H), 7.25 (m, 5 H), [R]20
D
) +11.8 (c ) 1.0, MeOH). Anal. Calcd for C27H46N2O3: C,
72.60; H, 10.38; N, 6.27. Found: C, 72.43; H, 10.46; N, 6.19.
Gen er a l P r oced u r e for th e Syn th esis of Ca r r ier s 9-11.
Rou te c: Sch em e . To 1.5 g (5.7 mmol) of Boc-L-Phe-OH and
6.0 mmol of NEt3 in 20 mL of CHCl3 was added 5.7 mmol of
isobutyl chloroformate at 0 °C. After stirring for 3 min, a well-
stirred solution of 7.0 mmol of the proper amino acid ester
hydrochloride (L-phenylalanine ethyl ester hydrochloride, L-
or D-leucine methyl ester hydrochloride) and 6.0 mmol NEt3
in 10 mL of CHCl3 was added at 0 °C. The ice bath was
removed, and the solution was stirred for 3 h at 25 °C. The
solution was concentrated in vacuo, and the residue was
dissolved in 25 mL of CHCl3, washed first with a dilute HCl
solution and then with a Na2CO3 solution, dried over Na2SO4,
and concentrated in vacuo. The solid residue was dissolved
in 10 mL of trifluoroacetic acid and stirred for 3 h. The
solution was concentrated in vacuo, and to the remaining oil
was added 50 mL of diethyl ether while stirring vigorously.
The resulting precipitate was collected by filtration, washed
with ether, and dried in vacuo. The thus obtained trifluoro-
acetate salts of the dipeptide esters were used without further
purification. The overall yield amounted to 60%.
H-L-P h e-L-P h e-OEt‚CF 3COOH: 1H-NMR (CDCl3) δ 1.15
(t, 3 H), 3.1 (d, 2 H), 3.4 (m, 2 H), 4.05 (q, 2 H), 4.58 (m, 1 H),
4.73 (m, 1 H), 7.25 (m, 10 H), 7.48 (d, 1 H), 8.6 (br s, 3 H).
H-L-P h e-L-Leu -OMe‚CF 3COOH: 1H-NMR (CD3OD) δ 0.9
(2d, 6 H), 1.7 (m, 3 H), 3.2 (m, 2 H), 3.75 (s, 3 H), 4.15 (2d, 1
H), 4.55 (t, 1 H), 7.35 (m, 5 H).
H-L-P h e-D-Leu -OMe‚CF 3COOH: 1H-NMR (CD3OD) δ 0.85
(2d, 6 H), 1.2 (m, 1 H), 1.5 (m, 2 H), 3.1 (m, 2 H), 3.72 (s, 3 H),
4.15 (t, 1 H), 4.4 (2d, 1 H), 7.35 (m, 5 H).
Rou te b: Sch em e 1, Used for Ca r r ier 6. To a vigorously
stirred solution of 3 g of L-Phe-OEt‚HCl in 50 mL of water
and 50 mL of CHCl3 was added a concentrated Na2CO3
solution. The CHCl3 layer was separated, dried over Na2SO4,
and concentrated in vacuo. To 2.0 g (10.4 mmol) of the thus
obtained H-L-Phe-OEt dissolved in 10 mL of ethanol was added
6.0 g (31 mmol) of 1-bromooctane, and the reaction mixture
was heated for 12 h at 75 °C. Subsequently 50 mL of CHCl3
was added, and the solution was washed with a Na2CO3
solution. The organic layer was separated, dried over Na2SO4
and concentrated in vacuo. The residue was dissolved in 10
mL of ethanol and heated for 12 h. This procedure was
repeated three more times. The crude product thus obtained
was purified by column chromatography (silica gel, CHCl3/
hexane ) 1:3). The N,N-dioctyl-L-phenylalanine ethyl ester
was hydrolyzed as described above. The final yield amounted
to 38%: 1H NMR (CD3OD) δ 0.9 (t, 6 H), 1.3 (m, 20 H), 1.65
(m, 4 H), 2.95 and 3.15 (2 m, 4 H), 3.3 (AB quartet, 2 H), 4.0
(t, 1 H), 7.3 (m, 5 H); mp ) 74 °C; [R]20D ) +1 (c ) 1.0, MeOH).
