Effect of structure in benzaldehyde oximes on the formation of aldehydes and nitriles under photoinduced electron-transfer conditions

Article abstract of DOI:10.1021/jo061325m

(Chemical Equation Presented) The mechanistic aspects of the photosensitized reactions of a series of benzaldehyde oximes (1a-o) were studied by steady-state (product studies) and laser flash photolysis methods. Nanosecond laser flash photolysis studies have shown that the reaction of the oxime with triplet chloranil (³CA) proceeds via an electron-transfer mechanism provided the free energy for electron transfer (ΔG_ET) is favorable; typically, the oxidation potential of the oxime should be below 2.0 V. Substituted benzaldehyde oximes with oxidation potentials greater than 2.0 V quench ³CA at rates that are independent of the substituent and the oxidation potential. The most likely mechanism under these conditions is a hydrogen atom transfer mechanism as this reaction should be dependent on the O-H bond strength only, which is virtually the same for all oximes. Product studies have shown that aldoximes feact to give both the corresponding aldehyde and the nitrile. The important intermediate in the aldehyde pathway is the iminoxyl radical, which is formed via an electron transfer-proton transfer (ET-PT) sequence (for oximes with low oxidation potentials) or via a hydrogen atom transfer (HAT) pathway (for oximes with larger oxidation potentials). The nitriles are proposed to result from intermediate iminoyl radicals, which can be formed via direct hydrogen atom abstraction or via an electron-transfer-proton- transfer sequence. The experimental data seems to support the direct hydrogen atom abstraction as evidenced by the break in linearity in the plot of the quenching rates against the oxidation potential, which suggests a change in mechanism. The nitrile product is favored when electron-accepting substituents are present on the benzene ring of the benzaldehyde oximes or when the hydroxyl hydrogen atom is unavailable for abstraction. The latter is the case in pyridine-2-carboxaldoxime (2), where a strong intramolecular hydrogen bond is formed. Other molecules that form weaker intramolecular hydrogen bonds such as 2-furaldehyde oxime (3) and thiophene-2-carboxaldoxime (4) tend to yield increasing amounts of aldehyde.

Full text of DOI:10.1021/jo061325m

Effect of Structure in Benzaldehyde Oximes on the Formation of
Aldehydes and Nitriles under Photoinduced Electron-Transfer
Conditions
H. J. Peter de Lijser,* Susan Hsu, Bernadette V. Marquez, Adriana Park,
Nawaporn Sanguantrakun, and Jody R. Sawyer
Department of Chemistry & Biochemistry, California State UniVersitysFullerton,
Fullerton, California 92834-6866
pdelijser@fullerton.edu
ReceiVed June 27, 2006
The mechanistic aspects of the photosensitized reactions of a series of benzaldehyde oximes (1a-o)
were studied by steady-state (product studies) and laser flash photolysis methods. Nanosecond laser flash
photolysis studies have shown that the reaction of the oxime with triplet chloranil (3CA) proceeds via an
electron-transfer mechanism provided the free energy for electron transfer (∆GET) is favorable; typically,
the oxidation potential of the oxime should be below 2.0 V. Substituted benzaldehyde oximes with
3
oxidation potentials greater than 2.0 V quench CA at rates that are independent of the substituent and
the oxidation potential. The most likely mechanism under these conditions is a hydrogen atom transfer
mechanism as this reaction should be dependent on the O-H bond strength only, which is virtually the
same for all oximes. Product studies have shown that aldoximes react to give both the corresponding
aldehyde and the nitrile. The important intermediate in the aldehyde pathway is the iminoxyl radical,
which is formed via an electron transfer-proton transfer (ET-PT) sequence (for oximes with low oxidation
potentials) or via a hydrogen atom transfer (HAT) pathway (for oximes with larger oxidation potentials).
The nitriles are proposed to result from intermediate iminoyl radicals, which can be formed via direct
hydrogen atom abstraction or via an electron-transfer-proton-transfer sequence. The experimental data
seems to support the direct hydrogen atom abstraction as evidenced by the break in linearity in the plot
of the quenching rates against the oxidation potential, which suggests a change in mechanism. The nitrile
product is favored when electron-accepting substituents are present on the benzene ring of the benzaldehyde
oximes or when the hydroxyl hydrogen atom is unavailable for abstraction. The latter is the case in
pyridine-2-carboxaldoxime (2), where a strong intramolecular hydrogen bond is formed. Other molecules
that form weaker intramolecular hydrogen bonds such as 2-furaldehyde oxime (3) and thiophene-2-
carboxaldoxime (4) tend to yield increasing amounts of aldehyde.
Introduction
mechanism for this transformation involves electron transfer,
followed by a proton transfer (ET-PT) and follow-up reactions,
including a second oxidation and nucleophilic attack by water,
to form the carbonyl compound (Scheme 1).1c
The photooxidation of oximes has been shown to yield the
corresponding ketones as the major product.1 The proposed
The proposed intermediate in the photooxidation reactions
of oximes is the corresponding iminoxyl radical, which is formed
via the ET-PT sequence. It must be noted that a hydrogen atom
transfer (HAT) pathway would also lead to the formation of
this reactive intermediate; however, this pathway was not
* To whom correspondence should be addressed. Phone: (714) 278-3290.
