Dimethyl malonate/LHMDS system as a new protocol for generating methyl formate anion (-COOMe) in the condensed-phase Dedicated to Professor Charles H. DePuy (1927-2013)

Article abstract of DOI:10.1016/j.tetlet.2016.02.090

The treatment of dimethyl malonate with LHMDS in anhydrous THF (condensed-phase) generates, in addition to the expected corresponding lithium enolate, methyl formate anion (or methoxycarbonyl anion, ^-COOMe) which can react with several electrophiles to give corresponding methoxycarbonyl derivatives by nucleophilic substitution reaction.

Full text of DOI:10.1016/j.tetlet.2016.02.090

Accepted Manuscript
Dimethyl malonate/LHMDS system as a new protocol for generating methyl
–
formate anion ( COOMe) in the condensed-phase
Valeria Di Bussolo, Salvatore Princiotto, Elisa Martinelli, Vittorio Bordoni,
Paolo Crotti
PII:
S0040-4039(16)30199-X
DOI:
http://dx.doi.org/10.1016/j.tetlet.2016.02.090
TETL 47365
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Accepted Date:
15 February 2016
24 February 2016
Please cite this article as: Bussolo, V.D., Princiotto, S., Martinelli, E., Bordoni, V., Crotti, P., Dimethyl malonate/
–
LHMDS system as a new protocol for generating methyl formate anion ( COOMe) in the condensed-phase,
Tetrahedron Letters (2016), doi: http://dx.doi.org/10.1016/j.tetlet.2016.02.090
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Dimethyl malonate/LHMDS system as a new protocol for generating
methyl formate anion (–COOMe) in the condensed-phase
Valeria Di Bussolo,* Salvatore Princiotto, Elisa Martinelli, Vittorio Bordoni, Paolo Crotti*
Dipartimento di Farmacia, Università di Pisa, Via Bonanno 33, 56126 Pisa, Italy
Abstract: The treatment of dimethyl malonate with LHMDS in anhydrous THF (condensed-phase)
generates, in addition to the expected corresponding lithium enolate, methyl formate anion (or
methoxycarbonyl anion, –COOMe) which can react with several electrophiles to give corresponding
methoxycarbonyl derivatives by nucleophilic substitution reaction.
Dedicated to Professor Charles H. DePuy (1927-2013)
Keywords: Methyl formate anion, Methoxycarbonyl anion, Dimethyl malonate, Condensed-phase,
Nucleophilic substitution.
–
Methyl formate anion (or methoxycarbonyl anion, COOMe) is known to be formed both in
the condensed- and gas-phase by deprotonation of methyl formate by another anion.1 In this
process, the electrophilic carbonyl carbon of methyl formate, is transformed into the nucleophilic
carbon of methyl formate anion, in a classical umpolung example. In the gas-phase, DePuy and coll.
found that formyl proton abstraction, with formation of –COOMe and deuterated ammonia, is one of
the primary reaction pathways promoted by reaction of deuterated methyl formate with amide ion
–
(NH2 ): a subsequent exothermic -elimination reaction leads to the more stable CO and MeO–.1a
Methyl formate anion was identified in the condensed-phase, as a consequence of the reaction
between Ni(CO)4 and CH3OK at 313-333 K. Under these conditions, Ni(CO)4 forms complexes
with the base with release of CO which, finally, reacts with CH3O– to give anion –COOCH3.2
In another example, methyl formate anion was demonstrated to be formed in the condensed-
phase in the course of the synthesis of -formyl acids. Actually, when the dianion of a carboxylic
_________________
*Corresponding authors. Tel.: +393472586140; e-mail addresses: valeria.dibussolo@farm.unipi.it
(V. Di Bussolo), paolo.crotti@farm.unipi.it (P. Crotti).
1

