Phenol-containing bis(oxazolines): synthesis and fluorescence sensing of amines

Article abstract of DOI:10.1016/j.tet.2006.09.060

The fluorescence sensing of primary amines as their neutral forms has been studied with bis(oxazolinyl)phenols (Me-BOP, Ph-BOP), which are efficiently synthesized starting from mesitylene in six steps and in overall 12-22% yields. The BOP sensors showed f

Full text of DOI:10.1016/j.tet.2006.09.060

Tetrahedron 62 (2006) 11645–11651
Phenol-containing bis(oxazolines): synthesis and fluorescence
sensing of amines
*
Yun Mi Chung, Balamurali Raman and Kyo Han Ahn
Department of Chemistry and Center for Integrated Molecular Systems, POSTECH, San 31, Hyoja-dong,
Pohang 790-784, Republic of Korea
Received 11 July 2006; revised 7 September 2006; accepted 19 September 2006
Available online 19 October 2006
Abstract—The fluorescence sensing of primary amines as their neutral forms has been studied with bis(oxazolinyl)phenols (Me-BOP,
Ph-BOP), which are efficiently synthesized starting from mesitylene in six steps and in overall 12–22% yields. The BOP sensors showed fluo-
rescence enhancement toward butylamine and several arylethylamines, whereas they showed fluorescence quenching toward secondary and
branched amines. The opposite fluorescence behavior is explained by an increased conformational restriction at the excited state, at which
a proton transfer complex between the host and guest forms that is stabilized in a tripodal hydrogen bonding mode. This is the first example
in which fluorescence enhancement is observed in amine sensing with phenolic fluorophores. Enantiomeric a-chiral organoamines were also
sensed with different fluorescent intensity changes by Ph-BOP, complementing the previous tris(oxazolines) that sense enantiomeric a-chiral
organoammonium ions.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
in which the acidic hydroxyl group is expected to act as a hy-
drogen bond donor toward amine guests. With this hydrogen
bond donor together with the two oxazoline hydrogen bond
Among various sensing techniques available for clinical,
biological, and environmental analyses, fluorescence sens-
ing has advantages of high sensitivity and compatibility
for the online and real-time analyses.1 Amines are an impor-
tant class of biologically active compounds among those to
be sensed by fluorescence analysis. Amines as their neutral
forms are generally sensed with fluorescence sensors based
on binol or other phenolic fluorophores, mostly in the fluo-
rescence quenching mode.1f,2 In some cases, amines are
sensed as their ammonium salts, which results in the fluores-
cence enhancement through the photo-induced electron-
transfer mechanism.3 Recently, we have introduced novel
tris(oxazoline) receptors 1, which selectively recognize cer-
tain organoammonium ions and also show unusual chiral
discrimination toward a-, b-, and a,b-chiral organoammo-
nium ions (Fig. 1).4 Interestingly, the receptors also sense
organoammonium ions with fluorescence enhancement.4d
The enhancement was attributed to the conformational
restriction of the tripodal ligands upon guest binding. As
an extension of this novel sensing mechanism, we were
interested in an analogous tripodal system that may recog-
nize organoamines in their neutral forms. Thus, we designed
a phenol-containing bis(oxazoline) system such as 2 (Fig. 1),
R
R
R
N
R
O
R
N
N
O
N
HO
N
O
O
O
2a: R = Ph (Ph-BOP)
2b: R = Me (Me-BOP)
1a: R = Ph (Ph-BTO)
1b: R = Me (Me-BTO)
Ph
Ph
N
O
N
MeO
O
3
R'
N
R'
N
R
R
δ+
R
R
O
N
N
H
H
H
H
δ-
O
N
O
N
H
HO
O
O
I
II
Keywords: Amine sensing; Fluorescence sensors; Hydrogen bonding; Chiral
sensing.
* Corresponding author. Tel.: +82 54 2792105; fax: +82 54 2793399;
e-mail: ahn@postech.ac.kr
Figure 1. Structures of tripodal oxazolines 1, phenol-containing bis(oxazo-
lines) 2, their amine inclusion complexes I (at the ground state) and II (at the
excited state), and a reference compound 3.
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.09.060

11646
Y. M. Chung et al. / Tetrahedron 62 (2006) 11645–11651
acceptors, the new receptors may recognize an amine guest
in a tripodal hydrogen bonding mode, which is reminiscent
of the tripodal hydrogen bonding between tris(oxazolines)
1 and organoammonium ions. The recognition of amines
by the phenol-containing receptors 2 in the tripodal fashion
is expected to generate a proton transfer complex such as II
at the excited state, because the phenol at the excited state
becomes much more acidic than its ground state (phenol:
pKa*¼4.1 vs pKa¼9.0).5 Thus, the proton transfer complex
is supposed to be conformationally more rigid compared
to its ground state complex I. Therefore, the receptor’s fluo-
rescence property is expected to change upon guest binding,
hopefully to a fluorescence enhancement mode. Described
herein are an efficient synthesis of bis(oxazolinyl)phenols
(BOPs) 2 and an investigation of their fluorescence behavior
toward various organoamines.
constructed by double condensation of amino alcohols
with nitrile group of 9 in the presence of cadmium acetate
as a mild Lewis acid catalyst10 to afford the BOP sensors 2
(Scheme 1).
