Rotational energy barrier of 2-(2′,6′-dihydroxyphenyl) benzoxazole: A case study by NMR

Article abstract of DOI:10.1021/jo201890f

2-(2′-Hydroxyphenyl)benzoxazole (HBO) derivatives represent an important class of luminescent materials, as they can undergo excited state intramolecular proton transfer (ESIPT). The material's ESIPT properties are dependent on the ratio of two different

Full text of DOI:10.1021/jo201890f

Article
pubs.acs.org/joc
Rotational Energy Barrier of 2-(2′,6′-Dihydroxyphenyl)benzoxazole:
A Case Study by NMR
Weihua Chen, Eric B. Twum, Linlin Li, Brian D. Wright, Peter L. Rinaldi, and Yi Pang*
Department of Chemistry & Maurice Morton Institute of Polymer Science, The University of Akron, Akron, Ohio 44325,
United States
S
* Supporting Information
ABSTRACT: 2-(2′-Hydroxyphenyl)benzoxazole (HBO) derivatives represent
an important class of luminescent materials, as they can undergo excited state
intramolecular proton transfer (ESIPT). The material’s ESIPT properties are
dependent on the ratio of two different rotamers, whose interconversion is
poorly understood. By using HBO derivative 4, the rotational energy barrier
of 2- (2′,6′-hydroxyphenyl)benzoxazole is determined to be 10.5 kcal/mol by variable-temperature NMR. Although a HBO deriv-
ative typically exhibits two rotamers with O···H−O (e.g., 1a) and N···H−O bonding (e.g., 1b), correlation of NMR with fluo-
rescence data reveals that the rotamer with N···H−O bonding is predominant in the solution.
Scheme 2. Schematic Illustration for ESIPT of HBO
Derivatives
INTRODUCTION
■
2-(2′-Hydroxyphenyl)benzoxazole (HBO, 1) has emerged to be
an interesting luminescent material that exhibits large Stokes’
shift arising from the excited-state intramolecular proton transfer
(ESIPT).1 The unique optical characteristics have resulted in
various applications including chemical sensors for zinc(II)2−5
and anions,6 light-emitting diode devices,7 and optical switching.8
Although the HBO molecule can exist in the intramolecular
hydrogen-bonded rotamers 1a and 1b (Scheme 1), only the
a
Scheme 1. Structures of HBO Derivatives
gives both enol and keto emission. Practically, it is desirable to
control the HBO in rotamer 1b for optimized keto emission, as
ESIPT is currently emerging as one of the appealing new mecha-
nisms for chemosensors.12 Unfortunately, our knowledge about
the relative population of these rotamers in the sample is very
limited, except a few scattered cases of studies being reported
from crystal structures of HBO derivatives.11,6 Although a fluo-
rescence study shows that the relative population of 1b in
solution increases at low temperature,13 few attempts have been
a
The thick arrows indicate the H-bond locations in 3 and 4. The
hydrogen bond lengths and bond angles of 3b and 4 are from their
corresponding crystal structures.
latter is thought to undergo ESIPT.9,10 X-ray diffraction reveals
that the two rotamers 1a and 1b exist in about a 1:1 ratio in the
crystalline state,11 with the hydroxyl group pointing to either
the N- or O-atom side of the oxazole ring. Significant interest
exists in elucidating the ESIPT process (Scheme 2) and to
discover the underlying principles that govern the rotamer ratio
1a and 1b and their conversion. Fluorescence of a HBO typically
Received: September 13, 2011
Published: November 21, 2011
© 2011 American Chemical Society
285
dx.doi.org/10.1021/jo201890f | J. Org. Chem. 2012, 77, 285−290

The Journal of Organic Chemistry
Article
made to characterize the rotamer population in solution. Lack
of a spectroscopic method to identify the specific rotamer(s), in
addition to the absence of a suitable model system, makes the
problem a challenging issue.
significant role in determining the rotamer ratio of HBO. The
crystal structure of 4 showed a planar geometry with concom-
itant N···H−O and O···H−O bonding. Analysis of hydrogen
bond length showed that the N···H distance in the N···H−O
bonding of 3 (1.815 Å) was shorter than that of 4 (1.903 Å), as
the competitive O···H−O bond weakened the N···H−O bond
in 4.
