Alkylation of 3,4-dibromo-4-methyltetrahydropyran with diethyl malonate as a key to understanding the electronic nature of chemo- and regioselectivity of molecules

Article abstract of DOI:10.1007/s11178-005-0136-8

The reaction of 3,4-dibromo-4-methyltetrahydropyran with diethyl malonate in the presence of sodium butoxide leads to formation of the corresponding cross-coupling product rather than of tetraethyl ethane-1,1,2,2-tetracarboxylate (product of dehydrodimerization of diethyl malonate) which is formed in the presence of sodium ethoxide. An explanation was proposed, which may be regarded as a key to understanding the nature of the driving force for one- and two-electron transfer, as well as chemo- and regioselectivity of organic molecules. 2005 Pleiades Publishing, Inc.

Full text of DOI:10.1007/s11178-005-0136-8

Russian Journal of Organic Chemistry, Vol. 41, No. 1, 2005, pp. 134 -140. Translated from Zhurnal Organicheskoi Khimii, Vol. 41, No. 1, 2005,
pp. 134-140.
Original Russian Text Copyright Ó 2005 by Gevorkyan, Arakelyan, Dzhandzhulyan, Mikaelyan, Petrosyan, Panosyan.
Alkylation of 3,4-Dibromo-4-methyltetrahydropyran
with Diethyl Malonate as a Key to Understanding the Electronic
Nature of Chemo- and Regioselectivity of Molecules
A.A. Gevorkyan1, A. S.Arakelyan1, Zh. L. Dzhandzhulyan1, A. R. Mikaelyan1,
K.A. Petrosyan1, and G.A. Panosyan2
1
Institute of Organic Chemistry, National Academy of Sciences of Armenia, Kanakertsi 167A, Erevan, 375091 Armenia
e-mail: agevork@sci.am
Molecular Structure Research Center, National Academy of Sciences of Armenia
2
Received March 2, 2004
Abstract—The reaction of 3,4-dibromo-4-methyltetrahydropyran with diethyl malonate in the presence of sodium
butoxide leads to formation of the corresponding cross-coupling product rather than of tetraethyl ethane-1,1,2,2-
tetracarboxylate (product of dehydrodimerization of diethyl malonate) which is formed in the presence of sodium
ethoxide. An explanation was proposed, which may be regarded as a key to understanding the nature of the
driving force for one- and two-electron transfer, as well as chemo- and regioselectivity of organic molecules.
implies the charge on an atom in a reagent, estimated in
the chemical bond ionic character units, as a measure of
the ability of a C–nucleofuge bond to undergo heterolysis
and the degree of extension of the C–nucleofuge bond in
the substrate as a measure of regio- and stereochemistry
[16–18]. In addition, it is believed that heterolytic disso-
ciation of a C–nucleofuge bond is a three- rather than
two-step process, each step corresponding to one type of
ion pairs. These types include contact, loose, and solvent-
separated ion pairs (IPc, IPl, and IPss), the two latter pos-
sessing electrophilic properties [16]. Each type of ion pairs
is characterized by specific regio- and stereochemical
transformations, regardless of the type of chemical bond
and transformation mode.
It was recently shown that the dehydrobromination–
substitution pattern (Scheme 1, path 1: I g II g III),
which is typical of reactions of 3,4-dibromo-4-methyltetra-
hydropyran (I) with a series of nucleophiles (such as
amines, phenols, alcohols, organic acid salts, etc.) [1–3],
is not followed in reactions with enolates derived from
b-dicarbonyl compounds [4]. In the latter case, dehydro-
dimers IV of b-dicarbonyl compounds (path 2) and
dihydropyran V are formed instead of cross-coupling
products III. It may seem that we revealed just one more
example of radical nucleophilic substitution whose gen-
eral relations have long been known [5–10].
However, it becomes clear that this fact also contains
important information on both driving forces of radical
nucleophilic substitution itself and chemo- and
regiochemistry of molecules on the whole [11–13]. This
information can be revealed by considering the problem
in terms of the ion-pair version of mechanisms of organic
reactions, which distinctly differentiates the roles of re-
agent and substrate [14–20]. Without going into details
(for more information, see [15–18]), we should note that
such refinement of the behavior of a reagent and a sub-
strate rules out possible arbitrary treatment of nucleo-
philic or electrophilic properties of reagents [11–13].
Another specific feature of the above version is that it
Success of the approach is achieved due to the fact
that the most important source of information on the na-
ture of the driving force of a reaction is the product struc-
ture. This belief originates from the assumption that struc-
ture of a molecule contains so much important and reli-
able information on the properties of its precursors (ion
pairs and molecules as a whole) that cannot be obtained
by any other method for studying mechanisms of organic
reactions.