2.2. Fluorescence studies
The UV–vis spectra of Ph-BOP 2a in acetonitrile showed ab-
sorption maxima at 196 and 276 nm at 0.1 mM concentra-
tion. The fluorescence emission spectra displayed emission
of the biphenyl nucleus at lmax¼300 nm when excited at
276 nm. Similarly, Me-BOP 2b showed absorption maxima
at 276 nm and the fluorescence emission at lmax¼305 nm
when excited at 275 nm in acetonitrile (at 0.1 mM concen-
tration). The emission wavelengths observed were shorter
than that of o-phenylphenol11 because little conjugation is
expected between the two phenyl rings.
2. Results and discussion
2.1. Synthesis
When Ph-BOP 2a (0.1 mM in CH3CN) was titrated with in-
creasing amount of 2-(3,4-dimethoxy)phenylethylamine
(Am4) as a guest at 25 ꢀC, about 2-fold increase in the fluo-
rescence intensity of the sensor was observed (Fig. 2a). This
is an interesting result because so far fluorescence sensing
of amines with phenol- or binaphthol-derived sensors re-
sulted in fluorescence quenching rather than enhancement.1f
Also, amines are efficient fluorescence quenchers of most
unsubstituted aromatic hydrocarbon fluorophores such as
anthracene, perylene, and carbazole.1b The fluorescence en-
hancement was further confirmed by titrating Ph-BOP 2a
with a non-fluorescent guest such as butylamine (Fig. 2b).
Similarly, Me-BOP 2b also gave a fluorescence enhance-
ment upon titrating with butylamine (Fig. 2c). Although
the enhancement is not large, as far as we know, this is the
first example of neutral amine sensing with phenol-contain-
ing sensors in the fluorescent enhancement mode. We eval-
uated other amine guests that have different structural
features, which are listed in Figure 3. The guests such as
2-phenylethylamine (Am1), 4-methoxyphenylethylamine
(Am3), and tryptamine (Am5) also exhibited fluorescence
enhancement. It should be noted that in the case of the guests
containing aromatic fluorophores such as Am1 and Am3–
Am5, the fluorescence from the guest itself overlapped
Our synthetic efforts started from commercially available
mesitylene. A controlled bromomethylation of mesitylene
with paraformaldehyde and HBr in acetic acid6 and subse-
quent nucleophilic substitution of the resulting bis(bromo-
methyl) compound 5 by cyanide afforded the bis(nitrile)
compound 6. Bromination of this compound with
7
Me3(PhCH2)NBr3/ZnCl2 gave the bromo compound 7. It
is worth mentioning that attempts to prepare 7 directly
from 2-bromomesitylene via the above sequence, namely
bromomethylation and substitution, failed as the bromo-
methylation of 2-bromomesitylene stopped at the mono-bro-
momethylated stage even under severe reaction conditions.
This could be probably owing to the deactivating effects of
the bromine and bromomethyl group toward the electro-
philic substitution on the benzene ring. Next, the Suzuki
coupling8 was employed to construct the carbon–carbon
bond of the biphenyl core. Thus, the coupling of 7 with
5,5-dimethyl-2-(2-methoxyphenyl)-1,3,2-dioxaborinane9
yielded methyl ether 8, which on subsequent deprotection
with BBr3 afforded 9. Finally, the oxazoline ring was
Br
Br
CN
CN
b)
a)
c)
4
5
6
CN
CN
OH
CN
CN
CN
CN
d)
e)
f)
2
O
R
B
O
OCH3
Br
HO
NH2
OCH3
7
8
9
f)
R
3
HO
NH2
Scheme 1. Synthesis of compounds 2 and 3. Reagents and conditions: (a) paraformaldehyde, 30% HBr in AcOH, AcOH, 80 ꢀC, 5 h, 95%; (b) NaCN, MeOH/
H2O (5:2), reflux, 5 h, 76%; (c) Me3(PhCH2)NBr3, ZnCl2, AcOH, 70 ꢀC, 6 h, 75%; (d) 5 mol % Pd(PPh3)4, K3PO4, DMF, 100 ꢀC, 24 h, 86%; (e) BBr3, CH2Cl2,
25 ꢀC, 48 h, 76%; (f) Cd(OAc)2, PhCl, reflux, 96 h, 58% (R¼Ph, 2a), 32% (R¼Me, 2b), 55% (R¼Ph, 3).