Conversion between 1a and 1b occurs via rotating the bond
between 2-hydroxyphenyl and oxazole units. Quantum calcula-
tions have predicted the interconversion barrier to be ∼2.5
kcal/mol (by CDDO/S-CI)13 or ∼15 kcal/mol (by ab initio
with the STO-3G basis set).14 The lack of convenient methods
to distinguish between the two different hydrogen-bonding
modes (i.e., O···H−O and N···H−O), in addition to their
reversible nature, makes it difficult to experimentally evaluate
the rotation barrier in solution. Extending our general interest
in exploring the fluorescence of ESIPT,4,6,15,16 we now report
the rotational barrier of 2-(2′,6′-dihydroxyphenyl)benzoxazole
4, which is determined by variable-temperature NMR. In
conjunction with crystal structure and fluorescence spectroscopy,
1H NMR of HBO compounds 1 and 3 also reveals that the
rotamer ratio found in the crystalline sample could be very
different from that in solution. This study thus sheds some light
on the identity of HBO rotamer in solution, as most ESIPT-
based dyes are expected to be used in the solvent.
Fluorescence of HBO 1 in solution is known to give enol and
keto emission, which are related to the relative population of 1a
and 1b. Only rotamer 1b can undergo ESIPT and contribute to
keto emission. In contrast, rotamer 1a and some “solvated 1b”
are thought to give enol emission. The fluorescence of 4 re-
vealed only keto emission in methylene chloride (Figure 2).
In methanol, negligible enol emission became visible from 4,
possibly due to the formation of “solvated HBO”. It was
assumed that the enol emission from 1 and 3 were mainly attri-
buted to the rotamers 1a and 3a, respectively. Although the
crystalline 3 was found to have no rotamer 3a, its enol emission
in solution was relatively higher in intensity than 1. Assuming
that the radiative decay of the enol and keto tautomers are
similar, the relative enol content of a HBO molecule could be
represented approximately by its enol emission intensity. The
higher enol emission thus indicated that the content of rotamer
3a in the solution of 3 was higher than that of 1a in the solution
of 1. The result also suggests that the rotamer ratio found in the
solid state could be very different from the ratio present in
solution.
RESULTS AND DISCUSSION
■
The HBO derivatives 3 and 4 were synthesized as reported
previously.15 Interestingly, the crystal structure of 3 revealed
only one rotamer with the hydroxyl group pointing to the
N-atom in the solid (Figure 1). The result is in sharp contrast
Rotational Energy Barrier by Variable-Temperature
1H NMR. Unlike 3, rotation of the hydroxyphenyl fragment
in 4 gives the same structure. This feature warrants that the
proton in the O−H···N bond has an equal population as the
proton in the O−H···O bond. It was assumed that the protons
in the O−H···N and O−H···O bonding structures have con-
1
siderably different chemical shifts, as H NMR is sensitive to
the hydrogen bonding environment.17 Equal proton popula-
tions in O−H···N and O−H···O environments, in addition to a
large difference in chemical shift, simplified the task of studying
the rate of rotation by low-temperature NMR.18
1
The H NMR spectrum of 4 in CDCl3 at 30 °C gave one
broad peak at ∼9.8 ppm (Figure 3a), which was attributed to
the phenolic protons. Apparently, at temperatures higher than
the coalescence temperature, “fast rotation” occurred and an average
signal was observed for the two protons in the different H-bonding
environments. As the temperature was decreased, the peak be-
came broad. Two peaks emerged when the temperature was below
−40 °C. In this case, at temperatures below the coalescence point,
the bond rotation became slow compared to the NMR time
scale. The resonance signals at ∼12.2 and 7.9 ppm were assigned
to O−H···N and O−H···O protons, respectively.
Spectral simulations (Figure 3b) were performed using
WINNMR-pro,19 which matched well with the experimental data.
