Selective synthesis of α-substituted β-keto esters from aldehydes and diazoesters on mesoporous silica catalysts

Article abstract of DOI:10.1016/j.tetlet.2008.05.077

The titled reactions are effectively catalyzed on mesoporous silica MCM-41 but not on amorphous silica prepared from the same starting materials as those of MCM-41. The ordered porous structure was essential to generate the acid catalysis. Planting of Al or Ti ion improved not only the catalytic activity but also the tolerance to basic functional groups of substrates. The catalytic activity of Al- or Ti-MCM-41 was better than that of SnCl₂ and NbCl₅ in the homogeneous phase. In particular, Al (or Ti)-MCM-41 was characteristically active for the reaction of aldehydes with α-substituted diazoesters to give 2-substituted-3-oxo propionic acid esters.

Full text of DOI:10.1016/j.tetlet.2008.05.077

Tetrahedron Letters 49 (2008) 4788–4791
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Selective synthesis of a-substituted b-keto esters from aldehydes
and diazoesters on mesoporous silica catalysts
*
Hiroaki Murata, Haruro Ishitani, Masakazu Iwamoto
Chemical Resources Laboratory, Tokyo Institute of Technology, 4259-R1-5 Nagatsuta, Midori-ku, Yokohama 226-8503, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The titled reactions are effectively catalyzed on mesoporous silica MCM-41 but not on amorphous silica
prepared from the same starting materials as those of MCM-41. The ordered porous structure was essen-
tial to generate the acid catalysis. Planting of Al or Ti ion improved not only the catalytic activity but also
the tolerance to basic functional groups of substrates. The catalytic activity of Al- or Ti-MCM-41 was
better than that of SnCl2 and NbCl5 in the homogeneous phase. In particular, Al (or Ti)-MCM-41 was
Received 3 April 2008
Revised 12 May 2008
Accepted 16 May 2008
Available online 22 May 2008
characteristically active for the reaction of aldehydes with
tuted-3-oxo propionic acid esters.
a
-substituted diazoesters to give 2-substi-
Ó 2008 Elsevier Ltd. All rights reserved.
b-Keto esters are one of the most important intermediates for
synthesis of useful organic compounds. Although many synthetic
routes have been proposed, there are only a few catalytic ap-
proaches.1 Acid-catalyzed reaction of aldehyde and diazoester is
regarded as a useful way to access b-keto esters because only nitro-
gen is generated as a by-product (Scheme 1). In 1909, Schlotter-
beck reported that this type of reaction proceeded at 333–363 K
without any catalysts; however, the examples were limited in
some halogenated aliphatic aldehydes or p-nitrobenzaldehydes
and the yields of desired products were less than 40% in most
cases.2 SnCl2 was found to be effective for the reaction of aldehydes
with diazoester;3 for example, with 10 mol % of SnCl2, hydrocinna-
maldehyde and ethyl diazoacetate were converted to the corre-
sponding b-keto ester in 86% yield. The SnCl2-catalyzed reaction
Catalyst
CO2R2
O
O
R1CHO
+
N2
+
N2
R1
OR2
1
2
3
nucleophilic addition
1,2-shift of H
O
H O
R1
Catalyst
OR2
O
N
CO2R2
+
N
R1
H
N
N
I
1,2-shift of R1
H
OH
CHO
R1 CO2R2
R1
CO2R2
4
Scheme 1. The reaction pathways of aldehydes and diazoesters.
* Corresponding author. Tel.: +81 45 924 5225; fax: +81 45 924 5228.
E-mail address: iwamoto@res.titech.ac.jp (M. Iwamoto).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.05.077

H. Murata et al. / Tetrahedron Letters 49 (2008) 4788–4791
4789
Table 1
was unfortunately not useful for the reaction of aromatic alde-
hydes.3 Recently, NbCl5 was reported as a new catalyst, on which
a variety of b-keto ester were obtained in 75–92% yield from ethyl
diazoacetate and aliphatic or aromatic aldehydes.4 Several hetero-
geneous catalysts were also investigated. Aromatic or aliphatic
aldehydes reacted with ethyl diazoacetate to yield b-keto ester in
75–85% yield on alumina,5 though the amount of alumina em-
ployed was very great. H-b-Zeolite6 and Cu-exchanged Mont
K107 were also found to be active for the reaction. The reflux con-
dition at 353 K was required on the catalysts to give satisfactory
yields in the reaction of aromatic or aliphatic aldehydes with ethyl
diazoacetate.
