Oxidative photo-decarboxylation in the presence of mesoporous silicas

Article abstract of DOI:10.1248/cpb.54.1571

FSM-16, a mesoporous silica, was found to catalyze oxidative photo-decarboxylation of α-hydroxy carboxylic acid, phenyl acetic acid derivatives and N-acyl-protected α-amino acids to afford the corresponding carbonyl compounds. Furthermore, FSM-16 proved to be re-usable by re-calcination at 450°C after the reaction.

Full text of DOI:10.1248/cpb.54.1571

November 2006
Chem. Pharm. Bull. 54(11) 1571—1575 (2006)
1571
Oxidative Photo-Decarboxylation in the Presence of Mesoporous Silicas
b
Akichika ITOH,*,a Tomohiro KODAMA,a Yukio MASAKI,a and Shinji INAGAKI
a Gifu Pharmaceutical University; Mitahora-higashi, Gifu 502–8585, Japan: and b Toyota Central R & D Labs., Inc.; Aichi
480–1192, Japan. Received July 12, 2006; accepted August 29, 2006
FSM-16, a mesoporous silica, was found to catalyze oxidative photo-decarboxylation of a-hydroxy car-
boxylic acid, phenyl acetic acid derivatives and N-acyl-protected a-amino acids to afford the corresponding car-
bonyl compounds. Furthermore, FSM-16 proved to be re-usable by re-calcination at 450 °C after the reaction.
Key words a-amino acid; a-hydroxycarboxylic acid; mesoporous silica; phenylacetic acid derivative; photo-decarboxylation
Recently, the importance of environmentally friendly Results and Discuusion
processes has been recognized over all fields of industry and,
Mesoporous silicas, FSM-16,5,6) MCM-417) and HMS,8)
needless to say, in the field of synthetic organic chemistry as were prepared according to the previous papers. Table 1
well.1) Easy work-up is an important factor for achieving an shows the ratios of the framework of mesoporous silicas,
environmentally benign process and the use of heterogeneous which were analyzed by 29Si-NMR spectra, and the pore
catalysts is advantageous. Among them, growing attention sizes of the silicas, which were measured by volumetric N2-
has been paid to porous solids, due to their potential utility as gas adsorption method.
new heterogeneous catalysts and providing selectivity arising
Table 2 shows the results of a study of reaction conditions
from the shape and size of pores. Mesoporous silicas, one of conducted with benzilic acid (1, 50 mg, 0.219 mmol) as a test
porous crystals, possess uniform meso pores, of which pore substrate in the presence of additives (100 mg) under external
sizes are 2—50 nm, and narrow pore-distribution and high
Table 1. The Ratio of the Framework and the Pore Size of the Silicas
surface area against the amorphous mesoporous materials,
such as silica gel or alumina.2) In the course of our study on
the utility of mesoporous silicas in organic synthesis,3,4) we,
at first, found that benzilic acid (1) was oxidatively decar-
boxylated under irradiation of UV to give benzophenone (2)
in the presence of FSM-16,5,6) which is one of typical meso-
porous silicas such as MCM-417) and HMS8) (Chart 1).9)
The photo-reactivity of microporous silicas2) which con-
tain transition metals also has been studied by many
groups10); however, little is known about that of the silica
itself.11—14) Similar to mesoporous silicas, there is only one
report on the photo-reactivity of the silica itself,15) and there
are no reports about an application of metal-free mesoporous
silicas as photo-catalyst to synthetic organic chemistry so
far.16) This is the driving force of our continued studies on
the generality of this reaction, and we have found a-hydroxy-
carboxylic acids, phenyl acetic acid derivatives,9) and N-acyl-
protected a-amino acids17) afforded the corresponding car-
bonyl compounds through an oxidative decarboxylation reac-
tion in the presence of FSM-16, under photo-irradiation. Al-
though thermal18—25) and photolytic26,27) oxidative decarboxy-
lation reactions of phenyl acetic acid derivatives have been
reported by several groups, they require heavy metals, which
are of environmentally high impact, as promoters to carry out
the reactions. Against this, our method is considered to be
environmentally benign due to both heavy-metal-free condi-
tions and easy work-up. Herein, we report our detailed study
on the scope, limitation and mechanism of this oxidative
photo-decarboxylation.