On the basis of the above concepts we were able to
reveal and solve many problems whose existence could
1070-4280/05/4101-0134Ó2005 Pleiades Publishing, Inc.

ALKYLATION OF 3,4-DIBROMO-4-METHYLTETRAHYDROPYRAN
135
not be surmised in terms of the modern organic chem-
istry principles [19, 20], especially in the given case, tak-
ing into account that dicarbonyl compounds VI under
analogous conditions (while being converted into sodium
enolate VII) give rise to normal alkylation products VIII.
touching one aspect of the problem, the others fall out
from the scope of attention, though just the latter should
be revised to understand the true cause for change in the
behavior of a molecule. Therefore, conclusions and pre-
dictions drawn on the basis of this concept are often con-
fusing rather than reliable. An illustrative example is the
suggestion to consider radical nucleophilic reactions to
be a specific kind of transformation without assigning them
to electrophilic or nucleophilic type, for supposedly there
is no clear differentiation between these terms. For the
same reason, rigorous classification of molecules into re-
agents and substrates [10] etc., seems to make no sense.
As a result, no answers were given so far to the follow-
ing questions: why some halogen derivatives react ac-
cording to the classical nucleophilic substitution scheme
Let us demonstrate once more the possibilities of the
proposed approach by revealing the reason for dehydro-
dimerization of sodium enolates by the action of dibromide
I as an example [4]. We start the discussion by noting
that since the discovery of nucleophilic radical substitu-
tion [5–10] some authors presumed classical nucleophilic
substitution to be a process including two steps of trans-
fer of one electron.
However, it became clear that such simplification of
the problem gives only an illusion of its solution. While
Scheme 1.
ꢁ$
1D<
5ꢅꢄ;
5ꢀ21D
,;
5&2&+ &ꢁ5ꢂ21D
5&2&+5ꢅꢄ&25
5&2&+ &25
5ꢅ
5
;
5ꢅ
;
9,
,;ꢀ,3
,;ꢀ,3
9,,,
9,,ꢀ$DꢁꢂE
1D<
ꢁ%
H
2
5
G
2
F
1D
+
5
+
D
1D
F
0H
0H
%U
%U
2
2
%U
5
%U
ꢃ$
ꢃ%
E
9,,ꢀ&DꢀꢂE
5&2&+&25
1D25ꢀ
5ꢅꢅꢄ ꢄ(W
5ꢃ,
2
,
2
,$ꢀ,3 ꢂ ꢂ'$&ꢃ,$
2
2
2
1D25ꢀ
0<ꢁ+<ꢂ
0H
%U
D
5
5
5
5
1D25ꢀ
5ꢀꢄ ꢄ%X
0H
ꢂ
%U
ꢁ&
0<ꢁ+<ꢂ
%U
E
2
5ꢃ,,
2
,9
0H
ꢀ$
2
0H
%U
0H
2
,ꢀ,3
<
%U
D
1D25ꢀ
2
0H
%U
2
0<ꢁ+<ꢂ
E
%U
,,
,,,Dꢁꢂ,,,E
9
ꢀ%
2
,ꢃ,3
HY = HNAlk2, H2NAlk; MY = MOAlk, MOAr, MOAc, M[CH(COR)2]; III, R = OEt (a), OBu (b); M = Na, K; X = Hlg;
IV, VII, R = OEt (a), Me (b).
RUSSIAN JOURNALOF ORGANIC CHEMISTRY Vol. 41 No. 1 2005

136
GEVORKYAN et al.
(paths 1 and 3) while reactions of others under the same
conditions involve formation of radical species (path 2)?
Which factors (and according to which scheme) are re-
sponsible for such difference in the properties of similar
molecules?
ticipation of co-reagent VII) been capable of playing this
role (reaction a), such reaction, as well as in the pres-
ence of co-reagents like MY (HY) (reaction b), would
lead to formation of monobromide II. This is not the case.
Compounds IV and V are formed instead. Their precur-
sors could be only radical species R-I and R-II, respec-
tively.Although this version is justified to a sufficient ex-
tent, it is not irreproachable as well, for it does not an-
swer the fundamental question of organic chemistry:
which forces and in which way induce homolytic disso-
ciation of the C–nucleofuge bond in one molecule but
heterolytic dissociation of the same bond in the others?
In order to elucidate the reason for the above ambigu-
ity and to answer the above questions, we focused on
some relevant data [14, 15, 21, 22]. The most important
of these are the following. First, the reason for elimina-
tion of hydrogen halide according to the Zaitsev rule is
not the conjugation effect but emergence of electrophilic
species [22]; second, elimination of a hydrogen atom
capable of departing as hydride ion from secondary
g-halo ethers cannot be prevented even in superbasic me-
dium [14].
