Photochemical acetalization of carbonyl compounds in protic media using an in situ generated photocatalyst

Article abstract of DOI:10.1021/jo0485886

Carbonyl compounds are conveniently converted into their corresponding dimethyl acetals in good yields and short reaction times by means of a photochemical reaction in methanol with a catalytic amount of chloranil (2,3,5,6-tetrachloro-1,4-benzoquinone, CA) as the sensitizer. Using aldehydes gives better results than using ketones, which also tend to form enol ethers as side products. These results are similar to those of simple acid-catalyzed acetalization reactions, suggesting the involvement of a photochemically generated acid. On the basis of steady state and laser flash photolysis data the reaction is proposed to involve the in situ generation of a photocatalyst (2,3,5,6-tetrachloro-1,4-hydroquinone, TCHQ) via reaction of CA with the solvent. The acetalization process is initiated by ionization of TCHQ, followed by loss of a proton to the solvent or the carbonyl, which starts a catalytic reaction. The photocatalyst is regenerated via a disproportionation reaction.

Full text of DOI:10.1021/jo0485886

Photochemical Acetalization of Carbonyl Compounds in Protic
Media Using an in Situ Generated Photocatalyst
H. J. Peter de Lijser* and Natalie Ann Rangel
Department of Chemistry and Biochemistry, California State University Fullerton,
Fullerton, California 92834-6866
pdelijser@fullerton.edu
Received August 12, 2004
Carbonyl compounds are conveniently converted into their corresponding dimethyl acetals in good
yields and short reaction times by means of a photochemical reaction in methanol with a catalytic
amount of chloranil (2,3,5,6-tetrachloro-1,4-benzoquinone, CA) as the sensitizer. Using aldehydes
gives better results than using ketones, which also tend to form enol ethers as side products. These
results are similar to those of simple acid-catalyzed acetalization reactions, suggesting the
involvement of a photochemically generated acid. On the basis of steady state and laser flash
photolysis data the reaction is proposed to involve the in situ generation of a photocatalyst (2,3,5,6-
tetrachloro-1,4-hydroquinone, TCHQ) via reaction of CA with the solvent. The acetalization process
is initiated by ionization of TCHQ, followed by loss of a proton to the solvent or the carbonyl, which
starts a catalytic reaction. The photocatalyst is regenerated via a disproportionation reaction.
Introduction
widespread application, most likely because it is consid-
ered an abnormality caused by (acidic) impurities. To
achieve good conversions and yields for this type of
reaction, a photosensitized process generating an acidic
species would be desirable. Quinones are commonly used
as photosensitizers, and irradiation of these types of
compounds in hydroxylic solvents results in the formation
of hydroquinones (hydroxyarenes). These photoproducts
may undergo photochemical reactions themselves as well.
One particularly interesting aspect of certain hydroxy-
arenes, as first noted by Fo¨rster,5a is that their excited
states (S1) are (much) more acidic than the groundstate
(S0).5 For example, for 1-naphthol (1) the pKa values are
9.4 (S0) and -0.2 (S1).5d Species that exhibit such behavior
are commonly referred to as photoacids. They are used
in a variety of processes including polymerization, the
degradation of polymers, transformation of functional
groups, photoresists, and optical lithography.6 In recent
years many different types of photoacids have been
Carbonyl compounds are essential building blocks of
organic chemistry because of their chemical versatility.
Often this functional group must be protected while
carrying out a series of steps, and acetalization is one of
the preferred methodologies.1 Most often these reactions
are acid-catalyzed, but this does have its drawbacks as
other functionalities in the molecule may be acid-sensi-
tive. As a result, many other methodologies have been
developed and the search for new and improved methods
continues.1,2 Photochemistry provides an elegant way to
access intermediates and pathways that are often not
available or difficult to reach when using ground-state
reactions. For example, the development of photoremov-
able groups has been and still is of great interest.3
Several photochemical methods for the deprotection of
carbonyl compounds have also been reported,4 but few if
any methods for the photochemical protection of carbonyl
compounds are known. Although the formation of acetals
from carbonyl compounds upon irradiation in hydroxylic
solvents has been reported, this process has not found
(5) (a) Fo¨rster, Th. Z. Elektrochem. Angew. Phys. Chem. 1950, 54,
531. (b) Ireland, J. F.; Wyatt, P. A. H. Adv. Phys. Org. Chem. 1976,
12, 131. (c) Lukeman, M.; Wan, P. In CRC Handbook of Organic
Photochemistry and Photobiology, 2nd ed.; Horspool, W., Lenci, F., Eds.;
CRC Press: Boca Raton, FL, 2004; Chapter 39. (d) Pines, E. In The
Chemistry of Phenols; Rappoport, Z., Ed.; John Wiley & Sons: New
York, 2003; pp 491-527. (e) Ortica, F.; Scaiano, J. C.; Pohlers, G.;
Cameron, J. F.; Zampini, A. Chem. Mater. 2000, 12, 414. (f) Andraos,
J.; Barclay, G. G.; Medeiros, D. R.; Baldovi, M. V.; Scaiano, J. C.; Sinta,
R. Chem. Mater. 1998, 10, 1694. (g) Scaiano, J. C.; Barra, M.; Sinta,
R. Chem. Mater. 1996, 8, 161. (h) Tolbert, L. M.; Haubrich, J. E. J.
Am. Chem. Soc. 1994, 116, 10593. (i) Wan, P.; Shukla, D. Chem. Rev.
