Improving the gelation properties of 3,5-diaminobenzoate-based organogelators in aromatic solvents with additional aromatic-containing pendants

Article abstract of DOI:10.1016/j.tet.2006.10.066

A family of 3,5-diaminobenzoate derivatives containing different Cbz-protected α- or β-amino side pendant chains and different aromatic-containing ester functionalities was prepared. It was found that the additional aromatic rings in the Cbz- and aromatic

Full text of DOI:10.1016/j.tet.2006.10.066

Tetrahedron 63 (2007) 363–373
Improving the gelation properties of 3,5-diaminobenzoate-based
organogelators in aromatic solvents with additional
aromatic-containing pendants
a
a
a
b
Hak-Fun Chow,a, Jie Zhang, Chui-Man Lo, Siu-Yin Cheung and Ka-Wai Wong
*
aDepartment of Chemistry andThe CenterofNovel FunctionalMolecules, The Chinese University of Hong Kong, Shatin, NT, Hong Kong
bDepartment of Physics, The Chinese University of Hong Kong, Shatin, NT, Hong Kong
Received 25 August 2006; revised 12 October 2006; accepted 25 October 2006
Available online 13 November 2006
Abstract—A family of 3,5-diaminobenzoate derivatives containing different Cbz-protected a- or b-amino side pendant chains and different
aromatic-containing ester functionalities was prepared. It was found that the additional aromatic rings in the Cbz- and aromatic-containing
ester moieties significantly improved the gelation properties of the resulting organogelators in aromatic solvents. Infrared and circular dichro-
ism spectroscopy revealed that both H-bonding and p–p aromatic stacking interactions were the main driving forces for gelation.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
other hand, if the binding forces among the organogelators
are weak, dissolution or weak gelation would result. Among
the various driving forces, ionic, hydrogen bonding, p–p
stacking, and hydrophobic interactions are the most com-
monly encountered ones. However, the latter two are some-
times too subtle to be characterized experimentally.
Low molecular weight organogelators are an interesting
class of molecules that possess many applications in medi-
cine, environmental science, separation technology, and
catalysis.1 These small molecules are capable of forming
a non-covalently linked three-dimensional network structure
with a specific solvent system that result in the immobiliza-
tion of solvent molecules.
Our group recently reported a series of a-amino acid-based
dendrons 1 and 2 and showed that they are powerful organo-
gelators with minimum gelation concentration (MGC) as
low as 2 mg/mL.2 We also demonstrated for the first time
that their gelation strength was strongly dependent on the na-
ture of the focal point functionality (i.e., R1) and the amino
acid side chain residues (i.e., R2–R7). In our previous study,
A key to the design of organogelators is how to fine-tune the
strength of the inter-molecular driving forces that hold them
together. If the binding interactions are too strong, precipita-
tion or crystallization, rather than gelation will occur. On the
BocHN
NHBoc
O
R5
R6
BocHN
O
NHBoc
R7
O
O
NH
HN
R4
HN
NH
NH
HN
R3
BocHN
R2
NHBoc
R3
O
O
O
O
O
O
R2
HN
NH
HN
NH
CO2R1
CO2R1
1
2
Keywords: Amino acids; Organogelators; H-bonding; p–p Aromatic stacking.
* Corresponding author. Tel.: +852 26096341; fax: +852 26035057; e-mail: hfchow@cuhk.edu.hk
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.10.066

364
H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
we focused our attention only on three different kinds of
a-amino acids, namely, alanine (Ala), phenylalanine (Phe),
and valine (Val) (i.e., R2–R7¼Me, Bn or i-Pr). Due to the
ready availability of the 20 naturally occurring amino acids,
it would be of particular interest to investigate other amino
acids on the gelation properties.3 In this paper, we report
an optimization study on the gelation property of C2-
symmetrical 3,5-bis(N-protected diamino) benzoates 3 by
employing different amino acid residues as pendant side
chains. It was found that among all the amino acids studied,
phenylalanine and valine derivatives proved to possess the
best gelating power. We also examined the effect of the car-
boxylic acid focal point functionality (i.e., R1) on gelation
while keeping the amino acid side chains as valine (i.e.,
R2¼i-Pr). As a result of this optimization study, several
stronger organogelators (MGC down to 0.8 mg/mL) were
found. We also gathered concrete experimental evidence to
support that both p–p aromatic stacking, in addition to
H-bonding interactions, were the driving forces behind their
excellent gelation ability.
Our immediate goal was then to develop an efficient syn-
thetic protocol to assemble the various Cbz-protected amino
acids and aromatic-containing alcohols to the central 3,5-di-
aminobenzoic acid core.
2.2. Synthesis
The general synthetic route to the target compounds was
shown in Scheme 1. 3,5-Di-(tert-butyloxycarbonylamino)-
benzoic acid 54 was reacted with benzyl bromide 6a in the
presence of K2CO3 and 18-crown-6 to give the correspond-
ing benzyl ester 7a in 61% yield. The Boc protecting groups
were then removed by trifluoroacetic acid (TFA) to furnish
the diaminoester 8a5 in 67% yield. Finally, coupling of 8a
with various Cbz-protected amino acids or Cbz-protected
b-alanine in the presence of 2-ethoxy-1-ethoxycarbonyl-
1,2-dihydroquinoline (EEDQ) gave the target compounds
aa-9a or b-Ala-10, respectively, in yields ranging from
45% to 86%. The benzyl ether of the tyrosine derivative
[i.e., Tyr(OBn)-9a], on the other hand, was prepared from
Tyr-9a using Williamson ether synthesis.
BocHN
Bn
NHBoc
Bn
ProHN
R2
NHPro
R2
O
O
O
O
For the introduction of the aromatic-containing ester func-
tionality, compound 5 was converted to various esters 7b–
7d by coupling with various aromatic-containing alcohols
6b–6d in the presence of dicyclohexylcarbodiimide
(DCC), 1-hydroxybenzotriazole (HOBt), and 4-dimethyl-
aminopyridine (DMAP) in 57–84% yield. The two Boc
groups were removed under acidic conditions to furnish the
3,5-diaminobenzoates 8b–8d. Coupling of 8b–8d with Cbz-
protected valine produced the target organogelators Val-9b–
Val-9d bearing different aromatic-containing ester group.
An ethyl ester Val-9e was also prepared as a reference com-
pound. A list of the target compounds is shown in Table 1.
HN
NH
HN
NH
CO2R1
CO2H
3
4
2. Results and discussion
2.1. Structural variations and optimizations
The best organogelator originated from our initial study
was the Boc-protected phenylalanine carboxylic acid 4
(MGC¼2 mg/mL in nitrobenzene).2a We envisaged that in-
sertion of additional aromatic moieties into this lead mole-
cule would further enhance its gelation ability through
additional p–p interactions with aromatic solvents. Exami-
nation of the structure of compound 4 revealed that there
were three possible positions to accommodate the extra
aromatic moieties (Fig. 1).
2.3. Characterization
2.3.1. NMR spectroscopy. The structural identities of all in-
termediates and target compounds were characterized by 1H
and 13C NMR spectroscopy. The 1H NMR spectra of all tar-
get organogelators were composed of two characteristic
features. One was a set of invariable signals originated from
the common 3,5-diaminobenzoic acid central core that was
shared among the gelators. Hence, the central aromatic
proton signals appeared as a triplet (dw8.3) and a doublet
(dw8.0) for all the target compounds, while that of the anilide
NH and Cbz-NH resonated at d 10.3 and 7.6, respectively
(Fig. 2). The signals of the benzylic and aromatic protons
of the Cbz protecting groups were at d 5.0 and d 7.0–7.5, re-
spectively. The other characteristic feature was a set of vari-
able ‘fingerprint’ signals due to the protons originated from
the different amino acid side chain and the focal point ester
functionality. For example, the 1H NMR spectrum of
(a) Changing of the tert-butyloxylcarbonyl (Boc) NH-pro-
tecting group to an aromatic-containing NH-protecting
group such as benzyloxycarbonyl (Cbz).
(b) Modifying the carboxylic acid functionality to an aro-
matic-containing carboxylic ester.