acid was treated with ethyl formate, in addition to the desired acyl substitution product (the -
formyl derivative), a competitive, undesired, acid-base reaction occurred with formyl proton
abstraction. The consequent formation of methyl formate anion was demonstrated by quenching the
reaction mixture with 3HCl with the obtainment of corresponding 3H-labelled ethyl formate.3
In the course of a program directed to evaluate the regio- and stereoselectivity of the
nucleophilic addition of metal enolates of methylene active carbonyl compounds (C-nucleophiles)
to glycal-derived vinyl epoxides, the reaction of D-galactal derived epoxide 14 with metal enolates
of dimethyl malonate was initially considered.
The addition of epoxide 1 to a THF solution of 1:1 dimethyl malonate/LHMDS (3 equiv),
reasonably containing the corresponding lithium enolate 2, afforded, after 4 h at room temperature,
a crude reaction product consisting of cis-2,5-disubstituted-2,5-dihydrofuran 35 (35%) and an 1,4-
addition product (65%) in which the residue of the starting nucleophile, the di-(methoxycarbonyl)-
methyl group [–CH(COOMe)2], turned out to be reduced to a simple methoxycarbonyl group (–
COOMe) (Scheme 1). In other words, the –CHCOOMe portion of the original enolate/nucleophile 2
1
appeared lost. All the collected H NMR evidences unexpectedly indicated for this major reaction
product the structure of methyl -glycosylcarboxylate derivative 4, subsequently confirmed by an
accurate examination of the corresponding 4-O-acetyl derivative 4-Ac (Scheme 1).
Scheme 1
OH
O
COOMe
BnO
O
/ LHMDS
MeOOC
COOMe
HC
O
BnO
CH(COOMe)2
BnO
+
THF
OMe

RO
O
+
-
O Li

1
3
2
COOMe
4, R = H
4-Ac, R = Ac
The formation of methyl -glycosylcarboxylate derivative 4 pointed to the presence in the
reaction mixture, under the reaction conditions used, of an unexpected nucleophilic species, as
methoxycarbonyl anion (–COOMe). Actually, control experiments appropriately carried out
indicated that the mixture of 2,5-dihydrofuran 3 and methyl -glycosylcarboxylate 4 is stable
under the reaction conditions and that methyl -glycosylcarboxylate 4 is formed only when
LHMDS is used as the base for generating the metal enolate species: the use of weaker bases as t-
BuOK and t-BuOLi was unsuccessful. These observations let us think that lithium enolate of
dimethyl malonate 2, formed, as usual, by reaction of equimolar amounts of dimethyl malonate and
LHMDS, could partially undergo a further deprotonation by the strong base to give dianion 5. A
2

–
subsequent -elimination process leads to carbene-anion species 6 and nucleophile COOMe, as
the corresponding lithium salt LiCOOMe (7) (Scheme 2).
Scheme 2
OMe
OMe
COOMe
COOMe
O
OMe
-
O Li
LHMDS
THF
LHMDS
THF
+
LiCOOMe
+
-
O Li
-
O
Li
Li+
+
-
C
C
Li+
+
HC
-C
OMe
COOMe
COOMe
6
7
2
5
While carbene-anion 6 probably decomposes, nucleophilic attack of LiCOOMe to epoxide 1
leads to methyl -glycosylcarboxylate 4 as the only addition product, in which the configuration
() of the obtained C-glycoside is the same as that () of the starting epoxide 1, in a typical
example of syn-1,4-addition process (Schemes 1 and following Scheme 3).
The completely 1,4-regio- and -stereoselective formation of methyl -glycosyl carboxylate
4 is in accordance with the typical behavior of D-galactal-derived vinyl epoxide 1, in addition
reactions by a nucleophile, as LiCOOMe, bearing the highly coordinating lithium cation.4b,c
Actually, by the occurrence of a coordination with the oxirane oxygen through the metal ion, as
shown in structure 1’ (Scheme 3), nucleophile LiCOOMe is brought on the  face of the epoxide
and, in this way, it is correctly oriented for an entropically favored conjugated attack to vinyl C(1)
carbon from the same side to give -glycosyl derivative 4, as experimentally found (route a,
Scheme 3). Evidently, the strongly associated nature of the new nucleophilic species LiCOOMe
allows only attack through a coordination with oxirane oxygen by lithium cation to give only the
corresponding syn-1,4-addition product 4 (coordination product)6,7 while its reactivity as a free,
not coordinated nucleophile is weakened. This could be the reason why the corresponding
regioisomeric anti-1,2-addition product 8 (route b) and stereoisomeric anti-1,4-addition product 4
(route c) (non-coordination products) 6,7 are not observed with this nucleophile under the described
reaction conditions (Scheme 3).
3