Determination of the coalescence temperature (Tc = −20 °C =
253K), in addition to the difference in chemical shift of the two
Figure 1. ORTEP plot of the crystals 3 (top) and 4 (bottom).
to crystalline 1, where two rotamers are found in 1:1 ratio.11
The results indicated that the crystal packing could play a
286
dx.doi.org/10.1021/jo201890f | J. Org. Chem. 2012, 77, 285−290

The Journal of Organic Chemistry
Article
Figure 2. Normalized emission of 1 (HBO), 3 (MHBO), and 4 (DHBO) in CH2Cl2 (a) and methanol (b).
Figure 3. (a) 400 MHz variable-temperature 1H NMR spectra of 4 (concertration: 25 mM in CDCl3). The hydrogen-bonded protons were observed
by starting from the fast exchange at 30 °C, passing through the coalescence temperature near −20 °C, and ending in the slow exchange at −40 °C
or below. The signal at 7.25 ppm was attributed to residual CHCl3 in the deuterated solvent. (b) Spectral simulations from dynamic NMR.
H-bonded protons, allowed calculation of the Gibbs free activa-
tion energy from the Eyring equation:
Information), it was estimated that kc could be determined with
a relative error of 10%.
Chemical Shift of O−HN and O−HO. Two
phenolic protons of 4 at different chemical shifts (at 7.9 and
12.2 ppm) were attributed to different hydrogen-bonding environ-
ments, i.e., O···H−O and N···H−O bonding, respectively.
Compound 3 can be used as a model compound to aid the assign-
ments. As shown in Figure 1, only rotamer 3b with N···H−O
bonding was observed in the crystal structure of 3. The 1H NMR
spectrum of 3 in CDCl3 revealed a sharp peak at 12.7 ppm (Figure
S1, Supporting Information), attributed to the proton in the
N···H−O bonding rotamer. Therefore, the resonances of 4 at
12.2 and 7.9 ppm could be attributed to the N···H−O and
O···H−O protons, respectively. The assignment was consistent
with the chemical shift reported in benzofuran 5, which can
be regarded as a close model compound for the proton in the
O···H−O bonding rotamer.20
⧧
ΔG = 2.303RT(10.319 + log T − log k )
c
c
=0.0191 × T (9.97 + log T − log (ν − ν ))
c
c
A
B
=0.0191 × 253[9.97 + log(253)
− log[400 × (12.2 − 7.9)]]
=44.25 kJ/mol = 10.59 kcal/mol
where νA and νB are chemical shifts in Hz, R is the gas constant,
Tc is the coalescence temperature (in K), and kc is the rate con-
stant at the coalescence temperature for interchange of rotamers
A and B. The rotational barrier was assumed to be associated
with breaking the intramolecular hydrogen bonds which hold
two aromatic fragments in coplanarity. Observation of the
same line broadening pattern at lower concentration (3.1 mM
in CDCl3) at variable temperature (Figure S6, Supporting
Information) further confirmed that the intermolecular hydro-
gen bonding was not a significant factor in determining the
rotation barrier. When the spectral data at coalescence was
compared with the simulated spectra (Figure S9, Supporting
287
dx.doi.org/10.1021/jo201890f | J. Org. Chem. 2012, 77, 285−290

The Journal of Organic Chemistry
Article
Scheme 3. Enol Rotamers of HBO in the Ground State
Figure 4. 400 MHz 1H NMR of 1 (concentration: 3.1 mM in CDCl3)
at 30 °C.
phenolic and Hb protons of 1 were as narrow as the solvent peak,
in sharp contrast to the significant peak broadening observed
1
from the H NMR spectrum of 4. Lack of enol emission in the
nonpolar solvent (e.g., CH2Cl2),15 in addition to the chemical
shift position, further supports the assignment of the signal at
∼11.45 ppm to 1b in chloroform. The results showed that the
tautomer ratio of 1a:1b (≈ 1:1) determined in the crystal
of 111 does not represent the ratio present in the solution
(predominantly 1b).
