Structure-reactivity relationships in (2-hydroxyethyl)benzophenone photoremovable protecting Groups

Article abstract of DOI:10.1016/j.tet.2010.02.087

A detailed study of substituent effects on the photochemical conversion of esters of (2-hydroxyethyl)benzophenone to the carboxylic acid was performed with the aim of improving on the reactivity of the parent, which was first reported decades ago. Over 20

Full text of DOI:10.1016/j.tet.2010.02.087

Tetrahedron 66 (2010) 3147–3151
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Structure-reactivity relationships in (2-hydroxyethyl)benzophenone
photoremovable protecting Groups
*
Michael C. Pirrung , Biswajit Gopal Roy, Surendra Gadamsetty
Department of Chemistry, University of California, Riverside, CA 92521-0403, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A detailed study of substituent effects on the photochemical conversion of esters of (2-hydroxy-
ethyl)benzophenone to the carboxylic acid was performed with the aim of improving on the reactivity of
the parent, which was first reported decades ago. Over 20 derivatives were prepared, 10 of which exhibit
some level of photochemical reactivity, but none of them is appreciably better than the original. The
reaction did not respond to a thioxanthone as a sensitizer or the inclusion of base to facilitate elimi-
nation. Solvent effects on the reaction also seem to be small, with the exception of the perfluorphenyl
analog, which is totally unreactive in hexanes but behaves normally in methanol or acetonitrile. These
neglected protecting groups may still prove useful in organic synthesis, with fairly rapid deprotection
even with long-wavelength irradiation.
Received 14 December 2009
Received in revised form 22 February 2010
Accepted 23 February 2010
Available online 26 February 2010
Keywords:
Benzophenone
Photoenolization
Dienol
Ó 2010 Elsevier Ltd. All rights reserved.
Elimination
1. Introduction
energy decreases. MeNPoc in particular has the drawback of ca.
90% cycle yields in DNA synthesis, but is the only photoremovable
Photochemically removable protecting groups are key ele-
ments of a variety of current technologies, including the fabrica-
tion of DNA microarrays. Many protecting groups for deoxyribose
hydroxyls have been investigated for this application. The two
most prominent are both nitroaromatics: the NPPoc group de-
veloped by Pfleiderer1 and the MeNPoc group developed by
Holmes.2 The latter derivatives are currently used for fabrication
of commercial Affymetrix microarrays, while the former are used
with the maskless array synthesizers as practiced by NimbleGen
and Febit. The NPPoc group undergoes a beta-elimination re-
action to give a hemi-carbonate that decarboxylates. The MeNPoc
group undergoes an internal redox reaction to release the hemi-
carbonate.
group that can be removed in the absence of solvent, which
provides virtues for large-scale manufacture. There are a limited
number3 of other fundamental organic compound frameworks
that exhibit sufficiently desirable photochemical traits to be ef-
fective as protecting groups. This work explored the photo-
chemical manifold of a neglected photoremovable protecting
group skeleton.
The photoenolization of aryl ketones bearing an ortho-alkyl
group that includes an alpha-hydrogen (Scheme 1) is an in-
teresting photochemical reaction that has an analogy in the pho-
tochemical behavior of the NPPoc group, yet it has been poorly
explored for deprotection compared to nitroaromatic groups. The
reaction proceeds via the triplet excited state, which is easily
reached owing to the rapid intersystem crossing known for aryl
ketones. This triplet can be converted to two stereoisomeric
dienols. The proximity of the hydrogen in the (Z)-dienol to the
terminal alkene facilitates an intramolecular transfer of this hy-
drogen, which converts it back to starting material. Consequently
this enol has a shorter lifetime than the (E)-dienol, which does not
There are many issues with nitrobenzyl photochemically re-
movable protecting groups, such as: their chemical yields are
most often <100%; they give a nitrosocarbonyl by-product with
a significant UV absorption and sometimes problematic chemical
reactivity; they are typically poor chromophores: their native
3 is
too low, and their native lmax is too blue, meaning that sub-
stituents must be added to shift the absorption to the red, out of
the range of the DNA being synthesized; consequently, they may
be subject to the common wavelength/reactivity compensation,
3
H
H
OH
R
O
O
hν
i.e., as lmax increases,
F decreases because the excited state
+
R
OH
R
R
ISC
Z
E
* Corresponding author. Fax: þ1 951 827 2749.