(c) Introduction of aromatic rings to the amino acid side
chains. This would involve the use of aromatic amino
acids such as phenylalanine and tyrosine, etc.
BocHN
Bn
NHBoc
CbzNH
O
O
other amino acid
side chains
Bn
HN
NH
CO2H
CO2(CH2)nAr
Figure 1. Structural modifications of the lead compound 4.

H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
BocHN NHBoc H2N NH2
365
BocHN
BocHN
H2N
NHBoc
BnBr 6a
K2CO3
CF3CO2H
CO2Bn
7a
CO2Bn
8a
CO2H
18-crown-6
5
O
O
O
NHCbz
R1 OH
HO
NHCbz
NHCbz
(6bꢀ6d)
HO
DCC, HOBt
DMAP
EEDQ
R2
EEDQ
CbzHN
R2
NHCbz CbzHN
R2
NHBoc
O
O
O
HN
NH
HN
NH
CO2R1
7bꢀ7d
CO2Bn
aa-9a
CO2Bn
β-Ala-10
CF3CO2H
b R1
c R1
=
=
) Ph
O
(CH2 3
CbzHN
Me2HC
NHCbz
CHMe2
NHCbz
CHMe2
EEDQ
CH2CH2
O
O
HO
NH2
HN
NH
CO2R1
8bꢀ8d, 8e
d R1
=
CH2
CO2R1
Val-9b–Val-9e
e R1 = Et
Scheme 1. General synthetic routes to target organogelators 9 and 10.
Table 1. List of target organogelators
Gln-9a were found at d 1.69–2.31 (CH2CH2), 6.81 (one of
the CONH2, the other NH was merged with the aromatic sig-
nals), 5.36 and 7.0–7.6 (benzyl ester). The switching of
a benzyl to other esters at R1 could also be reflected from
the resulting H NMR spectra. For example, the H NMR
spectrum of Val-9c showed, in addition to the isopropyl
side chain signals, the presence of the 2-(naphthalen-1-
yl)ethyl 1H signals at d 3.52, 4.58, and 7.2–8.3.
Compound
R1
R2
aa-9a
1
1
Gly-9a
Ala-9a
Val-9a
Phe-9a
Leu-9a
Ser-9a
—
—
—
—
—
—
H
Me
i-Pr
Bn
CH2(i-Pr)
CH2OH
Tyr-9a
Tyr(OBn)-9a
Gln-9a
Val-9b
Val-9c
Val-9d
Val-9e
b-Ala-10
—
—
—
CH2C6H4OH
CH2C6H4OBn
CH2CH2CONH2
—
—
—
—
—
The 13C NMR spectra of the target compounds also shared
a common spectral feature; the chemical shift values of the
carbon nuclei due to the common backbone skeleton were
nearly the same. Hence, the carbon signals of the Cbz groups
were found to situate at d 66 and 114–129, while the aro-
matic carbon signals of the 3,5-diaminobenzoate moiety
were found between d 130 and 140. Three sets of C]O sig-
nals located atwd 156, 165, and 171 were found. The signal
at the most upfield position could be assigned to the Cbz car-
bamate C]O while the intermediate signals could be attrib-
uted to the benzoate C]O. On the other hand, the most
downfield signal could be assigned to the anilide C]O.
(CH2)3Ph
(CH2)2(2-Np)
CH2C6H4Ph
Et
—
compound Val-9a showed the presence of the valine isopro-
pyl side chain as a doublet at d 0.91 and a multiplet at d 1.91–
2.11, together with the benzyl ester signals at d 5.35 and
7.0–7.6. Likewise, the corresponding feature signals of
1
Similar to the case in the H NMR spectroscopic analysis,
the chemical identities of the amino acid side chains could
be ascertained by their respective ‘fingerprint’ 13C NMR
signals.
δ ~ 7.6
δ ~ 5.0
PhH2CO
O
O
δ ~ 8.3
H
N
δ 7.0–7.5
NH
R2
2.3.2. Mass spectroscopic analysis. The structures of all
target organogelators were characterized by FAB or positive
ion EI mass spectroscopy. In addition to molecular ions in
O
O
O
R2
HN
N
H
the form of M+ or (M+H)+, dimeric M2 or (M2+H)+ ions
+
were also found (Fig. 3), suggesting the presence of molec-
ular aggregates in the solution states. The exact mass of
the molecular ion peak matched well with the theoretical
value.
δ ~ 10.3
CO2R1
δ ~ 8.0
Figure 2. Common 1H NMR spectral features shared by the target organo-
gelators.

366
H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
91
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
154
653 (M + H)+
1305 (M2 + H)+
0
100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400
m/z
Figure 3. Mass spectrum of compound Ala-9a.
2.4. Gelation properties
that the third aromatic ring further improved the p–p inter-
actions with the solvents.
The gelation properties of the target compounds were exam-
ined in various solvents and, as expected, strong gels were
formed mainly in aromatic solvents. The gelation effect of
changing the Boc into the Cbz protecting group was dra-
matic. Hence, the Cbz-protected valine ethyl ester Val-9e
was a much powerful organogelator (MGC¼3 mg/mL) as
compared to the previously reported Boc-protected analogue
3 (Pro¼Boc, R1¼Et, R2¼i-Pr) (MGCw100 mg/mL) (Table
2). This finding suggested that the two added phenyl rings
in the Cbz moieties enhanced the gelation property by in-
creasing p–p interactions with the aromatic solvents. None-
theless, it was noted that the gels so formed were translucent
(TG) and fragile. Addition of the third aromatic ring by re-
placing the ethyl ester moiety in Val-9e with a benzyl ester
gave compound Val-9a, which had similar MGC values
(3–10 mg/mL) as the corresponding ethyl ester. However,
the gels formed were mostly transparent (CG) and had
good physical strength. This finding therefore suggested
The effects of the various amino acid side chains on the ge-
lation properties were also examined (Table 3 and Fig. 4).
For compounds containing side chains that are highly hydro-
philic (Gln-9a and Tyr-9a) or bearing small alkyl groups
(Gly-9a, Ala-9a, and b-Ala-10), they possessed poor solu-
bility in aromatic solvents and tended to precipitate and
did not form gels. For Leu-9a having highly hydrophobic
side chains, it dissolved well in aromatic solvents and again
failed to form gels. More interestingly, compounds with aro-
matic side chains such as Phe-9a and Tyr(OBn)-9a were
good organogelators (MGCw0.8–3 mg/mL). The effect of
protecting the phenolic functionality in Tyr-9a as the benzyl
ether was obvious. The resulting compound Tyr(OBn)-9a
now became less hydrophilic and contained two extra aro-
matic rings, and was a much better organogelator. To our
surprise, the moderately hydrophobic compound Val-9a
(MGCw3–10 mg/mL) and the highly hydrophilic Ser-9a
(MGCw1–10 mg/mL) were also excellent organogelators.
The organogelating power of such amino acid containing
compounds had been shown previously to be highly depen-
dent on the amino acid compositions.2a We believed that the
amino acid side chain, in addition to interacting with the gel-
ating solvents, also played a crucial role in determining the
packing arrangement of the organogelators inside the gels.
Table 2. Effects of additional aromatic rings on gelation propertiesa
Solvent
3 (Pro¼Boc, R1¼Et, R2¼i-Pr)b Val-9e Val-9a
Benzene
Toluene
o-Xylene
Anisole
o-Dichlorobenzene PG
Nitrobenzene TG (90)
TG (100)
TG (80)
TG (80)
TG (90)
P
CG (3)
TG (3) CG (3)
TG (3) CG (4)
TG (3) TG (5)
TG (3) CG (6)
CG (3) CG (10)
Having studied the effect of the amino acid side chains, we
then turned our attention to the effect of modifying the ester
moiety and Val-9a was chosen as our lead molecule even
though Phe-9a possessed slightly better gelation power in
some selected solvents. This was because Phe-9a inadver-
tently exhibited poor solubility in some aromatic solvents
The values given in parentheses are the minimum concentration (mg/mL) to
achieve gelation at 25 ꢀC.
CG—transparent gel; PG—partial gel; TG—translucent gel; P—precipi-
a
tation.