Silica
HMS
MCM-41
FSM-16
Q2
Q3
Q4
Pore size (nm)
4
8
2
27
9
28
69
83
70
2.4
2.9
2.8
Table 2. Study of Reaction Conditions for Oxidative Photo-Deacrboxyla-
tion of a-Hydroxycarboxylic Acid
Time
(h)
Recovery Yield of
Entry
Additive
Solvent
of 1 (%) 2 (%)a)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
—
MCM-41
HMS
H-Y
Na-Y
H-ZSM-5
SiO2
—
5
5
5
5
5
5
5
4
3
5
5
5
24
5
5
5
5
5
5
5
47
35
32
18
56
21
ꢀ1
11
19
18
73
10
78
25
44
66
56
30
54
67
34
49
Acetone
MeCN
MeOH
H2O
50
36b)
12c)
65
Toluene
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
88
80
63
75d)
ꢀ1e)
79f)
9
55
35
20
20
66
23
10
Al2O3
a) All yields are for pure, isolated products. b) 17% of 1,1-diphenylethyleneglycol
was obtained. c) 11% of benzopinacol was obtained. d) The reaction was carried
out with 50 mg of FSM-16. e) The reaction was carried out in the dark. f ) The re-
action was carried out under irradiation of 100 W lamp.
Chart 1
∗ To whom correspondence should be addressed. e-mail: itoha@gifu-pu.ac.jp
© 2006 Pharmaceutical Society of Japan

1572
Vol. 54, No. 11
irradiation by a 100 W or 400 W high-pressure mercury lamp droxypalmitic acid (8), which are secondary a-hydroxy car-
in an aerobic atmosphere. The temperature of the final stage boxylic acids, gave a mixture of the corresponding aldehyde
of this reaction was about 50 °C. We do not have any direct and carboxylic acid (entries 2, 3). Surprisingly, 2-phenylpro-
evidence; however, we believe an effective wavelength of the pionic acid (11) and 4-methoxyphenylacetic acid (12), which
light is 365 nm, which is an emission line of the strongest possess no hydroxyl group at the a-position, also afforded
and longest wave length in an ultraviolet region irradiated acetophenone (4: 47%), and a mixture of the corresponding
from a high-pressure mercury lamp, since passing through degraded aldehyde (6: 30%) and carboxylic acid (7: 44%),
water, a test tube and a cooling jacket, which are made of respectively (entries 4, 5). On the other hand, palmitic acid
Pyrex glass, and air are necessary for the light to effect this (13), which is a simple carboxylic acid, was unfortunately in-
reaction. Among the solvents examined, hexane was found to tact under the conditions, and only the starting material was
afford the best results (entries 1—7).29) Lowering of the recovered quantitatively even after 48 h (entry 6).
yields was observed when the reaction was carried out in
As can be seen from Table 3, the substituent at the a-posi-
shorter time or with a smaller amount of FSM-16 (entries tion of acids is thought to play an important role in this reac-
8—10). Without irradiation, benzophenone (2) was scarcely tion because usual aliphatic carboxylic acids were intact
obtained even in the presence of FSM-16, while the reaction under the reaction conditions. Accordingly, a study of the
proceeded smoothly under irradiation even with a 100 W substituent at the a-position is important to clarify the mech-
lamp (entries 11, 12). Without additives, only a trace amount anism and to apply this reaction to organic synthesis. Thus,
of the product 2 was obtained even after longer reaction time we next studied photo-decarboxylation of a-amino acids,
(24 h), and 78% of the starting benzilic acid (1) was recov- which possess an amino group at the a-position. At first,
ered (entry 13). MCM-41 and HMS, which are other meso- glycine (14) and alanine (15) were examined for this photo-
porous silicas, and H-Y, Na-Y and H-ZSM-5, which are typi- decarboxylation reaction. A suspension of 14 or 15 (50 mg)
cal microporous zeolites, showed photo-reactivity to some and FSM-16 (100 mg) in dry hexane (7 ml) was irradiated at
extent; however, they were not so effective as FSM-16 (en- room temperature in Pylex glassware with a 400-W high-
tries 14—18). Silica gel and alumina, which are amorphous pressure mercury lamp externally for 60 h; however, the
porous materials, afforded 2 in only low yields (entries 19, starting materials were recovered in both cases. Similarly
20).
with N-benzylglycine (16), only the starting material was re-
Table 3 shows the results for oxidative photo-decarboxyla- covered. On the other hand, N-benzoylglycine (17a) afforded
tion of several substrates. Other a-hydroxycarboxylic acids N-formylbenzamide (18a) in 45% yield after 60 h irradiation.