Yields of b-keto esters on M41-catalyzed reactions of aldehydes and ethyl diazoace-
tate
Catalyst
O
O
R1CHO
CO2Et
2a
+
N2
Solvent
Temp., 14 h
R1
OEt
1
3
Entry
R1
Catalyst
Solvent
Temp (K)
Yield (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
nPr (1a)
nPr
M41
M41
M41
M41
M41
M41
M41
M41
M41
M41
M41
M41
M41
Octane
298
298
298
298
298
298
298
353
273
298
298
298
298
298
55a
Toluene
Chlorobenzene
CH2Cl2
ClCH2CH2Cl
THF
CH3CN
ClCH2CH2Cl
CH2Cl2
ClCH2CHClCH2Cl
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
51a
nPr
62a
nPr
82a
nPr
78a
In the above study on SnCl2,3 BF3, ZnCl2, ZnBr2, AlCl3, GeCl2, and
SnCl4 were also examined as a catalyst and good yields were re-
ported only on SnCl2, GeCl2, and BF3. The activity of AlCl3 and SnCl4
was again reported by Hossain et al. to be low.8 In the zeolites cat-
alysts, the activity was in the order H-b > H-Y > H-ZSM-5.6 The re-
ports in homogeneous and heterogeneous phases might suggest
that strong acids would not be suitable for this type of reaction.
On the other hand, La(OTf)3 was reported to catalyze the reaction
of aldehydes and diazoester to produce epoxides through the Dar-
zens-type reaction.9 All the reports indicate that the selection of
the acidity and metal ion is a key to carry out the two successive
reactions for the b-keto ester formation, the nucleophilic addition
of diazoester and the hydride transfer in the resulting intermediate
(Scheme 1).
The novel acidic properties of mesoporous silica material MCM-
41 (M41) have been reported from the present10 and the other
groups.11 The acidity is known to be not strong but unique to pro-
mote various selective catalyses. Our efforts were therefore de-
voted to clarify the catalytic activity of M41 for the b-keto ester
formation from aldehydes and diazoesters. In this Letter, we will
report the effect of mesopores on the catalytic activity, its enhance-
nPr
11a
nPr
49a
nPr
30a
nPr
23a
Me (1b)
>99a
79a,b
82c
nHex (1c)
Ph(CH2)2 (1d)
cHex (1e)
nPr
73c
d
SiO2
Tracea
Effect of solvents, reaction temperature and sorts of aldehydes (Typical reaction
conditions: M41, 50 mg; 1, 0.50 mmol; 2a, 0.75 mmol; solv., 2.5 mL; 14 h.).
a
GC Yield.
b
The reaction was carried out for 1 h.
Isolated yield.
Prepared in vigorous stirring from colloidal silica and C12H25N(Me)3Br without
c
d
hydrothermal treatment. The BET surface area was 122 m2 gÀ1. 415 mg of the SiO2
was used in the catalytic run.
pane was the suitable solvent resulting in the high yield.15 The
yield of reaction product of cyclohexanecarboxaldehyde (1e) (en-
try 13) was somewhat smaller than those of the linear aliphatic
aldehydes. It would be due to the steric hindrance around carbonyl
carbon.
To reveal the effect of pore structure of M41, SiO2 was sepa-
rately prepared under the vigorous stirring by using the same
raw materials as those employed in the preparation of M41. The
components in obtained SiO2 were the same as those of M41 with-
in the experimental errors, and the major difference between M41
and SiO2 was the presence/absence of pore structures. This SiO2
was entirely inert for the present reaction (entry 14). It should
be noted in the entry 14 that the BET surface area of SiO2 employed
was 122 m2 gÀ1 and therefore 415 mg of the sample was used as
the catalyst to unify the surface areas applied for the reaction.