Fax: (714) 278-531.
(1) (a) de Lijser, H. J. P.; Fardoun, F. H.; Sawyer, J. R.; Quant, M. Org.
Lett. 2002, 4, 2325. (b) de Lijser, H. J. P.; Kim, J. S.; McGrorty, S. M.;
Ulloa, E. M. Can. J. Chem. 2003, 81, 575. (c) Park, A.; Kim, J. S.; Kosareff,
N. M.; de Lijser, H. J. P. Photochem. Photobiol. 2006, 82, 110.
10.1021/jo061325m CCC: $33.50 © 2006 American Chemical Society
Published on Web 09/06/2006
J. Org. Chem. 2006, 71, 7785-7792
7785

de Lijser et al.
SCHEME 1
TABLE 1. Summary of Oxidation Potentials, Ionization Potentials, Quenching Rates, and Steady-State Photolysis Data of Meta- and
Para-Substituted Benzaldehyde Oximes
Epa (V)
∆GET (kcal mol-1
)
IP (kcal mol-1
)
kqd (M-1 s-1
)
% Ce
% Nf
% Af
N/Ag
b
c
oxime
X
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1o
p-OCH3
m-OCH3
p-CH3
m-CH3
p-F
p-Cl
p-H
m-Cl
m-F
m-CN
m-NO2
m-CF3
p-CN
p-CF3
p-NO2
1.45
1.77
1.77
1.84
1.84
1.85
1.90
1.99
2.01
2.10
2.14
2.17
2.19
2.22
-16.1
-8.8
-8.8
-7.1
-7.1
-6.9
-5.8
-3.7
-3.2
-1.2
-0.2
+0.5
+0.9
+1.6
184.44
189.60
191.29
193.95
196.36
196.54
195.76
199.34
200.71
203.51
209.51
205.09
203.27
207.01
211.93
1.02 × 1010
7.24 × 109
5.98 × 109
5.86 × 109
3.98 × 109
3.46 × 109
6.58 × 108
2.96 × 108
3.00 × 108
1.55 × 108
4.68 × 107
1.58 × 108
6.27 × 107
1.58 × 108
1.25 × 108
51
74
32
28
17
26
23
21
35
28
28
18
46
17
35
28
44
13
12
9
13
11
11
23
19
20
8
23
30
20
16
8
13
12
11
12
8
1.2
1.5
0.6
0.8
1.1
0.9
0.9
1.0
2.0
2.3
2.2
0.8
1.5
1.0
1.4
9
10
19
8
27
8
21
15
a Oxidation (peak) potentials measured by cyclic voltammetry (0.1 M tetraethylammonium perchlorate in CH3CN, Ag/AgCl electrode). b Calculated using
the Weller equation (ref 10): ∆GET ) 23.06[E°x - Ered - ET] kcal/mol. c Ionization potentials calculated by AM1. Listed value is that of the syn isomer.
d Quenching rate of CA determined by laser flash photolysis. e Conversion was calculated on the basis of the GC-FID peak area of the oxime before and
3
after photolysis. Product yields (% nitrile and % aldehyde) determined by calibrated GC/FID. g Ratio of nitrile and aldehyde products.
f
believed to be important under those conditions.1c A similar
mechanism may be involved in the enzymatic oxidation of
oximes; possible pathways for the detoxification of xenobiotics
by the cytochrome P450 enzymes include electron transfer
initiated processes as well as reactions via hydrogen atom
abstraction.2 Enzymatic oxidation and photooxidation of oximes
yield the same products as shown by a recent study on the use
of hydroxyimines as prodrugs for carbonyl-containing drugs
such as ketoprofen and nabumetone.3 Oxime radical cations and
iminoxyl radicals also are the proposed intermediates in the nitric
oxide synthase (NOS) catalyzed metabolism of L-arginine to
L-citrulline.4,5 The similarities between the P450 and NOS
systems have been reported.6 Furthermore, iminoxyl radicals
have also been proposed as intermediates in the formation of
nitric oxide from oximes.7,8 Interestingly, very little is known
about the structure and reactivity of these reactive intermediates.
the same as that leading from the ketoxime to the corresponding
ketone (Scheme 1); however, the pathway and intermediates
involved in the formation of the nitrile are not yet known. It is
interesting to note that the enzymatic oxidation of aldoximes was
reported to yield only the corresponding nitriles.9 The goal of the
work described in this paper is to learn more about the possible
pathways and intermediates involved in the photooxidation of
aldoximes. We have undertaken an extensive study involving the
steady-state (SS) and laser flash photolysis (LFP) of meta- and
para-substituted benzaldehyde oximes (1a-o) using triplet sensi-
tizers. The photochemical experiments were complemented by
electrochemical data as well as semiempirical (AM1) and density
functional theory (DFT) calculations on the neutral compounds,
the radical cation species, and some of the proposed intermediates.