To further confirm the assignment for the rotamer 1b, the
model compound 7 (2-bromo-6-(4-methylbenzo[d]oxazol-2-yl)-
phenol) was prepared by using the procedure reported previously.22
The phenolic proton of 7 (Figure 5) gave a narrow resonance
at ∼11.5 ppm, in sharp contrast to that of 4 which produced a
Spectroscopic Identification of Rotamer and Enol and
Keto Emission. As shown in Scheme 2, the HBO molecule is
driven to the excited state upon absorption of photons. The
excited molecule can then return to its ground state to give enol
emission, or undergo ESIPT to give keto emission. The enol can
exist in different rotamers 6a and 6b (Scheme 3). It is generally
agreed that only rotamer 6b can undergo the ESIPT process
to give the keto emission. In the excited state of DHBO 4, the
deactivation pathways are present for both enol emission and
ESIPT, making the molecule a suitable probe to examine the com-
petitive process between enol and keto emission. Exclusive keto
emission was observed from DHBO 4 in CH2Cl2 (Figure 2a),
hexanes, and CH3CN,15 showing that the ESIPT of 6b was a
more competitive process than the radiative deactivation of
both enols 6a and 6b. In a protic solvent such as methanol, very
weak emission was observed from the methanol solution of 4
(Figure 2b), suggesting the presence of solvated enol species 6c
(in addition to regular enol 6a and 6b). Predominant keto emis-
sion from 4, however, indicated that the ESIPT process of 6b is
a much more competitive process, as ESIPT is a fast process, with
the rate constant of ∼150 fs being reported for HBO.21
Correlation between HBO structures and their enol/keto emis-
sion requires development of a strategy to identify the specific
rotamers present in solution. The compound 3 serves as a suitable
model, which is structurally similar to 4 but with only one hydroxy
group to give two rotamers. The phenolic proton signal at 12.7
1
ppm in H NMR indicated that 3 was present mainly in the
form of rotamer 3b. Negligible enol emission from 3 in CH2Cl2
(Figure 2a) was consistent with the assumption that rotamer 3b
was major. Although the keto emission remained to be pre-
dominant in methanol (a protic solvent), 3 gave noticeable enol
emission (Figure 2). Since the enol emission from 6b (= 3b if
R = −OCH3) was less competitive (as seen from 4), the observed
increase in enol emission could be mainly attributed to the
solvated enol 6c.
A similar trend has been observed in the emission of 1, which
reveals the predominant keto emission in a nonpolar solvent
such as hexane and CH2Cl2 (Figure 2a), but gives noticeable
enol emission in a protic solvent such as methanol.15 The
1H NMR of 1 in CDCl3 revealed the phenolic proton signal at
∼11.45 ppm (Figure 4), whose intensity and peak width was
not affected by varying the temperature between +50 and
−50 °C (Figure S7, Supporting Information). The peaks of
Figure 5. (Top) Structure of rotamers 7a and 7b, with proton−proton
1
distances marked on the arrows. (Bottom) 400 MHz H NOESYspec-
trum of 7′-MeHBO 7 in CDCl3 shows correlation between phenolic
proton Ha and aromatic proton Hb.
288
dx.doi.org/10.1021/jo201890f | J. Org. Chem. 2012, 77, 285−290

The Journal of Organic Chemistry
Article
broad −OH resonance (Figure 3). The 2D NOESY spectrum
showed that the −OH group (Ha at 11.55 ppm) had NOESY
correlations with the methyl group resonance (Hh at 2.63 ppm)
and the adjacent aromatic proton resonance (Hb at 7.02 ppm).
The observed NOESY correlation between −OH and methyl
protons confirmed the presence of the cis-enol form 7b in
solution. In rotamer 7a, the proton−proton distance between the
−OH (Ha) and methyl (Hh) hydrogens are calculated to be 6.6 Å,
which is too large to produce NOESY correlations.23 It should be
noted that the phenolic protons Ha in 7a and 7b exhibit NOE
effects on different protons. In the rotamer 7b, the Ha has an
NOE effect on the methyl proton Hh and aromatic proton Hb.