Scheme 1.
E-mail address: michael.pirrung@ucr.edu (M.C. Pirrung).
0040-4020/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.02.087

3148
M.C. Pirrung et al. / Tetrahedron 66 (2010) 3147–3151
have such favorable proximity. This energy-wasting reversion to
starting material reduces the product quantum yield and the rate
of any reactions involving these dienols. One example is available
for the use of photoenolization to trigger photoremoval of a pro-
tecting group by beta-elimination. In 1976 work of Ullman and
Tseng,4 a benzophenone bearing an ethylene linker and a leaving
group was shown to undergo photochemical deprotection
(Scheme 2). This is a triplet reaction, as naphthalene quenches
OH
1. PhMgBr
2. H3O+
1. OsO4
O
2. NaIO4
O
Ph
1. PhMgBr
2. H3O+
3. Zn(BH4)2
Ph
OH
R
R
O
O
O
Ph
a reactive triplet state that has a 1 ms lifetime. The product
quantum yield Fprod correlates with leaving group ability, and the
photoenolization quantum yield Fphotenol appears to be w0.45,
close to the Fprod of 0.47 when the leaving group is the very re-
active tosylate (rather than the carbamate shown). Therefore,
photoenolization is rate-determining with good leaving groups,
OH
O2CAd
COCl
Et N / CH Cl /
or
R
R
O
O
Δ
2
3
2
Ar
Ar
pyridine / 50 °C
and FphotenolhFprod. It is conceivable that the 0.45 Fphotenol re-
Scheme 3.
flects only the formation of the (E)-dienol, with the (Z) rapidly
reverting to starting material. With poorer leaving groups, elimi-
nation is rate-determining, and some dienol reverts to starting
All of the prepared compounds that exhibited photoreactivity in
hexane (and two that do not) are collected in Chart 1, and the
specific routes and procedures for their syntheses are provided in
the Supplementary data. General procedures representing each of
the steps in Scheme 3 are provided in the Experimental section. All
of the compounds prepared were directly compared against the
parent chromophore reported by Ullman in the form of compound
1. For comparison to a photochemically removable group already
used in DNA synthesis, the adamantanoyl derivative of (o-nitro-
phenyl)propanol (2) was also prepared. Absorption properties of
compounds in Chart 1 are given in Table 1. This includes absorption
at 366 nm, our desired reaction wavelength, as well as at 300 nm,
which was examined when it became apparent that simple modi-
fications to the parent chromophore would not lead to large leaps
in efficiency. Many other analogs were prepared that showed no
photoreactivity, and some of their structures and absorption
properties are given in the Supplementary data. For all compounds
prepared in this work, their UV spectra were recorded in hexane
solvent.
material, so Fprod<Fphotenol. This is the case when a carbamate is
the leaving group, as shown. The product quantum yield for this
reactant could be marginally increased (to 0.28) by the inclusion of
a weak base in the reaction mixture, which presumably facilitates
the elimination step at the expense of reversion to starting
material.
hν
254 nm
H
O
N
or
Ar
Pyrex-filter Hg
H3N
O +
O
O
Ar
MeOH
or
91%
Ph
MeCN
Ph
Scheme 2.
Another example of this type of reaction with an acetophenone
derivative has been reported.5 This reactant releases a carboxylic
acid in respectable isolated yield upon irradiation. Its reaction is
quenched by oxygen, as expected for a triplet reactive excited state.
However, the simple acetophenone chromophore is far less ab-
sorbing than the benzophenone used by Ullman, and the reaction is
very slow.
