Condensation of 2-pyridylmethyllithium nucleophiles and pyridine electrophiles as a convenient synthetic route to polydentate chelating N-donor ligands

Article abstract of DOI:10.1023/A:1016023103898

Condensation of 2-pyridylmethyllithium or (6-methyl-2-pyridyl)methyllithium nucleophiles and pyridine, 2-picoline, or 4-tert-butylpyridine as electrophiles leads to new polydentate N-donor ligands, methyl-, tert-butyl-substituted tripyridinedimethanes, or to tripyridinedimethane itself, in good or high yields. Depending on the reagent ratio, solvent used, and reaction conditions, the corresponding intermediate dipyridinemethanes can be minor by-product or major products of the condensation. In contrast to 2-pyridylmethyllithium, lithiated 2-isopropylpyridine does not react with pyridine electrophiles. Vice versa, nucleophilic substitution at the C(2)-pyridine carbon of 2,2-bis(2-pyridyl)propane with 2-pyridylmethyllithium takes place to produce products of condensation of 2-isopropylpyridine and dipyridylmethyllithium. DFT calculations of the Gibbs free energies of reactions combined with pK_a values of the CH-acids involved help to explain the reactivity observed.

Full text of DOI:10.1023/A:1016023103898

Chemistry of Heterocyclic Compounds, Vol. 38, No. 4, 2002
CONDENSATION OF 2-PYRIDYL-
METHYLLITHIUM NUCLEOPHILES
AND PYRIDINE ELECTROPHILES AS A
CONVENIENT SYNTHETIC ROUTE TO
POLYDENTATE CHELATING N-DONOR LIGANDS*
1
1
1
2
1
N. Vedernikov , R. Miftakhov , S. V. Borisoglebski , K. G. Caulton , B. N. Solomonov
Condensation of 2-pyridylmethyllithium or (6-methyl-2-pyridyl)methyllithium nucleophiles and
pyridine, 2-picoline, or 4-tert-butylpyridine as electrophiles leads to new polydentate N-donor ligands,
methyl-, tert-butyl-substituted tripyridinedimethanes, or to tripyridinedimethane itself, in good or high
yields. Depending on the reagent ratio, solvent used, and reaction conditions, the corresponding
intermediate dipyridinemethanes can be minor by-product or major products of the condensation. In
contrast to 2-pyridylmethyllithium, lithiated 2-isopropylpyridine does not react with pyridine
electrophiles. Vice versa, nucleophilic substitution at the C(2)-pyridine carbon of 2,2-bis(2-
pyridyl)propane with 2-pyridylmethyllithium takes place to produce products of condensation of
2
-isopropylpyridine and dipyridylmethyllithium. DFT calculations of the Gibbs free energies of
reactions combined with pK values of the CH-acids involved help to explain the reactivity observed.
a
Keywords: 4-tert-butylpyridine, dipyridinemethanes, 2,6-lutidine, (6-methyl-2-pyridyl)methyllithium,
2
-picoline, pyridine, 2-pyridylmethyllithium, tripyridinedimethanes, C–C bond cleavage, DFT
calculations, condensation, Gibbs free energy of reaction, nucleophilic substitution.
Ligands can play an extremely important role in determining the "face" of a transition metal atom in its
complexes. So, chelating tridentate N-donor ligands and, first of all, facially chelating ones, turned out to be
6
very effective in stabilizing alkylhydrido complexes of d platinum metals [1-4]. Azaheterocycles are especially
useful in this respect, in particular, owing to their resistance to oxidation and low basicity. Polypyridines [5] and
derivatives of tris(2-pyridyl)methane [6] are the best known polydentate chelating ligands derived purely from
pyridine, which have only limited coordination abilities and form correspondingly mer- (tri- and higher
polypyridines) or fac(tris(2-pyridyl)methane ligands) chelates. Poly(2,6-pyridine)polymethanes (Scheme 1),
which belong to other group of polypyridine ligands, are conformationally more flexible and therefore can be
expected to be more variable in their coordination behavior. Unfortunately, till now they were much less
accessible and rarely used ligands [7-13]. Here we report a convenient synthetic route to (substituted)
tripyridinedimethanes and dipyridinemethanes, which can lead to extending the chemistry of this interesting
_
*
_
______
Dedicated to M. G. Voronkov.