Data taken from Ref. 2a.
b

H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
Table 3. Gelation properties of various Cbz-protected benzyl benzoatesa
367
Compound
Benzene
Toluene
o-Xylene
Anisole
o-Dichlorobenzene
Nitrobenzene
Gly-9a
Ala-9a
Val-9a
Phe-9a
Leu-9a
Ser-9a
NS
NS
P
CG (3)
TG (2)
S
OG (7)
NS
CG (2)
NS
NS
NS
P
CG (4)
CG (0.8)
S
CG (1)
NS
CG (2)
NS
TG (10)
P
S
P
P
P
TG (27)
CG (3)
P
TG (50)
CG (10)
OG (50)
S
TG (10)
TG (10)
S
TG (5)
P
S
TG (4)
TG (3)
CG (35)
NS
CG (6)
CG (3)
S
CG (6)
TG (10)
CG (3)
NS
S
OG (7)
NS
CG (3)
NS
Tyr-9a
Tyr(OBn)-9a
Gln-9a
b-Ala-10
TG (6)
P
P
TG (10)
OG (14)
The values given in parentheses are the minimum concentration (mg/mL) to achieve gelation at 25 ꢀC.
CG—transparent gel; OG—opaque gel; TG—translucent gel; NS—insoluble; S—soluble; P—precipitation.
a
Figure 4. (a) Gels formed from Phe-9a at 5 mg/mL (left to right: KMnO4 solution, toluene, p-xylene, o-xylene, m-xylene, and o-dichlorobenzene); (b) trans-
parent gels formed from Tyr(OBn)-9a at 5 mg/mL (left to right: KMnO4 solution, o-xylene, toluene, m-xylene, o-dichlorobenzene, and p-xylene); (c) transpar-
ent gels formed from Val-9a at 5 mg/mL (left to right: KMnO4 solution, p-xylene, o-xylene, o-dichlorobenzene, and m-xylene).
such as benzene and anisole. Therefore, addition of more
aromatic rings into the system would most likely further de-
crease the solubility. Hence, choosing valine as the amino
acid residue, we replaced the benzyl alcohol moiety with
aromatic-containing alcohols such as 3-phenylpropan-1-ol,
1-naphthaleneethanol, and biphenyl-4-methanol and ob-
tained compounds Val-9b, Val-9c, and Val-9d, respectively.
The gelation properties of the resulting compounds were
tabulated in Table 4. As it turned out, none of these com-
pounds showed superior gelating ability as compared to
the parent benzyl ester Val-9a. The gels formed were either
translucent or opaque with higher MGC values. It therefore
appeared that this part of the molecule could not be altered
without having negative impact on the gelation properties.
(Pro¼Boc, R1¼Et, R2¼i-Pr) and recorded its FTIR spectra
in both the solution and gel state at a concentration of
100 mg/mL in toluene. It was found that there was little
difference between the solution and gel FTIR spectra. In
particular, the N–H (3409 and 3311 cmꢁ1) and the C]O
stretching frequencies (1726 and 1677 cmꢁ1) remained un-
changed when switching from the solution to the gel states,
suggesting that the extent of H-bonding did not have much
difference between the two states in toluene. However, this
finding was not surprising in view of the fact that this com-
pound had already been shown to form dimeric species via
H-bonds in nonpolar CDCl3 solutions,2a and hence there
was little change of the stretching frequency upon gelation
in nonpolar toluene solutions.
2.5. Gelation mechanism
A more interesting finding was revealed by studying the cir-
cular dichroism (CD) spectral behavior of a 3 mg/mL sam-
ple of Tyr(OBn)-9a in benzene at different temperatures
(Fig. 5). Prior to gelation (temperature range¼60–75 ꢀC),
The gelation mechanism of the organogelators was investi-
gated by a number of techniques. Due to the rapid formation
of gels from Phe-9a, Tyr(OBn)-9a, and Val-9a, we were un-
able to obtain their solution FTIR spectra at concentrations
>20 mg/mL. On the other hand, the signal intensities of the
FTIR spectra at an organogelator concentration of <10 mg/
mL were too weak to be observed even after solvent signal
subtraction. In the end, we chose a weaker organogelator 3
one positive (lmax¼329 nm) and one negative (lmax
¼
313 nm) Cotton effects, with decreasing molar ellipticities
at higher temperatures, were observed. After gelation (tem-
perature range¼20–40 ꢀC), these signals were blue-shifted
to lmax¼326 and 308 nm, respectively, indicating that the
energy gaps of the electronic transitions were widened
Table 4. Gelation properties of Cbz-protected valine-based benzoates containing different aromatic ester pendant groupsa
Compound
Benzene
Toluene
o-Xylene
Anisole
o-Dichlorobenzene
Nitrobenzene
Val-9a
Val-9b
Val-9c
Val-9d
CG (3)
TG (5)
TG (4)
TG (10)
CG (3)
TG (10)
NS
CG (4)
TG (5)
TG (20)
OG (5)
P
CG (6)
CG (10)
P
TG (6)
TG (20)
TG (20)
TG (10)
TG (10)
TG (13)
CG (11)
CG (11)
TG (10)
The values given in parentheses are the minimum concentration (mg/mL) to achieve gelation at 25 ꢀC.
CG—transparent gel; OG—opaque gel; TG—translucent gel; NS—insoluble; P—precipitation.
a

368
H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
1500
solution state: 60–75 °C
60 °C
65 °C
1000
500
0
70 °C
75 °C
-500
-1000
-1500
40 °C
35 °C
31 °C
20 °C
gelation state: 20–40 °C
290
300
310
320
330
340
350
Wavelength (nm)
Figure 5. Temperature-dependent CD spectra of Tyr(OBn)-9a in benzene (3 mg/mL).
upon gelation. This was obviously due to a change of p–p
interactions between the organogelator and the aromatic
solvent (i.e., benzene) on going from the solution to the
gel state. In contrast, the CD spectrum of the weak organo-
gelator, Boc-protected 3 (Pro¼Boc, R1¼Et, R2¼i-Pr), did
not show any shift of lmax value from the solution to the
gel state, suggesting that p–p interactions contributed sig-
nificantly to the gel stability of Cbz-protected organgelators.
2.6. Electron microscopy
The morphology of the gel structure was examined by scan-
ning electron microscopy (SEM). All SEM images showed
the presence of dense structures without clear, well-defined
microscale architecture (Fig. 6). Although in some occasions
we were able to locate fine fibers with diametersw0.1 mm
protruding out from the dense particles, the existence of
Figure 6. SEM images of freeze-dried samples of (a) Phe-9a (1 mg/mL in o-xylene), (b) Ser-9a (1.5 mg/mL in o-xylene), and (c) Tyr(OBn)-9a (2 mg/mL in
o-xylene).

H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
369
chiral morphology could not be observed. Such densed
structures were most likely formed from the collapse of
the gel microstructure during the freeze-drying process.
1H NMR: 1.46 (18H, s, C(CH3)3), 5.33 (2H, s, CH2),
7.30–7.49 (5H, m, ArH), 7.72 (2H, s, ArH), 7.97 (1H, s,
ArH), 9.53 (2H, s, NHAr). 13C NMR: 28.1, 66.1, 79.2,
112.5, 113.0, 127.9, 128.1, 128.5, 130.2, 136.2, 140.3,
152.7, 165.6. MS (FAB): 442 (M+, 62%), 663 (M3
,
2+
+
3. Conclusions
43%), 885 [(M2+H)+, 6%], 1326 (M3 , 2%). Anal. Calcd
for C24H30N2O6: C, 65.14; H, 6.83; N, 6.33. Found: C,
65.24; H, 6.89; N, 6.22.
Starting from an ethyl 3,5-bis(Boc-protected aminoacyl-
amido)benzoate as a lead compound, a series of 3,5-bis-
(Cbz-protected aminoacylamido)benzoates bearing different
a- or b-amino side chain residues and different aromatic-con-
taining ester pendant groups were prepared. The additional
three aromatic rings in the Cbz- and aromatic-containing
ester moieties significantly improved the gelation properties
(MGC decreased fromw100 to 0.8 mg/mL) of the resulting
organogelators in many aromatic solvents. FTIR and CD
spectroscopy revealed that both H-bonding and p–p aro-
matic stacking were the driving forces for gelation.