Results and Discussion
In our previous work, we noticed that the photooxidation of
aldoximes yielded both the corresponding aldehyde and the nit-
rile.1a The pathway leading to the aldehyde product is most likely
A. Oxidation and Ionization Potentials of Meta- and Para-
Substituted Benzaldehyde Oximes. The measured oxidation
(peak) potentials (Ep) for the meta- and para-substituted ben-
zaldehyde oximes are listed in Table 1. The oxidative scans
are broad and irreversible in every case. No reliable data could
be obtained for oxime 1o. The data shows the expected trend
in which electron-accepting substituents increase the observed
oxidation potential whereas electron-donating substituents have
the opposite effect. Electron transfer is expected to take place
with most aldoximes when using chloranil (CA; E*red ) 2.15
(2) Guengerich, F. P. In Cytochromes P450: Metabolic and Toxicological
Aspects; Ioannides, C., Ed.; CRC Press: Boca Raton, 1996; p 55.
(3) Kumpulainen, H.; Ma¨ho¨nen, N.; Laitinen, M.-J.; Jaurakkaja¨rvi, M.;
Raunio, H.; Juvonen, R. O.; Vepsa¨la¨inen, J.; Ja¨rvinen, T.; Rautio, J. J. Med.
Chem. 2006, 49, 1207.
(4) (a) Cai, T.; Xian, M.; Wang, P. G. Bioorg. Med. Chem. Lett. 2002,
12, 1507. (b) Sanakis, Y.; Goussias, C.; Mason, R. P. Petrouleas, V.
Biochemistry 1997, 36, 1411. (c) Halliwell, B.; Gutteridge, J. M. C. Free
Radicals in Biology and Medicine, 3rd ed.; Oxford University Press: Oxford,
1999. (d) Rosen, G. M.; Tsai, P.; Pou, S. Chem. ReV. 2002, 102, 1191. (e)
Jousserandor, A.; Boucher, J.-L.; Henry, Y.; Niklaus, B.; Clement, B.;
Mansuy, D. Biochemistry 1998, 37, 17179.
V) as the sensitizer based on the calculated free energy (∆GET
)
using the Weller equation.10 The ionization potentials (IP) of
the benzaldehyde oximes were determined by theoretical
methods (AM1). Both syn and anti isomers were considered.
The geometry of the optimized neutral compound was used as
(5) Cho, J. Y.; Dutton, A.; Miller, T.; Houk, K. N.; Fukuto, J. M. Arch.
Biochem. Biophys. 2003, 417, 65.
(6) See for example: Miller, R. T. Curr. Drug. Metab. 2004, 5, 535.
(7) Wang, P. G.; Xian, M.; Tang, X.; Wu, X.; Wen, Z.; Cai, T.; Janczuk,
A. J. Chem. ReV. 2002, 102, 1091.
(8) Koikov, L. N.; Alexeeva, N. V.; Lisitza, E. A.; Krichevsky, E. S.;
Grigoryev, N. B.; Danilov, A. V.; Severina, I. S.; Pyatakova, N. V.; Granik,
V. G. MendeleeV Commun. 1998, 165.
(9) (a) De Master, E. G.; Shirota, F. N.; Nagasawa, H. T. J. Org. Chem.
1992, 57, 5074. (b) Boucher, J.-L.; Delaforge, M.; Mansuy, D. Biochemistry
1994, 33, 7811.
(10) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 59.
7786 J. Org. Chem., Vol. 71, No. 20, 2006

Formation of Aldehydes and Nitriles under Electron-Transfer Conditions
FIGURE 1. Plot of the measured oxidation potentials (Ep) of a series
of meta- and para-substituted benzaldehyde oximes against the calcu-
lated (AM1) ionization potentials (IP).
a starting point for a geometry optimization of the corresponding
radical cation species. The results show that the syn isomers
are of lower energy than the corresponding anti isomers but
the differences are small (never more than 2 kcal/mol). Previous
work has shown that the calculated IP values are very similar
when using either DFT or AM1.1b,c A plot of the measured
oxidation potentials against the calculated IP (AM1) gives an
excellent correlation (r2 ) 0.92; Figure 1). The highest IP was
calculated for 1o, which is consistent with the fact that no
reliable oxidation potential could be observed for that oxime.
These results confirm that the measured oxidation (peak)
potentials are an accurate parameter for testing whether the
reactions proceed via an initial electron-transfer step.
Further analysis of the electrochemical data revealed that they
also correlate well with several polar substituent constants (σ-
values) such as σpol11 (r2 ) 0.82) and σmb12 (r2 ) 0.84), but not
with radical substituent constants such as σJJ (r2 ) 0.04) or
12
FIGURE 2. Plots of (A) the correlation between the measured
oxidation potentials (Ep) of a series of substituted benzaldehyde oximes
and the measured quenching rates of triplet chloranil and (B) the
correlation between the calculated (AM1) ionization potentials and the
measured quenching rates of triplet chloranil. The obvious break in
both plots appearing around 2 V or 200 kcal/mol presumably suggests
a change in mechanism.