1993, 93, 571. (j) Tolbert, L. M.; Haubrich, J. E. J. Am. Chem. Soc.
1990, 112, 8163. (k) Wehry, E. L.; Rogers, L. B. J. Am. Chem. Soc.
1965, 87, 4234.
(1) (a) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 3rd ed.; Wiley-Interscience: New York, 1998. (b) Meskens,
F. A. J. Synthesis 1981, 601.
(2) (a) Larock, R. C. Comprehensive Organic Transformations, 2nd
ed.; Wiley-VCH: New York, 1999. (b) Perio, B.; Dozias, M.-J.; Jac-
quault, P.; Hamelin, J. Tetrahedron Lett. 1997, 38, 7867; Srivastava,
N.; Dasgupta, S. K.; Banik, B. K. Tetrahedron Lett. 2003, 44, 1191;
De, S. K. Tetrahedron Lett. 2004, 45, 2339.
(3) (a) Haley, M. F.; Yates, K. J. Org. Chem. 1987, 52, 1817. (b) de
Lijser, H. J. P.; Fardoun, F. H.; Sawyer, J. R.; Quant, M. Org. Lett.
2002, 4, 2325. (c) Yang, Y.; Zhang, D.; Wu, L.-Z.; Chen, B.; Zhang,
L.-P.; Tung, C.-H. J. Org. Chem. 2004, 69, 4788.
(4) For recent reviews on photoremovable groups, see: (a) Pelliccioli,
A. P.; Wirz, J. Photochem. Photobiol. Sci. 2002, 1, 441. (b) 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., Lenci,
F., Eds.; CRC Press: Boca Raton, FL, 2004; pp 69-1-69-46.
(6) (a) Shirai, M.; Tsunooka, M. Prog. Polym. Sci. 1996, 21, 1. (b)
Wallraff, G. M.; Hinsberg, W. D. Chem. Rev. 1999, 99, 1801. (c) Aoki,
A.; Ghosh, P.; Crooks, R. M. Langmuir 1999, 15, 7418. (d) Wu, H.;
Gonsalves, K. E. Adv. Funct. Mater. 2001, 11, 271. (e) Serafinowski,
P. J.; Garland, P. B. J. Am. Chem. Soc. 2003, 125, 962.
10.1021/jo0485886 CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/26/2004
J. Org. Chem. 2004, 69, 8315-8322
8315

de Lijser and Rangel
TABLE 1. Summary of Results for the Photolysis of
Benzaldehyde in Protic Solventsa
investigated. Much of the research has focused on the
development, behavior, and physical properties of so-
called “super” photoacids such as 5,8-dicyano-2-naphthol
(2; pKa* ≈ -4.5).7
time
conversion acetal
entry sensitizer (min)b
solvent
(%)c
(%)c
1d
2d
3d
4
CA
60
30
15
30
30
30
30
30
30
30
30
30
MeOH
96
97
96
97
97
89
89
82
63
82
78
74
100
100
100
100
100
100
100
100
100
100
100
100
CA
MeOH
CA
MeOH
CA
MeOH
5e
6f
7
CA
MeOH
CA
MeOH
CA
4:1 MeCN-MeOH
EtOH
8
9
10
11
12
CA
CA
4:1 MeCN-EtOH
4:1 MeCN-MeOH
4:1 MeCN-MeOH
4:1 MeCN-MeOH
A second option for generating acidic species would be
via ionization of a neutral substrate. Removal of a single
electron from a molecule results in the weakening of a
chemical bond, and the pKa of the radical cation is often
several orders of magnitude smaller than that of the
neutral species.8 One potentially attractive set of candi-
dates would be the phenols, as they can be generated
from quinones and are known to undergo fast depro-
tonation upon ionization.9
XA
TCHQ
DCNQ
a Reactions were carried out in Pyrex glass using a Rayonet
photochemical reactor equipped with sixteen 350 nm light bulbs;
the benzaldehyde concentration was 0.015 M unless indicated
otherwise; the sensitizer concentration was 0.005 M unless
indicated otherwise. b Irradiation time. c Conversions and yields
were determined by calibrated GC/FID. d [Sensitizer] ) 0.015 M.
e [Benzaldehyde] ) 0.050 M. f [Benzaldehyde] ) 0.500 M.
Recently we found that irradiation of carbonyl com-
pounds in the presence of quinone sensitizers in protic
solvents resulted in the formation of acetals.10 These
results are not consistent with the known photochemistry
of either the carbonyl compounds or the quinones. To find
out whether these observations were due to impurities
or other artifacts, we have undertaken a more extensive
investigation on which we here report. The results are
most consistent with the in situ formation of a photo-
catalyst, which initiates the acetalization process. The
major advantage of this system is that the photocatalyst
is regenerated continuously.
obtained when using ethanol as the solvent (or as a
mixture with MeCN; entries 8 and 9), but using i-PrOH
gives very poor results. When using ethylene glycol as
the solvent some acetal is observed; however, there was
significant overlap of the reactants and the solvent on
the GC, making a detailed analysis impossible.
Other photocatalysts that were tested include duro-
quinone (DQ), 1,4-naphthoquinone (NQ), phenanthrene-
quinone (PQ), xanthone (XA), 2,3-dichloro-1,4-naphtho-
quinone (DCNQ), and tetrachloro-p-hydroquinone (TCHQ).