4.3. Benzyl 3,5-diaminobenzoate (8a)5
A mixture of compound 7a (11.90 g, 271 mmol) and TFA
(41.75 mL, 542 mmol) in dry CH2Cl2 (150 mL) was stirred
at 25 ꢀC and the progress of deprotection was monitored
by TLC. After complete deprotection (24 h), the solvent
was evaporated on a rotary evaporator and the residue re-dis-
solved in EtOAc (50 mL). The solution was neutralized to
pH¼8 by NaHCO3 solution. The organic layer was collected
and the aqueous layer was extracted with EtOAc (2ꢂ50 mL).
The combined organic layers were washed with saturated
NaHCO3 solution (100 mL), water (100 mL), and saturated
NaCl solution (100 mL) successively, dried (MgSO4), fil-
tered, and evaporated in vacuo to give a pale brown solid
that was chromatographed on silica gel (eluent: hexane/
EtOAc¼1:1) to afford the title compound 8a (4.36 g, 67%)
as a pale orange solid. Mp: 106–107 ꢀC. Rf: 0.32 (hexane/
4. Experimental
4.1. General
THF was distilled from sodium benzophenone ketyl and
CH2Cl2 from CaH2 prior to use. Silica gel for flash chroma-
tography is Macherey Nagel 60M (230–400 mesh) silica gel.
CbzNH-L-amino acids (CbzNH-aa-CO2H) and other re-
agents were used as supplied from Aldrich or Sigma. All
reactions were conducted under dry N2 unless otherwise
stated. All NMR spectra were recorded in DMSO-d6 (dried
1
EtOAc¼1:1). H NMR: 5.01 (4H, s, NH2), 5.24 (2H, s,
CH2), 6.03 (1H, t, J¼2.0, ArH), 6.46 (2H, d, J¼2.0, ArH),
7.30–7.46 (5H, m, ArH). 13C NMR (some overlapping aro-
matic C signals): 65.6, 103.7, 103.8, 128.0, 128.5, 130.6,
136.5, 149.4, 166.7. MS (EI): 242 (M+, 91%). HRMS (EI):
C14H14N2O2 requires 242.1050; found: 242.1048.
˚
€
over molecular sieve 4 A) on a Bruker DPX300 spectrome-
ter at 300 MHz for H and 75.5 MHz for 13C nucleus at
1
25 ꢀC unless otherwise stated. The residual proton or carbon
signals of DMSO-d6 (dH¼2.50, dC¼39.52) were used as in-
ternal references. All chemical shifts were reported in parts
per million (d) and coupling constants in hertz. Positive ion
EI and FAB spectra were carried out on a Thermo Finnigan
MAT 95XL mass spectrometer. Melting points were mea-
sured on an Electrothermal IA9100 Digital Melting Point
Apparatus and were uncorrected. IR spectra were recorded
4.4. General procedure for the preparation of aa-9a
and b-Ala-10
EEDQ (2.25 g, 9.08 mmol) was added to a stirred solution of
benzyl 3,5-diaminobenzoate 8a (1.00 g, 4.13 mmol) and
Cbz-protected amino acid (9.08 mmol) in THF at 20 ꢀC.
The mixture was kept at 20 ꢀC for 24 h and the reaction
was worked up according to the following procedure.
€
on a Bruker Vertex 70 FT-IR spectrophotometer. Optical
rotations were taken on a Perkin–Elmer 341 Polarimeter at
589 nm and at 20 ꢀC, in a solvent mixture of 1,2-dichloro-
ethane/HOAc (v/v¼95/5) unless otherwise specified. CD
spectra were recorded on a JASCO J-715 spectropolarimeter
connected to a NESLAB RTE-211 temperature controller.
Elemental analyses were performed at MEDAC Ltd., Brunel
Science Center, Cooper’s Hill Lane, Englefield Green,
Egham, Surrey TW20 0JZ, UK.
4.4.1. Gly-9a. After stirring for 24 h, a white precipitate was
formed. It was collected by suction filtration and washed
with THF, EtOAc, Et2O, and hexane successively. The prod-
uct was dried in vacuo to give a white powder (2.04 g, 79%).
1
Mp: 220–221 ꢀC. Rf: 0.18 (hexane/EtOAc¼1:1). H NMR:
3.81 (4H, d, J¼6.0, NCH2), 5.05 (4H, s, CH2Ph), 5.35
(2H, s, CO2CH2Ph), 7.15–7.52 (15H, m, ArH), 7.58 (2H, t,
J¼6.0, CbzNH), 7.96 (2H, s, ArH), 8.27 (1H, s, ArH),
10.24 (2H, s, NHAr). 13C NMR (some overlapping aromatic
C signals): 44.4, 65.8, 66.6, 114.5, 115.0, 127.2, 127.9,
128.3, 128.5, 128.7, 130.6, 136.2, 137.2, 139.9, 156.9,
165.6, 168.6. MS (FAB): 625 [(M+H)+, 6%]. HRMS
(FAB): C34H33N4O8 requires 625.2293; found: 625.2303.
4.2. Benzyl 3,5-di-(tert-butyloxycarbonylamino)-
benzoate (7a)
A mixture of 3,5-di-(tert-butyloxycarbonylamino)benzoic
acid 54 (20.00 g, 56.8 mmol), K2CO3 (19.60 g,
141.9 mmol), benzyl bromide 6a (8.78 mL, 73.8 mmol),
and 18-crown-6 (5 mg) in acetone (150 mL) was heated to
reflux for 24 h. The mixture was filtered and the filtrate
was evaporated in vacuo to give a crude oil that was chroma-
tographed on silica gel (eluent: hexane/EtOAc¼6:1) to give
the title compound 7a (15.32 g, 61%) as a creamy white
solid. Mp: 158–159.5 ꢀC. Rf: 0.26 (hexane/EtOAc¼5:1).
4.4.2. Ala-9a. The solvent was evaporated in vacuo and the
residue was re-dissolved in EtOAc and washed with satu-
rated Na2CO3, 10% citric acid, saturated Na2CO3 solution,
water, and saturated NaCl solution successively. The organic
layer was dried (MgSO4) and filtered. The filtrate was evapo-
rated in vacuo to yield an oil that was chromatographed on
silica gel (eluent: hexane/EtOAc¼2:1 gradient to 1:1) to

370
H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
produce the target compound as a pale yellow solid. Recrys-
tallization of the solid from a mixture of THF and hexane
gave pure Ala-9a (1.21 g, 45%) as a creamy white solid.
Mp: 176–177 ꢀC. Rf : 0.29 (hexane/EtOAc¼1:1). [a]D
4.4.6. Ser-9a. The solvent was evaporated in vacuo and the
product was precipitated in hot hexane and filtered. Recrys-
tallization of the solid from a mixture of THF and hexane af-
forded the title compound (2.15 g, 76%) as a creamy white
solid. Mp: 196–197 ꢀC. Rf : 0.62 (EtOAc). [a]D ꢁ9.2 (c
1
ꢁ59.6 (c 1.0). H NMR: 1.29 (6H, d, J¼7.2, CH3), 4.18
1
(2H, quintet, J¼7.2, NCH), 5.02 (2H, d, J¼12.6, CHHPh),
5.04 (2H, d, J¼12.6, CHHPh), 5.36 (2H, s, CO2CH2Ph),
6.95–7.57 (15H, m, ArH), 7.64 (2H, d, J¼6.9, CbzNH),
7.97 (2H, s, ArH), 8.32 (1H, s, ArH), 10.25 (2H, s, NHAr).
13C NMR (some overlapping aromatic C signals): 17.9,
60.0, 65.5, 66.5, 114.5, 114.9, 127.0, 127.8, 128.3, 128.4,
128.6, 130.3, 136.1, 137.0, 139.8, 155.9, 165.4, 172.0. MS
(FAB): 653 [(M+H)+, 12%], 1305 [(M2+H)+, 3%]. Anal.
Calcd for C36H36N4O8: C, 66.25; H, 5.56; N, 8.58. Found:
C, 66.07; H, 5.61; N, 8.50.
