Oxygenation of benzyldimethylamine by singlet oxygen. Products and mechanism

Article abstract of DOI:10.1021/ol047876l

(Chemical Equation Presented) A product study of the reaction of benzyldimethylamine (1) with thermally and photochemically generated ¹O₂ in MeCN was carried out. Benzaldehyde and N-benzyl-N-methylformamide are the reaction products, oxygenation representing ca. 9% of the overall quenching of ¹O₂ by 1. The temperature effect and the intermolecular and intramolecular kinetic deuterium isotope effects were also determined. It is suggested that the products derive from an intracomplex hydrogen atom transfer in a reversibly formed charge-transfer complex.

Full text of DOI:10.1021/ol047876l

ORGANIC
LETTERS
2
004
Vol. 6, No. 25
791-4794
Oxygenation of Benzyldimethylamine by
Singlet Oxygen. Products and
Mechanism
4
Enrico Baciocchi,*,† Tiziana Del Giacco, and Andrea Lapi†
‡
Dipartimento di Chimica, UniVersit a` “La Sapienza”, P.le A Moro 5,
0
0185 Rome, Italy, and Dipartimento di Chimica and Centro di Eccellenza Materiali
InnoVatiVi Nanostrutturati (CEMIN), UniVersit a` di Perugia, Via Elce di sotto 8,
6123, Perugia, Italy
0
enrico.baciocchi@uniroma1.it
Received October 14, 2004
ABSTRACT
A product study of the reaction of benzyldimethylamine (1) with thermally and photochemically generated 1
Benzaldehyde and N-benzyl-N-methylformamide are the reaction products, oxygenation representing ca. 9% of the overall quenching of O
O
in MeCN was carried out.
2
1
2
by
1. The temperature effect and the intermolecular and intramolecular kinetic deuterium isotope effects were also determined. It is suggested
that the products derive from an intracomplex hydrogen atom transfer in a reversibly formed charge-transfer complex.
The ability of tertiary amines to quench O
2
(1∆
, in a very efficient way is well recognized.
g
), henceforth
the other hand, it is doubtful that such a mechanism may be
1
1
1
indicated as O
2
2
general, given the low reduction potential of O (about 0.1
3
The quenching is predominantly of a physical nature,
particularly with aromatic amines, but there are several
reports indicating that chemical quenching is also possible.
V vs SCE, in MeCN), which makes a full-fledged electron
transfer process highly endergonic, especially with aliphatic
2
1
amines. Another important point is that in most studies O
2
is produced by a sensitized photooxygenation with the
However, despite the large number of papers dealing with
this subject, very little information is presently available
concerning the reaction products as well as their mechanism
of formation. An electron-transfer mechanism has been
consequent possibility of a competition between the oxidation
1
of the substrate by O
2
(type II mechanism) and that by the
excited sensitizer (type I mechanism). Thus, so far, product
2f,h,i
studies under rigorously controlled conditions are lacking.
proposed in several cases,
substantiated only with NADH and aromatic amines of very
low oxidation potential (<0.5 V vs SCE, in water). On
which has been experimentally
1
2c
In view of the continuous interest in O reactivity with
2
2
d,g
organic compounds, we now report on a detailed product
1
study of the reaction of benzyldimethylamine (1) with O
2
1
2
in MeCN where O has been generated both thermally and
*
Phone: (+39) 06-49913711. Fax: (+39) 06490421.
Universit a` “La Sapienza”.
Universit a` di Perugia.
†
photochemically. The photoxygenation of 1 sensitized by
‡
methylene blue or rose bengal has already been studied by
(
1) Schweitzer, C.; Schmidt, R. Chem. ReV. 2003, 103, 1685.
4
Inoue and co-workers. Competition between type I and type
(2) (a) Smith, W. F., Jr. J. Am. Chem. Soc. 1972, 94, 186. (b) Bellus,
D.; Lind, H. Chem. Commun. 1972, 1199. (c) Peters, G.; Rodgers, M. A.