Y. M. Chung et al. / Tetrahedron 62 (2006) 11645–11651
11647
OMe
OCH3
OCH3
(a)
(b)
(c)
10
8
HN
2.1
1.8
1.5
1.2
0.9
6
NH2
Am1
NH2
NH2
Am4
NH2
Am5
NH2
Am2
Am3
4
2
0
1
3
4
[Am4]/[2a]
2
NH2
0
CO2Me
R/S-Am10
H2N
OH
NH2
R/S-Am9
Me
NHMe
Am6
NH2
NH2
Am7
R/S-Am11
Am8
300
350
400
450
500
wavelength (nm)
Figure 3. Chemical structures of the amine guests studied.
4
3
2
1
0
1.5
1.4
1.3
1.2
1.1
1.0
270
300
330
360
390
0
1
2
3
4
wavelength (nm)
[Am1]/[2a]
Figure 4. A plot of F/F0 versus [Am1]/[2a].
4
3
2
1
0
In contrast to the b-arylethylamine or butylamine guests
(Am1–Am5), interestingly, fluorescence quenching was
observed for other types of guests such as Am6–Am11.
For example, titrations of Ph-BOP 2a with Am6 resulted
in fluorescence quenching (Fig. 5), and a similar level of
quenching was observed in the case of other guests. How
can we rationalize such an opposite result depending on
the guests? We suppose that the fluorescence enhancement
or quenching observed is dependent on the guest’s binding
affinity, which is related to the fluorescence enhancement
mechanism. Previously we have shown that the ammonium
300
350
400
450
wavelength (nm)
5
4
3
2
1
0
Figure 2. Fluorescence changes measured in CH3CN at lex¼276 nm: (a) ti-
tration of Ph-BOP 2a (0.1 mM) with Am4 (0.0–4.0 equiv from the bottom)
(inset: dependence of fluorescence intensity (F/F0) depending on the ratio
[Am4]/[2a]); (b) titration of Ph-BOP 2a (0.1 mM) with Am2 (0.0–6.2 equiv
from the bottom); (c) titration of Me-BOP 2b (0.1 mM) with Am2 (0.0–
10 equiv from the bottom) with lex¼275 nm.
with that of the host–guest complex. However, the fluores-
cence enhancement due to the complexation was apparent,
as inferred from a non-linear increase of the fluorescence
intensity in the plot of F/F0 versus [G]0/[H]0 (Fig. 4). The
fluorescence intensity increases rather rapidly up to the
equivalent point and then slows down upon titration of 2a
with Am1, which can be interpreted in this way: at the initial
stage, the fluorescence increase results from both the host–
guest complex and guest, while in the later stage, the in-
crease is mainly due to the guest added.
300
350
400
450
500
wavelength (nm)
Figure 5. Fluorescence changes upon titration of Ph-BOP 2a (0.1 mM) with
Am6 (0.0–6.0 equiv from the top) in acetonitrile at lex¼276 nm.

11648
Y. M. Chung et al. / Tetrahedron 62 (2006) 11645–11651
ions of linear primary amines Am1–Am5 are bound more
strongly to the tris(oxazoline) receptors 1 than those of
branched or secondary amines Am6–Am11, because the
former guests experience less steric strain from the host’s
4-oxazolinyl substituents (Ph or Me in the case of Ph-BTO
and Me-BTO, respectively) in the inclusion complexes
than the latter guests.4 Similarly, the steric strain between
BOPs 2 and the guests can be expected under the tripodal hy-
drogen bonding mode (Fig. 6), and under this reasoning, we
can expect that BOPs 2 would experience less steric strain in
binding the linear organoamines than the others such as the
branched guests. In the case of the secondary amines such
as Am6, BOPs 2 cannot form the tripodal hydrogen bonding
suggested. The molecular interactions between Ph-BOP 2a
hydrogen bonding is not strong enough, we may not expect
the enforced conformational restriction and thus little fluo-
rescence enhancement would result. Instead, we may ob-
serve fluorescence quenching because generally amines are
fluorescence quenchers toward the phenol-based fluoro-
phores. In this way, the observed fluorescence behavior of
BOPs toward the amines can be explained.
To get a better understanding for the structural factors that
are responsible for the fluorescence enhancement observed,
we further examined the fluorescence behavior with refer-
ence compounds. We performed the fluorescence titration
of Am1 with the reference compound 3, in which the pheno-
lic hydroxyl group is protected as its methyl ether, under
otherwise identical conditions. As expected, fluorescence
quenching was observed in this case upon increasing the
amount of Am1 (Fig. 7a), which indicates that the hydroxyl
group in BOPs 2 is essential for the fluorescence enhance-
ment. As discussed above, the hydroxyl group of Ph-BOP
2a and an amine guest form a proton transfer complex at
the excited state, which is stabilized through the hydrogen
bonding by the oxazoline ligands. When the hydroxyl group
is blocked, no such proton transfer complex would form, and
thus no fluorescence enhancement is expected, as observed
in the case of 3.