To 3.0 mmol of the thus obtained dialkyl amino acids and
6.0 mmol of NEt3 in 20 mL of CHCl3 was added 1 equiv of
isobutyl chloroformate at 0 °C. After stirring for 3 min, a well-
stirred solution of 4.5 mmol of the proper amino acid ethyl
ester hydrochloride (glycine, L-leucine, or L-phenylalanine ethyl
ester hydrochloride) and 5 mmol of NEt3 in 15 mL of CHCl3
was added at 0 °C. The ice bath was removed, and the reaction
mixture was stirred for 3 h at room temperature. The reaction
mixture was concentrated in vacuo and the residue dissolved
in 40 mL of CHCl3. This solution was washed with Na2CO3,
dried over Na2SO4, and concentrated in vacuo. The residue
was purified by column chromatography (silica gel, CHCl3).
The thus obtained N,N-dialkyl dipeptide ethyl esters were
hydrolyzed with NaOH in 25 mL of ethanol/water ) 10:1. To
this solution was added 50 mL of water and the pH was
adjusted to 6 with HCl. Compounds 6-8 precipitated from
solution at pH ) 6 and were collected by filtration. In the
case of compounds 1-5 the solution was concentrated in vacuo
and the residue dissolved in CHCl3. The solid material was
removed by filtration, and the CHCl3 solution was concen-
trated in vacuo to yield the pure products. The overall yield
amounted to 80%. Compound 1-5, 7, and 8 are very viscous
oils; only 6 is a white solid.
N,N-Dieth ylglycylglycin e (1): 1H-NMR (CD3OD) δ 1.3 (t,
6 H), 3.2 (q, 4 H), 3.82 and 3.87 (2s, 4 H). Anal. Calcd for
C8H16N2O3: C, 51.05; H, 8.57; N, 14.88. Found: C, 50.91; H,
8.69; N, 14.43.
N,N-Dibu tylglycylglycin e (2): 1H-NMR (CD3OD) δ 0.9 (t,
6 H), 1.3 (m, 4 H), 1.7 (m, 4 H), 2.87 (t, 4 H), 3.6 and 3.75 (2s,
4 H). Anal. Calcd for C12H24N2O3: C, 58.99; H, 9.90; N, 11.47.
Found: C, 58.33; H, 10.00; N, 11.21.
N,N-Dih exylglycylglycin e (3), N,N-Dioctylglycylgly-
cin e (4), a n d N,N-Did od ecylglycylglycin e (5). The 1H-
NMR spectra of these compounds closely resemble the spec-
trum of 4 with exception of the signal at 1.3 ppm which has
an intensity of 12 H, 20 H, and 36 H, respectively. Anal.
Calcd for C16H32N2O3: C, 63.96; H, 10.74; N, 9.32. Found: C,
64.09; H, 10.61; N, 9.39. Anal. Calcd for C20H40N2O3: C,
67.37; H, 11.31; N, 7.86. Found: C, 67.14; H, 11.48; N, 7.79.
Anal. Calcd for C28H56N2O3: C, 71.74; H, 12.04; N, 5.98.
Found: C, 71.34; H, 12.11; N, 5.90.
The thus obtained dipeptides were dissolved in 30 mL of
water and 30 mL of CHCl3. To this vigorously stirred mixture
was added a Na2CO3 solution. The CHCl3 layer was separated,
dried over Na2SO4, and concentrated in vacuo; 1.0 g of the
residue and 2 mL of 1-bromooctane were dissolved in 4.0 mL
of ethanol and heated for 12 h at 70 °C. The reaction mixture
was concentrated in vacuo; the residue was dissolved in 20
mL of CHCl3, washed with Na2CO3, dried over Na2SO4, and
concentrated in vacuo. The residue was purified by column
chromatography (silica gel, CH2Cl2). In the case of the phe-
nylalanylphenylalanine dipeptide, the crude product was again
dissolved in 4.0 mL of ethanol and heated for a further 12 h
at 70 °C. This procedure was repeated three more times. The
purification was performed by column chromatography (silica
gel, CH2Cl2/petroleum ether ) 1:1). The thus obtained dipep-
tide esters were hydrolyzed with NaOH in 20 mL of ethanol/
water (9:1); 40 mL of water was added, and the pH was
adjusted to 6 with a 1.0 M HCl solution. For 10 and 11 the
precipitate was collected by filtration, washed with water, and
dried in vacuo. In the case of 9 the resulting emulsion was
extracted with 25 mL of CHCl3. The CHCl3 layer was sepa-
rated and concentrated in vacuo leaving 9 as a viscous oil.