This is in sharp contrast to Ha in the rotamer 7a, which has an
NOE effect on two aromatic protons Hb and He. Absence of a
cross peak between the resonances of Ha and He suggests that the
rotamer 7a was either not present in the solution or at a
concentration below the detection limit. The assumption (i.e., the
absence of 7a) is in agreement with the sharp signal of Hd, which
is expected to have different chemical shifts in the rotamers 7a and
7b. Therefore, the compound exists mainly as the rotamer 7b in
the solution. Since the phenolic protons in both 1 and 7 have very
similar chemical shift (δ ≈ 11.5 ppm), we can conclude that the
rotamer 1b is the predominant form in solution.
It should be pointed out that the rotational energy barrier
was determined from 4, where a second −OH was introduced
to facilitate the detection of both N···H−O and O···H−O
hydrogen-bonded protons. The presence of the second hydro-
gen bonding, i.e., O···H−O in 4, could further restrict the
rotation of benzoxazole. Therefore, the rotational energy barrier
from 4 (∼10.59 kcal/mol) could be higher than that from
mono(HBO) 1. The results clearly indicated that the previous
estimation of the rotational barrier for 1 (∼15 kcal/mol by ab
initio with STO-3G basis set) was incorrect.14 Deviation of the
rotational barrier from 1 is dependent on the impact of the
additional hydrogen bonding O···H−O in 4. The crystal
structure of 4 showed that the N···H distance in the N···H−O
bond was 1.903 Å, which was shorter than the O···H distance
(1.996 Å) in O···H−O bond. The rotational barrier of 4 could
be largely determined by the stronger N···H−O bond, as the
hydrogen bond O···H−O is relatively weak. It should also be
noted that the impact of the O···H−O bond on the rotational
barrier of 4 was partially offset by the weakened N···H−O
bond, as the N···H distance in 4 (1.903 Å) became significantly
longer than the N···H distance in 3 (1.815 Å) (see Scheme 1).
Therefore, the rotational barrier of 4 could still be a close estima-
tion for that of 1.
Figure 6. Temperature dependence of phenolic proton of 1 (in
CDCl3) and 7 (in CDCl2CDCl2).
the O−H···N hydrogen bond of 2-diethylaminomethyl-3,4,6-
trichlorophenol.24
CONCLUSION
We have designed a HBO derivative 4 as a model compound
■
to determine the rotational barrier between different rotamers.
1
By acquiring H NMR spectra at variable temperature, the
rotational barrier is calculated to be about 10.59 kcal/mol.
1
From X-ray crystal structure and solution H NMR of 3, the
chemical shift at ∼12 ppm is assigned to the rotamer with
O−H···N hydrogen bonding (not O−H···O bonding). Identi-
fication of a specific rotamer is further confirmed by using model
compound 7, whose NOESY spectrum contains detectable sig-
nals from only one rotamer in the solution. Although the two
HBO rotamers 1a and 1b are found in 1:1 ratio in the crystalline
state, the chemical shift of phenolic proton and predominant
keto emission in fluorescence suggests that 1b is the major
component in solutions of nonprotic solvent. On the basis of
1H NMR analysis, only rotamer 1b is detected in chloroform
solution, which is very different from the rotamer ratio of 1:1
observed in the crystalline state. The study further reveals that
the solvated structure could play a significant role in the enol
emission.
EXPERIMENTAL SECTION
■
All reagents were purchased from commercial suppliers and used with-
out further purification. Compound 1, 2-(2′-hydroxyphenyl)benzoxazole
(HBO), was synthesized following ref 25. Compound 4 was synthesized
as described in our previous report.15 Chromatographic purifications
were carried out with silica gel of mesh 200−300. NMR spectra were
Effect of Temperature on Hydrogen Bonding. In order
to obtain additional information about the rotamer conversion,
1
collected on a 300 or 400 MHz spectrometers in CDCl3 (CDCl3: H
7.27 ppm, 13C 77.2 ppm). UV−vis spectra were acquired on a diode-
array spectrometer. Emission spectra were obtained on a fluorescence
spectrometer. Mass spectra were determined on time-of-flight (TOF)
mass spectrometers equipped with MALDI ion sources.