The result indicates that first the Al impurity contained in M41 is
not the origin of the catalytic activity, and second the presence of
ment by addition of metal ion, and finally the reactions of
tuted diazoesters.
a-substi-
M41 was prepared in the reported manner by using
12H25N(CH3)3Br as the template and colloidal silica as the silica
C
source.12 The BET surface area and the BJH pore diameter were
1010 m2 gÀ1 and 2.12 nm, respectively. The hexagonal structure
of resulting M41 was confirmed from the XRD pattern and the
Si/Al atomic ratio was 237, in which the origin of Al is impurity
of the raw material, colloidal silica. Metal ion-planted M41s (M-
M41s) were prepared by the template-ion exchange (TIE) meth-
od.13 M41 and M-M41s were used in the catalytic reaction after
evacuation at 373 K for 1 h.
We first examined the effect of solvent in the reaction of
0.5 mmol of n-butanal (1a) and 0.75 mmol of ethyl diazoacetate
(2a) on 50 mg of M41.14 The reaction easily proceeded at room
temperature. As shown in Table 1, halogenated aromatic or ali-
phatic hydrocarbons such as chlorobenzene, dichloromethane,
and 1,2-dichloroethane induced better yields than hydrocarbons
including octane and toluene, and the best yield of 82% was at-
tained by using dichloromethane (entry 4). Polar solvents, THF
and CH3CN, were not effective. The effectiveness of halogenated
hydrocarbon as a solvent in this reaction was already recognized
in the previous reports.3,4,6,8 Next the effect of reaction tempera-
ture was studied on M41 in halogenated hydrocarbon solvents.
Upon temperature rising to 353 K in CH2ClCH2Cl, the yield became
lower to 30% from 78% at 298 K (entries 5 and 8). On the other
hand, at 273 K the yield was only 23% in CH2Cl2 (entry 9). All
experiments were hereinafter carried out at room temperature in
CH2Cl2.
Table 2
Change in the catalytic activity of M41 with the loading of metal ionsa
Catalyst
(4 mol% as metal)
O
O
p
-TolCHO
CO2Et
2a
+
N2
p
-Tol
OEt
CH2Cl2
273 K, 14 h
3f
1f
Entry
Catalyst
[M]/mmol gÀ1 Si/M
Conv. of Yieldb Select
1fb (%)
(%)
(%)
Weight/
mg
1
2
3
4
5
M41
Al-M41 0.33
Ti-M41 0.35
0.065
[Al]237 50
63
94
84
82
79
53
86
75
45
72
84
91
89
55
91
35
35
42
40
53
50
74
61
Sc-M41 0.27
La-M41 0.29
The reactivity of other aliphatic aldehydes such as acetoalde-
hyde (1b), n-heptanal (1c), and hydrocinnamaldehyde (1d) was
examined and corresponding b-keto esters could be obtained in
good to excellent yields (entries 10–12). For 1b, 1,2,3-trichloropro-
a
Typical reaction conditions: 1f, 0.50 mmol; 2a, 0.75 mmol; CH2Cl2, 2.5 mL;
273 K; 14 h.
b
Determined by GC analysis.

4790
H. Murata et al. / Tetrahedron Letters 49 (2008) 4788–4791
mesoporous structures is significant for the appearance of
catalysis.