Scheme 3
oxirane oxygen-nucleophile
coordination
syn-1,4-addition product
(coordination product)
+
O
+
Li
-
OMe
route a
O
COOMe
O
O
OBn
BnO
BnO
HO
LiCOOMe
THF
a
O
1

O
3
1'
4
b

1
c
O
(methyl -glycosyl carboxylate)
route b
route c
+
-
MeO

Li
O
O
COOMe
BnO
HO
BnO
HO
4
COOMe
8
(anti-1,2-addition product)
(anti-1,4-addition product)
not obtained
not obtained
In a preliminary confirmation of the formation of nucleophile LiCOOMe (7) in a mixture with
lithium enolate 2, under the above-described protocol, some electrophiles different from the
originally used epoxide 1 were checked. The use of benzyl bromide (9a), allyl bromide (9b), 2-
methylallyl bromide (9c), benzoyl chloride (9d) and p-methoxyphenacyl bromide (9e) in the
reaction with dimethyl malonate/LHMDS reagent system (1 equiv) led to an 1:1 (in the case of 9a
and 9e), 1.2: 1 (in the case of 9d) and 1.7: 1 mixture (in the case of 9b and 9c) of the two
corresponding substitution products, the expected malonyl derivatives 10a-e and the “unexpected”
methoxycarbonyl derivatives 11a-e (45-89% conversion, Scheme 4), accompanied by unreacted
dimethyl malonate (55-11%) (1H NMR). In all cases, the crude reaction mixtures were subjected to
preparative TLC and malonyl derivatives 10a-e8a-e and methoxycarbonyl derivatives (methyl esters
11a-c,9a-c methyl -keto ester 11d9d and methyl -keto ester 11e9e) were separated and identified by
1H NMR and/or comparison with reported data (see Supplementary data).8,9
4

Scheme 4
PhCH2CH(COOMe)2
+
PhCH2COOMe
11a
1 : 1
10a
+ CH2=CHCH2COOMe
CH2=CHCH2CH(COOMe)2
PhCH2Br (9a)
68%
10b 1.7 : 1
11b
CH2=CH-CH2Br
(9b)
79%
OMe
OLi
COOMe
COOMe
LHMDS
THF
CH2=C(Me)CH2Br (9c)
+ LiCOOMe
+
CH2=C(Me)CH2COOMe
CH2
CH2=C(Me)CH2CH(COOMe)2
HC
89%
7
COOMe
11c
2
10c
1.7 : 1
PhCOCl (9d)
45%
p-OMe-C6H4-COCH2Br
(9e)
PhCOCH(COOMe)2
+
PhCOCOOMe
50%
10d
1.2 : 1
11d
p-OMe-C6H4-COCH2CH(COOMe)2
+
p-OMe-C6H4-COCH2COOMe
11e
10e 1 : 1
The interesting results obtained with dimethyl malonate/LHMDS system prompted us to
check the possible extension of the new protocol to other symmetric methylene active
compounds. In particular, because interested in the possibility of generating other unusual
nucleophiles as benzoyl anion 141c and the particularly important N,N-dimethylcarbamoyl anion
15, useful in carbamoylation reactions,10 we directed our attention to dibenzoyl methane (12)
and N,N,N’,N’-tetramethylmalondiamide (13) (Scheme 5).
Scheme 5
Ph
COPh
COPh
COPh
O
LHMDS
THF
BnBr
20%
+
-
O Li
Li+
-
PhCH2
C
HC
COPh
Ph
COPh
14
12
16
18
(only substitution product)
no evidence
of formation
NMe2
CONMe2
CONMe2
13
CONMe2
PhCH2
O
LHMDS
THF
BnBr
99%
+
-
Li+
-
O Li
CONMe2
17
C
HC
CONMe2
NMe2
15
19
no evidence
of formation
(only sustitution product)
Several reaction conditions were checked by using benzyl bromide as the electrophile in
the reaction with 12/LHMDS and 13/LHMDS systems (different reaction temperature, different
benzyl bromide/12- or 13-LHMDS ratio). Unfortunately, the common addition product, 2-
5