_________________________________________________________________________________________
1
2
Kazan State University, Kazan 420008, Russia; e-mail: ave@ksu.ru. Department of Chemistry, Indiana
University, Bloomington, IN 47405, USA; e-mail: caulton@indiana.edu. Translated from Khimiya
Geterotsiklicheskikh Soedinenii, No. 4, pp. 471-481, April, 2002. Original article submitted July 6, 2001.
4
06
0009-3122/02/3804-0406$27.00©2002 Plenum Publishing Corporation

class of azaheteropolycycles. The synthetic method suggested in this work is based on a version of the
Chichibabin reaction, which implies hydride nucleophilic substitution at the most electrophilic carbon atoms in
position 2 or 6 of the pyridine ring with organyllithiums as nucleophiles and leads to 2-alkyl- or aryl- or
2
,6-dialkyl- or diaryl-substituted pyridines [13]:
+
+
LiH
H2
a
LiCHRR1
CHRR1
N
N
b
N
R1
R
(1)
Li
LiCHRR1
(
2)
1
1
N
CHRR1
R RHC
N
CHRR
It is reasonable to expect that in the case of 2-pyridylmethyllithium (LiPic) or its analogs chosen as
1
nucleophilic reagents (LiCHRR ), this condensation could lead to dipyridinemethanes (Eq. (1), LiPic instead of
1
LiCHRR ) and further to tripyridinedimethanes (Eq. (2)). Indeed, some examples of such condensation of LiPic
as a nucleophile and pyridine as electrophile leading to dipyridinemethane with 37% yield (DPM, Scheme 1)
[
14] or with 2-picoline leading to 6-methyl-2-pyridylpyridine (Me-DPM, Scheme 1) and 2-(6-methyl-2-
pyridylmethyl)-6-(2-pyridylmethyl)pyridine (Me-TPDM, Scheme 1) [15] are known. No systematic study
devoted to synthesis of ligands of the mentioned class has been reported yet.
Scheme 1
Representatives of dipyridylmethanes (dipyridinemethanes) and di(pyridylmethyl)pyridines
(
tripyridinedimethanes) synthesized
N
N
N
N
N
N
N
N
DPM 1
6-Me-DPM 2
6,6'-Me -DPM 3
2
N
N
N
N
6
-Me-TPDM 5
TPDM 4
t-Bu
t-Bu
N
N
N
N
N
N
6
,6'-Me -TPDM 6
4,4'-t-Bu -TPDM 7
2
2
In this work condensation of pyridine and its easily available derivatives, 2-picoline, 4-tert-
butylpyridine, and 2,6-lutidine, leading to seven poly-(2,6)-pyridinemethanes (Scheme 1) and three
corresponding intermediate dipyridinemethanes has been studied. To simplify their notation we will use the
4
07

further abbreviations DPM for dipyridinemethane with substituent symbols indicated just before this
abbreviation and the notation TPDM for tripyridinedimethane also with foregoing indication of symbols of
substituents at two terminal pyridine rings.
RESULTS AND DISCUSSION
Dipyridylmethanes are formed according to Eq. (3):
2
0–120 °C
(
3)
Li+
48–2 h
R1
R1
R2
N
N
R2
N
N
1
1
2
1
2
1
3
R
= R2 = H; 2 R = H, R = Me;
R = R = Me
We have found that TPDM and compounds derived from it by substitution of hydrogen atoms in
positions 6 or 4 of the two terminal pyridine rings with alkyl radical can be obtained in good or high yields
according to Eqs. (4) and (5):
1
10–160 °C
+
(4)
22
0 min – 6 h
R
1
Li
1
R
N
N
N
N
N
R
R
4
R = R1 = H; 5 R1 = H, R = R2 = Me
R2
R2
R2
1
10–160 °C
+
Li
(5)
2
–8 h
Li
R1
R1
N
N
R1
N
N
N
R1 = Me, R2 = H; 7 R1 = H, R = t-Bu
2
6
Reactions (4) and (5) proceed under significantly more severe conditions than those for reaction (3).