1.0, DMSO). H NMR: 3.52–3.77 (4H, m, CH2OH), 4.22
(2H, q, J¼9.0, NCH), 4.91–5.13 (6H, m, CH2Ph and
CH2OH), 5.35 (2H, s, CO2CH2Ph), 7.01–7.56 (17H, m,
ArH and CbzNH), 8.00 (2H, s, ArH), 8.32 (1H, s, ArH),
10.27 (2H, s, NHAr). 13C NMR (some overlapping aromatic
C signals): 57.9, 61.7, 65.6, 66.4, 114.6, 115.0, 127.0, 127.8,
128.2, 128.4, 128.6, 130.2, 136.0, 137.0, 139.6, 156.0, 165.4,
169.6. MS (FAB): 685 [(M+H)+, 9%]. HRMS (FAB):
C36H37N4O10 requires 685.2504; found: 685.2507.
4.4.7. Tyr-9a. The solvent was evaporated in vacuo and the
product was precipitated in hot hexane and filtered. Recrys-
tallization of the solid from a mixture of THF and hexane
produced the target compound (2.73 g, 79%) as an ivory
white solid. Mp: 216–218 ꢀC. Rf : 0.25(hexane/EtOAc¼1:1).
4.4.3. Val-9a. The solvent was evaporated in vacuo and the
residue recrystallized from a mixture of THF/hexane to
produce the title compound Val-9a (2.40 g, 82%) as a white
solid. Mp: 196–197 ꢀC. Rf : 0.57 (hexane/EtOAc¼1:1). [a]D
1
1
ꢁ30.1 (c 1.0). H NMR: 0.91 (12H, d, J¼6.3, CH3), 1.91–
[a]D +71.7 (c 1.0, DMSO). H NMR: 2.64–2.81 (2H, m,
2.11 (2H, m, Me2CH), 3.98 (2H, t, J¼7.8, NCH), 5.04
(4H, br s, CH2Ph), 5.35 (2H, s, CO2CH2Ph), 7.04–7.58
(17H, m, ArH+CbzNH), 7.98 (2H, s, ArH), 8.34 (1H, s,
ArH), 10.31 (2H, s, NHAr). 13C NMR (some overlapping
aromatic C signals): 18.5, 19.2, 30.2, 61.2, 65.5, 66.4,
114.4, 114.8, 127.0, 127.7, 128.2, 128.3, 128.6, 130.3,
136.0, 137.0, 139.5, 156.3, 165.3, 170.9. MS (FAB): 708
CHHArOH), 2.81–2.98 (2H, m, CHHArOH), 4.21–4.45
(2H, m, NCH), 4.97 (4H, s, CH2Ph), 5.36(2H, s, CO2CH2Ph),
6.66 (4H, d, J¼8.4, ArH), 7.12 (4H, d, J¼8.4, ArH), 7.21–
7.53 (15H, m, ArH), 7.66 (2H, d, J¼8.1, CbzNH), 7.97
(2H, s, ArH), 8.31 (1H, s, ArH), 9.21 (2H, s, ArOH), 10.35
(2H, s, NHAr). 13C NMR: 36.7, 57.5, 65.4, 66.5, 114.4,
114.9, 121.5, 126.6, 126.8, 127.6, 127.8, 128.29, 128.34,
128.6, 129.5, 130.2, 136.0, 137.0, 139.6, 155.9, 156.0,
(M+, 22%), 1416 (M2 , 3%). Anal. Calcd for C40H44N4O8:
+
+
C, 67.78; H, 6.26; N, 7.90. Found: C, 67.54; H, 6.35; N,
7.72.
165.4, 171.1. MS (FAB): 837 [(M+H)+, 6%], 1533 (M2 ,
1%). HRMS (FAB): C48H45N4O10 requires 837.3130; found:
837.3121.
4.4.4. Phe-9a. The solution was concentrated under reduced
pressure and the residue was recrystallized from THF/hex-
ane to give the title compound (2.82 g, 85%) as a white solid.
Mp: 133–135 ꢀC. Rf : 0.63 (hexane/EtOAc¼1:1). [a]D +25.1
(c 1.0). 1H NMR: 2.74–2.94 (2H, m, CH2Ph), 2.94–3.14 (2H,
m, CH2Ph), 4.28–4.51 (2H, m, NCH), 4.97 (4H, s, CH2Ph),
5.37 (2H, s, CO2CH2Ph), 6.99–7.58 (25H, m, ArH), 7.76
(2H, d, J¼8.1, CbzNH), 7.98 (2H, s, ArH), 8.34 (1H, s,
ArH), 10.42 (2H, s, NHAr). 13C NMR (some overlapping
aromatic C signals): 37.3, 57.1, 65.4, 66.5, 114.5, 114.9,
126.4, 126.8, 127.6, 127.8, 128.1, 128.3, 128.6, 129.2,
130.3, 136.0, 136.9, 137.8, 139.6, 156.0, 165.3, 170.9. MS
4.4.8. Gln-9a. The reaction mixture became turbid and the
pale yellow product was collected by suction filtration. Re-
crystallization from a mixture of THF and hexane afforded
the target compound (2.72 g, 86%) as a pale yellow solid.
Mp: 207–208 ꢀC. Rf : 0.66 (acetone/EtOAc¼1/2). [a]D
ꢁ7.6 (c 1.0, DMSO). 1H NMR: 1.69–2.03 (4H, m,
CH2CH2CON), 2.03–2.31 (4H, m, CH2CON), 4.11 (2H, q,
J¼6.8, NCH), 5.03 (4H, s, CH2Ph), 5.36 (2H, s, CO2CH2Ph),
6.81 (2H, s, CONHH), 7.04–7.55 (17H, m, CONHH+ArH),
7.63 (2H, d, J¼7.5, CbzNH), 7.98 (2H, s, ArH), 8.36 (1H, s,
ArH), 10.32 (2H, s, NHAr). 13C NMR (some overlapping
aromatic C signals): 27.5, 31.6, 55.4, 65.6, 66.5, 114.5,
115.0, 127.0, 127.8, 128.3, 128.4, 128.7, 130.3, 136.1,
137.0, 139.7, 156.1, 165.4, 171.1, 173.6; MS (FAB):
(FAB): 805 [(M+H)+, 10%], 1609 (M2 , 1%). HRMS
+
(FAB): C48H45N4O8 requires 805.3232; found: 805.3243.
767 [(M+H)+, 13%], 1533 (M2 , 1%). Anal. Calcd for
+
4.4.5. Leu-9a. The solvent was evaporated in vacuo and the
residue was purified by column chromatography on silica gel
(eluent: hexane/EtOAc¼3:1) to afford the title compound
(2.22 g, 73%) as a creamy yellow solid. Mp: 78–79 ꢀC. Rf :
C40H42N6O10: C, 62.65; H, 5.52; N, 10.95. Found: C,
62.33; H, 5.61; N, 10.47.
1
0.27 (hexane/EtOAc¼5:2). [a]D ꢁ30.9 (c 1.0). H NMR:
4.4.9. b-Ala-10. The solvent was evaporated in vacuo and
the product was precipitated in hot hexane and collected
by suction filtration. Recrystallization of the solid from
a mixture of THF and hexane afforded the title compound
(2.26 g, 84%) as a white powder. Mp: 173–174 ꢀC. Rf :
0.18 (hexane/EtOAc¼1:2). 1H NMR (acetone-d6): 2.65
(4H, t, J¼6.5, CH2CH2CON), 3.49 (4H, q, J¼6.4,
CbzNHCH2), 5.05 (4H, s, CH2Ph), 5.35 (2H, s, CO2CH2Ph),
6.46 (2H, br s, NH), 7.18–7.54 (15H, m, ArH), 8.04 (2H, s,
ArH), 8.32 (1H, s, ArH), 9.41 (2H, s, NHAr). 13C NMR
(some overlapping aromatic C signals): 36.8, 37.0, 65.3,
66.4, 114.2, 114.6, 127.4, 127.8, 128.2, 128.4, 128.6,
0.90 (6H, d, J¼6.0, CH3), 0.91 (6H, d, J¼6.3, CH3), 1.33–
1.84 (6H, m, CCH and CH2), 4.07–4.34 (2H, m, NCH),
5.05 (4H, s, CH2Ph), 5.37 (2H, s, CO2CH2Ph), 7.00–7.54
(15H, m, ArH), 7.63 (2H, d, J¼7.8, CbzNH), 8.01 (2H, s,
ArH), 8.40 (1H, s, ArH), 10.34 (2H, s, NHAr). 13C NMR
(some overlapping aromatic C signals): 21.5, 23.0, 24.4,
40.6, 54.0, 65.5, 66.5, 114.6, 114.9, 127.0, 127.7, 128.3,
128.4, 128.6, 130.3, 136.0, 137.0, 139.7, 156.1, 165.4,
171.9. MS (FAB): 737 [(M+H)+, 11%], 1473 (M2 , 5%).
+
Anal. Calcd for C42H48N4O8: C, 68.46; H, 6.57; N, 7.60.