benzyl-1,3-diphenyl-1,3-propandione (18)11 (20% conversion and presence of unreacted 12,
80%) and benzyl- N,N,N’,N’-tetramethylmalondiamide (19)12 (99% conversion) from 12 and 13,
respectively (1H NMR), were the only substitution products to indicate that in these cases the
corresponding lithium enolates 16 and 17 were the only nucleophiles present in the reaction
mixtures (Scheme 5).13
All these results would indicate that the behavior of dimethyl malonate in the presence of
LHMDS is unique and, contrary to our expectations, the dimethyl malonate/LHMDS protocol
cannot be extended to other methylene active compounds and, as a consequence, cannot be
considered as a general protocol for the generation of unusual nucleophilic species. At the
moment, we don’t see the reason and don’t have an explanation for this behavior apparently
limited to dimethyl malonate.
Typical procedure for the generation of methyl formate anion from dimethyl
malonate/LHMDS system and its reaction with an electrophile. Reaction of dimethyl
malonate/LHMDS system with benzyl bromide. A 1 M LHMDS solution in anhydrous THF (1.0
mL, 1.0 mmol) was treated at 0°C with a solution of dimethyl malonate (0.132 g, 1.0 mmol) in
anhydrous THF (3.0 mL) and the reaction mixture was stirred at the same temperature for 1 h. A
solution of benzyl bromide (0.171 g, 0.12 mL, 1.0 mmol) in anhydrous THF (0.5 mL) was
added dropwise at 0°C and the reaction mixture was stirred for 18 h at room temperature.
Dilution with Et2O and evaporation of the washed (saturated aqueous NaCl) organic solution
afforded a crude reaction mixture (0.160 g) consisting of a 35:33:32 mixture of dimethyl
benzylmalonate (10a), methyl phenylacetate (11a) and unreacted dimethyl malonate (1H NMR)
which was subjected to preparative TLC by using a 9:1 hexane/AcOEt mixture as the eluant.
Extraction of the two most intense bands (the faster moving band contained 10a) afforded pure
dimethyl benzylmalonate (10a) (0.048 g) and methyl phenylacetate (11a) (0.029 g).8a,9a
Acknowledgements
This work was supported by the University of Pisa (University Funds 2013-2015). Professor
Gloria Uccello-Barretta and Dr Federica Balzano of the Dipartimento di Chimica e Chimica
Industriale of the University of Pisa are gratefully acknowledged for NMR experiments.
6

Supplementary data
Supplementary data (experimental details for all reactions and corresponding products)
associated with this article can be found in the online version, at doi:
References and notes
1.
a) DePuy, C. H.; Grabowski, J. J.; Bierbaum, V. M.; Ingemann, S.; Nibbering, N. M. M. J.
Am. Chem. Soc. 1985, 107, 1093-1098. b) Ingemann, S.; Nibbering, N. M. M. J. Org. Chem.
1985, 50, 682-689; c) Chandrasekhar. J.; Andrade, J. G.; von Rague Schleyer, P. J. Am.
Chem. Soc. 1981, 103, 5612-5614 and references therein.
2.
3.
4.
Ohyama, S. Appl. Catal. A 2001, 220, 235-242.
Koch, G. K.; Kop, J. M. M. Tetrahedron Lett. 1974, 15, 603-606.
a) Di Bussolo, V.; Caselli, M.; Pineschi, M.; Crotti, P. Org. Lett. 2002, 4, 3695-3698. b) Di
Bussolo, V.; Caselli, M.; Pineschi, M.; Crotti, P. Org. Lett. 2003, 5, 2173-2176. c) Di
Bussolo, V.; Caselli, M.; Romano, M. R.; Pineschi, M.; Crotti, P. J. Org. Chem. 2004, 69,
8702-8708. d) Di Bussolo, V.; Checchia, L.; Romano, M. R.; Pineschi, M.; Crotti, P. Org.
Lett. 2008, 10, 2493-2496.
5.
6.
The regio- and stereoselective formation of 2,5-disubstituted-2,5-dihydrofuran systems, as 3,
by reaction of glycal-derived vinyl epoxides, as 1, with metal enolates of methylene active
compounds, will be the subject of a forthcoming paper from our laboratory.
In accordance with previous results with glycal-derived epoxides,7 the simplified
nomenclature of coordination product is given to syn-1,4-addition product 4, because
supposed to be formed through an oxirane oxygen-nucleophile coordination (route a,
Scheme 3). Analogously, anti-1,2-addition product 8 and anti-1,4-addition product 4, even
if not obtained, are simply identified as non-coordination products because they could be
formed only by attack of a free, non-coordinated nucleophile at C(3) (route b) and C(1)
carbon (route c) of epoxide 1, respectively (Scheme 3).
7.
8.
a) Di Bussolo, V.; Frau, I.; Favero. L.; Uccello-Barretta, G.; Balzano, F.; Crotti, P.
Tetrahedron 2015, 71, 6276-6284. b) Di Bussolo, V.; Favero, L.; Romano, M. R.; Pineschi,
M.; Crotti, P. Tetrahedron 2008, 64, 8188-8201.
a) Guyon, C.; Duclos, M.-C.; Sutter, M.; Métay, E.; Lemaire, M. Org. Biomol. Chem. 2015,
7