1
and amines have been studied by H NMR titrations. Al-
though the binding induced chemical shifts were small, cer-
tain proton peaks (particularly, the oxazoline ring protons)
shifted to downfield upon guest binding (Dd¼0.025 ppm
when 2 equiv of Am1 were added).
A direct and straightforward consequence of our presump-
tion should be that phenol 9, in which two oxazoline ligands
are replaced by nitrile groups, should show fluorescence
quenching toward amines, as the nitrile groups cannot act
as hydrogen bond acceptors to stabilize the host–guest com-
plex. Indeed, fluorescence quenching was observed when
phenol 9 was titrated with Am1 under otherwise identical
conditions (Fig. 7b).
These observations clearly augment our presumption to be
true and also highlight the key role of the oxazoline ligands
in the fluorescence enhancement observed for amines Am1–
Am5. Moreover, the absorption spectra measured during the
titration of Ph-BOP 2a with Am1 showed no appreciable
change, which indicates that the enhanced fluorescence is
not due to an increased absorption.
Figure 6. An energy minimized structure of the inclusion complex between
Ph-BOP 2a and Am4.
As noted above, the acidity of the phenolic ligand would in-
crease at the excited state and thus proton transfer complexes
would form when the receptors bind amines. The proton
transfer complexes should be tighter and thus conformation-
ally more rigid than the host–guest complexes at the ground
state. This enhanced conformational restriction is believed
to contribute to the fluorescence enhancement observed.
Such an explanation is based on the tripodal hydrogen bond-
ing mode between the host and guest. When the tripodal
The fluorescence enhancement resulting from the conforma-
tional restriction alone, however, seems to be moderate, such
as a few times enhancement as observed previously by us.4d
In this conformational restriction model, the hydrogen bond-
ing interactions are also believed to reduce the charge
(a)
(b)
5
5
4
3
2
1
0
4
3
2
1
0
300 350 400 450 500 550
wavelength (nm)
270
300
330
360
390
wavelength (nm)
Figure 7. Fluorescence changes measured in CH3CN: (a) titration of reference compound 3 (0.1 mM) with Am1 (0.0–4.2 equiv from the top) at lex¼284 nm;
(b) titration of phenol 9 (1.0 mM) with Am1 (0–10 equiv from the top) at lex¼287 nm.

Y. M. Chung et al. / Tetrahedron 62 (2006) 11645–11651
11649
separation of the excited state host–guest complex, thereby
suppressing some possible quenching processes.12 At this
point, it is premature to speculate on the nature of the
host–guest complex at the excited state, but it is certain
that the degree of the proton transfer between the phenolic
host and an amine is reduced by the ‘backside’ hydrogen
bonding provided by the two oxazoline ligands, which re-
sults in the fluorescence modulation of the phenolic fluoro-
phore toward amines, from quenching to enhancement, as
observed for the amines Am1–Am5.
a-substituents of a-chiral amines. The steric interactions
between the oxazolinyl phenyl substituents and the guest’s
a-substituents would be different between the enantiomers,
which would lead to different fluorescence properties, as
we pointed out above in the cases of different achiral amines.
Indeed, we were able to observe substantial differences in
the fluorescence behavior between the enantiomeric a-chiral
amines examined. For example, when Ph-BOP 2a was ti-
trated with each of the enantiomeric a-methylbenzylamines
(Am9), fluorescence quenching resulted in both cases but
with different magnitude. Plots of F/F0 depending on
[Am9]/[2a] values for each of the enantiomeric guests
show significant differences in the F/F0 values between the
enantiomers (Fig. 8). Similarly, the other two guests also
showed different fluorescence quenching behaviors depend-
ing on enantiomeric guests.
The association constants for the complex formation of
Ph-BOP 2a with some amine guests can be obtained from
the observed fluorescence data. On the assumption that
host 2a (H) and a guest amine (G) interact in a 1:1 fashion,
the non-linear least squares fit13 by the following equation14
gives the association constants K11,
1.0
F=F0 ꢁ ðkG=kHÞ½Gꢂ ¼ 1 þ ðk11=kHÞK11½Gꢂ=f1 þ K11½Gꢂg
(R)-Am9
(S)-Am9
where F0 is the fluorescence intensity of free host and F is
the observed fluorescence intensity during titration, which
can be expressed as F¼kG[G]+kH[H]+k11[HG]. The kG/kH
value can be obtained by measuring concentration-depen-
dent fluorescence changes for the guest and host separately
under the same conditions. Association constants thus ob-
tained were in the range of 102–103 Mꢁ1 (Table 1), which
are much less than that obtained with tris(oxazoline) recep-
tors 1 (log K11¼6.65 Mꢁ1 for n-BuNH+3, determined by
UV–vis titration).4a This is not an unexpected result because
0.9
0.8
0.7
0
2
4
6
8
10
[R/S-Am9]/[2a]
Figure 8. A plot of F/F0 versus [Am9]/[2a] depending on the enantiomeric
guest.