We aimed to modify the structure of the parent (2-hydroxy-
ethyl)benzophenone chromophore to increase its absorptivity at
the longer wavelengths more compatible with UV-absorbing bio-
molecules like proteins and nucleic acids and to enhance its
photochemical reactivity (both chemical and quantum yield) to
offer an alternative to nitrobenzyl systems for photochemical
deprotection.
O2CAd
O2CAd
O2CAd
O
O
NO2
Ph
Ph
1
2
3
CN
O2CAd
O2CAd
O2CAd
O2CAd
O2CAd
O
O
O
Br
NC
Ph
5
Ph
Ph
4
6
O2CAd
2. Results
O
O
O
Ph
Two principal methods were used to prepare the target (2-
hydroxyethyl)benzophenone derivatives (Scheme 3). The first is
essentially the route used by Ullman. An indanone is submitted to
phenyl Grignard addition and dehydration, and the resulting
indene is subjected to oxidative cleavage, reduction, and re-
oxidation. His route was modified in two ways: osmylation fol-
lowed by periodate cleavage is more convenient than ozonolysis,
and zinc borohydride6 is selective for reduction of the aldehyde
over the benzophenone and obviates the over-reduction and re-
oxidation step used by Ullman. The second method is direct
Grignard addition to an isochromanone. Once the alcohols were in
hand, they were converted to their adamantanoate esters by one of
two methods. Adamantanoic acid was chosen so that the chemical
yield of deprotection could be directly determined without risk of
loss of a volatile acid.
C6F5
7
C6H4OMe
Ph
9
8
O2CAd Ph
O2CAd
O
O
Cy2N
Ad
Ph
10
Ph
11
Chart 1.
The photochemical reactivity of these (hydroxyethyl)benzo-
phenone analogs was assessed by irradiating solutions at 1 mM in
hexane or acetonitrile for 30 min using the long-wavelength
phosphor lamps (ca. 366 nm) in a Rayonet photochemical reactor.
The extent of conversion was determined by the integration of the

M.C. Pirrung et al. / Tetrahedron 66 (2010) 3147–3151
3149
Table 1
substituted ring. This observation illustrates the difference between
structure-reactivity relationships in the nitrophenethyl groups and
the (hydroxyethyl)benzophenones. Likewise the ineffectiveness of
the addition of base, which was useful in the initial work on the
NPPoc group. Another illustration is the lack of any gain in photo-
reactivity from 5-phenyl substitution (11), whereas this modifica-
tion is very effective in nitrophenethyl groups.7
The tendency for the (Z)-dienols produced by photoenolization
to revert to starting material, as shown in their lifetimes, can be
modified by aromatic substituents.8 In particular, electron-
withdrawing influences (a cyano group or a pyridine nitrogen) in
conjugation with the enol lengthen its lifetime. Compounds 4 and 6
were designed to exploit this effect to increase the lifetime of the
dienol intermediates, shown in Chart 2. The fact that neither had
significantly improved performance may indicate that dienol life-
time does not limit the efficiency of this photochemical depro-
tection. It was likewise envisioned that increasing the acidity of the
photoenol through inductive effects could favor its ionization to the
dienolate, which would be more effective in displacing the leaving
group; hence, 7 was prepared. Again, the lack of a benefit of this
change suggests that the dienol acidity is not a key determinant of
photochemical reaction efficiency.
Absorption and photochemical properties of compounds in Chart 1
Compound 3366 3300
% Conversion % Conversion % Conversion
(hex)
(acn)
(MeOH, 300 nm)
1
2
3
4
5
6
7
8
77
501
65
63
100
0
72
62
0
261 1048 100
34
94
83
90
45
98
0
494
916
688
889
53
53
54
0
51
49
49
41
70
64
76
63
28
100
39
59
62
60
49
84
69
67
58
d
d
d
100 1526 69
93 1323 59
114 1087 54
174 1662 64
9
10
11
1þIPTX
2þIPTX
8þIPTX
d
d
d
d
d
d
d
d
d
carbinol CH2 in the starting materials versus the vinyl CH2 in the
styrene product in the NMR spectrum of the crude reaction mix-
ture. These experiments were performed in triplicate. In selected
cases, this conversion was verified by product isolation, purifica-
tion, and styrene yield determination. A preparative irradiation was
conducted on 1 and the styrene product was isolated in 58% yield,
nearly the % conversion found from NMR. This indicates that the
styrene is stable under these reaction conditions and that the NMR
integration is a fair indication of the conversion. Subsequent ex-
periments also used the 300 nm lamps in the Rayonet with
methanol as solvent.
