4-Chlorophenol in Cyclodextrins
FULL PAPER
trinsic signal of the corresponding cyclodextrin solution. The intensity of
the induced signal was a function of the CD concentration and was ana-
lyzed by means of a 1:1 association model described by a nonlinear equa-
tion.[13]
complexed with 1, respectively, and corresponds to the yield
of 8. Because the conversion of carbene 3 to radical 8 within
the cavity is presumably complete and the disappearance of
8 by reduction to 9 is not faster than its formation (see
above), a fraction of the initially formed aryl cation must
undergo H transfer from the CD at a rate comparable to
that at which deprotonation to 3 and formation of 8 occur
(ꢂ5ꢁ107 sꢀ1). Production of 9 directly from 2 via the radical
cation 9+C in the presence of CD is indeed supported by the
presence both of a shoulder at l = 420–430 nm in the spec-
tra of Figure 4 and Figure 5 and of the absorption tail at l
= 510–550 nm in Figure 6, which can be assigned to the rad-
ical cation, a species endowed by a large absorption band in
the 380–550 nm region.[10,28]
Monitoring the photoreaction by UV: The time course of the photodeha-
logenation was followed by variations in the UV absorption of 2-mL sam-
ples of 3.14ꢁ10ꢀ4 m solutions of 1 irradiated at 282 nm in a 1 cm cell. The
irradiation setup consisted of
ramHBO, 200 W) coupled with a high-intensity Bausch&Lomb grating
monochromator (1350 groovesmmꢀ1
blazed at 300 nm) with slits of
3 mm (bandwidth 18 nm). The intensity of the incident light was ꢁ1ꢁ
1014 photonssꢀ1
a high-pressure mercury lamp (Os-
,
.
Monitoring the photoreaction by HPLC and quantum yield measure-
ments: Quantum yield measurements were carried out with an optical
bench fitted with a focalized high-pressure mercury arc (150 W) and an
interference filter (transmission maximum at 281 nm). 1.5ꢁ10ꢀ3 m solu-
tions of 1 in spectrophotometric cells were used. The course of the reac-
tion was monitored by HPLC (AQUASIL1 C18 (250ꢁ4.6 mm) column,
MeCN/water (40:60), flux 1 mLminꢀ1) with UV detection at l = 270 nm.
The conversion was limited to <20%. The conversion of the starting ma-
terial was linear with time within this limit and beyond in the presence of
additives, while in neat water, the rate of reaction rapidly dropped for
conversions of <10%. This was attributed to the formation of photo-
products strongly absorbing at the exciting wavelength. Reliable meas-
urements could only be obtained in this case by the use of 7ꢁ10ꢀ4 m solu-
tions.
Conclusion
The present study demonstrates that the 4-hydroxyphenyl
cation 32 and carbene 33 are in equilibrium and are relatively
persistent in water. Specific reagents can selectively trap
these intermediates. Thus, water is a suitable solvent to con-
duct reactions involving either one of such highly activated
species, which are smoothly generated under photochemical
conditions, in particular for synthetically appealing cationic
arylation of alkenes and, reasonably, of other p nucleo-
philes.[29]
On the other hand, cyclodextrins efficiently scavenge and
reduce the intermediates. When cyclodextrin concentrations
are chosen in such a way as to associate more than 80% of
ground state 1, the largest fraction of the chlorophenol is re-
duced within the cavity and none of the above traps inter-
fere with the reduction. The large time-window involved in
the reaction in water (10 ns scale, photochemical fragmenta-
tion and protic equilibrium; 100 ns, trapping by oxygen and
alkenes; 1 ms, reaction with the starting material) is short-
ened to a few tens of nanoseconds.