From the results listed in Table 1 (entries 10-12) it can
be seen that under a set of standard conditions most of
these sensitizers give results similar to CA; however,
using DQ, NQ, and PQ did not result in any conversion
of the carbonyl compound or in acetal formation. The
results are reproducible as long as an appropriate pho-
tocatalyst is present in the mixture. In the absence of
CA or any other photocatalyst, virtually no acetal is
formed and the carbonyl conversion is significantly lower.
The solubility of CA in MeOH is low, and the majority
can be recovered afterward. TCHQ dissolves readily in
MeOH and can be detected afterward. A trace of CA was
also detected. Irradiation of the aldehyde and CA in the
presence of a base (sodium bicarbonate) does not result
in any reaction, suggesting the involvement of a photo-
generated acid. The reaction also proceeds when using
420 nm light bulbs but is somewhat less efficient (52%
yield after 30 min). Finally, irradiation of a mixture
containing the acetal and CA in 4:1 MeCN-MeOH does
not result in any reaction, i.e., the process is not revers-
ible under the conditions of the reaction. It must be noted
that small amounts of water do influence the reaction
significantly. For example, photolysis of a solution con-
taining the acetal and CA in MeCN did result in the
formation of the aldehyde. This is consistent with a report
by Sankararaman,11 who reported that DDQ can be used
for the photoinduced deprotection of thioacetals and
ketals. Also, when benzaldehyde and CA were irradiated
in 95% EtOH, only a small amount of the acetal was
formed and most of the benzaldehyde was recovered.
Results and Discussion
Steady-State Photolysis of Aldehydes and Ke-
tones with Chloranil in Protic Solvents. Photolysis
of O-benzylacetophenone oxime and chloranil (CA) in
acetonitrile results in the formation of acetophenone and
benzaldehyde. When using a polar protic solvent such as
methanol, no benzaldehyde is observed but rather benzyl
alcohol as well as traces of benzaldehyde dimethylac-
etal.10 Initially, benzyl alcohol was thought to be the
product of a photochemical reaction of benzaldehyde with
methanol; however, this was shown not to be true in a
separate experiment. Interestingly, irradiation (60 min)
of a solution of benzaldehyde and CA in methanol results
in the quantitative formation of the dimethyl acetal
(Table 1, entry 1). Further studies on this reaction have
shown that even a catalytic amount (1 mol %) of CA
(entry 6) can achieve a significant conversion of the
aldehyde and yield of the acetal. The reaction can be
completed in as little as 15 min when using benzaldehyde
and a higher concentration of CA. Similar results are
(7) See for example: (a) Tolbert, L. M.; Solntsev, K. M. Acc. Chem.
Res. 2002, 35, 19. (b) Solntsev, K. M.; Huppert, D.; Agmon, N. J. Phys.
Chem. A 1999, 103, 6984. (c) Solntsev, K. M.; Huppert, D.; Tolbert, L.
M.; Agmon, N. J. Am. Chem. Soc. 1998, 120, 7981.
(8) (a) Nicholas, A. M. de P.; Arnold, D. R. Can. J. Chem. 1982, 60,
2165. (b) Bordwell, F. G.; Cheng, J.-P.; Bausch, M. J. J. Am. Chem.
Soc. 1988, 110, 2872. (c) Bordwell, F. G.; Cheng, J.-P. J. Am. Chem.
Soc. 1991, 113, 1736.
(9) (a) Gadosy, T. A.; Shukla, D.; Johnston, L. J. J. Phys. Chem. A
1999, 103, 8834. (b) Ganapathi, M. R.; Naumov, S.; Hermann, R.;
Brede, O. Chem. Phys. Lett. 2001, 337, 335.
(10) de Lijser, H. J. P.; Tsai, C.-K. J. Org. Chem. 2004, 69, 3057.
(11) Mathew, L.; Sankararaman, S. J. Org. Chem., 1993, 58, 7576.
8316 J. Org. Chem., Vol. 69, No. 24, 2004

Photochemical Acetalization of Carbonyl Compounds
TABLE 2. Results from Photosensitized Reactions of
Aldehydes in Methanola
TABLE 3. Summary of Results for the Photolysis of
Cyclohexanone in Protic Solventsa
conversion acetal yield
conver-
enol
entry
aldehyde
(%)b
(%)b
sensi- time
sion
acetal ether
entry tizer (min)b
1d
2
3
4
solvent
(%)c
(%)c
(%)c
1
2
benzaldehyde
89
96
97
94
81
73
62
93
47
98
100
100
100
100
100
100
100
100
100
100
hexanal
CA
30
30
30
30
MeOH
MeOH
4:1 MeCN-MeOH
MeOH
90
95
44
74
92
85
>99
73
8
15
<1
26
3
cyclohexanecarboxaldehyde
3,3-dimethylbutyraldehyde
phenylacetaldehyde
o-tolualdehyde
2,6-dimethylbenzaldehyde
dodecanal
CA
4
CA
5
TCHQ
6
a Cyclohexanone concentration was 0.015 M; the sensitizer
concentration was 0.005 M unless indicted otherwise. b Irradiation
time. c Conversions and yields were determined by calibrated GC/
FID. d [Sensitizer] ) 0.015 M.
7
8
9
trans-cinnamaldehyde
hydrocinnamaldehyde
10
a [CA] ) 0.005 M; [aldehyde] ) 0.015 M; all solutions were
irradiated for 30 min. b Conversions and yields were determined
by calibrated GC/FID.