CN
O2CR
O2CR
OH
Ph
O2CR
OH
O
NC
C6F5
Ph
Chart 2.
None of these (hydroxyethyl)benzophenone derivatives
approached the reactivity of the NPPoc chromophore 2. The best
reactant observed was 8, which is only a slight improvement on
1. Its enhanced absorption due to the p-methoxy group is likely
the reason for this observation.
Sensitization of the reaction with isopropylthioxanthone was
also examined for compounds 1, 2, and 8, at 366 nm in acetonitrile.
The two (hydroxyethyl)benzophenone derivatives showed reduced
conversion. For all compounds that underwent photoreactions,
reactions were attempted in hexane with the inclusion of 1 mM
diisopropylethylamine, but it had no effect on the conversion.
Compound 8 represents the only example among the many
strategies attempted to provide even a modest benefit. It is also
interesting that there is little enhancement in photoreactivity by
moving the irradiation wavelength deeper into the UV absorption
bands of these materials.
One typically expects that a change in a chromophore can modify
the nature of its excited state, both in terms of its multiplicity and
electronic configuration. Though this work did not address multi-
plicity, we intrinsically assumed that these are triplet reactions. It is
crucial for the reaction of interest here that the excited state of the
Preparative irradiation of
1 for 2 h at 366 nm gave ada-
ketone be n,
the benzylic hydrogen. In benzophenones with a variety of sub-
stituents (H, Cl, OMe), the triplet state is n, in both polar and non-
polar solvents. With strong donor substituents (e.g., p-dimethyl-
aminobenzophenone), there is a lower energy excited state
p
*
, since its radical-like character favors abstraction of
mantanecarboxylic acid in 75% isolated yield.
p
*
3. Discussion
p,
p
*
Many aspects of the results are perplexing. Comparison of the
photochemically reactive compounds (Chart 1) to those that are not
reactive (Chart S1) reveals the dilemma. Methoxy substitution at
the para position of the phenyl ring (8) gives the best compound,
but methoxy substitution at the 4-position of the substituted ring
(S4) abolishes photoreactivity. Bromination at the 5-position of the
substituted ring (5) is tolerated, but bromination at the 4-position
(S5) abolishes photoreactivity. Phenyl substitution at the 4-position
of the substituted ring (11) gives a compound with the greatest
absorption of all compounds prepared, but this does not translate
to improved conversion. Unusual behavior is exhibited by the
perfluorophenyl analog 7, as it is unreactive in hexanes but reacts
normally in polar solvents. Some observations are understandable.
Introduction of a methyl group at the benzylic position to give
compound 3 was expected to stabilize the photoenol via alkyl
substitution, much as was noted with the aci-nitro intermediate in
the photochemical elimination of nitrophenylpropyl derivatives.7
configuration and a concomitant low reactivity toward C–H bonds in
polar solvent. These traits are onlypartiallyexpressed here. With the
addition of a methoxy substituent, the photoreaction is unaffected,
but only when it is added to the proper ring. With the addition of
a dimethylamino substituent (S8), the photoreaction is abolished in
either polar or non-polar media. The lack of reactivity of the naph-
thyl and biphenyl analogs S6 and S7 is presumably because these
long conjugated
p
-systems favor
p
,
p excitedstates, rendering them
*
unable to follow the mechanistic pathway of 1.