13, 7067-7075. b) Kammerer, C.; Prestat, G.; Gaillard, T.; Madec, D.; Poli, G. Org. Lett
2008, 10, 405-408. c) Laidig, G. J,; Hegedus, L. S. Synthesis 1995, 527-532. d) Lu, W.; Li,
Y.; Wang, C.; Xue, D.; Chen, J.-G.; Xiao, J. Org. Biomol. Chem. 2014, 12, 5243–5249. e)
New compound: see Supplementary data.
9.
a) Harker, W. R.R.; Emma L.; Carswell, E. L.; David R.; Carbery, D. R. Org. Lett. 2010, 12,
3712-3715. b) Ramachandran, P. V.; Nicponski, D.; Kim, B. Org.Lett 2013, 15, 1398-1401.
c) Duhaime, R. M.; Lombardo, D. A.; Skinner, I. A.; Weedon, A. C. J. Org. Chem. 1985,
50, 873-879. d) Moriyama, K.; Takemura, M.; Togo, H. Org.Lett. 2012, 14, 2414-2417. e)
Borowiecki, P.; Bretner, M. Tetrahedron: Asymmetry 2013, 24, 925-936.
10.
11.
Tsuda, T.; Miwa, M.; Saegusa, T. J. Org. Chem. 1979, 44, 3734-3736.
Gogoi, P.; Kotipalli, T.; Indukuri, K.; Bondalapati, S.; Saha, P.; Saikia, A., K. Tetrahedron
Letters 2012, 53, 2726–2729.
12.
13.
New compound: see Supplementary data.
In an alternative rationalization, both the usual lithium enolates 16 and 17 and the unusual
benzoyl 14 and dimethylcarbamoyl anion 15 are formed, but 16 and 17 could be decidedly
more reactive nucleophiles than 14 and 15 (at least under the reaction conditions used) to the
point that only the corresponding SN2 products 18 and 19 are formed and found in the
reaction mixtures.
8

Ref. TETL 47365
Dimethyl malonate/LHMDS system as a new protocol for generating
methyl formate anion (–COOMe) in the condensed-phase
Valeria Di Bussolo,* Salvatore Princiotto, Elisa Martinelli, Vittorio Bordoni, Paolo Crotti*
Dipartimento di Farmacia, Università di Pisa, Via Bonanno 33, 56126 Pisa, Italy
Captions to Schemes 1-5
OH
O
COOMe
BnO
O
/ LHMDS
MeOOC
COOMe
HC
O
BnO
CH(COOMe)2
BnO
+
THF
OMe

RO
O
+
-
O Li

1
3
2
COOMe
4, R = H
4-Ac, R = Ac
Scheme 1. Reaction of vinyl epoxide 1 with dimethyl malonate/LHMDS system.
OMe
OMe
COOMe
COOMe
O
OMe
-
O Li
LHMDS
THF
LHMDS
THF
+
LiCOOMe
+
-
O Li
-
O
Li
Li+
+
-
C
C
Li+
+
HC
-C
OMe
COOMe
COOMe
6
7
2
5
9