Found: C, 68.60; H, 6.56; N, 7.51.

H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
371
130.3, 136.1, 137.3, 139.9, 156.2, 165.5, 169.7. MS (FAB):
653 [(M+H)+, 24%]. Anal. Calcd for C36H36N4O8: C, 66.25;
H, 5.56; N, 8.58. Found: C, 66.09; H, 5.54; N, 8.50.
Rf : 0.29 (hexane/EtOAc¼4:1). 1H NMR: 1.48 (18H, s,
C(CH3)3), 3.50 (2H, t, J¼6.9, NpCH2), 4.55 (2H, t, J¼6.9,
CO2CH2), 7.39–7.64 (4H, m, ArH), 7.74 (2H, d, J¼1.5,
ArH), 7.82 (1H, d, J¼7.8, ArH), 7.86 (1H, t, J¼1.5, ArH),
7.93 (1H, d, J¼8.1, ArH), 8.22 (1H, d, J¼8.1, ArH), 9.55
(2H, s, NHAr). 13C NMR: 28.1, 31.6, 64.7, 79.2, 112.6,
112.9, 123.6, 125.6, 125.7, 126.2, 127.1, 127.4, 128.6,
130.3, 131.5, 133.4, 133.8, 140.2, 152.7, 165.7. MS
(FAB): 506 (M+, 20%), 1013 [(M2+H)+, 2%]. Anal. Calcd
for C29H34N2O6: C, 68.76; H, 6.76; N, 5.53. Found: C,
68.41; H, 6.87; N, 5.51.
4.4.10. Tyr(OBn)-9a. A mixture of Tyr-9a (2.00 g,
2.39 mmol), K2CO3 (1.65 g, 12.0 mmol), benzyl bromide
6a (0.71 mL, 5.97 mmol), and 18-crown-6 (5 mg) was re-
fluxed in a mixture of acetone (80 mL) and DMF (5 mL)
for 24 h. The mixture was filtered and the filtrate evaporated
in vacuo to give an oil that was purified by column chroma-
tography on silica gel (hexane/EtOAc¼2:1 gradient to 2:3)
to afford a pale yellow solid. Recrystallization from a mix-
ture of EtOAc and Et2O gave the target compound (1.82 g,
75%) as white needles. Mp: 150–151 ꢀC. Rf : 0.28 (hexane/
4.5.3. Biphenyl-4-methyl 3,5-di-(tert-butyloxycarbonyl-
amino)benzoate (7d). Starting from compound 5 (4.50 g,
12.8 mmol) and biphenyl-4-methanol 6d (2.82 g,
15.3 mmol), the title compound 7d was obtained as a white
solid (1.86 g, 82%) after column chromatography on silica
gel (eluent: hexane/EtOAc¼9:2). Mp: 150–152 ꢀC. Rf :
0.23 (hexane/EtOAc¼9:2). 1H NMR: 1.46 (18H, s,
C(CH3)3), 5.38 (2H, s, ArCH2), 7.33–7.59 (5H, m, ArH),
7.63–7.73 (4H, m, ArH), 7.75 (2H, d, J¼1.8, ArH), 7.99
(1H, t, J¼1.8, ArH), 9.55 (2H, s, NHAr). 13C NMR: 28.1,
65.9, 79.2, 112.5, 112.9, 126.7, 126.8, 127.6, 128.6, 129.0,
130.2, 135.3, 139.8, 140.0, 140.3, 152.7, 165.6. MS
(FAB): 518 (M+, 13%). HRMS (FAB): C30H34N2O6 requires
518.2411; found: 518.2411.
1
EtOAc¼2:1). [a]D +14.9 (c 1.0). H NMR: 2.68–2.87 (2H,
m, CHH), 2.87–3.12 (2H, m, CHH), 4.21–4.48 (2H, m,
NCH), 4.97 (4H, s, CH2Ph), 5.05 (4H, s, CH2Ph), 5.36
(2H, s, CO2CH2Ph), 6.92 (4H, d, J¼9.0, ArH), 7.06–7.57
(29H, m, ArH), 7.70 (2H, d, J¼8.0, CbzNH), 7.97 (2H, s,
ArH), 8.34 (1H, s, ArH), 10.39 (2H, s, NHAr). 13C NMR:
36.5, 57.3, 65.4, 66.5, 69.2, 114.4, 114.9, 126.8, 127.6,
127.7, 127.8, 128.3, 128.4, 128.6, 129.9, 130.3, 136.0,
136.9, 137.2, 139.6, 156.0, 157.0, 165.3, 171.0; MS
(FAB): 1017 [(M+H)+, 5%]. HRMS (FAB): C62H57N4O10
requires 1017.4069; found: 1017.4086.
4.5. General procedure for the synthesis of 3,5-di-(tert-
butyloxycarbonylamino)benzoates (7b–7d)
4.6. General procedure for the synthesis of 3,5-di-
aminobenzoates (8b–8d)
A mixture of compound 5 (1.0 equiv), the alcohol 6b–6d
(1.2 mol equiv), DMAP (0.1 mol equiv), HOBt (2 mol
equiv), and DCC (3 mol equiv) was stirred in dry THF at
25 ꢀC for 24 h. The white precipitate was filtered and the fil-
trate was evaporated under reduced pressure. The residue
was taken up in EtOAc, washed with citric acid solution,
Na2CO3 solution, water, and saturated NaCl solution succes-
sively. The organic layer was dried (MgSO4), filtered, and
evaporated in vacuo to give the crude product. The crude
product was further purified as described in the following
text.
A mixture of the Boc-protected benzoates 7b–7d and TFA
(25 mol equiv) was stirred in CH2Cl2 at 25 ꢀC for 3 h. The
solvent was evaporated on a rotary evaporator and the resid-
ual was neutralized by NaHCO3 solution until pH¼8 at 5 ꢀC.
The residue was taken up in EtOAc (50 mL) and extracted
with ethyl acetate (2ꢂ50 mL). The combined organic layers
were washed with water (100 mL) and then saturated brine
(100 mL), dried (MgSO4), filtered, and evaporated in vacuo
to give the crude product. The crude product was purified as
described in the following text.
4.5.1. 3-Phenylpropyl 3,5-di-(tert-butyloxycarbonyl-
amino)benzoate (7b). Starting from compound 5 (4.41 g,
12.5 mmol) and 3-phenylpropan-1-ol 6b (2.05 g,
15.0 mmol), the title product 7b was obtained as a white solid
(1.34 g, 84%) after column chromatography on silica gel
(eluent: hexane/EtOAc¼5:1). Mp: 117–119 ꢀC. Rf : 0.21
4.6.1. 3-Phenylpropyl 3,5-diaminobenzoate (8b). Starting
from compound 7b (0.69 g, 1.47 mmol), the title product
was obtained as a brown solid (0.33 g, 83%) after column
chromatography on silica gel (eluent: hexane/EtOAc¼1:1).