1
7 was further examined by H NMR at variable temperature
(Figure S8, Supporting Information). In deuterated 1,1,2,2-
tetrachloroethane solvent, the compound is predominantly in
the form of rotamer 7b, on the basis of the chemical shift of
phenolic proton Ha at 11.66 ppm (at 25 °C). As the tempera-
ture was increased, the signal of Ha gradually shifted upfield to
11.4 ppm at 140 °C (Figure S7, Supporting Information), while
the phenyl protons were not shifted. Plot of temperature versus
the chemical shift exhibited very good linear correlation for
both 7 and 1 (Figure 6), showing that no dramatic chemical event
occurred within the temperature range (−50 to 140 °C) examined.
It appeared that the hydrogen bonding was strengened at the lower
temperature, which led to gradual downfield shift of the phenolic
proton signal. The similar temperature effect has been reported in
Compound 3. 2,6-Dimethoxybenzaldehyde (300 mg, 1.8 mmol)
and 2-aminophenol (197 mg, 1.8 mmol) were refluxed in toluene
(20 mL) for 8 h. The reaction mixture was cooled to room tempera-
ture, and the precipitate was filtered to provide the Schiff base as a
1
reddish powder: H NMR (CDCl3 300 MHz) δ 9.14 (s, 1H), 7.32
(m, 2H), 7.14 (t, 1H, J = 8.1 Hz), 6.96 (d, 1H, J = 7.8 Hz), 6.86 (t, 1H,
J = 7.8 Hz), 6.63 (d, 2H, J = 7.8 Hz), 3.90 (s, 6H). The Schiff base
intermediate was transferred to a 25 mL round-bottomed flask equipped
with a side arm, a magnetic stirring bar, and a connecting tube. Then
Pd(OAc)2 (2.3 mg, 0.01 mmol), Cs2CO3 (130 mg, 0.4 mmol), and
10 mL DMF were added. The mixture was stirred at room tempera-
ture for 5 min and then warmed to 80 °C, while the oxygen gas was
289
dx.doi.org/10.1021/jo201890f | J. Org. Chem. 2012, 77, 285−290

The Journal of Organic Chemistry
Article
bubbled into the flask below the surface of the liquid. The progress of
the reaction was monitored by 1H NMR spectroscopy, which revealed
the disappearance of the imine resonance at around δ 8.8 ppm when
the reaction was complete. Upon completion of the reaction, the
mixture was poured into 20 mL of water. The precipitate was collected
by vacuum filtration and washed with 5 mL of water. The solid was
redissolved in 20 mL of dichloromethane, washed with 1% EDTA
aqueous solution (to remove the palladium catalyst), and then washed
with water. The organic layer was dried over anhydrous Na2SO4.
Removal of solvent afforded 2-(2,6-dimethoxyphenyl)benzoxazole-
(DMHBO) as white solid (391 mg, 85% from 2,6-dimethoxybenzal-
dehyde): 1H NMR (CDCl3, 300 MHz) δ = 11.59 (s, 1H), 8.13 (d, 1H,
J=2.4 Hz), 7.51 (m, 3H), 7.32 (m, 4H), 7.02 (d, 1H, J = 8.7 Hz), 2.63
(s, 3H); 13C NMR (CDCl3, 75 MHz) δ = 161.2, 157.6, 149.5, 137.7,
136.6, 136.0, 129.3, 126.6, 119.6, 119.4, 118.9, 112.5, 111.4, 111.1,
22.0. Boron tribromide (0.1 mL of 1 M BBr3 in hexanes, 0.1 mmol)
was added to a stirring mixture of DMHBO (50 mg, 0.2 mmol) in
anhydrous CH2Cl2 (5 mL), and the reaction was stirred at room
temperature under a nitrogen atmosphere. After 48 h, the reaction was
quenched with MeOH (1 mL), and the reaction mixture was extracted
with EtOAc (10 mL) and washed with H2O (2 × 5 mL) and brine
(5 mL). The organics were then filtered, dried over Na2SO4, and con-
centrated on a rotatory evaporator. Purification over a silica column
with (4:1 hexane/EtOAc) afforded compound 3 as a white solid (39
mg, 82%). The obtained pure product was recrystallized from hexane/
CH2Cl2 (2:1) to give white needle-like crystals (mp 180−181 °C): 1H
NMR (CDCl3 300 MHz) δ 12.90 (s, 1H), 7.68 (m, 2H), 7.27 (m,
3H), 6.77 (d, 1H, J = 9 Hz), 6.53 (d, 1H, J = 9 Hz, 1H), 4.01 (s, 3H);
13C NMR (CDCl3, 75 MHz) δ 163.2, 161.1, 159.4, 138.6, 133.5, 125.1,
124.9, 118.7, 111.0, 110.3, 102.1, 101.1, 56.3; HRMS (m/z) [M + H]+
calcd for C14H11NO3 242.0817, found, 242.0816; IR (KBr) νmax(cm−1)
3034 (w), 2898 (w), 1638 (m), 1594 (m), 1543 (s), 1505 (s), 1284
(s), 1226 (s), 1055 (m), 998 (m), 821 (s).