We succeeded in answering this question by calling
our attention to the following two facts. First, radical spe-
cies R-I and R-II are also formed as a result of exten-
sion and rupture of the C–Br bond, although this scheme
requires smaller energy than does heterolytic dissocia-
tion. Second, electrophilic power of a species increases
in parallel with the degree of extension of the C–nucleo-
fuge bond. From the latter statement it follows that the
degrees of extension of the C–nucleofuge bond in
monohaloalkanes IX and the C–Br bond in dibromide I
should be different. The extension of the C–nucleofuge
bond in IX should be greater since dibromide I contains
two g-alkoxy substituents (one of which also possesses
a b-bromine atom) hampering heterolytic dissociation.
Therefore, ion pair derived from dibromide I should be
less electrophilic than that formed from haloalkanes IX.
We can state that ion pair IA-IPl undergoes one-electron
oxidation just because it is less electrophilic than ion pairs
derived from IX. Presumably, this is the reason why the
latter are capable of effecting two- rather than one-elec-
tron transfer (transfer of anionic ligand), which is inter-
preted as nucleophilic substitution.
Taking into account the above stated, let us to deter-
mine those properties of reagent and substrate, which
control one- and two-electron transfer, in particular the
formation of dimeric molecules IV. It is known that com-
pounds IV are also formed in the oxidation of enolates
derived from b-dicarbonyl compounds with iodine, silver
oxide, copper(II) trifluoromethanesulfonate, etc. [23]. The
specificity of the described reaction is that dibromide I
acts here as oxidant, while with other nucleophiles it re-
acts according to the dehydrobromination–substitution
pattern to afford products III. How can this occur? The
simplest answer is that the oxidant is not dibromide I it-
self but one of its electrophilic ion pairs, IA-IPl or IA-
IPss. Nevertheless, it cannot be regarded as an answer to
the above questions; this is only a description of the cause
of one-electron transfer in some molecules but not of the
cause of behavior of other related molecules which are
not involved in analogous reaction.
We succeeded in noticing that information necessary
for solution of this problem is contained in the mere fact
of alteration of the reaction direction upon replacement
of reagent of one type by another. This obviously indi-
cates that emergence of ion pairs responsible for product
formation involves both base (NaOR) and reagent mol-
ecule (reactions a and b). Formalistically, this resembles
a trimolecular process of dissociation of Swain’s
electrofuge–nucleofuge bond [24] occurring with partici-
pation of the solvent and reagent. Apart from similarity,
the proposed version is characterized by a radical differ-
ence. It does not imply stabilization of already developed
cationoid species by the solvent but participation of the
latter (or a co-reagent) just in the extension of the
electrofuge–nucleofuge bond in the substrate.
In fact, even g-haloalkyl ethers react mainly accord-
ing to the electrophilic substitution pattern (see above)
under conditions less favorable for heterolysis (in a con-
centrated solution of a base), whereas monohaloalkanes
IX which are more prone to heterolytic dissociation give
rise to electrophilic ion pairs under more favorable condi-
tions. This also follows from the fact that base-catalyzed
alkylation of b-dicarbonyl compounds with haloalkanes
IX is sometimes accompanied by formation of dialkylated
products in an amount comparable to the monoalkylated
product [25, 26].
Although it is difficult to disagree with the above stated,
we cannot believe it to answer the raised questions. The
reason is that the assumption involving formation of
IA-IPl ion pair with reduced electrophilicity contradicts
the Pauling principle which implies ionic character of
In fact, had the NaOR base alone (i.e., without par-
RUSSIAN JOURNALOF ORGANIC CHEMISTRY Vol. 41 No. 1 2005

ALKYLATION OF 3,4-DIBROMO-4-METHYLTETRAHYDROPYRAN
137
a chemical bond [27, 28]. Here, the main point is the fol-
lowing.Although the sodium atom in the O–Na fragment
possesses a greater positive charge (+2.51), it induces
lesser extension of the C–Br bond (interaction b) than
does the N–H proton which bears a charge of +0.84 (the
ionic orders of the O–Na and N–H bonds are 2.51 and
0.84, respectively). Estimation of atomic charges in chemi-
cal bond ionic character units (hereinafter, c.b.i.c.) can
be regarded as valid, for these values correlate with those
determined by electronic spectroscopy [28]. In addition,
a direct relation was revealed between the charges on
atoms in the reagent and degree of extension of the
C–nucleofuge bond in the substrate [17–20]. Therefore,
to draw a final conclusion on the nature of this anomaly it
was necessary to elucidate (1) whether such reduction in
atomic properties is possible or not and (2) if it is possible,
why it is observed for the enolate sodium atom of b-di-
carbonyl compounds and not for its unidentate analogs?
hydroxide in ethanol (in essence, by sodium hydroxide
and water) gave no desired result. Under these condi-
tions, dibromide I almost failed to react and was recov-
ered from the reaction mixture.