(3, 5, 8) also afforded the corresponding products in the These results suggest that the basicity of the nitrogen atom
same manner as 1. Acetophenone (4) was obtained in 78% inhibits the reaction, and it is thought to be necessary to
yield as a sole product when atrolactic acid (3) was used as a mask the lone pair with protecting groups in order for the re-
substrate (entry 1). 4-Methoxymandelic acid (5) and 2-hy- action to proceed (Chart 2).
Table 3. Oxidative Photo-Deacrboxylation of Several Substrates with
FSM-16
Entry
Substrate
Time (h) Products (% yield)a)
Chart 2
1
2
12
12
Table 4. Study of Reaction Conditions for Oxidative Photo-Deacrboxyla-
tion of N-Protected Amino Acid
3
15
Recovery
Yield of
Entry
Additive
Solvent
of 17b (%) 18b (%)a)
4
5
60
36
1
2
3
4
5
6
7
8
9
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-16
FSM-8
FSM-12
—
Et2O
44
43
20
17
0
Quant.
62
4
39
23
51
50
70
Toluene
Acetone
MeCN
Hexane
Hexane
Hexane
Hexane
Hexane
0b)
3
76
0
6
48
No reaction
Quant.
a) All yields are for pure, isolated products. b) The reaction was carried out in the
dark.
a) All yields are for pure, isolated products.

November 2006
1573
Table 5. Oxidative Photo-Deacrboxylation of N-Protected Several Amino Table 6. Re-use of FSM-16
Acids with FSM-16
Cycle of FSM-16
1st
87
2nd
91
3rd
86
4th
86
5th
81
Yield of 2 (%)a)
Substrate
R
Time
(h)
Product
Yield
(%)a)
Entry
P
a) All yields are for pure, isolated products.
1
2
3
4
5
6
7
8
Ph
Ph
Ph
H
Me
i-Bu
Ph
Me
H
17a
17b
17c
17d
17e
17f
60
36
48
36
36
60
60
36
18a
18b
18c
18d
18e
18f
45
70
54
74
51
66
55
45
Table 7. Effect of Oxygen
Ph
BnO
BnO
Bn
H
Me
17g
17h
18g
18h
Yield of 2 (%)a)
t-BuO
Time
(h)
Under N2
Under O2
Under air
a) All yields are for pure, isolated products.
3
5
5
7
61
73
69
88
Table 4 shows the results of our study of reaction condi-
tions conducted with N-benzoylalanine (17b) as a test sub-
strate. At first, the reactivity of 17b in several organic sol-
vents was examined, and hexane proved to be a better solvent
a) All yields are for pure, isolated products.
than other typical ones (entries 1—5). The results that with- viding 81% yield of 2 even after being re-used 5 times
out irradiation or FSM-16, only the starting material was re- (Table 6).
covered (entries 6, 9) show the necessity of both irradiation
Reaction Mechanism In our study with a-hydroxycar-
and FSM-16 for this reaction. Furthermore, the pore size of boxylic acids, the real oxidant is assumed to be the oxygen in
the silica seems to be important for this reaction. N-Benzoyl- air. Table 7 shows the results of a study of the effect of oxy-
alanine (17b) was irradiated for 36 h in the presence of FSM- gen. Only a small amount of product 2 was obtained under
8 (Dꢀ1.5 nm; D, pore diameter) and FSM-12 (Dꢁ2.0 nm). N2, while under O2 the yield of 2 was comparable to that
Although a large amount of the starting 17b was recovered under air. Since no acceleration of this reaction was observed
with FSM-8, 76% of N-acetylbenzamide (18b) was obtained under O2, a high concentration of oxygen was not required;
in the presence of FSM-12 (entries 7, 8). These results show however, oxygen proved to be an important factor for this
that FSM-16 (Dꢁ2.7 nm) is superior for the substrates of reaction.
similar or larger size than that of 17b to other FSMs of small
Among the typical solvents examined, the order of solubil-
pore size, and that the reaction field is actually inside the ity of oxygen is hexaneꢂtolueneꢂacetoneꢂacetnitrile.29) Al-
pores of FSMs, because the composition and the chemical though the yield of 2 correlated to the order, the yields of
properties of FSMs are basically the same.