AUTHOR INFORMATION
Corresponding Author
*E-mail: yp5@uakron.edu.
■
ACKNOWLEDGMENTS
■
This work was supported by The University of Akron and
Coleman Endowment. We also thank the National Science
Foundation (CHE-9977144) for funds used to purchase the
NMR instrument used in this work and for partial support
(DMR-0905120) to E.T. and L.L. We thank two reviewers for
very helpful suggestions.
REFERENCES
■
(1) Williams, D. L.; Heller, A. J. Phys. Chem. 1970, 74 (26), 4473−
4480.
(2) Taki, M.; Wolford, J. L.; O’Halloran, T. V. J. Am. Chem. Soc.
2004, 126 (3), 712−713.
(3) Ohshima, A.; Momotake, A.; Arai, T. Tetrahedron Lett. 2004, 45,
9377−9381.
(4) Xu, Y.; Pang, Y. Chem. Commun. 2010, 46, 4070−4072.
(5) Xu, Y.; Pang, Y. Dalton Trans. 2011, 40 (7), 1503−1509.
(6) Chu, Q.; Medvetz, D. A.; Pang, Y. Chem. Mater. 2007, 19 (26),
6421−6429.
(7) Vazquez, S. R.; Rodriguez, M. C. R.; Mosquera, M.; Rodriguez-
Prieto, F. J. Phys. Chem. A 2007, 111 (10), 1814−1826.
(8) Zhang, G.; Wang, H.; Yu, Y.; Xiong, F.; Tang, G.; Chen, W. Appl.
Phys. B: Lasers Opt. 2003, 76, 677−681.
(9) Woolfe, G. J.; Melzig, M.; Schneider, S.; Doerr, F. Chem. Phys.
1983, 72 (2), 213−221.
(10) Das, K.; Sarkar, N.; Majumdar, D.; Bhattacharyya, K. Chem.
Phys. Lett. 1992, 198 (5), 443−448.
(11) Tong, Y. P. Acta Crystallogr. Sect. E 2005, 61 (9), o3076−o3078.
(12) Wu, J.; Liu, W.; Ge, J.; Zhang, H.; Wang, P. Chem. Soc. Rev.
2011, 40, 3483−3495.
(13) Das, K.; Sarkar, N.; Ghosh, A. K.; Majumdar, D.; Nath, D. N.;
Bhattacharyya, K. J. Phys. Chem. 1994, 98, 9126−9132.
(14) Nagaoka, S.; Itoh, A.; Mukai, K. J. Phys. Chem. 1993, 97, 11385−
11392.