J. Biochim. Biophys. Acta 1981, 637, 43. (d) Manring, L. E.; Foote, C. S.
J. Phys. Chem. 1982, 86, 1257. (e) Saito, I.; Matsuura, T.; Inoue, K. J. Am.
Chem. Soc. 1983, 105, 3200. (f) Haugen, C. M.; Bergmark, W. R.; Whitten,
D. G. J. Am. Chem. Soc. 1992, 114, 10293. (g) Darmanyan A. P.; Jenks,
W. S.; Jardon, P. J. Phys. Chem. A 1998, 102, 7420. (h) Bernstein, R.;
Foote, C. S. J. Phys. Chem. A 1999, 103, 7244. (i) Cocquet, G.; Rool, P.;
Ferroud, C. J. Chem. Soc., Perkin Trans. 1 2000, 14, 2277.
II oxygenation mechanisms was suggested.
Reactions with Thermally Generated O . O was
2 2
generated by heating the endoperoxide of 1,4-dimethylnaph-
1
1
(3) Sawyer, D. T.; Chiericato, G., Jr.; Angelis, C. T.; Nanni, E. J., Jr.;
Tsuchiya, T. Anal. Chem. 1982, 54, 1720.
(4) Inoue, K.; Saito, I.; Matsuura, T. Chem. Lett. 1977, 607.
1
0.1021/ol047876l CCC: $27.50
© 2004 American Chemical Society
Published on Web 11/11/2004

An additional and important observation was that the number
of micromoles of products formed is almost independent of
Scheme 1
the substrate concentration (Table 1A, entries 1-3). This
1
indicates that substantially all O
2
formed reacts with 1. Thus,
by considering the molar ratio between the endoperoxide
1
2
used and 1 (10) and the fact that the yield in O from the
6
endoperoxide is 25%, we can estimate that the contribution
of chemical quenching to the overall quenching (chemical
1
+
physical quenching) of O
2
by 1 in MeCN is about 9%.
By time-resolved luminescence at 1270 nm, it was deter-
5
thalene (2). A 10:1 excess of 2 was added to 1 mL of a
.01 M solution of 1 in MeCN and the resulting solution
mined that for 1 the overall quenching occurs with a rate
0
7
-1 -1 7
constant (k
q
) of 9.7 × 10 M s ; hence, the rate constant
left at 40 °C in the dark for 4 h. The reaction mixture was
examined by GC and GC-MS. All 2 was converted to 1,4-
dimethylnaphthalene, and two reaction products were ob-
served (comparison with authentic specimens), namely,
benzaldehyde (3) and N-benzyl-N-methylformamide (4). The
/4 molar ratio was 1.4 ( 0.2, and the overall yield of
products was around 22%, with respect to the starting amine
Scheme 1, Table 1A, entry 1). The mass balance was
7 -1 -1
for product formation is ca. 1 × 10 M s .
The kinetic deuterium isotope effect (KDIE) of the
oxygenation reaction was measured by using (benzyl-d
2
)-
, forms
) and N-methyl-N-(benzyl-d )-
). The intramolecular KDIE, [(k /k
.5 ( 0.6] was obtained by comparing the 3-d /4-d
ratio obtained from 1-d with the 3/4 molar ratio observed
for the reaction of 1, of course assuming that the reactivity
of the N-methyl group is the same in 1 and 1-d . For the
1
dimethylamine (1-d
benzaldehyde-R-d
formamide (4-d
2
), which, by reaction with O
(3-d
2
1
1
2
3
2
H
D
)
intra
)
3
1
2
molar
(
2
excellent (95 ( 5%).
2
intermolecular KDIE [(k
molar mixture of 1 and 1-d
molar ratio between 3 and 3-d
of our knowledge, these are the first values of KDIE
H
/k
D
)
inter ) 2.93 ( 0.02], an equi-
1
Table 1. Benzyldimethylamine Oxygenation by Thermally and
Photochemically Generated Singlet Oxygen in MeCN
2
was reacted with O
2
and the
1
was measured. To the best
products (µmol)b
reaction
conditions
1
determined in the oxygenation of amines by O
are summarized in Table 2A.