N,N-Dioct yl-L-p h en yla la n yl-L-p h en yla la n in e (9): 1H-
NMR (CD3OD) δ 0.9 (t, 6 H), 1.3 (m, 24 H), 2.55 (m, 4 H), 3.0
(d, 2 H), 3.15 (m, 2 H), 3.78 (t, 1 H), 4.5 (dd, 1 H), 7.2 (m, 10
H), yield ) 13%; [R]20
) +2.8 (c ) 1.0, MeOH). Anal. Calcd
D
for C34H52N2O3: C, 76.08; H, 9.76; N, 5.22. Found: C, 75.92;
H, 9.81; N, 5.15.
N-Octyl-L-ph en ylalan yl-L-leu cin e (10): 1H-NMR (CD3OD)
δ 0.95 (m, 9 H), 1.3 (m, 10 H), 1.65 (m, 5 H), 2.8 (m, 2 H), 3.2
(AB quartet, 2 H), 3.95 (t, 1 H), 4.38 (m, 1 H), 7.35 (m, 5 H),
yield ) 14%; mp ) 204 °C; [R]20D ) -9 (c ) 0.3, MeOH). Anal.
Calcd for C23H38N2O3: C, 70.73; H, 9.81; N, 7.17. Found: C,
70.51; H, 9.88; N, 7.20.
N,N-Dioctyl-L-ph en ylalan ylglycin e (6): 1H-NMR (CD3OD)
δ 0.9 (t, 6 H), 1.3 (m, 20 H), 1.65 (m, 4 H), 3.05 (m, 4 H), 3.15
(d , 2 H), 3.65 (AB quartet, 2 H), 3.93 (t, 1 H), 7.25 (m, 5 H),
mp ) 105 °C; [R]20 ) +30.6 (c ) 1.0, MeOH). Anal. Calcd
D

Efficient and Highly Selective Cu(II) Transport
J . Org. Chem., Vol. 62, No. 16, 1997 5599
N-Octyl-L-ph en ylalan yl-D-leu cin e (11): 1H-NMR (CD3OD)
δ 0.75 (2d, 6 H), 0.85 (m, 1 H), 0.95 (t, 3 H), 1.4 (m, 12 H), 1.7
(m, 2 H), 3.05 (t, 2 H), 3.15 (AB quartet, 2 H), 3.95 (dd, 1 H),
the analysis. The micromoles of metal ion were determined
with a Perkin Elmer 360 atomic absorption instrument using
a multielement lamp.
4.15 (dd, 1 H), 7.3 (m, 5 H); yield ) 25%; mp ) 210 °C; [R]20
D
Ack n ow led gm en t. The authors thank Mr. E. Cas-
tiglione for technical assistance. Financial support by
the Ministry of the University and Scientific and Tech-
nological Research (MURST) and by the European
Economic Community for Dr. M. C. Cleij’s fellowship
under the network Science Programme, Contract SC1*-
CT92-0764, is gratefully acknowledged.
) +79 (c ) 0.4, MeOH). Anal. Calcd for C23H38N2O3: C, 70.73;
H, 9.81; N, 7.17. Found: C, 70.61; H, 9.84; N, 7.21.
Tr a n sp or t Exp er im en ts. These experiments have been
performed using the apparatus and solutions described in the
previous section. The water phases were mechanically stirred
with two rotating Teflon bars, while the chloroform solution
was magnetically stirred at 400 rpm. At time intervals 100
µL portions of source and receiving phase were withdrawn,
diluted with 1.9 mL of water, and subsequently subjected to
J O9703257