4. Conclusion
While it has not been possible by straightforward manipulation
of subsitutents to substantially improve upon the photochemical
performance of (2-hydroxyethyl)benzophenone esters, photo-
chemical behavior that bears further investigation has been dis-
covered. Among the interesting observations is a profound solvent
effect on the photochemistry of perfluorophenyl compound 7 that
could be profitably exploited in selective deprotection photo-
chemistry.
However, compound
3 is totally unreactive, which can be
explained by its much-reduced absorption. Evidently, steric in-
teractions between the methyl group and the adjacent benzoyl
group cause the latter to rotate out of conjugation with the

3150
M.C. Pirrung et al. / Tetrahedron 66 (2010) 3147–3151
5. Experimental
5.4. General procedure for the zinc borohydride reduction to
give a (2-hydroxyethyl)benzophenone. (2-(1-Hydroxypropan-
2-yl)phenyl)(phenyl)methanone (S13)
5.1. General
All reagents were purchased from Aldrich or Acros Organics and
used without further purification. All moisture-sensitive reactions
were performed in oven-dried glassware under a nitrogen or argon
atmosphere. Dry solvents were from a Glass Contour Solvent Pu-
rification System. Flash column chromatography was carried out on
EM Reagents Silica Gel 60 (230–400 mesh). Thin-layer chroma-
tography (TLC) was performed using Merck EMD TLC Silica Gel 60
F254 plates. 1H NMR spectra were obtained on a Varian MVX-300
(300 MHz), or a Varian INOVA 400 (400 MHz) spectrometer.
Chemical shifts are reported in parts per million using tetrame-
Freshly fused ZnCl2 (2.0 g, 15 mmol) was taken up in THF
(15 mL), added to NaBH4 (solid, 1.7 g, 45 mmol), and allowed to stir
at rt for 24 h under nitrogen. The mixture was filtered quickly and
the cooled (ꢂ10 ꢁC) filtrate (1 mL) was directly added to a cooled
(ꢂ10 ꢁC) THF solution of aldehyde (500 mg, 2.10 mmol) with
constant stirring. After 10 min at ꢂ10 ꢁC, the reaction was
quenched by the slow addition of cold water. The mixture was
evaporated to dryness and the residue was partitioned between
ethyl acetate and water. The organic phase was washed with brine,
dried and evaporated to give a white semisolid, which on purifi-
cation by silica gel chromatography (ethyl acetate/hexanes 2:23)
gave the title alcohol (332 mg, 66%) as a colorless gum. 1H NMR
thylsilane (d
0.0) as an internal standard. 13C NMR spectra were
obtained on a Varian MVX-300 (75 MHz) or a Varian INOVA 400
(100 MHz) spectrometer and chemical shift were reported in parts
(300 MHz, CDCl3):
d
1.40 (d, J¼6.6 Hz, 3H), 2.48 (d, J¼7.5 Hz, 1H),
per million using CDCl3
(d
77.0) as an internal standard. IR spectra
2.80 (s, 1H), 3.11–3.16 (m, 1H), 3.87 (t, J¼7.5 Hz, 1H), 6.92 (d,
were obtained using a Perkin–Elmer Spectrum One FT-IR spec-
trometer. UV spectra were obtained using Cary 50 UV–vis–NIR
spectrophotometers from Varian, Inc.
J¼7.5 Hz, 1H), 7.11–7.37 (m, 8H). 13C NMR (75 MHz, CDCl3):
d 16.0,
43.3, 88.8, 123.8, 125.4, 126.6, 127.5, 127.8, 128.2, 129.0, 129.9, 131.1,
143.8, 144.4, 145.7. IR (film): 3366, 2969, 2926, 1602, 1493, 1475,
1446, 1173, 1107 cmꢂ1. HRMS (ESI): calcd for C16H17O2 (MþHþ):
241.1229; found: 241.1233.