Scheme 2. Temptative rationalization of the formation of lithium methyl formate (LiCOOMe, 7) by
treatment of dimethyl malonate with LHMDS.
oxirane oxygen-nucleophile
coordination
syn-1,4-addition product
+
O
(coordination product)
+
Li
-
OMe
route a
O
COOMe
O
O
OBn
O
BnO
BnO
HO
LiCOOMe
THF
a
1

O
3
1'
4
b

1
c
O
(methyl -glycosyl carboxylate)
route b
route c
+
-
MeO

Li
O
O
COOMe
BnO
HO
BnO
HO
4
COOMe
8
(anti-1,2-addition product)
(anti-1,4-addition product)
not obtained
not obtained
Scheme 3. Completely 1,4-regio- and syn-stereoselective addition of LiCOOMe to vinyl epoxide 1.
PhCH2CH(COOMe)2
+
PhCH2COOMe
11a
1 : 1
10a
+ CH2=CHCH2COOMe
CH2=CHCH2CH(COOMe)2
PhCH2Br (9a)
68%
10b 1.7 : 1
11b
CH2=CH-CH2Br
(9b)
79%
OMe
OLi
COOMe
COOMe
CH2
LHMDS
THF
CH2=C(Me)CH2Br (9c)
+ LiCOOMe
+
CH2=C(Me)CH2COOMe
CH2=C(Me)CH2CH(COOMe)2
HC
89%
7
COOMe
11c
2
10c
1.7 : 1
PhCOCl (9d)
45%
p-OMe-C6H4-COCH2Br
(9e)
PhCOCH(COOMe)2
+
PhCOCOOMe
50%
10d
1.2 : 1
11d
p-OMe-C6H4-COCH2CH(COOMe)2
+
p-OMe-C6H4-COCH2COOMe
11e
10e 1 : 1
10

Scheme 4. Products obtained by reaction of dimethyl malonate/LHMDS system with electrophiles (benzyl-,
allyl-, phenacyl bromide, benzoyl chloride).
Ph
COPh
COPh
COPh
O
LHMDS
THF
BnBr
20%
+
-
O Li
Li+
-
PhCH2
C
HC
COPh
Ph
COPh
14
12
16
18
(only substitution product)
no evidence
of formation
NMe2
CONMe2
CONMe2
13
CONMe2
PhCH2
O
LHMDS
THF
BnBr
99%
+
-
Li+
-
O Li
CONMe2
17
C
HC
CONMe2
NMe2
15
19
no evidence
of formation
(only sustitution product)
Scheme 5. Products obtained by reaction of dibenzoylmetane/LHMDS and N,N,N’,N’-
tetramethylmalondiamide/LHMDS systems with benzyl bromide.
11

Dimethyl malonate/LHMDS system as a new protocol for generating
methyl formate anion (–COOMe) in the condensed-phase
Valeria Di Bussolo,* Salvatore Princiotto, Elisa Martinelli, Vittorio Bordoni, Paolo Crotti*
Dipartimento di Farmacia, Università di Pisa, Via Bonanno 33, 56126 Pisa, Italy
Highlights
1. A new protocol for the formation of methyl formate anion in the condensed-phase is given.
2. The protocol is based on the treatment of dimethyl malonate with LHMDS.
3. A mechanism based on a double deprotonation of dimethyl malonate is proposed.
12

Dimethyl malonate/LHMDS system as a new protocol for
generating methyl formate anion (–COOMe) in the condensed-phase
Valeria Di Bussolo,* Salvatore Princiotto, Elisa Martinelli, Vittorio Bordoni,
Paolo Crotti*
Dipartimento di Farmacia, Università di Pisa, via Bonanno 33, 56126 Pisa, Italy
OMe
COOMe
COOMe
CH2
O
LHMDS
THF
RX
Li+
C
+
-
O
-
RCOOMe
+
+
CH-R
Li
HC
COOMe
OMe
COOMe
COOMe
RX = benzyl-, allyl-, phenacyl-bromide, benzoyl chloride
13