Both lithium hydride precipitation and dihydrogen evolution occur in the course of these condensations.
According to Eqs. (4), (5), the central pyridine ring of TPDM can originate from the electrophile, which
undergoes one (picoline) or two consecutive (pyridine) attacks of the organolithium compound (Eq. (4)) or from
the nucleophile (2,6-dilithiolutidine, Eq. (5)), which also serves twice in this role.
Lithiated picoline, LiPic (pK (PicH) 34), has been obtained by treatment of the corresponding CH-acid
a
with butyllithium in either diethyl ether or glyme solution. The preparation of dilithiated lutidine turned out to
be less convenient than the preparation of the monolithiated compound due to the incompleteness of
butyllithium protodemetallation. We have shown that dilithiated lutidine can be efficiently replaced by two
moles of LiLut. In all cases the ether solvent has been removed and the residual organolithium compound dried
in high vacuum. Nevertheless, ether coordinated to lithium atoms remained in this solid and was liberated only
in the course of its treatment with the electrophile–toluene mixture. The residual ether was then removed when
the resulting reaction mixture was heated. Depending on the solvent used for the lithiation, a variable amount of
the picoline to be metallated was transformed into the 6-butyl derivative in the course of a side reaction
according to Eq. (1), being higher in ether solution and negligible in glyme. Lutidine metallation was an almost
clean reaction. Ether bonded to organolithium nucleophile affected its reactivity, which was significantly higher
in the case of diethylether solvates. As a result, glyme-solvated organolithiums turned out to be excellent in the
synthesis of DPM, minimizing the formation of TPDM.
4
08

Table 1 includes results on the synthesis of DPM, Me-DPM, and Me -DPM under various conditions.
2
The mildest conditions, room temperature, have been used in the case of DPM. In order to increase the rate of
DPM formation, a 3-fold excess of pyridine was used (entry 1). The same reaction can be run with higher yield
at increased temperature (entry 2). For a 1:1 reagent ratio and the same reaction conditions as in the entry (2),
the yield of DPM is lower (entry 3). The most effective use of pyridine can be achieved if 2 moles of LiPic
prepared in glyme are used per one mole of pyridine (entry 4) with the main by-product being 2,2'-bipyridine.
The yield of DPM is significantly higher as compared with that reported earlier [14].
Following the same procedure (the use of LiPic prepared in glyme and either a 3-fold excess of PicH or
a 2-fold excess of the nucleophile), satisfactory yields of Me-DPM on either LiPic or PicH correspondingly
were obtained (entries 5, 6).
The main by-product observed in the case of the synthesis of Me
2
-DPM was Me-DPM. Its formation can
be explained by the proton transfer from the more acidic picoline to the anion of the less acidic lutidine:
+
+
Li
Li
N
N
N
N
leading to LiPic with its subsequent condensation with free picoline.