1
Mp: 80–81 ꢀC. Rf : 0.23 (hexane/EtOAc¼1:1). H NMR:
1.88–2.03 (2H, m, PhCH2CH2), 2.71 (2H, t, J¼7.7,
PhCH2CH2), 4.14 (2H, t, J¼6.5, ArCO2CH2), 5.00 (4H, s,
NH2), 6.03 (1H, t, J¼1.8, ArH), 6.45 (2H, d, J¼1.8, ArH),
7.17–7.36 (5H, m, ArH). 13C NMR: 30.9, 32.5, 64.2,
104.6, 104.7, 126.9, 129.3, 129.4, 131.8, 142.1, 150.4,
167.8. MS (FAB): 270 [(M+H)+, 98%]. Anal. Calcd for
C16H18N2O2: C, 71.09; H, 6.71; N, 10.36. Found: C,
71.12; H, 6.75; N, 10.01.
1
(hexane/EtOAc¼5:1). H NMR: 1.47 (18H, s, C(CH3)3),
1.90–2.07 (2H, m, PhCH2CH2), 2.74 (2H, t, J¼7.7,
PhCH2CH2), 4.21 (2H, t, J¼6.3, ArCO2CH2), 7.14–7.34
(5H, m, ArH), 7.75 (2H, d, J¼1.8, ArH), 7.97 (1H, t,
J¼1.8, ArH), 9.55 (2H, s, NHAr). 13C NMR: 28.1, 29.9,
31.4, 63.7, 79.2, 112.4, 112.9, 125.9, 128.4, 130.4, 140.2,
141.1, 152.7, 165.7. MS (FAB): 470 (M+, 70%), 940 (M2 ,
+
5%). Anal. Calcd for C26H34N2O6: C, 66.36; H, 7.28; N,
5.95. Found: C, 66.66; H, 7.47; N, 5.83.
4.6.2. 2-(Naphthalen-1-yl)ethyl 3,5-diaminobenzoate (8c).
Starting from compound 7c (2.96 g, 5.84 mmol), the target
compound 8c was obtained as a pale yellow solid (1.35 g,
75%) after column chromatography on silica gel (eluent:
hexane/EtOAc¼1:1). Mp: 124–128 ꢀC. Rf : 0.23 (hexane/
4.5.2. 2-(Naphthalen-1-yl)ethyl 3,5-di-(tert-butyloxycar-
bonylamino)benzoate (7c). Starting from compound 5
(2.00 g, 5.68 mmol) and 1-naphthaleneethanol 6c (1.17 g,
6.81 mmol), the target product 7c was obtained as a pale
yellow solid (1.63 g, 57%) after column chromatography on
silica gel (eluent: hexane/EtOAc¼4:1). Mp: 131–134 ꢀC.
1
EtOAc¼1:1). H NMR: 3.46 (2H, t, J¼6.9, NpCH2), 4.48
(2H, t, J¼6.9, ArCO2CH2), 4.99 (4H, s, NH2), 6.03 (1H, t,
J¼1.8, ArH), 6.42 (2H, d, J¼1.8, ArH), 7.42–7.49 (2H,

372
H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
m, ArH), 7.49–7.64 (2H, m, ArH), 7.78–7.87 (1H, m, ArH),
7.94 (1H, d, J¼7.8, ArH), 8.20 (1H, d, J¼8.4, ArH). 13C
NMR (some overlapping aromatic C signals): 31.6, 64.1,
103.7, 123.7, 125.6, 125.7, 126.3, 127.1, 128.6, 130.7,
131.6, 133.4, 134.0, 149.3, 166.8. MS (FAB): 306 (M+,
55%). Anal. Calcd for C19H18N2O2: C, 74.49; H, 5.92; N,
9.14. Found: C, 74.40; H, 6.00; N, 9.01.
125.71, 126.3, 127.2, 127.3, 127.8, 128.4, 128.6, 130.4,
131.5, 133.4, 133.8, 137.0, 139.4, 156.3, 165.5, 170.9; MS
(FAB): 773 [(M+H)+, 15%], 1544 (M2 , 2%). HRMS
+
(FAB): C45H48N4O8 requires 773.3545; found: 773.3542.
4.7.3. Val-9d. Starting from compound 8d (0.65 g,
2.04 mmol), the crude product was subjected to chromato-
graphic purification on silica gel (eluent: hexane/EtOAc¼
9:4). The pure product was obtained as a white solid
(0.94 g, 59%) after recrystallization from a mixture of THF
and hexane. Mp:184–187 ꢀC. Rf :0.64 (hexane/EtOAc¼1:1).
[a]D ꢁ23.6 (c 1.0). 1H NMR: 0.91 (12H, d, J¼6.3, CHCH3),
1.90–2.10 (2H, m, CHMe2), 3.98 (2H, t, J¼9.0, NCH), 5.04
(4H, s, PhCH2OCON), 5.40 (2H, s, ArCO2CH2), 7.10–7.42
(11H, m, ArH), 7.42–7.62 (6H, m, ArH and CbzNH), 7.63–
7.75 (4H, m, ArH), 8.00 (2H, s, ArH), 8.34 (1H, s, ArH),
10.32 (2H, s, NHAr). 13C NMR (some overlapping aromatic
C signals): 18.5, 19.2, 30.2, 61.2, 65.5, 66.2, 114.4, 114.8,
126.8, 126.9, 127.6, 127.8, 128.4, 129.0, 130.3, 135.2,
137.0, 139.5, 139.7, 140.2, 156.3, 165.4, 170.9. MS (FAB):
785 [(M+H)+, 10%], 1569 [(M2+H)+, 2%]. HRMS (FAB):
C46H48N4O8 requires 785.3545; found: 785.3544.
4.6.3. Biphenyl-4-methyl 3,5-diaminobenzoate (8d).
Starting from compound 7d (1.61 g, 3.10 mmol), the target
product 8d was obtained as a pale pink solid (0.75 g, 76%)
after column chromatography on silica gel (eluent: hexane/
EtOAc¼1:1). Mp: 162–165 ꢀC. Rf : 0.27 (hexane/
1
EtOAc¼1:1). H NMR: 5.03 (4H, s, NH2), 5.29 (2H, s,
ArCH2), 6.03 (1H, t, J¼1.8, ArH), 6.48 (2H, d, J¼1.8,
ArH), 7.29–7.41 (1H, m, ArH), 7.41–7.57 (4H, m, ArH),
7.57–7.76 (4H, m, ArH). 13C NMR: 65.3, 103.7, 103.8,
126.7, 126.8, 127.6, 128.7, 129.0, 130.5, 135.7, 139.8,
139.9, 149.5, 166.7. MS (FAB): 318 (M+, 35%). HRMS
(FAB): C20H18N2O2 requires 318.1363; found: 318.1362.
4.7. General procedure for the synthesis
of Val-9b–Val-9e
4.7.4. Val-9e. Starting from ethyl 3,5-diaminobenzoate6
(0.77 g, 4.27 mmol), the product was obtained as a white
solid (1.65 g, 60%) after recrystallization from a mixture
of THF and hexane. Mp: 225–229 ꢀC. Rf : 0.67 (hexane/
A mixture of the 3,5-diaminobenzoates 8b–8d or 8e,6 Cbz-
protected valine (2.3 mol equiv), and EEDQ (2.2 mol equiv)
in THF was stirred at 25 ꢀC for 24 h. The solvent was evapo-
rated in vacuo and the residual was taken up in EtOAc. The
organic layer was washed with Na2CO3 solution, citric acid
solution, Na2CO3 solution, water, and saturated NaCl solu-
tion successively. It was then dried (MgSO4), filtered, and
evaporated in vacuo to give the crude product. The crude
product was purified as described in the following text.
1
EtOAc¼1:1). [a]D +39.6 (c 1.0, DMSO). H NMR: 0.94
(12H, d, J¼6.5, CHCH3), 1.34 (3H, t, J¼7.0, CO2CH2CH3),
1.92–2.15 (2H, m, CHMe2), 4.01 (2H, t, J¼7.8, NCH), 4.34
(2H, q, J¼7.0, CO2CH2CH3), 5.06 (4H, s, PhCH2OCON),
7.06–7.46 (10H, m, ArH), 7.55 (2H, d, J¼8.2, CONH),
7.99 (2H, s, ArH), 8.29 (1H, s, ArH), 10.33 (2H, s, NHAr).