Seemingly, this failure indicates that the problem could
not be solved with the use of a base in which the counterion
is sodium. This also follows from published data accord-
ing to which the polarity of alcohols (and hence of the
corresponding alkoxides) decreases within a homologous
series [30]. By contrast, a radically different solution re-
sults from the new version of organic reaction mecha-
nisms [16–20]. In the framework of this version, the posi-
tive charge on the sodium atom in an alkoxide group and
therefore its ability to favor heterolytic dissociation in-
crease rather than decrease within a homologous series.
This information is usually revealed starting from deter-
mination of local polarity which, according to the elec-
tronegativities of atoms [17, 18, 27, 28], is induced by
hydrogen and oxygen atoms on each carbon atom of the
alkoxide group. It is assumed that the induced polarity
(charges on atoms are expressed in c.b.i.c. units) is then
transmitted along the atom chain toward the most elec-
tronegative atom of a functional group. This effect can
be estimated on a semiquantitative level if the algebraic
sum of charges induced on each atom in the chain is used
as a measure of polarity (Scheme 2). In order to simplify
the calculation, the loss in the polarity with increase in the
distance from a bond to the reaction center (functional
group) is neglected. Scheme 2 shows an appreciable in-
crease of the electron-donor power of the alkyl group in
the given alkoxide series (–0.16 < –0.86 < –2.26 for me-
thyl, ethyl, and butyl, respectively).As this polarity is trans-
mitted to the oxygen atom, we obtain a “refined” series
of ionic characters of the O–Na bonds (or charges on
the sodium atom in c.b.i.c. units; 1.335 < 1.685 < 2.385)
against the basis value equal to 2.51.
We have succeeded in answering these questions.
First, we have succeeded in findings arguments in sup-
port of the possibility of the presumed effect and, second,
we have succeeded in demonstrating that this effect re-
sults from the well-known intramolecular donor–accep-
tor interaction [8, 11–13] which in the given case occurs
between the sodium cation and carbonyl oxygen atom of
enolate VII (intramolecular interaction c).Analogous in-
fluence (reduction of the acceptor power of, e.g., alumi-
num atom in alanes) is observed even in going from
a solution in benzene to more basic diethyl ether [18, 29].
This effect is so strong that it leads to complete inversion
of stereoselectivity in the addition of alane at the carbo-
nyl group (88:12 against 17:83). It is assumed that analo-
gous reduction of the acceptor power of the enolate so-
dium atom precedes its contact with bromine electrons at
the C–Br bond (RONa···Br···NaOC=C). There are
published data on a wide variety of molecules which are
characterized by a similar solvent effect on the behavior
of other metal atoms [5–10, 26].
The application of such a simple (and inaccurate) pro-
cedure for estimation of the ionic character of a chemi-
cal bond and atomic charge is justified by the facts that
the obtained results are confirmed by other data [28] and
that they possess a predictive power [16–20]. Therefore,
we presumed that the difference between the acceptor
powers of sodium butoxide and sodium ethoxide (2.385
against 1.685 c.b.i.c. units) is too large to ensure
dehydrobromination–alkylation of dibromide I even with
b-dicarbonyl compounds.
The conclusion is obvious: the reason for one-electron
oxidation of enolates VII is not electrophilicity alone but
the reduced electrophilicity of ion pair IA due to low
degree of heterolysis of the C–Br bond in dibromide I. If
the proposed version correctly describes the nature of
driving forces governing the behavior of dibromide I, in-
crease in the ability of the medium to favor heterolytic
dissociation (e.g., by raising its polarity) should change
the reaction direction (chemoselectivity). However, con-
trary to expectations, replacement of ethanolic sodium
ethoxide by a more polar reagent, a solution of sodium
These theoretical predictions turned out to be valid. In
the reaction of dibromide I with diethyl malonate in
a solution of sodium butoxide in butyl alcohol we observed
RUSSIAN JOURNALOF ORGANIC CHEMISTRY Vol. 41 No. 1 2005

138
GEVORKYAN et al.