18b did not correlate to the order except hexane of the high-
Table 5 shows the results for the photo-decarboxylation est solubility of oxygen (Tables 2, 4). The reason is not clear;
with several N-acyl-protected a-amino acids. N-Benzoyl- however, concentration effect due to hydrophobicity of
(17a), N-benzyloxycarbonyl- (17f), and N-phenylacetyl- hexane is assumed to accelerate the photo-reaction in the
glycine (17g) took a longer time to complete the reaction pores of FSM-16s.
(entries 1, 6, 7). On the other hand, conversion of N-benzoyl-
Furthermore, an attempt was made to detect the intermedi-
(17b), N-benzyloxycarbonyl- (17e), and N-tert-butoxycar- ates to clarify the reaction pathway by terminating the reac-
bonylalanine (17h) to the corresponding imides completed in tion with 12 at 12 h; however, the possible intermediates, 5 or
relatively short time (entries 2, 5, 8). N-Benzoylleucine (17c) 4-methoxybenzylalcohol (19), were not detected by 400
afforded the product 18c in a moderate yield (entry 3). The MHz-NMR analysis (Chart 3, Eq. 1). On the other hand, 19
best yield in our study was attained by the reaction with N- was allowed to react under the same conditions as for 12, and
benzoylphenylglycine (17d) (entry 4). From these results, the found to afford 6 (9%) and 7 (3%), which were similar to
suitable protective group for each a-amino acid seems to be results for 12 (Chart 3, Eq. 2). This suggests the final oxi-
different. Although the substrate 17a is smaller than 17b, 17c dized products were provided by oxidation of a benzylalco-
and 17d, the yield was lower than that of the others (entries hol species as the possible intermediates.
1—4). Furthermore 17f and 17g, which are thought to be
Although the mechanism of this reaction is not clear yet,
more bulky than 17a, afforded 18f and 18g respectively in the active sites produced by dehydration from silanol groups
higher yields than 17a did. These results suggest that not on the wall of FSM-16 during calcination are thought to play
only the shape of the substrates, but also the electronic effect an important role in decarboxylation of the substrate. We
of the substituents participate in the direction of this reaction. present in Charts 4 and 5 what we assume is a mechanism of
Furthermore, the promoter FSM-16 was found to be recy- this oxidation, which is postulated by considering all of the
clable and useful as a photocatalyst. Thus, the recovered results mentioned above. At first, Chart 4 summarizes the
FSM-16, which was re-calcined at 450 °C for 4 h after the generation step of the active sites and initial photoexcitation
photo-reaction of 1, showed high activity in the reaction, pro- step of photo-decarboxylation. Non-bonding oxygen hole

1574
Vol. 54, No. 11
Chart 3
Chart 4
tionally simple from the viewpoint of the convenience of the
work-up in which recyclable FSM-16 can be removed only
by filtration of the reaction mixture.
Experimental
General Diethyl ether was freshly distilled from Na metal/benzophe-
none ketyl. All other dry solvents were obtained from Kanto Kagaku Co.,
Ltd. Other chemicals used were of reagent grade and were obtained from
Aldrich Chemical Co., Tokyo Kasei Kogyo Co., Ltd. and Wako Pure Chemi-
cal Industries, Ltd. H-Y, Na-Y and H-ZSM-5 were purchased from TOSOH
Co. SiO2 (230—400 mesh) and Al2O3 were purchased from Merck Co. and
Nacalai Tesque Inc. respectively. All reactions were carried out in a Pyrex
test tube under aerobic conditions, which was set up from the center of a
400-W high pressure mercury lamp to the distance of 37.5 mm. All of the
products are known compounds, and were identified by comparison of their
NMR spectra with those of authentic samples.