13
σrad (r2 ) 0.02). Similar results were obtained for a series of
substituted acetophenone oximes1b,c and suggest that the elec-
trochemical process results in the formation of cationic species
rather than radicals, consistent with a study by Tallec et al.14
B. Laser Flash Photolysis Studies: Quenching of Triplet
Chloranil. All substituted benzaldehyde oximes quenched triplet
chloranil (3CA); the quenching rate constants are listed in Table
1 and varied from a low of 4.68 × 107 M-1s-1 (1k) to a high
These observations suggest a change of mechanism as a result
of the structural differences. Triplet quinones can be quenched
by substrates via a number of different pathways (e.g., electron
transfer, energy transfer, hydrogen atom transfer, etc.). As
proposed before,1b a linear correlation between the measured
oxidation potentials and the quenching rates suggests an initial
electron-transfer step, which is the case for aldoximes with Ep
< 2 V. However, for aldoximes with Ep > 2 V the quenching
rate is independent of the substituent, and therefore, electron
transfer (ET) is most likely not involved. Other pathways that
may become dominant under these conditions are energy transfer
or hydrogen atom transfer (HAT). In the case of HAT, the
quenching rate would be determined by the O-H bond strength.
Bordwell and co-workers have reported that the O-H bond
strength in neutral oximes is virtually independent of the
of 1.02 × 1010 M-1 -1
s
(1a). A plot of the quenching rate against
the oxidation potentials shows a linear behavior for those oximes
with Ep < 2 V; however, the quenching rate levels off when
using oximes with Ep > 2 V (Figure 2A). A similar trend is
observed when using the IP data (Figure 2B). The quenching
of 3CA by a series of substituted acetophenone oximes did not
show this trend even though some of the measured oxidation
potentials were above 2 V.1b
(11) McDaniel, D. H.; Brown, H. C. J. Org. Chem. 1958, 23, 420.
(12) Jiang, X. K.; Ji, G. Z. J. Org. Chem. 1992, 57, 6051.
(13) (a) Creary, X. J. Org. Chem. 1980, 45, 280. (b) Creary, X.;
Mehrsheikh-Mohammadi, M. E.; McDonald, S. J. Org. Chem. 1987, 52,
3254.
(14) Benchariff, L.; Tallec, A.; Tardivel, R. Electrochim. Acta 1997, 42,
3509.
J. Org. Chem, Vol. 71, No. 20, 2006 7787

de Lijser et al.
TABLE 2. Comparison of Hammett Data Obtained from Laser
Flash Photolysis (LFP) Experiments on Meta- and Para-Substituted
Benzaldehyde and Acetophenone Oximes
meta
para
benz-
aldehyde
oxime
benz-
aldehyde
oxime
acetophenone
acetophenone
oxime
oxime
Frad
Fpol
-45
-10.7
4.2
0.94
5.75
-1.86
-3.1
0.90
-30
-7.1
4.2
0.94
3.16
-1.22
-2.6
0.95
0.87
-2.3
-0.38
0.84
0.44
-0.89
-0.49
0.93
F
rad/Fpol
r2
FJJ
Fmb
FJJ/Fmb
r2
0.83
0.56
-1.51
-0.5
0.90
-0.56
-1.0
0.91
substituent,15 whereas in oxime radical cations the O-H bond
strength is directly related to the substituent.16 These results
suggest that the quenching process changes from an ET process
to a HAT process if ET becomes energetically unfavorable. The
differences in reactivity between aldoximes and ketoximes may
be due to the presence of another abstractable hydrogen in
aldoximes. The bond strength of the aldehyde dC-H is
approximately 88 kcal/mol,17 which is almost identical to the
O-H bond strength in oximes (assuming that the dC-H bond
strength in oximes is similar to that in aldehydes).15,18 Such a
(competing) pathway is not available in acetophenone oximes.19
To see if the change in mechanism was directly related to
the substituents present on the ring, the quenching data was
analyzed in terms of different Hammett parameters. Both polar
(σpol and σmb) and radical (σrad and σJJ) substituent constants
were used. In all cases, dual parameter sets (polar + radical)
were required to obtain useful data (Table 2). For comparison,
the data from the ketoximes studies is also included.
Little difference is seen between the data set obtained from
a series of benzaldehyde oximes when compared to the data
set from the acetophenone oxime series.1b Despite the clear break
that was observed in Figure 2, analysis of the quenching data
in terms of polar and radical substituent constants does not lend
support for the involvement of different intermediates in the
formation of aldehydes and nitriles from aldoximes. However,
it must be noted that steady-state experiments (see below)
showed that in all cases both the nitrile and the aldehyde were
present in the product mixtures, suggesting that possibly more
than one pathway can be followed from the proposed intermedi-
ate, which would complicate the kinetic analysis of the data. A
better mechanistic picture would probably be available if a single
aldoxime would react exclusively to give the nitrile product.
FIGURE 3. Plot of the nitrile to aldehyde product ratio as a function
of the substituent in the DDQ-sensitized reactions of substituted
benzaldehyde oximes.