Scheme 2. Estimation of the acceptor power (charge, in c.b.i.c. units) of the sodium atom
in sodium methoxide, ethoxide, and butoxide
F/ Fꢉꢁꢎꢎ FꢀꢁꢄꢉF Fꢅꢁꢆꢇ
E/ Fꢉꢁꢎꢎ FꢅꢁꢆꢆF ꢀꢀꢁꢂꢄ
D/ FꢅꢁꢆꢆF FꢅꢁꢅꢀF Fꢀꢁꢉꢆ
+
+
+
ꢁD ꢁD
ꢁD
ꢀꢁꢂꢄ
E
ꢀꢁꢂꢄ
E
ꢀꢁꢂꢄ
E
ꢅꢁꢆꢇ
+
&
ꢀD
0H
+
&
+
+
&
ꢀD
ꢅꢁꢆꢇ
F
ꢅꢁꢆꢇ
F
D
&
2
1D
2
1D
0H
&
2
1D
F
ꢊꢇꢁꢃꢂꢆ
ꢅꢁꢃꢈ ꢊFꢇꢁꢉꢉꢆ
ꢉꢁꢉꢈ
ꢎꢁꢈꢈ ꢊFꢅꢁꢉꢂꢆ
GF F ꢀꢁꢇꢃ
GF F ꢀꢁꢂꢃ
GF F ꢅꢁꢅꢃ
GF FꢆFFꢋ ꢀꢁꢉꢆꢌFꢊFꢀꢁꢂꢄF F ꢀꢁꢂꢃ
HF Fꢋ ꢅꢁꢆꢇF FꢀꢁꢂꢃꢌF F ꢉꢁꢉꢈ
IF F ꢉꢁꢉꢈFꢍFꢅF F ꢇꢁꢃꢂꢆ
GF FꢄFFꢋ ꢀꢁꢉꢆꢌFꢊFꢀꢁꢂꢄF F ꢅꢁꢅꢃ
HF Fꢋ ꢅꢁꢆꢇF FꢅꢁꢅꢃꢌF F ꢎꢁꢈꢈ
IF F ꢎꢁꢈꢈFꢍFꢅF F ꢅꢁꢉꢂꢆ
GF FꢉFFꢋ ꢀꢁꢉꢆꢌFꢊFꢀꢁꢂꢄF F ꢀꢁꢇꢃ
HF Fꢋ ꢅꢁꢆꢇF FꢀꢁꢇꢃꢌF ꢀ ꢅꢁꢃꢈ
IF F ꢅꢁꢃꢈFꢍFꢅF F ꢇꢁꢉꢉꢆ
a, b, and c are the initial ionic characters of bonds and charges (in c.b.i.c. units) estimated from the tabulated electronegativities
of the hydrogen (2.20), carbon (2.55), and oxygen atoms (3.44) [27, 28]; d is the total charge induced on the carbon atom of
the C–O bond by the oxygen atom and hydrogen atoms of the entire alkyl group; e is the total charge induced by the alkyl
group and sodium atom on the oxygen atom; and f are the average (refined) charges on the sodium atom in alkoxides.
formation of cross coupling products as a mixture of di-
ethyl and dibutyl malonates IIIa and IIIb at a ratio of
1 : 9 with an overall yield of ~90%.
when a reaction mixture consisting of ethanol, sodium
ethoxide, and enolate derived from a dicarbonyl compound
gives rise to an electrophilic ion pair, the latter necessar-
ily attacks the alkoxide oxygen atom (i.e., oxygen atom in
the C–O–Na fragment). None of the other atoms in mol-
ecules present in the reaction medium, including the
enolate oxygen atom (see above) possesses such elec-
tron density. If a reagent is capable of overcoming
counteraction of substrate atoms (i.e., of the sodium and
C–O–Na carbon atoms in sodium ethoxide), a new chemi-
cal bond is formed (in our case, with the oxygen atom).
Only when electron transfer from the oxygen atom to
an atom in the reagent becomes impossible, either de-
composition of the complex occurs or the reaction center
in the reagent undergoes reorientation (certainly, the af-
finity of the reagent weakens upon complex formation)
toward other electron donors. The direction of attack by
the “weakened” reagent is determined by decreasing
power of electron donor. Therefore, the next electron-
donor center to which the attack by the reagent is di-
rected (interaction e) is p electrons of the enolate oxygen
atom (C=C–O–Na). In other words, the contact of the
reagent with electron-rich reaction center is completed
not by automatic replacement of the metal by electrophile
but only by formation of a donor–acceptor complex like
DAC-IA, as was shown with IA-IPl as an example. If
this interaction is not completed by electron transfer as
well, the electrophilic center is reoriented again (predomi-
nant interaction f in DAC-IA) toward energetically more
accessible electrons of the neighboring group, in our case
toward ð electrons of the C=C bond; a new stage of
comparison of the reagent and substrate powers begins
until charge deficit on the reagent be eliminated via for-
Finally, let us consider driving forces which control
chemoselectivity in a reaction medium consisting of nu-
merous molecules and ion pairs. The problem is that trans-
formations of these species follow paths contradicting
traditional concepts of organic chemistry. For example,
what is the reason for the fact that the reaction in a me-
dium consisting of an alcoholic solution of a haloalkane,
b-dicarbonyl compound sodium salt, and sodium alkoxide
involves enolate derived from the b-dicarbonyl compound
rather than alkoxide? This is the riddle which has not been
solved so far.At best, attempts were made to reveal some
relations for particular classes of molecules but not for
the phenomenon as a whole [31].