much higher than that of SnCl2 and NbCl5, and furthermore M41
and its modified ones required only mild reaction conditions such
as room temperature and short reaction time. Substituted diazo-
esters could also be used as substrates; as a result, 3-oxo-2-substi-
The reaction of p-tolualdehyde with ethyldiazoacetate 2a was
examined on M41 (Table 2) in more detail. The Hossain’s report8
indicated that Brønsted acids conducted the 1,2-shift of the aro-
matic group on the intermediate I to obtain 3-hydroxy-2-arylacry-
lates (Scheme 1, 4), whereas weak Lewis acids such as SnCl2 had
high selectivity to the synthesis of b-keto ester 3 though the yields
were low to moderate. As shown in Table 2, entry 1, M41 showed
high selectivity (84%) and moderate conversion (63%) in the reac-
tion of p-tolualdehyde (1f) with 2a. The catalytic activity of M41
for aromatic aldehyde was good but the possibility of its improve-
ment with addition of metal ion was investigated here. As summa-
rized in Table 2, the planting of Al, Ti, and La onto M41 greatly
improved the yields of 3f and further the selectivity also became
better from 84% to 89–91%. It would be worth to note that the
highest yield of 86% was obtained on Al-M41 and we needed
neither the reflux condition nor large amount of catalyst used5–7
(Table 2, entry 2).
The excellent catalytic activity of Al-M41 led us to investigate
systematically the reactions of various substrates (Table 3). Entries
116 and 2 concluded that the catalytic activity of Al-M41 was better
than or similar to that of M41 (Table 1) for the condensation of ali-
phatic aldehyde 1c or 1d with 2a. The aromatic aldehydes were
applicable partners in the b-keto ester synthesis on Al-M41 and
gave 3-oxo-3-aryl propionic acid esters in 50–86% yields (entries
Table 3
Synthesis of various substituted b-keto ester on MCM-41
O
O
R2
Catalyst
R1CHO
R1
OR3
+
N2 CO2R3
CH2Cl2
Temp., 14 h
R2
1
2
3
Entry
1d
Product
Temp (K)
Yield (%)
SnCl2
b
c
Al-M41a
NbCl5
34
>99,e >99f
O
O
298
298
273
298
76
85
41
30
n
-Hex
OEt
86,e,g >99f,g
80,e,h >99f,h
3c
O
O
2i
82
86
50
71
49
39
OEt
3d
O
O
3d
OEt
Me
3–5). In addition, the reactions of a-substituted diazoacetates with
3f
aldehydes afforded 3-oxo-2-phenyl propionic or 3-oxo-2-methyl
acid esters in good yields (entries 6–8).17 They can provide a direct
way for the catalytic synthesis of 3-oxo-2-substituted propionic
acid esters. 2-Pyridinecarboxaldehyde (1l) and 4-nitrobenzalde-
hyde (1m) did not give good yields on Al-M41 (entries 9 and 10),
while the use of Ti-M41 instead of Al-M41 improved the yields.
Although the active sites of M41, Al-M41, and Ti-M41 catalysts still
remain unclear, appropriate Lewis acidity of Ti ion would prevent
the substrates bearing basic functional groups from occasionally
coordinating to acid sites and deactivating the catalytic activities.
Finally two important findings should be noted. Firstly, the
reusability of the present catalyst was examined. The Al-M41 cat-
alyst was recovered by filtration after the reaction of 1c with 2a,
dried at 353 K for 1 h, and evacuated at 373 K for 1 h (the same
treatment as that in the 1st run). The results of 2nd and 3rd runs
are shown in Table 3, entry 1. The yields after 10 min were slightly
lower than that of 1st run, but those after 60 min were all >99%.
The results indicate that the present Al-M41 was reusable without
significant loss of the catalytic activity. Secondly, the catalytic
activity of Al-M41 (and Ti-M41) was compared with that of the re-
ported catalysts. SnCl32 and NbCl45 were employed here as the refer-
ence catalysts, and the results determined in this study are
summarized in Table 3. At first it should be noted that the results
in the present SnCl2-catalyzed reactions are in good agreement
with those of the literature3 (entries 2 and 5). The catalytic activity
of Al-M41 for the reaction of aliphatic aldehydes with 2a (entries 1
and 2) was greater than or almost the same as that of SnCl2. The
difference between the yields on the former and the latter became
larger in the reactions of aromatic aldehydes (entries 3–5). The cat-
alytic activity of SnCl2 for the syntheses of 3i–m, not reported in
the literature, was much lower than that of M-M41 as shown in en-
tries 6–10. In the case of NbCl5 (5 mol %), the yields of b-keto esters
were always low without exception. We should add that unfortu-
nately we could not reproduce the Yadav’s results4 in the present
experiments, though the reactions were performed under strict
anhydrous conditions by using purified NbCl5 and reagents.