TABLE 1. Results of Condensation of Pyridine Electrophiles with
2
-Pyridylmethyllithium Nucleophiles (Eq. (3), (4))
Reagents
Reaction
NMR yield/
Entry
Compound
Reagents*
LiPic / Py
LiPic / Py
LiPic / Py
LiPic / Py
LiPic / PicH
LiPic / PicH
ratio*2
t, °C
time, h isolated yield, %
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
1
1
1
1
2
2
3
3
3
3
4
5
6
6
7
1* / 3
20
12
4
4
2
48
2
3
6
6
25 / 16
41 / 30
33 / 25
65 / 45
41 / 30
50 / 40
32 / 20
44 / 35
55 / 45
56 / 45
63 / 50
60 / 50
80 / 65
77 / 65
65 / 50
1* / 3
110
110
120
20
125
160
160
160
120
110
170
160
145
115
1* / 1
2 / 1*3
1* / 3*
2 / 1*3
3
LiLut / LutH / PicH 1* / 1 / 1
LiLut / LutH / PicH 1* / 2 / 1
LiLut / LutH / PicH 1* / 4.5 / 1
LiLut / LutH / PicH 1* / 2 / 1 *4
1
8
1
1
1
1
1
LiLut / Py
LiPic / PicH
2 / 1*
2 / 1*
—*5
8
LiLut / LutH / PicH 2 / 0 / 1*
LiLut / LutH / PicH 2 / 0 / 1*
8
12
5
LiPic / t-BuPy
2 / 1*
_
*
______
Organolithium compounds were obtained by lithiation of the appropriate
methylpyridine with butyllithium in diethyl ether solution; Pic =
;
N
CH2
Lut =
, t-BuPy = 4-tert-butylpyridine.
Me
N
CH2
2
*
Asterisk indicates compound on which yield is calculated.
3
*
*
*
An organolithium compound was obtained in 1,2-dimethoxyethane solution.
No toluene as co-solvent was used.
4
5
Includes 20 min heating at 110°C for ether distillation for reactions
conducted at tempe-ratures higher than 20°C.
4
09

To shift the position of this equilibrium to the left, free lutidine was introduced into the reaction
mixtures. Increase of the LutH/LiLut ratio from 1 to 2-4.5 allowed us to obtain Me -DPM of higher purity and
in satisfactory yield (entries 7-9). Milder reaction conditions can be reached if no toluene as co-solvent is used
entry 10).
2
(
Thus, we can conclude that the simplest dipyridinemethanes can be obtained in satisfactory yields under
mild conditions with a properly chosen reagent ratio.
The greater part of the TPDM ligands listed in Table 1 can be obtained via either Eqs. (4) or (5). Results
optimized for one of the schemes possible for each of the TPDM compounds in Scheme 1 are given in Table 1.
All of the tripyridinedimethanes were synthesized using organolithium compounds prepared in diethyl ether.
When glyme was used instead of diethyl ether only small amounts of TPDM's were obtained with the major
products being the corresponding DPM's. Tripyridinedimethane itself can be synthesized in good yield under
milder conditions (entry 11) than those for Me-TPDM, in accordance with the higher electrophilicity of pyridine
as compared with that of picoline (entry 12). According to data in the entries 13-15 Me
t-Bu -TPDM, ligands were obtained in good or high yields under conditions similar to those for Me-TPDM.
Thus, reactions (1), (2) can be a general route for synthesis of dipyridinemethanes and
2
-TPDM and
2
tripyridinedimethanes from the correspondingly substituted simplest pyridines. In order to learn more about the
scope and limitations of this method, we have studied a more complicated nucleophile, 2-pyridyl-2-
propyllithium, and a more complicated electrophile, 2,2-bis(2-pyridyl)propane (DPP). The latter was obtained
by one-pot four-step synthesis by two stepwise deprotonations of DPM with butyllithium in glyme solution and
methylation of each carbanion formed with methyl iodide. It turned out that lithiated isopropylpyridine, which
can be effectively prepared in glyme but not in ether solution, does not react with isopropylpyridine even in 2 h
at 190°C:
o
1
90 C
+
no reaction
Li
2
h
Moreover, treatment of DPP with LiPic at ambient temperature overnight leads to complete conversion
of DPP with destruction of its skeleton and formation of products of condensation of 2-isopropylpyridine and
dipyridylmethane. This result is consistent with the scheme of nucleophilic substitution at the second ortho
carbon atom of the pyridine ring of DPP being an unusual example of aromatic nucleophilic substitution with
carbanion as a leaving group:
r.t.