Synthesis of FSM-16 Kanemite was prepared as paste by filtration of a
suspension of sodium silicate (50 g) and dist. H2O (500 ml), which was
stirred for 3 h at r.t. Kanemite (as paste) was added to an aqueous solution
(1 l) of hexadecyltrimetylammonium chloride (32 g, 0.1 mol), and stirred for
3 h at 70 °C. After adjusting the pH at 8.5 with 2 N dil. HCl, the suspension
was further stirred for 3 h at 70 °C. The white precipitate was filtered,
washed with dist. H2O, dried at 60 °C for 12 h, and calcined at 550 °C for
5 h.
Chart 5
center (NBOHC) (20) and Eꢃ center (21) are formed by de-
hydroxylation from the isolated hydroxyl groups of silanols
through be calcined at 450 °C.30) Oxygen is, then, adsorbed
on the Eꢃ center, and generates the peroxy radical (POR)
Synthesis of MCM-41 Tetraethylorthosilicate (20.8 g, 0.1 mol) was
(22), which we believe to be the active site of this decarboxy- added by pipette to a solution of hexadecyltrimetylammonium bromide
lation.31) Chart 5 shows a plausible path of this reaction. The
radical species 24 is thought to be generated by abstraction
of a hydrogen radical with 22 and decarboxylated to give the
(7.3 g, 0.02 mol) and conc. HCl (8.3 ml, 0.1 mol) in H2O (270 ml), and
stirred for 48 h at r.t. The white precipitate was filtered and washed with dist.
H2O. The solid was dried at 120 °C for 12 h, and calcined at 650 °C for 4 h.
Synthesis of HMS A ethanol solution of tetraethylorthosilicate (20.8 g,
radical species 25 (Eqs. 1, 2). Aldehyde 26 was afforded by
abstraction of a hydrogen radical with 20 (Eq. 3), and the hy-
droperoxide 23 was homolytically cleaved under photo-irra-
diation to 20 and hydroxyl radical 27, which reacts with 26 to
afford the corresponding carboxylic acid (Eq. 4).
Although we have not measured metal content in our syn-
thesized FSM-16, it is incontrovertible that the metals, which
came from a water glass (sodium silicate made by Tokuyama
Siltech Co.), made possible to effect the photo-catalytic ac-
tivity, since the water glass included metals such as Al
(85 ppm), Fe (30 ppm), Ti (30 ppm), Ca (10 ppm) and Mg
(7 ppm).32)
In conclusion, this new method of decarboxylation is
thought to be more environmentally benign than previous
methods since our method eliminates the use of heavy metals
and organic solvents of high environmental impact, such as
halogenated solvents or benzene, and the reaction is opera-
0.1 mol) was added to a suspension of dodecylamine (5.0 g, 0.027 mol) in
H2O (38 ml) under vigorous stirring for 1 h, and aged for 18 h at r.t. The
white precipitate was washed with ethanol and H2O, and calcined at 650 °C
for 4 h.
Typical Procedure of Photo-Decarboxylation A suspension of sub-
strate (50 mg) and FSM-16 (100 mg) in dry hexane (5 ml) was irradiated at
room temperature with a 400-W high-pressure mercury lamp externally for
the indicated time. FSM-16 was then filtered off and washed with ethyl ac-
etate, and the filtrate was concentrated under reduced pressure. Pure product
was obtained after purification by preparative TLC.
Typical Procedure of Re-use of FSM-16 A suspension of substrate
(50 mg) and FSM-16 (100 mg) in dry hexane (5 ml) was irradiated at room
temperature with a 400-W high-pressure mercury lamp externally for the
indicated time. FSM-16 was then filtered off and washed with ethyl acetate.
The used FSM-16 was collected and re-calcined at 450 °C for 4 h by an elec-
tric furnace before being re-used.
References and Notes
1) Sheldon R. A., Chemtech, 24, 38—47 (1994).
2) IUPAC recommends classification of pores to micropore (Dꢀ2 nm; D,

November 2006
pore diameter) and mesopore (2 nmꢀDꢀ50 nm), see: Everett D. H.,
1575
543 (2000).
18) Kim Y. I., Kim Y. H., Tetrahedron Lett., 39, 639—642 (1998).