TABLE 4. Results from Photosensitized Reactions of
Ketones in Methanola
enol
conversion acetal ether
entry
ketone (%)
(%)b
(%)b
100
(%)b
To determine the generality of this reaction, a number
of carbonyl compounds were tested under standard
conditions. The results for the aldehydes are listed in
Table 2, and those for the ketones are shown in Tables 3
and 4.
1a
1b
1c
2a
2b
2c
2d
2e
2f
2g
2h
2i
cyclopentanone
25
18
1-indanone
2-indanone
70
56
85
44
15
33
cyclohexanone
4-tert-butylcyclohexanone
2-methylcyclohexanone
1-decalone
95
91
67
None of the aldehydes or ketones tested, with the
exception of trans-cinnamaldehyde, absorb any light
above 300 nm. The only species absorbing light under
the conditions of the reaction (Pyrex tubes) is CA. It can
be seen that most aldehydes typically react fast and give
high yields of the corresponding acetal. The decreased
reactivity of o-tolualdehyde (entry 6) and 2,6-dimethyl-
benzaldehyde (entry 7) suggest that either steric effects
or electronic effects are important. Electronic effects are
known to be important in acetalization reaction, which
results in ketones being less reactive than aldehydes.
However, it remains to be seen whether the o-methyl
groups cause a steric effect and/or an electronic effect.
The results for trans-cinnamaldehyde (entry 9) are
interesting. Analysis of the product mixture revealed the
presence of both the cis- and the trans-aldehydes (1:10
ratio), as well as the cis- and the trans-acetals (1:1.2
ratio). It has been reported that irradiation of trans-
cinnamaldehyde in MeOH results in the formation of cis-
and trans-cinnamaldehyde as well as the cis- and trans-
dimethylacetals.12 However, in our laboratory, irradiation
of trans-cinnamaldehyde without CA present (in 4:1
MeCN-MeOH) resulted only in some trans-cis isomer-
ization. These results suggest that formation of the acetal
is an acid-catalyzed process rather than a photochemical
process. Adkins and Hartung reported that extended
conjugation results in a lower reactivity for acetaliza-
tion.13 Our results are in agreement with that observa-
tion; trans-cinnamaldehyde reacts significantly more
slowly than hydrocinnamaldehyde.
32
100
<5c
61
2-decalone
d
d
94
46
R-tetralone
<5c
44
â-tetralone
6
100
54
camphor
36
[3,2,1]-bicyclo-2-octanone
2-adamantanone
cycloheptanone
1-benzosuberone
cyclooctanone
72
2j
87
100
3a
3b
4
12
<5c
<5c
35
5
6
7a
7b
8
3-pentanone
dibenzyl ketone
acetophenone
2,2,2-trifluoroacetophenone
benzylacetone
10
28
<5c
30
a [CA] ) 0.005 M; [ketone] ) 0.015 M; irradiated for 30 min in
MeOH. b Conversions and yields were determined by calibrated
GC/FID. c Within the limits of experimental error; no conversion
of starting material. d Product mixture consisted of several com-
pounds. Minor product was the acetal and major product was the
enol ether, but quantification was not possible because of signifi-
cant peak overlap.
ated in the presence of a mixture containing the ketone
(5%), acetal (80%) and enol ether (15%) was obtained.
Not all ketones react under these conditions. From
Table 4 it can be seen that, even when using methanol
as the solvent, several ketones (1-decalone, R-tetralone,
1-benzosuberone, cyclooctanone, and 2,2,2-trifluoroac-
etophenone) do not react at all (no conversion is ob-
served). Several other ketones do react but do not form
the acetal in any significant amount. In some cases small
amounts of dimerized products (as determined by GC/
MS) were formed instead, but often no distinct product(s)
were observed. In contrast, certain ketones, including
cyclohexanone, 4-tert-butylcyclohexanone, and 2-ada-
mantanone give excellent results. Often, the formation
of an enol ether is also observed; only in the cases of
cyclopentanone, 2-methylcyclohexanone, camphor, and
2-adamantanone was the acetal the only product formed.
The most reactive substrates seem to be those with six
membered rings (entries 2a-2j). Steric effects are im-
portant, as can be seen from entries 2c, 2d, and 2f.
Rigidity of the carbon skeleton is not of major importance;
Different results are obtained when using ketones. A
first set of experiments with cyclohexanone gave results
similar to those observed for benzaldehyde (Table 3). The
conversion was much larger when using pure methanol
as the solvent rather than an acetonitrile-methanol
mixture. The corresponding enol ether was always present,
except when the reaction was carried out in 4:1 MeCN-
MeOH. When cyclohexanone dimethylacetal was irradi-
(12) Ceppan, M.; Fiala, R.; Brezova, V.; Panak, J.; Motlikova, V.
Chem. Pap. 1994, 48, 25.
(13) Hartung, W. H.; Adkins, H. J. Am. Chem. Soc. 1925, 47, 1368;
1927, 49, 2517.