+
+
Li 10 h
Li
N
N
N
N
N
N
condensation product
+
N
N
N
Li
In order to understand the driving force of the reactions studied in this work, we have conducted DFT
calculations for three model transformations (Eq. (6)-(8)). Standard Gibbs free energies for each of the reaction
are indicated:
4
10

(
(
6)
7)
+
+
CH4
+
+
+
H
2
N
N
N
N
N
H2
N
N
N
N
H2
(8)
+
i-Pr
N
0
∆
G 298 = 11.2 (6), 11.0 (7), 14.4 kcal/mol (8)
According to these results, pyridine alkylations with methane, 2-pyridylmethane, and 2-(2-pyridyl)-
propane are almost equally unfavorable. Calculation of Gibbs free energy changes for reactions involving the
corresponding organolithium compounds, methyllithium, 2-pyridylmethyllithium, and 2-pyridyl-2-
propyllithium, can be done by the combination of these values with the pK values of the alkylating reagent from
a
one side and the reaction product from the other (Eq. (9)):
+
LiCH3
+
LiH
(9)
N
N
G 0 (9) = ∆ G 0 (6) + 1.36 [pK (H ) – pK (CH )] = 11.2 – 27.2 = – 16.0 kcal/mol
∆
2
98
298
a
2
a
4
Here the values pK
a
(H
2
) = 35 for KH–18-crown-6 system in THF solution [16] and pK
a
(CH ) = 55 for
4
DMSO solution [17] were used. In fact, due to the lower solubility of LiH in most solvents as compared with
0
that of KH in the presence of 18-crown-6, dihydrogen can behave as a stronger acid and ∆G 298 (9) will be more
negative. So, increased acidity by going from methane to dihydrogen makes this reaction possible in accordance
with experimental observations [13] and can be considered as its main driving force. Another combination of
the reaction products is favorable but slightly less than in the case of LiH as a reaction product, again, in
accordance with experimental observations (Eq. (10)):
+
LiCH3
+
H2
(10)
Li
N
N
G 0 (10) =∆ G 0 (6) + 1.36 [pK (PicH) – pK (CH )] = 11.2 – 28.6 = – 17.4 kcal/mol
∆
298
298
a
a
4
The value pK
a
(PicH) = 34 [18] for lithium salt scale in THF solution was used.
In a similar way, Gibbs free energy changes for reactions (11) and (12) can be calculated:
+
+
LiH
(11)
Li
N
N
N
N
G 0 (11) =∆ G 0 (7) + 1.36 [pK (H ) – pK (PicH)] = 11.0 + 1.4 = 12.4 kcal/mol
∆
298
298
a
2
a
4
11

According to this result, no gain in acidity is observed here and the reactions are unfavorable almost to
the same degree as reaction (7). But in the case of the reaction products being dipyridylmethyllithium and
dihydrogen, this condensation can proceed in accordance with our observations:
+
+
H2
(12)
Li
N
N
N
N
Li
G 0 (12) =∆ G 0 (7) + 1.36 [pK (DPM) – pK (PicH)] = 11.0 – 21.8 = – 10.8 kcal/mol
∆
298
298
a
a
Here we have assumed that the influence of the 2-pyridyl substituents on methane acidity is additive and
difference for the pair PicH–DPM is almost the same as for methane–2-picoline, pK (DPM) = 18.
It is important to mention here that when the LiPic-to-pyridine ratio is 2:1, the condensation can be even
more favorable due to product stabilization via protonation of a second mole of LiPic with DPM produced
Eq. (13)):
the pK
a
a
(
+
2
+
(
13)
Li
–
LiH
N
N
N
N
N
Li
G 0 (13) =∆ G 0 (7) + 1.36 [pK (DPM) + pK (H ) – 2 pK (PicH)] = – 9.4 kcal/mol
∆
2
98
298
a
a
2
a
This provides a possible explanation to the highly effective use of a 2:1 ratio of LiPic to electrophile
which has been found experimentally and which leads to the best results for Me-DPM and Me
Entries 4, 6).