Preparative photolysis: A solution of 4-chlorophenol (100 mg, 0.78 mm)
and 2-propenol (750 mL, 15 mm) in water (15 mL) was poured into a
quartz tube. The tube was flushed with argon for 15 min, capped, and ir-
radiated with six phosphor-coated lamps (15 W, center of emission l =
310 nm). After 16 h, the reagent had been completely converted (HPLC)
and a single product accounted for ꢁ80% of the material (as determined
by comparison with a weighted solution of 10, see below). The solvent
was evaporated under low pressure and the residue chromatographed on
silica gel to give 32 mg (25%) of the main product. This was recognized
as 3-(4-hydroxyphenyl)-1,2-propandiol, based on the spectroscopic char-
acteristics and a comparison with reported data.[30,31] 1H NMR (CDCl3):
d = 2.67 (AB part of an ABX signal, 2H, J(A,B) = 15 Hz), 3.47 (AB
part of an ABX signal, 2H, J(A,B) = 12.5 Hz), 3.78 (X part of the previ-
ous signals, 1H), 6.75 (2H) and 7.07 (AB quartet, 4H); 13C NMR
(CDCl3): d = 38.6 (CH2), 65.0 (CH2), 73.3 (CH), 114.6 (CH), 129.3, 130.0
(CH), 155.2 ppm; IR (melt): n˜
=
3300 cmꢀ1; elemental analysis (%)
calcd for C9H12O3: C 64.27, H 7.19; found: C 63.9, H 7.3.
Flash photolysis: Transient absorption spectroscopy was carried out on a
setup for nanosecond absorption measurements described previously.[32]
The minimum response time of the detection system was ꢁ2 ns. The
laser beam from a JK-Lasers Nd-YAG, operated at l = 266 nm, pulse
width 20 ns FWHM, was focused on a 3 mm high and 10 mm wide rectan-
gular area of the cell, and the first 2 mm in depth were analyzed with a
right-angle geometry. The incident pulse energies used were
<17 mJcmꢀ2 (5 mJ per pulse). The bandwidth used for the transient ab-
sorption measurements was typically 2 nm (0.5 mm slit width). The spec-
tra were reconstructed point-by-point from time profiles taken every 5–
10 nm. The sample absorbance at 266 nm was typically 0.5–1 over 1 cm.
Oxygen was removed from the water by vigorously bubbling the solu-
tions with a constant flux of Ar, previously passed through an aqueous
trap to prevent evaporation of the sample. The solution, in a flow cell of
1 cm optical path, was renewed after each laser shot. The temperature
was 295ꢄ2 K. The detection system was perturbed from 290 nm to
380 nm by the emission of 1, generated by the laser excitation. These
problems were minimized by the use of neutral density filters at the en-
trance slit of the monochromator and by pulsing the 150 W high-pressure
Xe lamp with a high current (~200 mA for 1 ms) to increase the intensity
of the analyzing light. In spite of this, transient spectra in this wavelength
region were not significant before 20–35 ns from the end of the pulse.
The absorption signals were acquired and processed with a homemade
program and Asyst 3.1 (Software Technologies, Inc.). Nonlinear fitting
The high reductive power of the CD towards both the
aryl cation and the carbene disfavors the use of CDs as reac-
tion vessels for bimolecular processes, although an efficient
coinclusion of the trap could allow a guest–guest process to
compete efficiently with a reaction with the CD itself. In the
present study, the high solubility of 2-propenol in water pre-
vents such condition being achieved.
Experimental Section
General: 4-Chlorophenol (1) and 2-propenol were obtained from Al-
drich. a-CD and b-CD from SERVA (research grade) were used as re-
ceived. Water was purified by passing it through a Millipore MilliQ
system.
The UV absorption spectra were recorded with a Perkin Elmer320 spec-
trophotometer. Circular dichroism spectra were obtained with a JascoJ-
710 dichrograph.
Measuring the complexation by CD: Circular dichroism spectra of 4-
chlorophenol in water were recorded by titrating with cyclodextrins at 8–
10 different concentrations and by subtracting from each of them the in-
Chem. Eur. J. 2005, 11, 4274 – 4282
ꢃ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4281