Furthermore, if the mechanism were to switch from ET to HAT,
abstraction of the hydrogen from the OH group would yield an
iminoxyl radical, which, according to Scheme 1, would undergo
an electron transfer step and follow-up reactions to yield the
aldehyde product. This pathway would further cloud the
mechanistic aspects of these reactions. Since the quenching data
only involves the initial step of the reaction, we have opted to
analyze the aldoxime conversion as well as the product
distribution in terms of polar and radical Hammett parameters
in order to see if any trends would arise.
C. Hammett Studies of Aldehyde and Nitrile Formation
in the Steady-State Photolysis of Meta- and Para-Substituted
Benzaldehyde Oximes. To learn more about the intermediates
involved in the photosensitized reactions of benzaldoximes, a
mixture of the substituted benzaldehyde oxime and benzalde-
hyde oxime in acetonitrile was irradiated in the presence of a
triplet sensitizer (2,3-dichloro-5,6-dicyano-1,4-benzoquinone;
DDQ). As expected, both the corresponding aldehydes and
nitriles were formed in each case. The results from steady-state
(SS) photolysis experiments are listed in Table 1. A plot of the
nitrile/aldehyde ratio against the substituent shows that in general
more of the nitrile is formed when using electron-accepting
substituents (e.g., NO2, CN; Figure 3).
A plot of the nitrile-aldehyde product ratio against the
quenching rate (see the Supporting Information) does not show
an obvious trend; however, a plot of the ratio against the
calculated IP does seem to be in agreement with the data in
Figure 3. It is interesting to note that in general the oximes that
tend to form more of the nitrile upon irradiation are also those
whose rate constants for quenching 3CA are slower. As indicated
in the discussion above, it is suspected that these oximes react
with the triplet sensitizers via a different mechanism. Note,
however, that there are some exceptions. For example, al-
doximes 1a and 1b quench 3CA very rapidly, yet they produce
a significant amount of nitrile. Clearly, more than just a single
parameter must be taken into account when analyzing the data.
Most importantly, the measured quenching rates only represent
the initial step in the overall sequence, and a variety of follow-
(15) (a) Bordwell, F. G.; Zhang, S. J. Am. Chem. Soc. 1995, 117, 4858.
(b) Bordwell, F. G.; Zhao, Y.; Cheng, J.-P. J. Phys. Org. Chem. 1998, 11,
10.
(16) Bordwell, F. G.; Ji, G.-Z. J. Org. Chem. 1992, 57, 3019.
(17) (a) McMillan, D. F.; Golden, D. M. Ann ReV. Chem. 1982, 33, 493.
(b) Simoes, J. A. M.; Griller, D. Chem. Phys. Lett. 1989, 158, 175. (c)
Lund, H.; Daasbjerg, K.; Occhialini, D.; Pedersen, S. U. Russ. J. Electro-
chem. 1995, 31, 865.
(18) To the best of our knowledge, the bond strengths of the dC-H
bonds in benzaldehyde oximes have not been determined. When comparing
the dC-H bond lengths of aldehyde oximes and aldehydes (calculated by
DFT methods, B3LYP/6-31G*), it is shown that these are nearly identical
(0.001-0.002Å difference at most). On the basis of these calculated values
it is fair to conclude that the dC-H bond strengths in these molecules will
be similar.
(19) Hydrogen atom abstraction from the hydroxyl group can be avoided
by using O-alkyl aldoximes as shown by our studies on a series of aldoxime
ethers. The results of this investigation will be published later.
7788 J. Org. Chem., Vol. 71, No. 20, 2006

Formation of Aldehydes and Nitriles under Electron-Transfer Conditions
TABLE 3. Comparison of Hammett Data Obtained from Percent
Conversion in Steady-State (SS) Photolysis Experiments on Meta-
and Para-Substituted Benzaldehyde and Acetophenone Oximes
SCHEME 2
meta
para
benz-
aldehyde
oxime
benz-
aldehyde
oxime
acetophenone
acetophenone
oxime
oxime
Frad
-43
-8
-155
-28.7
5.4
1.10
-0.42
-2.60
0.90
1.26
-1.2
-1.05
0.98
Fpol
F
rad/Fpol
5.4
r2
0.93
2.72
-0.22
-12.4
0.42
0.93
FJJ
2.71
-0.43
-6.3
0.39
1.27
1.29
Fmb
FJJ/Fmb
r2
-0.17
-7.34
0.72
-0.69
-1.87
0.95
CHART 1. Aldoximes Used for Studies on the Effect of the
Presence of an Internal Base on the Formation of an
Iminoxyl Radical Intermediate
TABLE 4. Hammett Data Obtained from Analysis of Relative
Nitrile and Aldehyde Yields in the Steady-State (SS) Photolysis of
Meta- and Para-Substituted Benzaldehyde Oximes
F
rad/Fpol
r2
FJJ/Fmb
r2
nitrile
meta
para
meta
para
4.4
-3.2
5.5
0.92
0.81
0.90
0.91
-36.2
-9.4
-3.7
-5.6
0.09
0.58
0.80
0.70
up reactions can take place. The quenching process is a
kinetically controlled reaction, whereas product formation
(aldehyde-nitrile ratio) is more likely to be thermodynamically
controlled. This may lead to differences between the measured
kq values and the observed nitrile-aldehyde ratios, but at this
point is it uncertain why only benzaldoximes with certain
substituents would show such behavior.