Table 1. Association constants K11 and Stern–Volmer constants KSV deter-
mined from the fluorescence titration of Ph-BOP 2a with the amines in
acetonitrile at 25 ꢀC
Therefore, Ph-BOP 2a can be used as a fluorescence sensor
for the monitoring of enantiomeric purity of certain amines
in their neutral forms, which complements the enantiomeric
sensing of a-chiral organoammonium ions by its tris(oxazo-
line) analogue 1a.
Am1
2100
Am2
330
Am3
2440
Am4
3200
Am5
7490
K11 (Mꢁ1
)
Am6
1290
Am7
930
Am8
1080
(R)-Am9
1290
(S)-Am9
770
KSV (Mꢁ1
)
neutral hydrogen bonding in the case of BOPs 2 should be
much weaker than the charged hydrogen bonding between
an ammonium ion and BTOs 1. The association data indicate
that the fluorescence enhancement is related to the associa-
tion constant: as in the case of the tris(oxazoline) receptor
Ph-BTO 1, the b-aryl-substituted linear amine guests
(Am1 and Am3–Am5), which show larger fluorescence en-
hancement compared to an aliphatic guest Am2, also gave
larger association constants. In the case of amine guests
Am6–Am10, which are believed to have poor binding affin-
ities, we were not able to determine meaningful association
constants due to large errors in the non-linear regression
process. Instead, the Stern–Volmer constants were obtained
for these weakly binding amine guests, which showed linear
relationships for the equation: F0/F¼1+KSV[G].1b
3. Conclusion
In conclusion, we have synthesized the phenol-containing
bis(oxazolines) as novel fluorescence sensors toward pri-
mary amines. The sensors showed fluorescence enhance-
ment toward some linear primary amines and fluorescence
quenching toward branched and secondary amines. The
observed fluorescence enhancement is ascribed to the con-
formational restriction and modulation of charge separation
of the excited state host–guest complex through hydrogen
bonding by the well-directed oxazoline ligands. Although
the observed fluorescence enhancement is moderate, it
should be noted that this is the first example of amine sens-
ing by phenolic fluorophores in the fluorescence enhance-
ment mode. We have also shown that Ph-BOP 2a is
capable of enantio-discriminating a-chiral amines and thus
may be useful for the determination of their enantiomeric
purities. A further study on the structure modification of
the present system toward a colorimetric sensor of amines
is underway.
2.3. Enantio-discrimination
We reasoned that Ph-BOP 2a might discriminate a-chiral
amines because the two phenyl-substituted oxazolines could
provide a pseudo-C2-symmetric chiral environment toward

11650
Y. M. Chung et al. / Tetrahedron 62 (2006) 11645–11651
4. Experimental
(5 mL) was stirred at 100 ꢀC for 24 h. The resulting mixture
was filtered through Celite, washed with hot ethyl acetate,
and then the filtrate was concentrated to yield a black residue.
Purification by column chromatography gave 8 (0.26 g,
86%): mp 134 ꢀC; 1H NMR (300 MHz, CDCl3, Me4Si)
d 2.0 (s, 6H), 2.1 (s, 3H), 3.7 (s, 7H), 7.0–7.1 (m, 3H), 7.3–
7.4 (m, 1H); 13C NMR (75 MHz, CDCl3, Me4Si) d 17.5,
18.7, 19.6, 56.1, 111.6, 118.2, 121.6, 126.9, 129.7, 130.0,
131.4, 135.0, 136.6, 139.0, 157.1; MS (EI) m/z 304.21
(M+); HRMS calcd for C20H20N2O (M+) required 304.1576,
found 304.1580. Anal. Calcd for C20H20N2O: C, 78.92; H,
6.62; N, 9.20. Found: C, 78.48; H, 6.61; N, 9.30.
4.1. General
2,4-Bis(bromomethyl)-1,3,5-trimethylbenzene,6
methyl-2-(2-methoxyphenyl)-1,3,2-dioxaborinane,9
5,5-di-
and
benzyl(trimethyl)ammonium tribromide7 were prepared ac-
cording to the literature procedures. All other chemicals
were commercially available and used without further purifi-
cation. The solvents for dry reactions were dried with appro-
priate desiccants and distilled prior to use. Unless otherwise
mentioned, all NMR spectra were recorded in CDCl3 solu-
tion containing tetramethylsilane as internal standard. Chem-
ical shifts are reported in d unit. For column chromatography
silica gel of 230–400 mesh was used. Fluorescence spectra
were recorded on Photon Technical International Fluores-
cence system. Fluorescence experiments were carried out
in 10-mm quartz cuvette at room temperature. Fluorescence
titrations were performed at 0.1 mM concentration and the
required guest solutions (10–30 mM) are prepared from ap-
propriate amount of amines in 0.1 mM host solution. Both
excitation and emission slit widths were 2 nm.