J. Org. Chem, Vol. 69, No. 24, 2004 8317

de Lijser and Rangel
even though 2-adamantanone gives the best results in
terms of acetal formation, cyclohexanone itself reacts
rapidly and forms the acetal in good yield. Other more
rigid structures such as the decalones, camphor, and
bicyclo[3.2.1]octan-2-one react poorly or give large amounts
of the enol ether. The observed reactivity of camphor and
bicyclo[3.2.1]octan-2-one is interesting and could possibly
be a result of the presence of the five- and seven-member
rings in these compounds. Substrates with five-mem-
bered rings are less reactive compared to six-membered
rings. Cyclopentanone reacts to give the acetal as the only
product albeit in low yield. Attaching an aromatic ring
in the â-position increases the reactivity, but now the
product mixture contains both the acetal and the enol
ether in a 1:1 ratio. These trends are similar to those
observed for cyclohexanone and the tetralones. Sub-
strates with seven-membered rings do not react; the
exception would be bicyclo[3.2.1]octan-2-one, although
this is more likely to be considered a substituted six-
membered ring. These results are in reasonable agree-
ment with those from the acid-catalyzed acetalization of
carbonyl compounds reported by Brown14 and Djerassi.15
In both studies it was observed that five- and seven-
membered rings react more slowly than six-membered
rings.
Steric effects are of major importance in these reac-
tions, as seen from the results of the reaction of aceto-
phenone. No reaction occurred, which is a dramatic
difference compared to benzaldehyde. It is well-known
that the acid-catalyzed acetalization of ketones is slower
than that of aldehydes, although the reaction can be
forced in the direction of the acetal.16 For example,
Roelofsen and van Bekkum reported that the acetaliza-
tion of ketones required larger amounts of the acid
catalyst in order to proceed.17 We have not yet been able
to increase the yield of acetal for unreactive ketones by
increasing the amount of CA. Irradiation of mixtures
containing acetophenone and various amounts of CA
(0.005, 0.015, and 0.025 M) did not result in the formation
of the acetal. The significant (steric) effect of a methyl
group can also be seen from the reactivity of 3-pentanone,
dibenzyl ketone, and benzylacetone. Even though the
phenyl group is further away from the carbonyl or even
completely absent, still no significant reaction can be
observed. This is consistent with a study by Wiberg et
al., which concluded that alkyl substitution stabilizes a
carbonyl group and destabilizes the acetal.18a
acetophenone does not yield any acetal, confirming the
predominance of steric effects in these reactions.
Mechanistic Interpretation. The most likely expla-
nation for these results is the in situ photogeneration of
an acidic species, which catalyzes the acetalization.
Others investigators have also observed the formation
of acetals in the photolysis of carbonyl compounds in
protic solvents.19,20 It these particular cases it was
proposed that the acetal is formed by means of an acid-
catalyzed process due to impurities in the solvent or via
a photoacid generation process. The formation of the
acetal was usually suppressed when using base-washed
glassware or when the reaction was carried out in the
presence of sodium bicarbonate.19a Kim and co-workers
observed the formation of 3-methoxycyclohex-2-en-1-one
from photolysis of 1,3-cyclohexanedione and p-benzo-
quinone in methanol and proposed an enol radical cation
intermediate.19c Hill and co-workers observed the forma-
tion of acetals upon photolysis of R-aryloxyacetones;
however, the reactions were not selective and the yields
were generally low.20 Clearly, the results reported here
cannot be due to impurities as the presence of a photo-
catalyst gives completely different (clean and reproduc-
ible) results. More evidence for the involvement of a
photogenerated acidic species came from an experiment
in which a sample was irradiated for only 1 min and then
placed in the dark. Analysis of the sample before and
after storage in the dark clearly shows that the reaction
continued on, although it was less efficient compared to
a sample that was irradiated for 30 min. An interesting
observation was the fact that almost identical results are
obtained when using CA or TCHQ, suggesting that the
latter is formed from CA and acts as the acid. This is
consistent with earlier observations. It was reported that
irradiation of CA in methanol or ethanol results in the
formation of TCHQ (major) and trichlorohydroxy-p-
benzoquinone (minor).21 Another important piece of
information is the fact that the reaction only requires a
catalytic amount of CA. We will discuss three mechanistic
schemes, each of which can explain these results.
The results presented in the tables above are very
similar to those observed for the classic acid-catalyzed
acetalization of carbonyl compounds.1 As such, a simple
mechanism generating an inorganic acid such as HCl
must be considered first. It has been reported that
photolysis of CA in aldehydes results in the formation of
acylhydroquinones and generates hydrogen chloride
(Scheme 1).22
In addition to the steric effects, we were also interested
to see if electronic effects were important for the reactions
of ketones. It is known that the introduction of electron-
accepting groups destabilize the carbonyl group and favor
formation of the acetal.18b Preliminary experiments sug-
gest that electronic effects are not predominant in these
reactions. For example, irradiation of 2,2,2-trifluoro-
In order for HCl to be formed, the acyl radical must
add to a chlorine-substituted ring carbon. Studies on CA
and different aldehydes have shown, however, that the
major product forms via reaction at the oxygen to produce
an ester. For example, the reaction of CA with acetalde-
hyde produces tetrachloroquinol monoacetate (73%) and
trichloroacetylquinol (1%).22 The reaction of CA with
benzaldehyde reportedly produced only the benzoate
(14) Brown, H. C.; Brewster, J. H.; Shechter, H. J. Am. Chem. Soc.
1954, 76, 467.
(15) Djerassi, C.; Mitscher, L. A.; Mitscher, B. J. J. Am. Chem. Soc.
1959, 81, 947.
(16) For examples, see: (a) Dauben, W. G.; Gerdes, J. M.; Look, G.
C. J. Org. Chem. 1986, 51, 4964. (b) Otera, J.; Mizutani, T.; Nozaki,
H. Organometallics 1989, 8, 2063. (c) Thurkauf, A.; Jacobson, A. E.;
Rice, K. C. Synthesis 1988, 233.