2
-DPM (Table 1,
Finally, considering pyridine condensation with 2-pyridyl-2-propyllithium, since no acidic hydrogens
are present at the bridging carbon of DPP, we can calculate the Gibbs free energy change only for the case
where LiH and DPP are the condensation products (Eq. (14)):
(
14)
+
+
LiH
Li
N
N
N
N
G 0 (14) =∆ G 0 (8) + 1.36 [pK (H ) – pK (i-PrPy)] = 14.4 – 1.4 = 13.0 kcal/mol
298
298 a 2 a
∆
In this calculation it is assumed that each methyl group at the acidic α-carbon atom increases the pK
a
value of PicH by 1 unit [19], and thus we obtain an estimate of the pK
a
of 2-isopropylpyridine, pK (i-PrPy) ~ 36.
a
In accordance with this calculated result and with our experimental observation, no condensation of
2
-pyridyl-2-propyllithium with 2-isopropylpyridine is possible. Vice versa, nucleophilic cleavage of the
C(propanediyl)–C(pyridyl) bond in DPP with LiPic is very favorable due to the formation of the lithium salt of
DPM, which is more acidic than PicH:
4
12

+
+
(
15)
Li
N
N
N
N
N
N
Li
G 0 (15) = ∆ G 0298 (7) – ∆ G 0298(8) + 1.36 [pK
(DPM) – pK
(PicH)] =
∆
a
a
2
98
=
– 3.4 – 21.8 = – 25.2 kcal/mol
This is in full agreement with our experiment.
EXPERIMENTAL
All manipulations were carried out under purified argon using the standard Schlenk technique. Diethyl
ether, glyme, and toluene from Aldrich were distilled under argon over sodium – benzophenone. Pyridine,
2
-picoline, 2,6-lutidine, and 4-tert-butylpyridine from Aldrich were distilled under argon atmosphere over
calcium hydride. All the liquids were stored in gas-tight bulbs. All other reagents, 10 M butyllithium solution in
hexane, and methyl iodide were used as purchased from Aldrich. 2-Isopropylpyridine was synthesized according
1
to a slightly modified procedure [20]. H NMR spectra were recorded on a Varian Gemini 300 spectrometer
1
1
(
H 300 MHz). H NMR chemical shifts are relative to the residual solvent resonance as internal standard. High-
resolution mass spectra were obtained on a Kratos MS80 RFAQQ instrument.
The theoretical calculations in this work were performed using the density functional theory (DFT)
method [21], specifically functional PBE [22], implemented in an original program package "Priroda" [23]. In
the PBE calculations, relativistic Stevens–Bash–Krauss (SBK) effective core potentials (ECP) [24-26]
optimized for DFT calculations were used. The basis set was the 311-split for main group elements with one
additional polarization p-function for hydrogen, and an additional two polarization d-functions for elements of
higher periods. Full geometry optimization was performed without constraints on symmetry. For all the species
under investigation, frequency analysis was carried out. All minima were checked in the absence of imaginary
frequencies.
Preparation of 2-Pyridylmethyllithium and 6-Methyl-2-pyridylmethyllithium. A flame-dried
3
00 ml Schlenk flask connected to a vacuum-argon line and equipped with a Teflon valve was filled with
purified argon and the teflon stopcock was replaced by a rubber septum. In the flask 30 ml of diethyl ether or
glyme per each 0.1 mol of pyridine derivative to be lithiated was added through a Teflon cannula. Then
-picoline or 2,6-lutidine (0.1 or 0.2 mol) was introduced with a syringe. The flask was placed then into an ice
2
bath and 10 M butyllithium solution in hexane (10.0 ml, 0.1 mol per each 0.1 mol of 2-picoline or 2,6-lutidine to
be metallated) was added dropwise with stirring by hand. After the addition of butyllithium was complete, the
flask with the dark red solution has been removed from the bath. The rubber septum was replaced by the teflon
stopcock and the flask left to stay at ambient temperature for 1 h. Diethyl ether (or glyme) was removed in
vacuum and the yellow crystalline residue was dried at 0.1 Torr during 1 h.