Reactions with Photochemically Generated O
2
. The results
a
entry
3
4
3/4
8
1
(
A) oxygenation by thermally generated O2
1
2
. Having
1
2
3
4
5
1.3 (0.1)
1.2 (0.1)
1.4 (0.1)
1.2 (0.1)
1.4 (0.1)
0.93 (0.04)
0.89 (0.05)
1.07 (0.07)
0.94 (0.04)
1.1 (0.1)
1.4 (0.2)
1.3 (0.2)
1.3 (0.2)
1.3 (0.1)
1.3 (0.2)
unequivocally established the properties of the chemical
-
2
1 (2 × 10 M)
1
quenching of O
2
by 1 in MeCN, we considered it worthwhile
-
2
1 (5 × 10 M)
BQ (2.5 mM)
TBP (5 mM)
to check if the same results would be obtained by using
1
photosensisitized generation of O
2
. Tetraphenylporphyrin
(
TPP) was used as the sensitizer, since its singlet excited
1
(
B) oxygenation by photochemically generated O2
9
state is a weaker oxidant (Ered ) 0.81 V vs SCE) than the
excited states of methylene blue and rose bengal used by
Inoue. This was expected to give us more chances to obtain
an exclusive type II mechanism of photooxygenation.
Actually, when 1 (0.01 M) was irradiated (external
irradiation) for 30 min at 450-600 nm in the presence of
6
7
8
5.6 (0.4)
54 (4)
35 (4)
3.8 (0.2)
43 (3)
28 (3)
1.5 (0.2)
1.3 (0.2)
1.3 (0.2)
25 °Cc
-48 °Cc
a
-2
If not specified, the inital concentrations are: 1 (10 M), 2 (0.1 M),
TPP (10 M). Reactions with 2 were carried out at 40 °C for 4 h. With
TPP, at 25 °C for 30 min if not specified. Calculated by GC analysis.
Average of at least three determinations. The error (standard deviation) in
-
4
b
-
4
c
TPP (1 × 10 M) in O
2
-saturated MeCN, 3 and 4 were
the last significant digit is given in parentheses. Volume: 50 mL. Reaction
time: 10 min.
obtained in a ratio of 1.5 ( 0.2 (mass balance ) 98%) which
is the same, within experimental error, as that found in the
reaction with the endoperoxide (Table 1B, entry 6).10 Still
As control experiments showed that no reaction occurs
and 1 under the reaction conditions, it is clear
that some chemical quenching actually occurs in the reaction
more significantly, when the reaction of 1-d
gated, the values of KDIE were very close to those obtained
with thermally generated O
are already strong evidence in favor of a predominant
2
was investi-
3
between O
2
1
2
(Table 2B). While these results
1
2
of 1 with O in MeCN and that 3 and 4 are the products
formed in this process. It was also found that the extent of
chemical quenching is unaffected by the presence of 2,4,6-
tri-tert-butylphenol (TBP), entry 5 in Table 1A, as well as
by the presence of benzoquinone (BQ), entry 4 in Table 1A.
(
6) G u¨ nther, G. S.; Lemp, E. M.; Zanocco, A. L. J. Photochem. Photobiol.
1
5a
A 2002, 151, 1. However, Turro et al. report a 76% yield of O2 in dioxane.
7) Determined by laser flash photolysis experiments (see Supporting
Information).
(
(
8) In view of the largely predominant physical quenching, kq for 1-d2
7
-1 -1
(
5) (a) Turro, N. J.; Chow, M. F. J. Am. Chem. Soc. 1981, 103, 7218.
(9.9 × 10
M
s ) is the same as that for 1, within the experimental
(
b) Adam, W.; Prein, M. Acc. Chem. Res. 1996, 29, 275. (c) Greer, A.;
error.
Vassilikogiannakis, G.; Lee, K.-C.; Koffas, T. S.; Nahm, K.; Foote, C. S.