4.1.4. [(5-Cyanomethyl-20-hydroxy-2,4,6-trimethyl)bi-
phenyl-3-yl]acetonitrile (9). To a solution of 8 (0.3 g,
1.0 mmol) in dichloromethane was charged BF3$OEt2
(8.0 mL, 8.0 mmol) (1.0 M in dichloromethane) at 0 ꢀC,
and the mixture was allowed to stir at room temperature for
48 h. An aqueous workup and subsequent purification by col-
umn chromatography gave 9 (0.22 g, 76%); mp 202 ꢀC. 1H
NMR (300 MHz, CDCl3, Me4Si) d 2.1 (s, 6H), 2.5 (s, 3H),
3.8 (s, 4H), 6.9–7.0 (m, 3H), 7.28–7.31 (m, 1H); 13C NMR
(75 MHz, CDCl3, Me4Si) d 17.5, 18.5, 19.7, 116.4, 117.7,
121.9, 127.3, 128.1, 130.2, 130.8, 136.3, 136.4, 137.7; MS
(EI) m/z 290.19 (M+); HRMS calcd for C19H18N2O (M+)
required 290.1419, found 290.1420. Anal. Calcd for
C19H18N2O: C, 78.59; H, 6.25; N, 9.65. Found: C, 76.75;
H, 6.19; N, 9.68.
4.1.1. [(3-Cyanomethyl-2,4,6-trimethyl)phenyl]aceto-
nitrile (6). A mixture of 2,4-bis-(bromomethyl)-1,3,5-tri-
methylbenzene (5) (3.06 g, 10 mmol), sodium cyanide
(1.47 g, 30 mmol) in water (20 mL) and methanol (50 mL)
was refluxed for 5 h. The solvent was evaporated and diluted
with cold water (100 mL) to give the product as a precipitate,
which was filtered and washed with water (1.5 g, 76%). An
analytically pure sample was obtained by recrystallization
from dichloromethane/hexane: Rf ¼0.3 (dichloromethane/
hexane¼4/1); mp 169 ꢀC; 1H NMR (300 MHz, CDCl3,
Me4Si) d 2.3 (s, 6H), 2.4 (s, 3H), 3.7 (s, 4H), 7.0 (s, 1H);
13C NMR (75 MHz, CDCl3, Me4Si) d 16.5, 18.4, 117.3,
126.5, 131.1, 135.6, 136.9; MS (EI) m/z 198.10 (M+);
HRMS calcd for C13H14N2 (M+) required 198.1157, found
198.1155. Anal. Calcd for C13H14N2: C, 78.75; H, 7.12; N,
14.13. Found: C, 78.84; H, 7.20; N, 13.89.
4.1.5. [20,40,60-Trimethyl-30,50-bis(4-phenyl-4,5-dihydro-
oxazol-2-yl)methyl]biphenyl-2-ol (Ph-BOP, 2a). A mixture
of bis(nitrile) 9 (0.29 g, 1.0 mmol), (S)-(+)-2-phenylglycinol
(0.54 g, 4.0 mmol), and Cd(OAc)2 (0.013 g, 0.25 mmol) in
chlorobenzene was refluxed for 4 days while a mild stream
of nitrogen was bubbled through the reaction mixture.
Evaporation of the solvent gave a black residue, which
was purified by column chromatography (gradient elution:
10%/60% ethyl acetate in hexane) to give 2a (0.3 g,
1
58%): mp 90 ꢀC; [a]D25 +87.3 (c 0.01, CHCl3); H NMR
(300 MHz, CDCl3, Me4Si) d 2.05 (s, 3H), 2.08 (s, 3H), 2.4
(s, 3H), 3.8 (s, 4H), 4.0–4.1 (m, 2H), 4.6–4.7 (m, 2H), 5.0–
5.1 (m, 2H), 6.9–7.0 (m, 4H), 7.1–7.3 (m, 12H); 13C NMR
(75 MHz, CDCl3, Me4Si) d 17.1, 18.0, 19.0, 29.9, 69.3,
69.4, 75.2, 75.5, 116.2, 120.1, 126.5, 127.6, 127.7, 128.1,
128.8, 128.89, 128.93, 130.2, 131.1, 135.3, 136.5, 142.3,
142.5, 154.1, 167.7; MS (EI) m/z 530.39 (M+); HRMS calcd
for C35H34N2O3 (M+) required 530.2569, found 530.2560.
Anal. Calcd for C35H34N2O3: C, 79.22; H, 6.46; N, 5.28.
Found: C, 77.07; H, 6.69; N, 5.18.