The mechanism for the formation of the aldehydes is assumed
to be similar to the pathway involving the formation of
acetophenones from acetophenone oximes (Scheme 1). The
central intermediate in this pathway is the iminoxyl radical.
Oxidation of this radical followed by hydrolysis results in the
formation of the corresponding carbonyl compound. We have
proposed that oxidation of the iminoxyl radical involves the
ground-state sensitizer, which would allow the aldehyde to be
formed via two separate pathways; this would explain the
formation of the aldehyde in the case of aldoximes with high
oxidation potentials (Scheme 2).
What remains uncertain, however, is whether the iminoxyl
radical intermediate is also responsible for the formation of the
nitrile product or if a different intermediate (and pathway) is
involved. It is difficult to obtain any direct evidence for the
intermediates involved in these reactions; however, if, as
hypothesized above, a HAT process were more important for
the formation of the nitrile from benzaldehyde oximes, radical
effects would be expected to be more dominant in these
reactions.
The data obtained in these studies (Table 1) was analyzed in
terms of polar and radical substituent effects. First, the relative
conversion was used to see if a pattern would appear. The results
are listed in Table 3. For comparison, the data from a similar
study on acetophenone oximes is also listed.1c It can be seen
from this data set that there are some differences between the
aldehyde and ketone oxime series. In all cases, the radical effect
is significantly larger than the polar effect; on the basis of these
numbers, ionic (polar) effects seem less important in the
reactions of benzaldehyde oximes compared to acetophenone
oximes.
aldehyde
-2.1
the formation of both the aldehyde and the nitrile. However,
there are only small differences between the two data sets and
the errors do not allow for an accurate evaluation of the type of
intermediate involved in the nitrile formation, although the most
likely candidate would be a radical species. Due to the
differences observed between the results obtained in the
acetophenone oxime and benzaldehyde oxime studies, the
intermediate is most likely not the iminoxyl radical, but the exact
nature of the radical species remains uncertain.
D. Involvement of Iminoxyl Radicals in the Formation of
Nitriles from Aldoximes: Effect of an Added Base. To further
evaluate the involvement of iminoxyl radicals in the formation
of nitriles from aldoximes, we have studied the effect of an
added base on the reactions. The proposed mechanism for the
regeneration of carbonyl compounds from oximes via photo-
sensitized electron-transfer reactions involves deprotonation of
the oxime radical cation.1 The working hypothesis was that if
iminoxyl radicals are not involved in the formation of the
nitriles, favoring the iminoxyl radical formation would lead to
a decrease in nitrile yield. To confirm this pathway, and to
determine whether this reaction can be used more generally to
generate iminoxyl radicals selectively, we have studied the
influence of an added base on the photosensitized reactions of
oximes. For proton-transfer reactions involving radical cation
species a variety of bases can be used, but there is some
precedent for using amines such as pyridine and collidine.20 We
havefocusedonthreealdoximeswithabuilt-inbasicsite(pyridine-
2-carboxaldoxime (2), 2-furaldehyde oxime (3), and thiophene-
2-carboxaldoxime (4); Chart 1) which were all expected to
participate in the desired intramolecular proton transfer.
All oximes (0.015 M) were studied in the presence of
benzaldehyde oxime (1g; 0.015 M) itself, which served as an
internal reference. The results of these studies are listed in Table
5 and Figure 4.
A few interesting observations can be made. The relative
reactivity of the oximes increases from 2 < 3 < 4. As the
The data were also analyzed in terms of relative yield of the
corresponding aldehydes and nitriles. The results of these
analyses are listed in Table 4. Although the data is not always
clear (some correlations were poor), a similar trend seems to
arise; radical effects are more important than polar effects for
(20) (a) Arnold, D. R.; Mines, S. A. Can. J. Chem. 1987, 65, 2312. (b)
Arnold, D. R.; Mines, S. A. Can. J. Chem. 1989, 67, 689. (c) Mangion, D.;
Kendall, J.; Arnold, D. R. Org. Lett. 2001, 3, 45.
J. Org. Chem, Vol. 71, No. 20, 2006 7789

de Lijser et al.