J. Org. Chem. 2000, 65, 6876. (d) Ben-Shabat, S.; Itagaki, Y.; Jockusch,
S.; Sparrow, J. R.; Turro, N. J.; Nakanishi, K. Angew. Chem., Int. Ed. 2002,
(9) Darwent, J. R.; Douglas, P.; Harriman, A.; Porter, G.; Richoux, M.-
C. Coord. Chem. ReV. 1982, 44, 83.
(10) With longer reaction times, higher conversions can be reached but
significant overoxidation takes place. Thus, after 90 min of irradiation with
5% TPP, the conversion of 1 is 90% complete, but the yield of 3 + 4 is
only 50%.
4
1, 814. (e) Poon, T.; Turro, N. J.; Chapman, J.; Lakshminarasimhan, P.;
Lei, X.; Jockusch, S.; Franz, R.; Washington, I.; Adam, W.; Bosio, S. G.
Org. Lett. 2003, 5, 4951.
4792
Org. Lett., Vol. 6, No. 25, 2004

Table 2. Benzyldimethylamine and (Benzyl-d
2
)dimethylamine Oxygenation by Thermally and Photochemically Generated Singlet
Oxygen in MeCN
products (µmol)a,b
amine
3-d1
4-d2
3
4
kH/kD b
1
(
A) oxygenation by thermally generated O2
1
-d2
0.51 (0.01)
0.27 (0.01)
1.28 (0.01)
0.64 (0.02)
3.5 (0.6)c
1-d2 + 1
0.77 (0.01)
0.61 (0.02)
1.83 (0.03)
2.93 (0.02)d
1
(
B) oxygenation by photochemically generated O2
1
1
-d2
-d2 + 1
1.7 (0.2)
0.76 (0.05)
4.0 (0.1)
1.80 (0.03)
3.5 (0.8)c
3.06 (0.06)d
2.34 (0.05)
a
Calculated by GC and GC-MS analysis. In competitive oxygenations of 1 + 1-d2, the 3/3-d1 and 4/4-d2 ratios were determined by GC-MS analysis.
b
c
Average of at least three independent determinations. The error (standard deviation) in the last significant digit is given in parentheses. Determined from
d
the 3/4 and 3-d1/4-d2 ratios obtained in independent oxygenations of 1 and 1-d2. Determined by GC-MS analysis of benzaldehyde by the ratio of the
intensity of the molecular peaks m/z ) 106 and 107, corrected for the 13C contribution.
operation of the type II mechanism, further support in this
respect was found in experiments in the presence of DABCO
electron-transfer process,14 does not exert any appreciable
effect on the reaction yield (Table 1A, entry 4).15
1
11
(an efficient, exclusively physical, quencher of
O
2
).
There is a general agreement on the hypothesis that
1
Accordingly, a progressive decrease in the product yields
was observed by increasing the concentration of DABCO.
Moreover, the ratio between the product yields (3 + 4) in
the absence and in the presence of DABCO was linearly
correlated to the DABCO concentration. From this correla-
physical quenching of O
2
by amines involves the formation
of an exciplex with partial charge transfer (CT) character as
the reaction intermediate, and it has been suggested that such
an exciplex can also be the intermediate in the chemical
1
,2e,g,16
quenching.
That is, once formed, the exciplex can
8
-1
3
tion it was possible to calculate a value of 3.0 × 10 M
undergo intersystem crossing (isc) to give the amine and O
2
-1
1
s
as the rate constant for quenching of O
2
by DABCO in
(physical quenching) or be converted into products (chemical
quenching). In view of the almost negligible effect of the
temperature observed with respect to the rate of formation
MeCN, which resulted in agreement with literature values
(
For details, see Supporting Information).12
Having obtained overwhelming evidence that the type II
of 3 and 4, it is very likely that this mechanism also operates
1
mechanism is operating in the photoxygenation of 1 in the
presence of TPP, it was possible to study the temperature
effect on this reaction. To this purpose, irradiation of a 1/TPP/
in the reaction of 1 with O
2
in MeCN (Scheme 2, paths a
1
6c
and b + c).