In summary, mesoporous silica M41 and Al- or Ti-loaded M41
were highly active for the synthesis of b-keto esters from a wide
variety of aldehydes and diazoesters. The catalytic activity was
O
O
4i
OEt
MeO
3g
O
O
5d
6i
7i
298
298
298
78
41
59
49
5
20
6
OEt
3h
O
O
n
-Hex
OBn
3i
Ph
O
O
O
OBn
9
16
Ph
Me
Me
3j
O
8i
9i
OEt
298
298
298
55
48
10
17
12
14
11
Me
3k
O
O
19, 53j
16, 65j
OEt
N
3l
O
O
10i
OEt
O N
2
3m
a
Typical reaction conditions: Al-M41, 50 mg; 1, 0.50 mmol; 2, 0.75 mmol;
CH2Cl2, 2.5 mL; 14 h.
b
By using 10 mol % of SnCl2 as a catalyst.
By using 5 mol % of NbCl5 as a catalyst.
GC yield.
Reaction time, 10 min.
Reaction time, 60 min.
2nd run. Since the amount of Al-M41 recovered after the 1st run was 36 mg, the
c
d
e
f
g
following reaction conditions were applied: 1, 0.36 mmol; 2, 0.54 mmol, CH2Cl2,
1.8 mL.
h
3rd run. Recovered Al-M41, 25 mg; 1, 0.25 mmol; 2, 0.38 mmol; CH2Cl2, 1.3 mL.
i
Isolated yield.
j
By using 50 mg of Ti-M41 ([Ti] = 0.90 mmol gÀ1, Si/Ti = 16) as a catalyst.

H. Murata et al. / Tetrahedron Letters 49 (2008) 4788–4791
4791
11. (a) Itoh, A.; Kodama, T.; Maeda, S.; Masaki, Y. Tetrahedron Lett. 1998, 39, 9461–
9464; Itoh, A.; Kodama, T.; Masaki, Y. Chem. Pharm. Bull. 2007, 55, 861–864;
(c)(b) Yamamoto, T.; Tanaka, T.; Funabiki, T.; Yoshida, S. J. Phys. Chem. B 1998,
102, 5830–5839.
tuted propionic acid esters were directly obtained through the
reaction.
12. Abe, T.; Tachibana, Y.; Uematsu, T.; Iwamoto, M. J. Chem. Soc., Chem. Commun.
1995, 1617–1618.
13. Iwamoto, M.; Tanaka, Y. Catal. Surv. Jpn. 2001, 5, 25–36.
References and notes
14. Typical procedure for b-ketoester preparation:
a solution of aldehyde 1
1. Several preparation methods of b-keto esters such as Claisen condensations
were described below: (a) Benetti, S.; Romagnoli, R.; Risi, C. D.; Spalluto, G.;
Zanirato, V. Chem. Rev. 1995, 95, 1065–1114; (b) Smith, M. B.; March, J.
Advanced Organic Chemistry, 5th ed.; Benjamin: New York, 2001; pp 569–573;
(c) Iida, A.; Takai, K.; Okabayashi, T.; Misaki, T.; Tanabe, Y. Chem. Commun.
2005, 3171–3173; (d) Sato, T.; Itoh, T.; Fujiwara, T. Chem. Lett. 1982, 1559–
1560.
(0.50 mmol) with tetradecane (0.10 mmol, internal standard) in 2.0 mL
distilled CH2Cl2 was added to the evacuated M41s. Then, 0.75 mmol of
diazoester
2 with 0.5 mL of CH2Cl2 were dropwised into the suspension
(30 min). After 14 hour, the reaction media was diluted by CH2Cl2 and the
catalyst was removed by the filtration. Yield was determined by preparative
TLC or GC.