4.1.2. [(3-Bromo-5-cyanomethyl-2,4,6-trimethyl)phenyl]-
acetonitrile (7). To a suspension of bis(nitrile) 6 (0.99 g,
5.0 mmol) in AcOH (1.7 mL) were charged benzyl(tri-
methyl)ammonium tribromide (1.47 g, 5 mmol) and zinc
chloride (0.75 g, 5.5 mmol). The resulting mixture was
stirred at 70 ꢀC for 5 h, and then cooled and poured into
crushed ice containing 5% sodium hydrogen sulfate. The
precipitate was filtered, washed with excess water, and re-
crystallized from ethyl acetate/hexane to yield 7 (1.05 g,
76%) as white needles: Rf¼0.3 (dichloromethane); mp
2.6 (s, 3H), 3.8 (s, 4H); 13C NMR (75 MHz, CDCl3, Me4Si)
d 17.5, 20.5, 22.3, 117.3, 128.4, 128.7, 135.0, 138.0; MS
(EI) m/z 276.13 (M+), 278.12 (M++2); HRMS calcd for
C13H13BrN2 (M+) required 276.0262, found 276.0262.
Anal. Calcd for C13H13BrN2: C, 56.34; H, 4.73; N, 10.11.
Found: C, 55.84; H, 4.68; N, 10.05.
1
185 ꢀC; H NMR (300 MHz, CDCl3, Me4Si) d 2.4 (s, 6H),
4.1.6. [20,40,60-Trimethyl-30,50-bis(4-methyl-4,5-dihydro-
oxazol-2-yl)methyl]biphenyl-2-ol (Me-BOP, 2b). Pre-
pared from 9 as above in 32% yield: mp 133 ꢀC; [a]D25
+11.2 (c 0.01, CHCl3); 1H NMR (300 MHz, CDCl3,
Me4Si) d 1.2 (q, 6H), 2.0 (s, 6H), 2.3 (s, 3H), 3.7 (s, 4H),
3.78–3.83 (m, 2H), 4.07–4.12 (m, 2H), 4.3–4.4 (m, 2H),
6.9–7.0 (m, 3H), 7.2–7.4 (m, 1H); 13C NMR (75 MHz,
CDCl3, Me4Si) d 17.2, 18.3, 18.4, 22.17, 22.2, 30.3, 61.7,
61.8, 75.0, 116.7, 120.5, 128.7, 129.4, 130.7, 131.69,
131.73, 136.8, 154.7, 166.6; MS (EI) m/z 406.27 (M+);
HRMS calcd for C25H30N2O3 (M+) required 406.2256,
found 406.2253. Anal. Calcd for C25H30N2O3: C, 73.86;
H, 7.44; N, 6.89. Found: C, 69.94; H, 7.67; N, 6.77.
4.1.3. [(5-Cyanomethyl-20-methoxy-2,4,6-trimethyl)bi-
phenyl-3-yl]acetonitrile (8). A mixture of bromide 7
(0.28 g, 1.0 mmol), 5,5-dimethyl-2-(2-methoxyphenyl)-
1,3,2-dioxaborinane (0.60 g, 2.75 mmol), Pd(PPh3)4
(0.05 g, 5 mol %), and K3PO4 (1.16 g, 5.5 mmol) in DMF

Y. M. Chung et al. / Tetrahedron 62 (2006) 11645–11651
11651
1138; (e) Jennings, K.; Diamond, D. Analyst 2001, 126, 1063;
(f) Reetz, M. T.; Sostmann, S. Tetrahedron 2001, 57, 1515.
3. (a) de Silva, A. P. Angew. Chem., Int. Ed. Engl. 1990, 29, 1173;
(b) Ballardini, R.; Balzani, V.; Credi, A.; Gandolfi, M. T.;
Hibert, F. K.; Lehn, J.-M.; Pordi, L. J. Am. Chem. Soc. 1994,
116, 5741.
4. (a) Kim, S.-G.; Ahn, K. H. Chem.—Eur. J. 2000, 6, 3399; (b)
Kim, S.-G.; Kim, K.-H.; Jung, J.; Shin, S. K.; Ahn, K. H.
J. Am. Chem. Soc. 2002, 124, 591; (c) Ahn, K. H.; Ku, H.-Y.;
Kim, Y.-S.; Kim, S.-G.; Kim, Y. K.; Son, H. S.; Ku, J. K.
Org. Lett. 2003, 5, 1419; (d) Ahn, K. H.; Kim, S.-G.; Kim,
K.-H.; Kim, Y. K.; Shin, S. K. J. Am. Chem. Soc. 2003, 125,
13819; (e) Kim, Y. K.; Lee, Y.-H.; Lee, H.-Y.; Kim, K. M.;
Cha, G. S.; Ahn, K. H. Org. Lett. 2003, 5, 4003; (f) Kim, J.;
Kim, S.-G.; Seong, H. R.; Ahn, K. H. J. Org. Chem. 2005,
70, 7227; (g) Kim, J.; Raman, B.; Ahn, K. H. J. Org. Chem.