O
2
solution was carried out at 25 and -48 °C for 10 min.
The total amount of product (3 + 4) formed was 97 µmol at
5 °C and 63 µmol at -48 °C (Table 1B, entries 7 and 8).
No appreciable effect of the temperature upon k was
quenching by 1
Scheme 2
2
q
1
observed when rate measurements of O
were carried out at 20 °C and -15 °C.13
2
Reaction Mechanism. The experiments described above
clearly showed that 3 and 4 are the products formed by the
1
reaction of 1 with O
2
. Concerning their mechanism of
formation, the possibility of an autoxidation process pro-
1
2
moted by O is excluded by the observation that there is no
inhibition of the reaction by TBP (Table 1A, entry 5). An
electron transfer process involving the formation of a radical
cation, which then undergoes proton loss, is also highly
unlikely being largely endergonic (the difference in the
reduction potential between 1+• and O
1
is about 0.7 V). Also
2
inconsistent with this mechanism is the observation that
-
•
benzoquinone, a trap for O
2
that should form in the
(
(
11) Silverman, S. K.; Foote, C. S. J. Am. Chem. Soc. 1991, 113, 7672.
12) Wilkinson, F.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref.
Data 1995, 24, 663.
13) (a) These measurements were carried out by laser flash photolysis
It has been pointed out that a close analogy exists between
(
1
2
the properties of O (an electronically excited state) and
experiments in the presence of 1,3-diphenylisobenzofuran according to the
1
3b
8
13b
procedure described by Gorman et al. The values of kq are 1.3 × 10
those of ketone triplets. It therefore seems reasonable to
-1
-1
8
-1 -1
M
s
and 1.4 × 10 M
s
at 20 and -15 °C, respectively. (b) Gorman,
A. A.; Gould, I. R.; Hamblett, I.; Standen, M. C. J. Am. Chem. Soc. 1984,
suggest that 3 and 4 derive from an exciplex formed by 1
1
1
06, 6956.
and O
2
via an intracomplex hydrogen atom transfer similar
Org. Lett., Vol. 6, No. 25, 2004
4793

to that proposed for the reactions of triplet ketones with
with the stronger bond dissociation energy of the methylic
C-H bonds.
amines and alkylaromatics.17 In our case, the hydrogen is
21
1
transferred from the R-carbon to O
2
, leading to the carbon
However, it has to be noted that in the reaction of 1 with
triplet benzophenone, a completely different regiochemistry
is observed, as the hydrogen atom transfer concerns only
•
radicals 5 and 6 and HO
2
(Scheme 2, paths b and c). A
significant extent of charge transfer can take place in the
exciplex to give the transfer of hydrogen a partial character
of proton transfer.
22
the N-methyl group. It should, however, be considered that
the degree of charge transfer might be significantly higher
Whatever the detailed mechanism, the observation of a
clearly primary KDIE indicates that the hydrogen is trans-
ferred in the product-determining step. More significantly,
in the benzophenone/amine CT complex, since triplet ben-
1
zophenone has a much higher reduction potential than O
2
.
That in CT complexes the product distribution may depend
on the degree of charge transfer is shown by the finding that
only transfer of the benzylic hydrogen is found in the
photoreduction of triplet azoalkanes (much weaker oxidants
1
8
the very close values of (k
H D H D
/k )intra and (k /k )inter point to
a conversion of the exciplex into carbon radicals in a single
1
9
step, without the intervention of other intermediates.