5.2. General procedure for the dihydroxylation of an indene.
1-Phenyl-2,3-dihydro-1H-indene-1,2-diol (S9)
5.5. General procedure for organometallic addition to
To
a
solution of N-methylmorpholine-N-oxide (975 mg,
isochroman-1-ones to give a (2-hydroxyethyl)benzophenone
8.32 mmol) in 40:3:3 acetone, water, and tert-butanol (18 mL) was
added osmium tetroxide (28 mg, 0.11 mmol, 2 mol %). A solution of
3-phenyl-1H-indene (1.00 g, 5.55 mmol) was slowly added in ac-
etone (8 mL) and the reaction mixture was allowed to stir for 12 h
at rt. Any remaining osmium tetroxide was destroyed by the ad-
dition of NaBH4 (10 mg). The mixture was evaporated to dryness,
dissolved in ethyl acetate (30 mL), washed with brine (2ꢀ10 mL),
dried, and evaporated. The resulting brown crude material was
purified by silica gel column chromatography with ethyl acetate–
hexanes (1:9) to yield a colorless solid (225 mg, 51%). Mp 107 ꢁC.
To a solution of isochroman-1-one (200 mg, 1.35 mmol) in
5 mL THF was added phenylmagnesium bromide (3 M solution in
ether, 0.57 mL, 1.25 equiv). After stirring for 3 h, excess Grignard
reagent was destroyed by the addition of water, and the organic
solvents were evaporated under reduced pressure. The solid
white residue was dissolved in ethyl acetate, washed with water,
and brine, dried (anhydrous Na2SO4) and evaporated under re-
duced pressure. The resulting colorless gum was purified by silica
gel column chromatography (ethyl acetate/hexanes 3:22) to ob-
tain (2-hydroxyethyl)benzophenone (S10) (186 mg, 61%) as
1H NMR (300 MHz, CDCl3):
d
2.59 (d, J¼6.3 Hz, 3H), 3.00 (dd, J¼5.7,
15.9 Hz, 1H), 3.11 (s, 1H), 3.25 (dd, J¼6.6, 15.9 Hz, 1H), 4.45 (dd,
a colorless oil. 1H NMR (300 MHz, CDCl3):
d 1.57 (s, 1H), 2.96 (t,
J¼4.2, 8.1 Hz, 1H), 7.15 (d, J¼4.5 Hz, 1H), 7.28–7.36 (m, 8H). 13C
J¼6.0, 2H), 3.91 (m, 2H), 7.28–7.38 (m, 4 h), 7.42–7.53 (m, 3H),
NMR (75 MHz, CDCl3):
d
37.9, 81.1, 84.0, 120.2, 120.4, 121.4, 122.4,
7.58–7.64 (m, 1H), 7.81–7.84 (m, 1H). 13C NMR (75 MHz, CDCl3):
122.6, 123.2, 124.1, 135.9, 138.7, 140.1. IR (film): 3371, 2962, 2893,
d 36.5, 64.1, 125.8, 128.1, 128.7, 129.6, 130.8, 131.2, 131.3, 133.7,
1654, 1525, 1497, 1454, 1411, 1360, 1304, 1216, 1157, 1124 cmꢂ1
.
137.8, 138.7, 139.2, 199.1. This compound is known from Ullman,
HRMS (ESI): calcd for C15H13O2 (MꢂH): 225.0916; found:
of course.4
225.0923.