(17) Roelofsen, D. P.; van Bekkum, H. Synthesis 1972, 419.
(18) (a) Wiberg, K. B.; Morgan, K. M.; Maltz, H. J. Am. Chem. Soc.
1994, 116, 11067. (b) Bell, R. P. Adv. Phys. Org. Chem. 1966, 4, 1.
(19) (a) Yates, P. Pure Appl. Chem. 1968, 16, 93. (b) Coyle, J. D.
Introduction to Organic Photochemistry; John Wiley & Sons: New
York, 1989; p. 119. (c) Kim, S. S.; Chang, J. A.; Kim, A. R.; Mah, Y. J.;
Kim, H. J.; Kang, C. J. Photosci. 2000, 7, 111.
(20) (a) Dirania, M. K. K.; Hill, J. J. Chem. Soc. C 1968, 1311. (b)
Collier, J. R.; Dirania, M. K. M.; Hill, J. J. Chem. Soc. C 1970, 155. (c)
Dirania, M. K. K.; Hill, J. J. Chem. Soc. C 1971, 1213.
(21) Fisch, M.; Hemmerlin, W. M. Tetrahedron Lett. 1972, 3125.
(22) Bruce, J. M.; Ellis, J. N. J. Chem. Soc. C 1966, 1624.
8318 J. Org. Chem., Vol. 69, No. 24, 2004

Photochemical Acetalization of Carbonyl Compounds
SCHEME 1
SCHEME 2
ester.23 Although under our conditions, on the basis of
the concentrations of MeOH and the aldehyde used, this
reaction is unlikely, it must be noted that in order for
the acetalization to occur, only a catalytic amount of HCl
is required. It cannot be ruled out that when using CA
as the sensitizer side reactions occur that produce this
catalyst. Once it forms, it could easily be regenerated,
not requiring any further side reactions.
One observation that does not support this mechanism,
however, is the fact that almost identical results are
obtained when using xanthone (XA) instead of CA. XA
is not expected to form such acid catalysts. This observa-
tion, together with the argument that the concentration
of the aldehyde is more than 1600 times smaller than
that of the solvent, suggest that this mechanism is most
likely not valid.
A second possible mechanism, which incorporates the
observation that quinones are reduced to quinols upon
photolysis in hydroxylic solvents, is shown in Scheme 2.
Photolysis of CA in methanol results in the formation
of TCHQ.21 Excitation of this hydroquinone generates an
acidic species that protonates the carbonyl compound
(directly or indirectly via a proton carrier such as MeOH).
Further reaction of this protonated species with the
solvent yields the acetal.
There is ample precedent for the enhanced acidity of
phenolic species upon irradiation. For example, 2-cyano-
phenol has an excited-state pKa of 0.66 (compared to a
pKa of 6.97 in the groundstate).24 The pKa* of TCHQ is
not known, and therefore it remains to be seen whether
it is acidic enough to protonate either MeOH or a car-
+
bonyl group. The pKa value for MeOH2 is -2.2 (esti-
mated),25 whereas those of CdOH+ species are estimated
to be between -6 and -10,25,26 depending on the struc-
ture. For example, the conjugate acids of cyclohexanone
and benzaldehyde have pKa’s of -6.8 and -7.1, respec-
tively. On the other hand, it must be noted that weaker
acids are also capable of catalyzing the acetalization of
carbonyl compounds, although it often requires removal
of water from the mixture in order to push the equilib-
(24) Schulman, S. G.; Vincent, W. R.; Underberg, W. J. M. J. Phys.
Chem. 1981, 85, 4068.
(25) Arnett, E. M. Prog. Phys. Org. Chem. 1963, 1, 223.
(26) (a) Levy, G. C.; Cargioli, J. D.; Racela, W. J. J. Am. Chem. Soc.
1970, 92, 6238. (b) Yates, K.; Stewart, R. Can. J. Chem. 1959, 37, 664.
(c) Stewart, R.; Yates, K. J. Am. Chem. Soc. 1958, 80, 6355. (d) Deno,
N. C.; Gaugler, R. W.; Wisotsky, M. J. J. Org. Chem. 1966, 31, 1967.
(e) Lee, D. G. Can. J. Chem. 1970, 48, 1919. (f) Fischer, A.; Grigor, B.
A.; Packer, J.; Vaughan, J. J. Am. Chem. Soc. 1961, 83, 4208.
(23) Moore, R. F.; Wasters, W. A. J. Chem. Soc. 1953, 238.
J. Org. Chem, Vol. 69, No. 24, 2004 8319

de Lijser and Rangel
rium in the direction of the acetal. In another example,
N-hydroxybenzenesulfonamide (pKa ) 9.26) was used
successfully as a catalyst for the acetalization of p-
chlorobenzaldehyde, but its catalytic activity was sug-
gested to be a result of its combined nucleophilicity and
leaving group ability.27
To find out whether weaker photoacids such as 6-cy-
ano-2-naphthol, 6CN2 (pKa* ) 0.2), and 2-cyanophenol,
2CP (pKa* ) 0.66), can act as the photocatalyst, a MeOH
solution containing benzaldehyde and 6CN2 or 2CP was
irradiated for 30 min. Neither reaction yielded any acetal,
even when the concentration of the photoacid was
increased. These results suggest that the excited states
of 2CP and 6CN2 are not strong enough to protonate
either the solvent or the carbonyl. From these observa-
tions, it follows that (i) TCHQ is either a much stronger
photoacid than 2CP and 6CN2, or (ii) the reaction occurs
via a different mechanism. Because the pKa* of TCHQ
is unknown, we have focused our attention on a third
possible pathway, which involves an electron transfer
(ET) step.