Despite the apparent complexity of the problem, it can
be solved fairly simply on the basis of the above stated
principles. In this respect, the most important information
is obtained from the statement that a chemical reaction
occurs only when the affinity of a reagent exceeds the
force preventing rupture of the electrofuge–nucleofuge
bond in a substrate [32]. Otherwise, the interaction stops
at the stage of formation of a metastable donor–acceptor
complex like DAC-IA which occurs in equilibrium with
the initial reactants. Analysis of numerous data [5–10,
23–26] (including those given above) led us to the follow-
ing simple rules. Let us consider these rules as applied to
reactions where the reagent is an electrophile.
First, reactions of an electrophile (reagent), regard-
less of its power, always begin with attack on an electron
donor possessing the highest electron density. Therefore,
RUSSIAN JOURNALOF ORGANIC CHEMISTRY Vol. 41 No. 1 2005

ALKYLATION OF 3,4-DIBROMO-4-METHYLTETRAHYDROPYRAN
139
mation of a stable molecule. In the given series of elec-
tron donors, the greater ability of ð electrons of the C=C
bond to undergo cleavage results from the fact that the
ionic character of this bond is entirely determined by
a small difference in the electronic properties of its
atoms due to effect of substituents.
ethanol was distiled off, and the residue was cooled, neu-
tralized with hydrochloric acid, and extracted with diethyl
ether. The extract was dried over Na2SO4, the solvent
was distilled off, and the product was distilled under re-
duced pressure. Yield 4.2 g (70.5%), bp 125–133°C
(1 mm). The product solidified on storage to form color-
1
less crystals with mp 74–75°C. H NMR spectrum, d,
Nevertheless, difficulties appear even in the transfer
of such electrons of substrate to reagent.As shown above,
these difficulties arise in the case of weak but not strong
electrophiles. For that reason, the power of ion pairs de-
rived from haloalkanes IX is sufficient to ensure two-
electron transfer and formation of a new bond with par-
ticipation of both oxygen atom (O-alkylation products)
and carbon atom of the enolate moiety. This factor is also
responsible for the fact that O-alkylation occurs just with
those electrophiles in which the maximal extenstion of
the C–nucleofuge bond is achieved [11–13, 25, 26]. For
example, acid chlorides, a-chloro ethers, chlorosilanes,
and other related reagents, in which the electrofuge–
nucleofuge bond undergoes heterolytic dissociation
especialy readily, give rise mainly (or almost exclusively)
to O-alkylation products. This also follows from the data
on the regiochemistry of molecules affording radical nu-
cleophilic substitution reaction [5–10]. For instance, the
alkylation of 2-nitropropane sodium salt with p-nitrobenzyl
chloride gives C- and O-alkylation products (92 and 6%,
respectively), while an analogous reaction with benzyl
chlorides having no electron-acceptor nitro group leads
to almost exclusive formation of O-alkylation products
[33].
ppm: 1.27 t (12H, J = 7.1 Hz), 3.93 s (2H), 4.17 q (8H).
13C NMR spectrum, dC, ppm: 13.4 (Me), 50.6 (CH),
60.9 (CH2), 165.8 (C=O). The product was identical to
a known sample [34]. In addition, from the low-boiling
fraction we isolated 4-methyl-3,6-dihydropyran, bp 114–
116°C (680 mm).
3,4-Diacetyl-2,5-hexanedione (IVb). The reaction
was carried out following the above procedure with the
use of 1.85 g (0.08 mol) of sodium, 8 g (0.08 mol) of
acetylacetone, and 5.2 g (0.02 mol) of 3,4-dibromo-4-
methyltetrahydropyran. Yield 2 g (50.5%). bp 120–122°C
(18 mm). The product solidified to give colorless crystals
with mp 192–193°C. 1H NMR spectrum, d, ppm: 1.99 s
(12H), 12.74 s (2H). 13C NMR spectrum, dC, ppm: 22.81
(Me), 191.66 (C=O). The product was identical to that
described in [35].
3-Bis(ethoxycarbonyl)methyl- and 3-bis(butoxy-
carbonyl)methyl-4-methyl-3,6-dihydro-2H-pyrans
IIIa and IIIb.Asolution of sodium butoxide in 1-butanol
was prepared from 40 ml of 1-butanol and 3.68 g
(0.16 mol) of metallic sodium at 100°C. The solution was
cooled to 50–60°C, and 13 g (0.081 mol) of diethyl mal-
onate was addede. The mixture was stirred for 15 min,
heated to 70–75°C (the mixture became homogeneous),
and 10.4 g (0.04 mol) of 3,4-dibromo-4-methyltetra-
hydropyran was added dropwise over a period of 25 min.