13C NMR: 14.2, 18.5, 19.2, 30.2, 60.9, 61.2, 65.5, 114.2,
114.8, 127.0, 127.7, 128.3, 130.6, 137.0, 139.4, 156.3,
4.7.1. Val-9b. Starting from compound 8b (0.25 g,
0.92 mmol), the product was obtained as a brown solid
(0.47g, 69%) after recrystallization from a mixture of THF
and hexane. Mp: 180–183 ꢀC. Rf : 0.59 (hexane/EtOAc¼1:1).
[a]D ꢁ22.9 (c 1.0). 1H NMR: 0.92 (12H, d, J¼6.6, CHCH3),
1.92–2.12 (4H, m, PhCH2CH2 and CHMe2), 2.75 (2H, t,
J¼7.5, PhCH2CH2), 4.00 (2H, t, J¼7.5, NCH), 4.25 (2H, t,
J¼7.5, CO2CH2CH2), 5.04 (4H, s, PhCH2OCON), 7.05–
7.42 (15H, m, ArH), 7.54 (2H, d, J¼8.4, CbzNH), 8.00
(2H, s, ArH), 8.29 (1H, s, ArH), 10.31 (2H, s, NHAr). 13C
NMR (some overlapping aromatic C signals): 18.5, 19.2,
29.8, 30.2, 31.4, 61.2, 64.0, 65.5, 114.3, 114.9, 125.9, 127.7,
128.3, 130.5, 137.0, 139.4, 141.0, 156.3, 165.4, 170.9. MS
(FAB): 737 [(M+H)+, 12%], 1473 [(M2+H)+, 1%]. HRMS
(FAB): C42H48N4O8 requires 737.3545; found: 737.3541.
165.4, 170.9. MS (FAB): 646 (M+, 100%), 1292 (M2 ,
+
10%). Anal. Calcd for C35H42N4O8: C, 65.00; H, 6.55; N,
8.66. Found: C, 64.94; H, 6.63; N, 8.62.
4.8. FTIR experiments
Compound 3 (Pro¼Boc, R1¼Et, R2¼i-Pr, 100 mg) was dis-
solved in hot toluene (1.0 mL) and transferred to a KBr cell.
The IR spectra were recorded at t¼0, 5.5, and 11 min. At
t¼0, the sample was in the solution state and at t¼11 min
it became a gel. For background scanning, a drop of toluene
was added to the KBr cell. Background subtraction of both
the solvent and components in air was already done.
4.9. CD experiments
4.7.2. Val-9c. Starting from compound 8c (1.20 g,
3.92 mmol), the product was obtained as a white solid
(1.95 g, 64%) after recrystallization from a mixture of THF
and hexane. Mp: 233–239 ꢀC. Rf : 0.66 (hexane/EtOAc¼
Tyr(OBn)-9a (9 mg) was dissolved in hot benzene (3 mL)
and transferred to a quartz cell (1 cm path length). It was
allowed to cool down to allow for gel formation. The tem-
perature of the cell was then increased to the specified tem-
perature via the temperature controller. The measurement
was done from gel state (20–40 ꢀC) to solution state (60–
75 ꢀC) at increasing temperature.
1
1:1). [a]D ꢁ27.7 (c 1.0). H NMR: 0.92 (12H, d, J¼6.6,
CHCH3), 1.91–2.12 (2H, m, CHMe2), 3.52 (2H, t, J¼6.0,
PhCH2CH2), 4.00 (2H, t, J¼7.5, NCH), 4.58 (2H, t, J¼6.0,
PhCH2CH2), 5.05 (4H, s, PhCH2OCON), 7.20–7.41 (10H,
m, ArH), 7.41–7.64 (6H, m, ArH and CbzNH), 7.81 (1H,
d, J¼6.0, ArH), 7.93 (1H, d, J¼6.0, ArH), 7.98 (2H, s,
ArH), 8.17–8.27 (2H, m, ArH), 10.29 (2H, s, NHAr). 13C
NMR (some overlapping aromatic C signals): 18.5, 19.2,
30.2, 31.5, 61.2, 64.9, 65.5, 114.4, 115.0, 123.7, 125.65,
4.10. SEM experiments
Phe-9a (1 mg), Ser-9a (1.5 mg), and Tyr(OBn)-9a (2 mg)
were dissolved in o-xylene (1 mL) separately. The samples

H.-F. Chow et al. / Tetrahedron 63 (2007) 363–373
373
3. (a) For examples of low molecular weight amino acid containing
organogelators for organic solvents, see: Hanabusa, K.; Itoh, A.;
Kimura, M.; Shirai, H. Chem. Lett. 1999, 767–768; (b)
Hanabusa, K.; Matsumoto, M.; Kimura, M.; Kakehi, A.;
Shirai. J. Colloid Interface Sci. 2000, 224, 231–244; (c) Luo,
X.; Liu, B.; Liang, Y. Chem. Commun. 2001, 1556–1557; (d)
were freeze-dried on a Labconco Freezone 6LT Freezer Dry
System under 50ꢂ10ꢁ3 mbar at ꢁ47 ꢀC for 7 days. The
SEM images were taken by a Leo Scanning Electron Micro-
scope as described previously.2a
´
´
´
ˇ ´
´
´
Makarevic, J.; Jokic, M.; Peric, B.; Tomisic, V.; Kojic-Prodic,
Acknowledgements
ˇ
´
B.; Zinic, M. Chem.—Eur. J. 2001, 7, 3328–3341; (e) Suzuki,
M.; Waraksa, C. C.; Nakayama, H.; Hanabusa, K.; Kimura,
M.; Shirai, H. Chem. Commun. 2001, 2012–2013; (f) Becerril,
J.; Burguete, M. I.; Escuder, B.; Luis, S. V.; Miravet, J. F.;
Querol, M. Chem. Commun. 2002, 738–739; (g) Wang, G.;
Hamilton, A. D. Chem.—Eur. J. 2002, 8, 1954–1961; (h)
The work described in this paper was supported by a Direct
Grant (Account no. 2060245) and a Focused Investment into
Areas of Strength: Specialized Areas (Scheme B) by The
Chinese University of Hong Kong. H.-F.C. is a recipient of
the Croucher Senior Research Fellowship, Hong Kong
(2006–2007).
ˇ
´
´
´
´
Makarevic, J.; Jokic, M.; Frkanec, L.; Katalenic, D.; Zinic, M.
Chem. Commun. 2002, 2238–2239; (i) Becerril, J.; Burguete,
M. I.; Escuder, B.; Galindo, F.; Gavara, R.; Miravet, J. F.;
Luis, S. V.; Peris, G. Chem.—Eur. J. 2004, 10, 3879–3890; (j)
References and notes
ꢀ
Broose, N.; Barth, D.; Jamart-Gregoire, B. Tetrahedron Lett.
2004, 45, 9521–9524; (k) Suzuki, M.; Sato, T.; Kurose, A.;
Shirai, H.; Hanabusa, K. Tetrahedron Lett. 2005, 46, 2741–2745.
4. Kumar, A.; Meijer, E. W. Chem. Commun. 1998, 1629–1630.
5. Arya, P.; Kutterer, K. M. K.; Qin, H.; Roby, J.; Barnes, M. L.;
Lin, S.; Lingwood, C. A.; Peter, M. G. Bioorg. Med. Chem.
1999, 7, 2823–2833.
1. (a) For reviews on organogelators on organic solvents, see:
Terech, P.; Weiss, R. G. Chem. Rev. 1997, 97, 3133–3159; (b)
van Esch, J. H.; Feringa, B. L. Angew. Chem., Int. Ed. 2000,
39, 2263–2266; (c) Cronwald, O.; Snip, E.; Shinkai, S. Curr.
Opin. Colloid Interface Sci. 2002, 7, 148–156.
2. (a) Chow, H.-F.; Zhang, J. Chem.—Eur. J. 2005, 11, 5817–5831;
(b) Chow, H.-F.; Zhang, J. Tetrahedron 2005, 61, 11279–
11290.
6. Mong, T. K.-K.; Niu, A.; Chow, H.-F.; Wu, C.; Li, L.; Chen, R.
Chem.—Eur. J. 2001, 7, 686–699.