Flash photolysis studies have shown that irradiation
of chloranil (and other quinones) in hydroxylic solvents
results in the formation of the semiquinone radical, which
exists in equilibrium with the chloranil (quinone) radical
anion:28
FIGURE 1. Absorption spectrum of the semiquinone radical,
generated via nanosecond laser flash photolysis (355 nm) of a
solution containing CA and acetophenone oxime in MeOH.
SCHEME 3
In ethanol, the equilibrium favors the semiquinone
radical (K ) 1 × 10-6 M), and the rate constant for the
recombination reaction was estimated to be 7.4 × 109 M-1
s-1 28b
The decay of the semiquinone radical was proposed
.
to occur via a disproportionation reaction rather than
reaction with the solvent. The lifetime of the semiquinone
radical is on the order of milliseconds, which is consistent
with our own observations. The species observed upon
flash photolysis (355 nm) of a solution containing CA and
an electron donor (acetophenone oxime) in MeOH is long-
lived (no significant decay of the signal on the microsec-
ond time scale was observed; Figure 1). The signal is
attributed to that of the semiquinone radical, which is
formed via reaction of the radical anion with the sol-
vent.10 Considering these studies, the mechanism shown
in Scheme 3 can be proposed.
quinone radical (CA-H•) and the hydroxymethyl radical
(•CH2OH). Ground state CA is easily reduced (Ered
+0.02 V), and ET from the hydroxymethyl radical (Eox
)
)
-0.98 V)29 generates the CA radical anion (which reacts
rapidly with MeOH to form the semiquinone radical),10
as well as the conjugate acid of formaldehyde, a strong
acid. The acid generated this way can catalyze the acetal
formation. This mechanism is energetically feasible,30 but
it ignores the involvement of TCHQ, which was shown
to be an effective photocatalyst as well. This observation
is better explained by the second sequence of reactions.
Ionization of the hydroquinone (most likely via a photo-
In principle, two catalytic cycles are available. The first
step in both sequences involves formation of the semi-
(27) Hassner, A.; Wiederkehr, R.; Kascheres, A. J. J. Org. Chem.
1970, 35, 1962.
(28) (a) Bridge, N. K.; Porter, G. Proc. R. Soc. A 1958, 244, 259. (b)
Bridge, N. K.; Porter, G. Proc. R. Soc. Ser. A 1958, 244, 276. (c) Hudson,
A.; Lewis, J. W. J. Chem. Soc. B 1969, 531. (d) Hales, B. J.; Bolton, J.
R. Photochem. Photobiol. 1970, 12, 239.
(29) (a) Lilie, V. J.; Beck, G.; Henglein, A. Ber. Bunsen-Ges. Phys.
Chem. 1971, 75, 458. (b) Kolt, R. J.; Wayner, D. D. M.; Griller, D. J.
Org. Chem. 1989, 54, 4259. (c) Lund, T.; Wayner, D. D. M.; Jonsson,
M.; Larsen, A. G.; Daasbjerg, K. J. Am. Chem. Soc. 2001, 123, 12590.
8320 J. Org. Chem., Vol. 69, No. 24, 2004

Photochemical Acetalization of Carbonyl Compounds
FIGURE 3. Comparison of the spectrum obtained from
photolysis (355 nm) of TCHQ in MeCN (O) to that of the
semiquinone radical (0). The semiquinone radical was gener-
ated via photolysis of a solution containing CA and benzyl-
trimethylsilane in the presence of trifluoroacetic acid. The
spectra were recorded approximately 200 ns after the laser
pulse.
FIGURE 2. Comparison of the spectrum obtained from
photolysis (355 nm) of TCHQ in MeOH (0) to that of the
semiquinone radical (O). The spectra were recorded ap-
proximately 2.5 µs after the laser pulse and are normalized.
induced electron-transfer pathway with an appropriate
electron acceptor) would generate the TCHQ radical
cation, a strong acid.31 Deprotonation of this species
produces the semiquinone radical and yields the required
acid for the acetalization.
logical explanation would be ionization of TCHQ to form
the radical cation, which rapidly loses a proton to yield
the (long-lived) semiquinone radical. The radical cation
itself was not observed; however, this is easily explained
based on the fact that phenolic radical cations react very
rapidly with nucleophiles such as alcohols.9
A similar result was obtained by photolysis of TCHQ
in acetonitrile. The observed spectrum (Figure 3) is
identical to that of the semiquinone radical generated via
reduction of CA in the presence of trifluoroacetic acid.10
Even in a nonnucleophilic solvent such as MeCN the
deprotonation step is very fast, most likely as a result of
traces of water present in the solvent.
The results described above are most consistent with
a mechanism involving the two reaction sequences shown
in Scheme 3 operating side-by-side. Both sequences lead
to the formation of the semiquinone radical, which
disproportionates to yield TCHQ, the photocatalyst, as
well as CA, which can participate in a new cycle. Further
studies on the scope and mechanism of these reactions,
including LFP studies on the radical cations intermedi-
ates are underway.