Pure Appl. Chem., 31, 579—638 (1972).
19) Mohri K., Mamiya J., Kasahara Y., Isobe K., Tsuda Y., Chem. Pharm.
Bull., 44, 2218—2222 (1996).
3) Itoh A., Kodama T., Maeda S., Masaki Y., Tetrahedron Lett., 39,
9461—9464 (1998).
20) Komuro M., Nagatsu Y., Higuchi T., Hirobe M., Tetrahedron Lett., 33,
4949—4952 (1992).
21) Banerjee A., Hazra B., Bhattacharya A., Banerjee S., Banerjee G. C.,
Sengupta S., Synthesis, 1989, 765—766 (1989).
22) Fristad W. E., Klang J. A., Tetrahedron Lett., 24, 2219—2222 (1983).
23) Giordano C., Belli A., Citterio A., Minisci F., J. Chem. Soc. Perkin
Trans. 1, 1981, 1574—1576 (1981).
4) Itoh A., Kodama T., Masaki Y., Synlett, 1999, 357—359 (1999).
5) Inagaki S., Koiwai A., Suzuki N., Fukushima Y., Kuroda K., Bull.
Chem. Soc. Jpn., 69, 1449—1457 (1996).
6) Inagaki S., Fukushima Y., Kuroda K., J. Chem. Soc., Chem. Commun.,
1993, 680—682 (1993).
7) Kresge C. T., Leonowicz M. E., Roth W. J., Vartuli J. C., Beck J. S.,
Nature (London), 359, 710—712 (1992).
24) Santaniello E., Ponti F., Manzocchi A., Tetrahedron Lett., 21, 2655—
2656 (1980).
25) Dessau R. M., Heiba E. I., J. Org. Chem., 25, 3647—3649 (1975).
26) Habibi M. H., Farhadi S., Tetrahedron Lett., 40, 2821—2824 (1999).
27) Maki Y., Sako M., Oyabu I., Murase T., Kitade Y., Hirota K., J. Chem.
Soc., Chem. Commun., 1989, 1780—1782 (1989).
28) Silica consists of the unit of the composition shown with
Si(OSi)n(OH)4ꢄn, and, in general, shows this by Qn, see: Thomas J. M.,
Klinowski J., Adv. Catal., 33, 197 (1985).
8) Tanev P. T., Pinnavaia T. J., Science, 267, 865—867 (1995).
9) Itoh A., Kodama T., Inagaki S., Masaki Y., Org. Lett., 2, 331—333
(2000).
10) Anpo M., Yamashita H., “Surface Photochemistry,” ed. by Anpo M.,
John Wiley & Sons, London, 1996.
11) Yoshida H., Tanaka T., Matsuo S., Funabiki T., Yoshida S., J. Chem.
Soc., Chem. Commun., 1995, 761—762 (1995).
12) Ogata A., Kazusaka A., Enyo M., J. Phys. Chem., 90, 5201—5205
(1986).
29) The solubility of O2 in hexane is higher than that of other typical or-
ganic solvents, see: Che Y., Tokuda K., Ohsaka T., Bull. Chem. Soc.
Jpn., 71, 651—656 (1998).
13) Anpo M., Yun C., Kubokawa Y., J. Catal., 61, 267—269 (1980).
14) Morikawa A., Hattori M., Yagi K., Otsuka K., Z. Phys. Chem., 104,
309—320 (1977).
30) Inaki Y., Yoshida H., Yoshida T., Hattori T., J. Phys. Chem. B, 106,
9098—9106 (2002).
15) Yoshida H., Kimura K., Inaki Y., Hattori T., Chem. Commun., 1997,
129—130 (1997).
31) Yoshida H., Murata C., Inaki Y., Hattori T., Chem. Lett., 1998, 1121—
1122 (1998).
32) We used plastic-wear and ultra-pure water exclusively for synthesis of
FSM-16 to prevent contamination with metal ions from other sources.
16) For example of application of mesoporous silica as solid acid to or-
ganic synthesis, see: Ishitani H., Iwamoto M., Tetrahedron Lett., 44,
299—301 (2003).
17) Itoh A., Kodama T., Inagaki S., Masaki Y., Chem. Lett., 2000, 542—