Compound 7. 2-amino-3-methylphenol (25 mg,0.2 mmol) and
5-bromo-2-hydroxybenzaldehyde(40 mg, 0.2 mmol) were refluxed in
absolute ethanol (15 mL) for 5 h. The reaction mixture was cooled to
room temperature, and the precipitated Schiff base was filtered and
transferred to a 25 mL round-bottomed flask equipped with a side
arm, a magnetic stirring bar, and a connecting tube. Then Pd(OAc)2
(2.3 mg, 0.01 mmol), Cs2CO3 (130 mg, 0.4 mmol), and 10 mL of
DMF were added. The mixture was stirred at room temperature for
5 min and then warmed to 80 °C, while oxygen gas was bubbled into
the flask below the surface of the liquid. The progress of the reaction
was monitored by 1H NMR spectroscopy, which revealed the
disappearance of the imine resonance at around δ 8.8 ppm when
the reaction was complete. Upon completion of the reaction, the
mixture was poured into 20 mL of water. The precipitate was collected
by vacuum filtration and washed with 5 mL of water. The solid was
redissolved in 20 mL of dichloromethane, washed with 1% EDTA
aqueous solution (to remove the palladium catalyst), and then washed
with water. The organic layer was dried over anhydrous Na2SO4.
Removal of solvent afforded the compound 7 (56 mg, 92%). Crystals
(15) Chen, W.-H.; Pang, Y. Tetrahedron Lett. 2010, 51 (14), 1914−
1918.
(16) Chen, W.; Xing, Y.; Pang, Y. A Org. Lett. 2011, 13 (6), 1262−
1265.
(17) 1H NMR chemical shift for protons on oxygen and nitrogen can
be found, for example, in the following reference: Lambert, J. B.;
Shurvell, H. F.; Lightner, D. A.; Cooks, R. G., Organic Structural
Spectroscopy; Prentice: Upper Saddle River, NJ, 1998; pp 46−47.
(18) Bovey, F. A. In Nuclear Magnetic Resonance Spectroscopy;
Academic Press: New York, 1988; pp 255−324.
(19) Reich, H. J. J. Chem. Educ. Software 3D2, 1996
(20) Takeda, N.; Miyata, O.; Naito, T. Eur. J. Org. Chem. 2007, No. 9,
1491−1509.
(21) Hsieh, C. C.; Cheng, Y. M.; Hsu, C. J.; Chen, K. Y.; Chou, P. T.
J. Phys. Chem. A 2008, 112 (36), 8323−8332.
1
suitable for X-ray analysis were obtained from dichloromethane: H
NMR (CDCl3, 300 MHz) δ 11.59 (s, 1H), 8.13 (d, 1H, J=2.4 Hz),
7.51 (m, 3H), 7.32 (m, 4H), 7.02 (d, 1H, J = 8.7 Hz), 2.63 (s, 3H);
13C NMR (CDCl3, 75 MHz) δ 161.2, 157.6, 149.5, 137.7, 136.6, 136.0,
129.3, 126.6, 119.6, 119.4, 118.9, 112.5, 111.4, 111.1, 22.0; HRMS
(m/z) [M + H]+ calcd for C14H11BrNO2 303.9973, found 303.9978;
IR (KBr) νmax (cm−1) 3065 (w), 2925 (m), 1627 (m), 1584 (m), 1541
(s), 1483 (s), 1453 (s), 1278 (s), 1251 (s), 831 (m), 777 (s).
(22) Chen, W.; Pang, Y. Tetrahedron Lett. 2009, 50, 6680−6683.
(23) In practice, the cross-peak is unobservable when the proton−
proton distance exceeds about 5 Å. Lambert, J. B.; Shurvell, H. F.;
Lightner, D. A.; Cooks, R. G., Organic Structural Spectroscopy; Prentice
Hall: NJ, 1998; pp135−138.
(24) Sobczyk, L. Appl. Magn. Reson. 2000, 18, 47−61.
(25) Ohshima, A.; Momotake, A.; Nagahata, R.; Arai, T. J. Phys.
Chem. A 2005, 109 (43), 9731−9736.
ASSOCIATED CONTENT
■
S
* Supporting Information
1
1
Crystal structures of 3 and 4, H NMR spectrum of 3, and H
NMR of 1, 4, and 7 at variable temperatures. This material is
available free of charge via the Internet at http://pubs.acs.org.
290
dx.doi.org/10.1021/jo201890f | J. Org. Chem. 2012, 77, 285−290