The mixture was stirred for 5 h at that temperature, cooled,
neutralized with hydrochloric acid, and extracted with
ether. The extracts were washed with water and dried
over sodium sulfate, the solvent was removed, and the
residue was distilled under reduced pressure, a fraction
boiling in the temperature range from 160 to 175°C
(3 mm) being collected (about 16 g). Repeated fractional
distillation of that fraction gave 11.1 g (88.9%) of a prod-
uct with bp 168–175°C (3 mm) which, according to the
1H NMR data, was a mixture of 3-bis(ethoxy-
carbonyl)methyl- and 3-bis(butoxycarbonyl)methyl-4-
methyl-3,6-dihydro-2H-pyrans IIIa and IIIb at a ratio of
16:84.
The general applicability of the proposed approach and
its use in the solution of other theoretical and synthetic
problems will be reported elsewhere.
EXPERIMENTAL
The IR spectra were recorded on a Specord 75IR
spectrophotometer in mineral oil. The 1H and 13C NMR
spectra were obtained on a Varian Mercury-300 spec-
trometer (300 MHz for 1H) from solutions in DMSO-d6–
CCl4 (1:3) using TMS as internal reference.
1,1,2,2-Tetrakis(ethoxycarbonyl)ethane (IVa).
Diethyl malonate, 13 g (0.081 mol), was gradually added
to a solution of sodium methoxide prepared by standard
procedure from 15 ml of anhydrous ethanol and 1.85 g
(0.08 mol) of metallic sodium. The mixture was stirred
for 15 min and heated to 70–75°C, and 5.2 g (0.02 mol)
of 3,4-dibromo-4-methyltetrahydropyran was added over
a period of 1.5 h to the resulting suspension. The mixture
was stirred for 5 h at that temperature, about 10 ml of
1
Ester IIIa. H NMR spectrum, d, ppm: 0.94 t and
0.95 t (6H, CH3CH2, J = 7.3 Hz), 1.31 m and 1.45 m
(4H, CH3CH2, CH2), 1.54–1.66 m (4H, CH3CH2, CH2),
RUSSIAN JOURNALOF ORGANIC CHEMISTRY Vol. 41 No. 1 2005

140
GEVORKYAN et al.
13. Dneprovskii, A.S. and Temnikova, T.I., Teoreticheskie
osnovy organicheskoi khimii (Theoretical Foundations of
Organic Chemistry), Moscow: Khimiya, 1979.
14. Gevorkyan, A.A., Arakelyan, A.S., and Kosyan, S.M., Zh.
Org. Khim., 1987, vol. 23, p. 1660.
15. Gevorkyan,A.A.,Arakelyan,A.S., and Cockerill,A.F., Tet-
rahedron, 1977, vol. 53, p. 7947.
16. Gevorkyan, A.A., Arakelyan, A.S., Petrosyan, K.A., and
Margaryan, A.Kh., Russ. J. Gen. Chem., 2001, vol. 71,
p. 729.
1.69 q (3H, =CCH3, J = 1.9 Hz), 2.55 br.s (1H, =CCH),
3.51 d (1H, O=C–CH–C=O, J = 6.0 Hz), 3.84–4.17 m
(8H, OCH2), 5.48 br.s (1H, =CH). 13C NMR spectrum,
dC, ppm: 13.1 (CH3); 21.3 (=CCH3); 18.4, 29.8, and 29.9
(CH2, ester); 64.1 and 64.1 (OCH2, heteroring); 64.7 and
65.3 (OCH2, ester); 123.0 (=CH); 130.9 (MeC=C); 167.0
and 167.9 (C=O).
1
Ester IIIb. H NMR spectrum, d, ppm: 1.24 t and
1.27 t (6H, CH3CH2, J = 7.1 Hz), 1.69 q (3H, =CCH3,
J = 1.9 Hz), 3.51 br.s (1H, =CCH), 3.55 d (1H,
O=CCHC=O, J = 3.4 Hz), 3.84–4.17 m (8H, OCH2),
5.48 br.s (1H, =CH).
17. Gevorkyan,A.A.,Arakelyan,A.S., and Petrosyan, K.A., Russ.
J. Gen. Chem., 2002, vol. 72, p. 767.
18. Gevorkyan,A.A.,Arakelyan,A.S., and Petrosyan, K.A., Arm.
Khim. Zh., 2003, vol. 56, p. 136.
REFERENCES
19. Gevorkyan, A.A., Arakelyan, A.S., Obosyan, N.G., and
Esayan, V.A., Russ. J. Org. Chem., 2000, vol. 36, p. 293.
20. Gevorkyan, A.A., Arakelyan, A.S., Esayan, V.A., and
Petrosyan, K.A., Russ. J. Gen. Chem., 2000, vol. 70, p. 1654.