5.6. General procedure 1 for acylation with adamantanoyl
chloride. 2-(2-Nitrophenyl)propyl adamantanoate (2)
5.3. General procedure for the periodate cleavage of an
indene diol. 2-(2-Benzoylphenyl)propanal
To
a
solution of 2-(2-nitrophenyl)propanol (240 mg,
To a methanolic solution (30 mL) of 3-methyl-1-phenyl-2,3-
dihydro-1H-indene-1,2-diol (260 mg, 1.08 mmol) was added an
aqueous solution (6 mL) of NaIO4 (290 mg, 1.35 mmol). After
stirring at rt for 30 min, the reaction mixture was filtered and
the filtrate was concentrated by evaporation. The residue was
diluted with water (14 mL) and this solution was extracted with
ethyl acetate (2ꢀ15 mL). The combined organic extracts were
washed with brine, dried, and evaporated to give the title al-
dehyde (234 mg, 91%) as a colorless gum. 1H NMR (300 MHz,
1.32 mmol) in 15 mL dichloromethane in a screw-top vial under
a nitrogen atmosphere were added adamantanecarbonyl chloride
(392 mg, 1.98 mmol) and triethylamine (265 mg, 2.62 mmol). The
vial was closed tightly and heated at 50 ꢁC for 6 h. The solution
was washed with water (2ꢀ10 mL) and brine, dried, and evapo-
rated. The brown residue was purified by silica gel chromatog-
raphy (hexanes) giving the title ester (360 mg, 59%) as a light
brown transparent gum. 1H NMR (300 MHz, CDCl3):
d 1.36 (d,
J¼7.2 Hz, 3H), 1.64–1.73 (m, 6H), 1.76 (d, J¼2.7 Hz, 6H), 1.97 (s,
3H), 3.67–3.74 (m, 1H), 4.16–4.27 (m, 2H), 7.34 (t, J¼6.9 Hz, 1H),
7.34 (dt, J¼1.2, 7.8 Hz, 1H), 7.48 (dd, J¼1.2, 7.8 Hz, 1H), 7.57 (dt,
J¼1.2, 7.5 Hz, 1H), 7.34 (d, J¼1.5, 8.4 Hz, 1H). 13C NMR (75 MHz,
CDCl3):
1H), 7.28–7.85 (m, 9H), 9.76 (s, 1H). 13C NMR (75 MHz, CDCl3):
15.3, 27.6, 49.4, 126.9, 128.7, 128.8, 129.3, 129.5, 129.8, 130.6,
131.4, 133.8, 137.8, 137.9, 198.0, 200.9. IR (film): 3060, 2929,
d
1.43 (d, J¼6.9 Hz, 3H), 1.71 (s, 6H), 3.99 (q, J¼6.9 Hz,
d
CDCl3):
d 18.1, 28.1, 33.5, 36.6, 38.9, 40.9, 68.0, 124.3, 127.6, 128.6,
1704, 1658, 1595, 1578, 1485, 1447, 1315, 1267, 1152, 1122 cmꢂ1
.
132.7, 137.7, 150.7, 177.5. IR (film): 2905, 2851, 1724, 1523,
HRMS (ESI): calcd for C16H15O2 (MþHþ): 239.1072; found:
239.1080. This compound was not routinely purified before the
next step.
1452, 1353, 1226, 1183, 1103 cmꢂ1
. HRMS (ESI): calcd for
C20H25NNaO4 (MþNaþ): 366.1681; found: 366.1690. UV lmax
, 3
(hexane): 254, 3041.

M.C. Pirrung et al. / Tetrahedron 66 (2010) 3147–3151
3151
5.7. General procedure 2 for acylation of a (2-hydroxyethyl)-
Acknowledgements
benzophenone with adamantanoyl chloride. 2-
Benzoylphenethyl adamantanoate (1)
Financial support from Samsung is appreciated. Mass spectra
were obtained on an instrument purchased with NSF grant
CHE0541848. Dr. Goutam Biswas provided valued assistance in the
analysis of some samples.
To
a solution of (2-hydroxyethyl)benzophenone (400 mg,
1.77 mmol) in dry pyridine (2 mL) was added ada-
mantanecarbonyl chloride (525 mg, 2.65 mmol). After stirring at
50 ꢁC for 2 h, the pyridine was evaporated under reduced pres-
sure. The final traces of pyridine were removed by addition of
toluene and evaporation of the azeotrope. The residual brown
solid was dissolved in ethyl acetate (15 mL), washed with water
(10 mL) and brine (10 mL), dried, and evaporated to give
a brownish sticky residue that was purified by silica gel column
chromatography (2% ethyl acetate–hexanes) to give 1 (558 mg,
Supplementary data
Narrative of routes for the synthesis of compounds that are pho-
tochemically reactive. Experimental descriptions for the synthesis of
compounds 1 and 3–11. Structures and UV properties of prepared
compounds that are not photochemically reactive. 1H NMR and 13C
NMR spectra for 1–11 and compounds S15–S17, S19–S20, S26–S27,
and S29 (55 pages total). Supplementary data associated with this
article can be found in online version at doi:10.1016/j.tet.2010.02.087.