15. 3b was 65% yield when CH2Cl2 was used as a solvent.
16. Ethyl diazoacetate 2a was added at a stroke.
2. (a) Schlotterbeck, F. Chem. Ber. 1909, 42, 2565–2573; (b) Curtius, T.; Buchner, E.
Chem. Ber. 1885, 18, 2371–2377.
17. Physical data of representative compounds: 3i: 1H NMR (400 MHz; CDCl3/TMS) d
(ppm) 0.83 (t, 3H, J = 7.0), 1.14–1.21 (m, 6H), 1.45–1.57 (m, 2H), 2.08 (t, 0.6H,
J = 8.0), 2.44 (t, 1.4H, J = 7.2), 4.76 (s, 0.7H), 5.19 (s, 2H), 7.15–7.19 (m, 1H),
7.27–7.38 (m, 9H). 13C NMR (100 MHz, CDCl3/TMS): d (ppm) 13.98, 22.42,
23.56, 26.63, 28.51, 28.83, 31.39, 31.43, 32.80, 41.64, 64.92, 65.76, 67.26,
127.02, 127.14, 127.80, 128.05, 128.26, 128.35, 128.38, 128.56, 128.83, 129.47,
131.33, 132.50, 135.40, 168.45, 203.65. Anal. for C22H26O3 (338.19). Calcd: C,
78.07; H, 7.74; O, 14.18. Obsd: C, 77.67; H, 8.00; O, 14.89. Compound 3j: 1H
NMR (400 MHz, CDCl3/TMS): d (ppm) 2.27 (s, 3H), 5.16 (s, 2H), 5.64 (s, 1H),
7.10–7.40 (m, 12H), 7.81 (d, 2H). 13C NMR (100 MHz, CDCl3/TMS): d (ppm)
21.5, 60.41, 67.23, 126.32, 126.75, 127.27, 127.43, 127.90, 128.10, 128.24,
128.75, 129.13, 133.06, 135.56, 144.41, 168.77, 192.69. Anal. for C23H20O3
(338.19). Calcd: C, 80.21; H, 5.85; O, 13.94. Obsd: C, 79.30; H, 6.02; O, 16.23.
The amounts of carbon and hydrogen were analyzed in a usual manner with a
CHNS-932 system of LECO Corporation. The oxygen content was separately
determined with a CHNS-932 system using a VFT-900 furnace operated at
1573 K.
3. Holmquist, C. R.; Roskamp, E. J. J. Org. Chem. 1989, 54, 3258–3260.
4. Yadav, J. S.; Subba Reddy, B. V.; Eeshwaraiah, B.; Reddy, P. N. Tetrahedron 2005,
61, 875–878.
5. Dhavale, D. D.; Patil, P. N.; Mali, R. S. J. Chem. Res. (S) 1994, 152–153.
6. Balaji, B. S.; Chanda, B. M. Tetrahedron 1998, 54, 13237–13252.
7. Phukan, P.; Mohan, J. M.; Sudalai, A. J. Chem. Soc., Perkin Trans. 1 1999, 3685–
3689.
8. (a) Mahmood, S. J.; Hossain, M. M. J. Org. Chem. 1998, 63, 3333–3336; (b)
Dudley, M. E.; Morshed, M. M.; Brennan, C. L.; Islam, M. S.; Ahmad, M. S.; Atuu,
M.; Branstetter, B.; Hossain, M. M. J. Org. Chem. 2004, 69, 7599–7608.
9. Curini, M.; Epifano, F.; Marcotullio, M. C.; Rosati, O. Eur. J. Org. Chem. 2002,
1562–1565.
10. (a) Tanaka, Y.; Sawamura, N.; Iwamoto, M. Tetrahedron Lett. 1998, 39, 9457–
9460; (b) Iwamoto, M.; Yasuhiro, T.; Sawammura, N.; Namba, S. J. Am. Chem.
Soc. 2003, 125, 13032–13033; (c) Ishitani, H.; Iwamoto, M. Tetrahedron Lett.
2003, 44, 299–301.