21. Krebbenhofft, O., J. Org. Chem., 1979, vol. 85, p. 4285.
22. Gevorkyan,A.A., Arakelyan,A.S., and Kosyan, S.M., Arm.
Khim. Zh., 1988, vol. 41, p. 163.
23. Schmittel, M., Burghart, A., Malisch, W., Reising, J., and
Sollner, R., J. Org. Chem., 1998, vol. 63, p. 396.
24. Swain, C.G., J. Am. Chem. Soc., 1948, vol. 70, p. 119.
25. Cope,A.C., Holmes, H.L., and House, H.O., Organic Reac-
tions,Adams, R., Ed., NewYork: Wiley, 1957, p. 104.
26. Reutov, O.A. and Kurts, A.A., Usp. Khim., 1977, vol. 46,
p. 1964.
1. Gevorkyan, A.A., Kazaryan, P.I., Khizantsyan, N.M.,
Arakelyan, A.S., and Panosyan, G.A., Khim. Geterotsikl.
Soedin., 1981, no. 8, p. 1025.
2. Gevorkyan, A.A., Arakelyan, A.S., Avakyan, O.V., and
Obosyan, N.G., Russ. J. Org. Chem., 1998, vol. 34, p. 289.
3. Murza, M.M., Safina, D.M., and Kuvatov, Z.Kh., Khim.
Geterotsikl. Soedin., 1993, no. 12, p. 1628.
4. Gevorkyan, A.A., Arakelyan, A.S., Barsegyan, S.V.,
Petrosyan, K.A., and Panosyan, G.A., Russ. J. Org. Chem.,
2003, vol. 39, p. 1204.
5. Beletskaya, I.P. and Drozd, V.N., Usp. Khim., 1979, vol. 48,
p. 793.
6. Rossi, R.A. and de Rossi, R.H., Aromatic Substitution by
the SRN1 Mechanism, Washington, DC: Am. Chem. Soc.,
1983.
27. Pauling, L. and Pauling, P., Chemistry, San Francisco:
W.H. Freeman, 1975.
28. Huheey, J.E., Inorganic Chemistry: Principles of Structure
and Reactivity, New York: Harper and Row, 1983, 3rd ed.
29. Ashby, E.C. and Laemmle, J.T., Chem. Rev., 1975, vol. 75,
p. 521.
30. Osipov, O.A. and Minkin, V.I., Spravochnik po dipol’nym
momentam (Dipole Moment Handbook), Moscow: Vys-
shaya Shkola, 1965.
31. Archer, E.A., Gong Hegui, and Krische, M.J., Tetrahedron,
2001, vol. 57, p. 1139.
32. Gevorkyan, A.A., Arakelyan, A.S., and Petrosyan, K.A.,
Abstract Book, Enicolopov’s Readings, Yerevan, 2003,
p. 25.
33. Kornblum, N., Michel, R.E., and Kerber, R.C., J. Am. Chem.
Soc., 1966, vol. 88, p. 5662.
34. Weygand, C., Organisch-chemische Experimentierkunst,
Leipzig, Johann Ambrosius Barth, 1938. Translated under
the title Metody eksperimenta v organicheskoi khimii,
Moscow: Khimiya, 1952, vol. 2, p. 382.
7. Mekhanizmy geteroliticheskikh reaktsii (Mechanisms of
Heterolytic Reactions), Moscow: Nauka, 1976, vol. 8, p. 23.
8. Reutov, O.A., Beletskaya, I.P., and Sokolov, V.I., Mekhanizmy
reaktsii metalloorganicheskikh soedinenii (Mechanisms
of Reactions of Organometallic Compounds), Moscow:
Khimiya, 1972.
9. Todres, Z.V., Ion-radikaly v organicheskom sinteze (Radi-
cal Ions in Organic Synthesis), Moscow: Khimiya, 1986.
10. Okhlobystin, O.Yu., Perenos elektrona v organicheskikh
reaktsiyakh (Electron Transfer in Organic Reactions),
Rostov-on-Don: Rostov. Gos. Univ., 1974.
11. March, J., Advanced Organic Chemistry: Reactions, Mecha-
nisms, and Structure, New York: Wiley, 1985. Translated
under the title Organicheskaya khimiya, Moscow: Mir,
1987, vol. 1, p. 337; March, J., Advanced Organic Chemis-
try: Reactions, Mechanisms, and Structure, New York:
Wiley, 1994, 4th ed., chap. 8.
..
12. Becker, H., Einfuhrung in die Elektronentheorie
organisch–chemischer Reaktionen, Berlin: Wissen-
schaften, 1974, 3rd ed.
35. Organic Syntheses, Rabjohn, N., Ed., NewYork: Wiley, 1966,
collect. vol. 4, p. 869.
RUSSIAN JOURNALOF ORGANIC CHEMISTRY Vol. 41 No. 1 2005