81%). 1H NMR (300 MHz, CDCl3):
d 1.68–1.73 (m, 6H), 1.81 (d,
J¼2.7 Hz, 6H), 1.97 (s, 3H), 3.03 (t, J¼6.6 Hz, 2H), 4.25 (t, J¼6.6 Hz,
2H), 7.29–7.49 (m, 6H), 7.49–7.52 (m 1H), 7.71–7.74 (m, 2H). 13C
NMR (75 MHz, CDCl3):
d 27.8, 32.5, 36.4, 38.7, 40.5, 64.3, 125.8,
References and notes
128.3, 129.0, 130.2, 131.1, 133.1, 137.2, 137.6, 138.5, 177.3, 197.9. IR
(film): 2904, 2851, 1724, 1662, 1597, 1579, 1448, 1344, 1315, 1266,
1. Giegrich, H.; Eisele-Buhler, S.; Hermann, C.; Kvasyuk, E.; Charubala, R.; Pfleiderer,
W. Nucleosides Nucleotides 1998, 17, 1987–1996.
2. Holmes, C.P.; Solas, D.W.; Kiangsoontra, B. PCT Int. Appl., 1994, WO9410128.
Pease, A. C.; Solas, D.; Sullivan, E. J.; Cronin, M. T.; Holmes, C. P.; Fodor, S. P. A.
Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 5022–5026.
1248, 1182 cmꢂ1
.
HRMS (ESI): calcd for C26H29O3 (MþHþ):
389.2111; found: 389.2121, calcd for C26H28NaO3 (MþNaþ):
411.1931; found: 411.1940.
3. Pelliccioli, A. P.; Wirz, J. Photochem. Photobiol. Sci. 2002, 1, 441–458; Givens, R. S.;
Conrad, P. G., II; Yousef, A. L.; Lee, J.-I. In CRC Handbook of Organic Photochemistry
and Photobiology, 2nd ed.; Horspool, W. M., Lenci, F., Eds.; CRC: Boca Raton, FL,
2003; Chapter 69, pp 1–46; Conrad II, P. G.; Givens, R. S.; Weber, J. F. W.; Kandler, K.
Org. Lett. 2000, 2, 1545–1547.
4. Tseng, S.-S.; Ullman, E. F. J. Am. Chem. Soc. 1976, 98, 541–544.
5. Atemnkeng, W. N.; Louisiana, L. D., II; Yong, P. K.; Vottero, B.; Banerjee, A. Org.
Lett. 2003, 5, 4469–4471.
6. Ranu, B. C.; Chakraborty, R. Tetrahedron Lett. 1990, 31, 7663–7664; Narasimhan, S.;
Balakumar, R. Aldrichimica Acta 1998, 31, 19–26; Ranu, B. C. Synlett 1993, 885–892.
7. Buehler, S.; Lagoja, I.; Giegrich, H.; Stengele, K.-P.; Pfleiderer, W. Helv. Chim. Acta
2004, 87, 620–659.
5.8. Photochemistry
UV spectra were recorded in hexane. Photochemical reactivity
was assessed by making up a solution at 1 mM in hexane (also in
some cases including 1 mM diisopropylethylamine), acetonitrile or
methanol and irradiating each solution for 30 min using 366 or
300 nm lamps in a Rayonet photochemical reactor. The percent
conversion was determined by NMR analysis of the crude reaction
mixture after evaporation.
8. Koner, A. L.; Singhal, N.; Nau, W. M.; Moorthy, J. N. J. Org. Chem. 2005, 70, 7439–7442.