The presence of formaldehyde (or the dimethylacetal)
in the reaction mixture would provide evidence for this
mechanism. Unfortunately, neither formaldehyde nor its
dimethyl acetal can be detected by GC/FID under these
conditions because of overlap with other components in
the mixture. Instead, we tried a different approach and
used 1-hexanol as the solvent. According to Scheme 3,
1-hexanol would react to give hexanal, which should be
detectable under the conditions of the reaction. Irradia-
tion of a mixture containing benzaldehyde and CA in
1-hexanol did indeed result in the formation of hexanal.
This result is most consistent with the mechanism shown
in Scheme 3.
LFP studies on TCHQ in MeOH also support the
mechanism shown in Scheme 3. Photolysis of a solution
containing TCHQ in MeOH results in the formation of a
signal with maxima at 450 and 420 nm. Comparison of
this spectrum to that of the semiquinone radical (Figure
1) shows that they are virtually identical (Figure 2),
suggesting the formation of this radical species. The most
Conclusions
(30) The free energy for electron transfer (∆GET) is calculated by
the following equation: ∆GET ) Eox - Ered. Using the reported numbers
yields ∆GET ) -23 kcal mol-1. Note that the reported oxidation
potential for •CH2OH was measured in aqueous solution. In other
solvents (e.g., MeOH) this value is likely to be more positive; however,
the overall reaction is still expected to be favorable. This conclusion is
based on the reported oxidation potentials of (CH3)2C•OH in acetonitrile
(-0.60 V), a 3:1 mixture of 2-propanol-acetonitrile (-1.1 V), and water
(-1.30 V).29 Similar trends are expected for the hydroxymethyl radical.
(31) Although the pKa of TCHQ radical cation is not known,
Bordwell and Cheng8c have estimated the pKa values of similar species;
the radical cation of 3,5-dichlorophenol has a pKa of -14 and that of
3,4,5-trichlorophenol a pKa of -12.5. According to these numbers, we
expect the radical cation of TCHQ to be a strong enough acid to
protonate a carbonyl group.
Aldehydes can be conveniently converted into their
dimethyl acetals via a photochemical reaction with CA
in MeOH. Ketones are less reactive most likely because
of steric restraints. Some ketones were found to react
rapidly, but they often also produce the enol ether.
Although the actual mechanism remains uncertain, the
results described are most consistent with the formation
of a photocatalyst (TCHQ) from CA. The actual acid is
formed via ionization of TCHQ, which results in the
highly acidic TCHQ radical cation. Loss of a proton to
the solvent or the carbonyl species yields the semiquinone
J. Org. Chem, Vol. 69, No. 24, 2004 8321

de Lijser and Rangel
saturated) solution was irradiated in the Rayonet photo-
chemical reactor for 30 min. Afterward, the solution was
analyzed by GC/FID (three times), and the products were
identified by GC/MS. Conversion of the starting material and
product yields were determined by calibrated GC/FID. The
photolysis of benzaldehyde and CA in MeOH under argon
resulted in a similar conversion and product yield.
Laser Flash Photolysis Experiments. The apparatus
used for the laser flash photolysis (LFP) experiments was of
standard design,33 and the details have been described else-
where.34 Spectra were obtained by flash photolysis (355 nm,
10 Hz, 0.5-2 mJ/pulse; 4 ns pulse width) of argon-saturated
solutions (3 mL) in a glass cuvette containing the compound
of interest (OD ca. 0.5-1).
radical, which can regenerate the photocatalyst via a
disproportionation reaction. The advantage of this method
is that the photocatalyst is regenerated via a dispropor-
tionation reaction requiring only a catalytic amount of
CA (or the hydroquinone itself). The reaction can be done
at longer wavelengths, making this method attractive
and potentially useful for applications requiring lower-
energy radiation.
Experimental Section
Chemicals and Equipment. Methanol and acetonitrile
were of spectrophotometric grade. All carbonyl compounds
were of the highest purity. If necessary, they were distilled or
recrystallized. All quinones were recrystallized before use.
Samples were irradiated in a photochemical reactor, equipped
with 16 black light phosphor bulbs (λ ) 350 nm). The progress
of the reactions was followed by GC/FID, and the products
were identified by GC/MS. Conversion of the starting material
and product yields were determined by calibrated GC/FID.
Details of the experimental methods have been reported
elsewhere.32
Steady-State Photolysis Experiments. Appropriate
amounts of the carbonyl compound (0.015 M) and chloranil
(0.005 M) were weighed out and dissolved in 5 mL of solvent
in a Pyrex tube. A small amount of the solution was removed
and analyzed by GC/FID (three times). The remaining (air-
Acknowledgment. This work was supported by
Research Corporation (CC5107) and California State
University, Fullerton. Part of this work was supported
by the National Science Foundation under Grant
0097795. We are grateful to Professor I. R. Gould
(Arizona State University) for the use of his laser flash
photolysis equipment. We thank Mr. Ryan Shade for
his assistance with some of the experiments.
JO0485886
(33) Herkstroeter, W. G.; Gould, I. R. In Physical Methods of
Chemistry Series, 2nd ed.; Rossiter, B., Baetzold, R., Eds.; Wiley: New
York, 1993; Vol. 8, p 225.
(34) Lorance, E. D.; Kramer, W. H.; Gould, I. R. J. Am. Chem. Soc.
2002, 124, 15225.
(32) de Lijser, H. J. P.; Kim, J. S.; McGrorty, S. M.; Ulloa, E. M.
Can. J. Chem. 2003, 81, 575.
8322 J. Org. Chem., Vol. 69, No. 24, 2004