2006, 71, 38.
4.1.7. {[3,5-Bis(4-phenyl-4,5-dihydrooxazol-2-yl)-
methyl]-20-methoxy-2,4,6-trimethyl}biphenyl (3). Pre-
pared from 8 similarly as above in 55% yield: mp 93 ꢀC;
[a]2D5 +74.37 (c 0.01, CHCl3); 1H NMR (300 MHz, CDCl3,
Me4Si) d 2.06 (s, 3H), 2.07 (s, 3H), 2.6 (s, 3H), 3.7 (s,
3H), 3.9 (s, 4H), 4.0–4.1 (m, 2H), 4.56–4.62 (m, 2H), 5.1–
5.2 (m, 2H), 7.0–7.1 (m, 4H), 7.0–7.3 (m, 12H); 13C NMR
(75 MHz, CDCl3, Me4Si) d 17.4, 18.3, 30.3, 55.6, 69.7,
75.0, 110.9, 120.8, 126.8, 126.9, 127.3, 127.6, 128.1,
128.5, 128.8, 129.1, 130.3, 131.1, 131.3, 135.4, 136.1,
137.1, 142.8, 156.9, 167.4; MS (EI) m/z 545.25 (M++1);
HRMS calcd for C36H36N2O3 (M+) required 544.2726,
found 544.2729. Anal. Calcd for C36H36N2O3: C, 79.38;
H, 6.66; N, 5.14. Found: C, 79.23; H, 6.57; N, 5.13.
Acknowledgements
5. Iwanek, W.; Mattay, J. J. Photochem. Photobiol. A: Chem.
1992, 67, 209.
6. van der Made, A. W.; van der Made, R. H. J. Org. Chem. 1993,
58, 1262.
This work was supported by the SRC program of KOSEF
(R11-2000-070-070010).
7. Kajigaeshi, S.; Kakinami, T.; Moriwaki, M.; Tanaka, T.;
Fujusaki, S.; Okamoto, T. Bull. Chem. Soc. Jpn. 1989, 62, 439.
8. Watanabe, T.; Miyura, N.; Suzuki, A. Synlett 1992, 207.
9. Chaumeil, H.; Signorella, S.; Le Derian, C. Tetrahedron 2000,
56, 9655.
Supplementary data
Supplementary data associated with this article can be found
in the online version, at doi:10.1016/j.tet.2006.09.060.
10. Witte, H.; Seeliger, W. Angew. Chem., Int. Ed. Engl. 1972, 11,
287; Gerisch, M.; Krumper, J. R.; Bergman, R. G.; Tilley, T. D.
J. Am. Chem. Soc. 2001, 123, 5818.
References and notes
´
11. Garcia Reyes, I. F.; Llorent Martınez, E. J.; Ortega Barrales, P.;
Molina Dıaz, A. Anal. Bioanal. Chem. 2004, 378, 429. The
1. (a) de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.;
Huxley, A. J. M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E.
Chem. Rev. 1997, 97, 1515; (b) Lakowicz, J. R. Principles of
Fluorescence Spectroscopy; Kluwer Academic, Plenum: New
York, NY, 1999; (c) Valeur, B.; Leray, I. Coord. Chem. Rev.
2000, 205, 3; (d) Prodi, L.; Bolletta, F.; Montalti, M.;
Zaccheroni, N. Coord. Chem. Rev. 2000, 205, 59; (e)
Fabbrizzi, L.; Licchelli, M.; Rabaioli, G.; Taglietti, A. Coord.
Chem. Rev. 2000, 205, 85;(f) Pu, L. Chem. Rev. 2004, 104, 1687.
2. (a) Yorozu, T.; Hayashi, K.; Irie, M. J. Am. Chem. Soc. 1981,
103, 5480; (b) Grady, T.; Harris, S. J.; Smyth, M. R.;
Diamond, D. Anal. Chem. 1996, 68, 3775; (c) Pugh, V. J.;
Hu, Q.-S.; Pu, L. Angew. Chem., Int. Ed. 2000, 39, 3638;
(d) Beer, G.; Rurack, K.; Daub, J. Chem. Commun. 2001,
´
maxima excitation/emission wavelengths of an aqueous solu-
tion of o-phenylphenol are 243/350 nm.
12. It is well known that the proton transfer complex between a
phenolic host and an amine guest results in fluorescence
quenching. The hydrogen bonding stabilization of the complex
II by the oxazoline ligands certainly reduces its charge separa-
tion and thereby modulates its quenching processes.
13. Computer software program Origin 7.0 was used.
14. (a) Eric, J.; Davies, D.; Ripmeester, J. A. Comprehensive
Supramolecular Chemistry; Pergamon: Oxford, UK, 1996;
(b) Conners, K. A. Binding Constants; Wiley: New York,
NY, 1987.