SCHEME 3
TABLE 5. Summary of Data Obtained from Chloranil-Sensitized
Photolysis of Pyridine-2-carboxaldoxime (2), 2-Furaldehyde Oxime
(3), and Thiophene-2-carboxaldoxime (4) in the Presence of
Benzaldehyde Oxime (1g)
TABLE 6. Results from Theoretical Studies (B3LYP/6-31G*) on
the Different Conformations (A-C) of the Neutral and Radical
Cations of Pyridine-2-carboxaldoxime (2), 2-Furaldehyde Oxime (3),
and Thiophene-2-carboxaldoxime (4)
% conversion
% aldehyde
% nitrile
absolute energies
(Hartrees)
relative energies
(kcal/mol)
2
34
59
0.6
42
29
1.4
61
20
3.1
0
29
0.0
30
25
1.2
47
16
2.9
33
29
1.1
10
4
2.5
14
4
1g
neutral
RC
neutral
RC
IP (kcal/mol)
relative
3
1g
relative
4
2A -416.91125 -416.65160
2B -416.91033 -416.60738
2C -416.90498 -416.60349
3A -398.64152 -398.34480
3B -398.64889 -398.36098
3C -398.64625 -398.36483
4A -721.61572 -721.31785
4B -721.62539 -721.34475
4C -721.62801 -721.34646
0.00
0.58
3.93
4.62
0.00
1.66
7.71
1.64
0.00
0.00
27.74
30.19
12.57
2.41
0.00
17.95
1.08
162.92
190.09
189.18
186.18
180.65
176.58
186.90
176.10
176.66
1g
relative
3.5
0.00
CHART 2. Different Conformations of
Pyridine-2-carboxaldoxime (2) That Were Studied by
Theoretical (B3LYP/6-31G*) Methods. Similar
Conformations Were Studied for Oximes 3 and 4
of this hydrogen would result in the formation of an iminoyl
radical, which could react to give the observed nitrile. These
observations are in agreement with the results obtained from
DFT calculations. Three different conformers (A, B, and C;
Chart 2) of the neutral oximes 2, 3, and 4 and their radical
cations were optimized using the B3LYP/6-31G* basis set. The
results listed in Table 6 show that the lowest energy conformer
of 2 is the one with the hydroxyl hydrogen directed toward the
nitrogen of the pyridine ring (2A). The basicity of the nitrogen
is clear from the fact that in the optimized structure of the radical
cation the hydroxyl proton is actually transferred to the nitrogen.
Similar structures for 3 and 4 turn out to be less favorable; the
relative energies for 3A and 4A are 4.62 and 7.71 kcal/mol
higher than the lowest energy conformers, respectively. Simi-
larly, their radical cations are significantly less stable.
FIGURE 4. Relative oxime conversions, aldehyde yields, and nitrile
yields in the photolysis of oximes 2-4 with benzaldoxime (1g) as the
standard and chloranil as the sensitizer.
relative reactivity increases, the aldehyde and nitrile yields also
increase but the nitrile/aldehyde ratio decreases. The most
striking observation is the absence of any aldehyde when using
oxime 2. Increasing amounts of aldehyde are observed for the
furan (3) and thiophene (4) oximes. The trend observed in this
series (N < O < S) is in line with the expected basicity of
those atoms. Based on these results, we propose that in these
aldoximes the heteroatom acts as a base and forms a hydrogen
bond with the proton from the hydroxyl group. The stronger
the hydrogen bond, the less likely the hydrogen atom is to be
removed and the mechanism changes. Since the hydroxyl
hydrogen is “locked up” to some extent, the only other available
hydrogen in these substrates is the iminyl hydrogen. Removal
These results confirm that the tendency of forming a hydrogen
bond decreases in going from 2 > 3 > 4, which supports the
hypothesis that locking up the hydroxyl proton would presum-
ably lead to removal of an alternative hydrogen atom or proton
to yield an iminoyl radical (Scheme 3). It is uncertain whether
this process involves a hydrogen atom transfer (via reaction with
the triplet sensitizer) from the neutral oxime or a proton transfer
from the aldoxime radical cation (via reaction with the sensitizer
radical anion).
7790 J. Org. Chem., Vol. 71, No. 20, 2006

Formation of Aldehydes and Nitriles under Electron-Transfer Conditions
SCHEME 4
larger oxidation potential observed in those oximes, which would
make electron transfer less favorable. Formation of the nitrile
was the dominant pathway when the hydroxyl hydrogen was
not available for abstraction, such as in the case of pyridine-
2-carboxaldoxime (2), where a strong intramolecular hydrogen
bond is formed. Under these conditions it is proposed that the
only hydrogen available for abstraction is the iminyl hydrogen.
The strength of the hydrogen bond is determined by the basicity
of the heteroatom present in the aldoxime; formation of a
stronger hydrogen bond (i.e., presence of a more basic hetero-
atom) leads to a larger amount of nitrile.
Experimental Section
Materials. All aldoximes used in this study are known com-
pounds. Benzaldehyde oxime, m-nitrobenzaldehyde oxime, and
p-nitrobenzaldehyde oxime were commercially available. All other
oximes were prepared from reaction of the corresponding aldehyde
with hydroxylamine hydrochloride using standard published pro-
cedures.21,22 Acetonitrile (spectrophotometric grade) was used as
received.
Steady-State Photolysis Experiments. Appropriate amounts of
the substituted benzaldehyde oxime (0.015 M), benzaldehyde oxime
(0.015 M), and DDQ (0.005 M) were weighed out and dissolved
in 5 mL of solvent. For experiments where oxygen was to be
excluded, the solution was purged with argon for 15 min prior to
photolysis. The solution was placed in a Pyrex tube and irradiated
in a Rayonet RPR-100 photochemical reactor, equipped with 16
RPR-3500A (black light phosphor) bulbs (λ ) 350 nm) for 30 min.
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. The products were confirmed by comparison with authentic
(commercially available) samples.