Interestingly, the observed products suggest that the carbon
23
than benzophenone) by 1.
radicals 5 and 6 follow different pathways. A possible
A final notation is that the significant chemical quenching
•
hypothesis is that, as shown in Scheme 2, 5 reacts with HO
2
24
observed with 1 (E° ) 0.82 V vs SCE in MeCN) contrasts
to form a hydroperoxide from which 4 can be formed,
with the behaviors of aromatic amines. These amines can
•
whereas 6 is oxidized by HO
2
to a carbocation that in several
have k
q
values higher than that of 1 but exhibit partial
O and when an electron
steps is converted into benzaldehyde. Consistent with this
proposal is the observation that in the photooxygenation of
chemical quenching only in H
2
2
e,g
transfer is possible (E° e 0.5 V vs SCE). Thus, it would
seem that with aromatic amines, hydrogen transfer in the
CT complex cannot compete with isc. Only when the electron
transfer is possible can conversion of the CT complex to a
couple of radical ions compete with isc.
1
2 2
, H O is formed in an amount equivalent to that of
2
0
benzaldehyde. The different behaviors of 5 and 6 may be
due to the fact that 6 should be more easily oxidizable than
5, due to the presence of the phenyl group, which, with
respect to hydrogen, is expected to stabilize a carbocation
more than a carbon radical. Alternatively (or in addition),
steric effects that may make radical coupling more difficult
in 6 than in 5 might come into play.
At any rate, the 3/4 product ratios observed (Table 1)
2
indicate that transfer of hydrogen from the benzylic CH is
Acknowledgment. MIUR, University “La Sapienza” of
Rome and University of Perugia are thanked for the financial
support. Thanks are also due to Prof. F. Elisei (Universita`
di Perugia) for his advice concerning the time-resolved
experiments.
favored about 4 times (after statistical correction) with respect
to hydrogen tranfer from the methyl groups, which is in line
Supporting Information Available: Experimental details
1
(
14) Manring, L. E.; Kramer, M. K.; Foote, C. S. Tetrahedron Lett. 1984,
5, 2523.
15) (a) It has also been reported that no formation of O2 is observed
and measurement of the quenching rate constant of O
2
by
2
1
and the effect of DABCO in the TPP-sensitized photo-
-
•
(
1
15b
oxygenation of 1. This material is available free of charge
via the Internet at http://pubs.acs.org.
in the reaction between 1 and O2. In this study, however, the reaction
products were not investigated. (b) Saito, I.; Matsuura, T.; Inoue, K. J.
Am. Chem. Soc. 1981, 103, 188.
(
16) (a) Young, R. H.; Martin, R. L.; Feriozi, D.; Brewer, D.; Kayser,
R. Photochem. Photobiol. 1973, 17, 233. (b) Clennan, E. L.; Noe, L. J.;
Wen, T.; Szneler, E. J. Org. Chem. 1989, 54, 3581. (c) Gorman, A. AdV.
Photochem. 1992, 17, 217. (d) Martin, N. H.; Allen, N. W., III; Cottle, C.
A.; Marschke, C. K., Jr. J. Photochem. Photobiol. A 1997, 103, 33.
OL047876L
-
1
(21) (a) Bond dissociation energy is 92.6 against 89.1 kcal mol for
21b
(
17) Wagner, P.; Park, B.-S Org. Photochem. 1991, 11, 227.
the N-benzylic C-H bond. (b) Dombrowski, G. W.; Dinnocenzo, J. P.;
Farid, S.; Goodman, J. L.; Gould, I. R. J. Org. Chem. 1999, 64, 427.
(22) Cohen, S. G.; Stein, N. M. J. Am. Chem. Soc. 1971, 93, 6542.
(23) Adam, W.; Moorthy, J. N.; Nau, W. M.; Scaiano, J. C. J. Am. Chem.
Soc. 1997, 119, 6749.
(18) Actually, the two values are within the experimental error. It should
also be considered that (kH/kD)intra was obtained by product ratios measured
in separate oxidations and is therefore less accurate than (kH/kD)inter.
(
19) Gollnick, K.; Lindner, J. K. E. Tetrahedron Lett. 1973, 21, 1903.
(
20) Amount of H2O2 was quantitatively determined by titration with
(24) Hall, L. R.; Iwamoto, R. T.; Hanzlik, R. P. J. Org. Chem. 1989,
54, 2446.
iodide ion (see Supporting Information).
4794
Org. Lett., Vol. 6, No. 25, 2004