Although there is no direct evidence, the results seem to favor
the HAT process. First, the quenching data suggests a change
in mechanism (from ET to HAT) when using electron-
withdrawing groups; this change in mechanism corresponds to
larger amounts of nitrile being produced, suggesting that these
two observations are connected. Second, the results from
theoretical studies are not consistent with a radical cation
intermediate. In order for a deprotonation reaction to occur, one
would expect significant charge build-up at the iminyl proton.
To determine the charge and spin distributions in aldoxime
radical cations, m-cyanobenzaldehyde oxime (1j) was chosen
as the species of interest due to the fact that it reacted under
steady-state conditions to give the largest amount of nitrile.
Optimization of the structure (two conformations each of the
syn and anti isomers) by means of DFT (B3LYP/6-31G*) did
not show any significant charge buildup at the iminyl proton.
The largest charge density is in fact on the hydroxyl proton.
The results from our studies presented here suggest that
aldoximes can react to give both aldehydes and nitriles under
photoinduced electron transfer conditions. The proposed inter-
mediates in these reactions are iminoxyl radicals (aldehydes)
and iminoyl radicals (nitriles) as summarized in Scheme 4. We
are currently further investigating the structure and reactivity
of the potential reactive intermediates in these reactions.
Electrochemistry. The oxidation potentials of the oximes were
determined by cyclic voltammetry at a scan rate of 100 mV/s. The
reported potentials were referenced to the Fc+/Fc couple. All
measurements were carried out under an argon atmosphere.
Laser Flash Photolysis. The apparatus used for the laser flash
photolysis (LFP) experiments was of standard design,23 and the
details have been described elsewhere.24 The quenching rates were
obtained as follows. An MeCN (spectrophotometric grade) solution
containing chloranil (CA; OD ∼ 0.5-1) in a glass cuvette was
purged with argon for about 5 min. The sample was subjected to
the laser pulse (355 nm, 10 Hz, 0.5-2 mJ/pulse; 4 ns pulse width),
Conclusions
3
and the decay of CA at 510 nm was observed. Small amounts
(10-25 µL) of the quencher (∼0.015 M oxime standard solutions
in MeCN) were added to the solution after which the decay was
measured. The quenching rate was obtained from a plot of the
measured decay rates against the quencher concentration.
Benzaldehyde oximes react to give aldehydes and nitriles
when irradiated in acetonitrile in the presence of quinone
sensitizers. These results are best explained by two different
pathways that involve different intermediates. When the ener-
getics for electron transfer are favorable, the preferred pathway
involves electron transfer, followed by proton transfer to form
an iminoxyl radical. Follow-up reactions eventually lead to the
formation of the corresponding aldehyde. This pathway is
assumed to be identical to that proposed for ketoximes under
similar conditions. When electron transfer becomes energetically
unfavorable, a hydrogen atom transfer mechanism becomes
dominant. In this case, competition between hydrogen atom
abstraction from the hydroxyl group (to form an iminoxyl
radical) or from the iminyl carbon (to form an iminoyl radical)
is observed. The former process also leads to the formation of
the aldehyde, but the latter pathway leads to the production of
the nitrile. Electron-accepting substituents present on the
Computational Methods. Semiempirical (AM1)25 and DFT
(B3LYP)26 calculations were performed with Spartan 2004,27
(21) Vogel, A. I.; Furniss, B. S.; Hannaford, A. J.; Smith, P. W. G.;
Tatchell, A. R. Vogel’s Textbook of Practical Organic Chemistry; Long-
man: New York, 1989.
(22) Kad, G. L.; Bhandari, M.; Kaur, J.; Rathee, R.; Singh, J. Green
Chem. 2001, 3, 275.
(23) 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.
(24) Lorance, E. D.; Kramer, W. H.; Gould, I. R. J. Am. Chem. Soc.
2002, 124, 15225.
(25) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J.
Am. Chem. Soc. 1985, 107, 3902.
(26) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Stephens, P.
J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem. 1994,
98, 11623.
benzene ring in benzaldehyde oximes favor the formation of
the corresponding nitriles. This is thought to be a result of the
(27) Spartan ‘04, Wavefunction, Inc., Irvine CA.
J. Org. Chem, Vol. 71, No. 20, 2006 7791

de Lijser et al.
installed on a PowerMac G4. For the DFT calculations, the 6-31G*
basis set28 implemented within the program was used.
some of these experiments. We are grateful to Professor I. R.
Gould (Arizona State University) for the generous use of his
laser flash photolysis equipment.
Acknowledgment is made to the donors of the Petroleum
Research Fund, administered by the American Chemical Society.
Part of this work was supported by the National Science
Foundation under Grant Nos. 0097795 and 0354159. We thank
Mr. Van Buzzo and Mr. Franklin Vinson for their help with
Supporting Information Available: Figures containing certain
data analyses, the best fit-data of single-parameter and dual-
parameter Hammett analyses described in this work, as well as
Cartesian coordinates for isomers A-C of compounds 2-4. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(28) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; John Wiley and Sons: New York, 1986.
JO061325M
7792 J. Org. Chem., Vol. 71, No. 20, 2006