Condensation of 2-Pyridylmethyllithium, (6-Methyl-2-pyridyl)methyllithium, and 2-Pyridyl-2-
propyllithium with either Pyridine, 2-Picoline, 4-tert-Butylpyridine, or 2,2-Bis(2-pyridyl)propane
(
General Procedure). Into a Schlenk flask containing 0.1 or 0.2 mol of the appropriate 2-pyridylmethyllithium
nucleophile suitable pyridine derivatives in amounts indicated in Table 1 were introduced with a syringe.
Toluene (30 ml per each 0.1 mol of butyllithium used) was added, then via cannula if necessary. The rubber
septum was replaced by a Teflon stopcock and the flask was either left to stay at room temperature or connected
to a condenser with receiver and immersed for 20 min into silicon oil bath heated to 110°C to distill the residual
ether off and to let the greater part of the dihydrogen evolve. Usually 10-20 ml of liquid was collected. Then if
heating at higher temperatures was needed, the Teflon valve was closed and the temperature of the silicone oil
bath was raised. At the end of the synthesis, the reaction mixture turned purple-red. It was cooled in an ice bath
4
13

and methanol (8 ml per each 0.1 mol of organolithiums used) was added with a syringe dropwise to decompose
the lithium hydride precipitated. After addition of water (8-9 ml per each 0.1 mol of organolithiums used), the
yellow liquid could be easily separated by filtration from the white precipitate of hydrated lithium hydroxide.
The liquid was dried over solid potassium hydroxide and fractionally distilled first at ambient temperature to
recover the excess of pyridine compound if it was used and then under vacuum.
Condensation of 2,2-Bis(2-pyridyl)propane with 2-pyridylmethyllithium. The amounts of the
reagents were tenfold less as compared with those given in the general procedure above; 0.01 mol of
2
-pyridylmethyllithium and 0.01 mol of 2,2-bis(2-pyridyl)propane were used. The reaction mixture was left to
stay for 10 h at ambient temperature. After the usual work-up procedure, according to NMR analysis, it
contained no starting DPP but some amounts of DPM (see below) and 2-isopropyl-substituted DPM derivatives
(
the resonance of the i-Pr group was not equivalent to that of free 2-isopropylpyridine [20]).
Condensation of 2-Pyridyl-2-propyllithium with 2-Isopropylpyridine. The amounts of the reagents
were tenfold less as compared with those given in the general procedure above; 0.01 mol of 2-pyridyl-2-
propyllithium and 0.01 mol of 2-isopropylpyridine were used. The reaction mixture was heated 2 h at 190°C.
According to NMR analysis, after the usual work-up it contained the starting 2-isopropylpyridine only.
2
,2'-Dipyridylmethane, DPM (1). The reagent ratio and yield of the condensation product are
1
indicated in Table 1. Bp 95-98°C at 0.7 Torr (cf. 104-106°C/0.2 Torr [27]). H NMR (25°C, CDCl
3
), δ, ppm:
4
.31 (2H, s, CH
2
); 7.10 (2H, dd, J = 7.5, 5.0, C(5)H); 7.24 (2H, d, J = 7.8, C(3)H); 7.58 (2H, dt, J = 1.8, 7.6,
C(4)H); 8.50-8.56 (2H, m, C(6)H).
6
-Methyl-2,2'-dipyridylmethane, Me-DPM (2). The reagent ratio and yield of the condensation
1
product are indicated in Table 1. Bp 97-102°C at 0.7 Torr (cf. 114-116°C/0.5 Torr [27]). H NMR (25°C,
CDCl
3
), δ, ppm: 2.51 (3H, s, CH
3
); 4.29 (2H, s, CH
2
); 6.96 (1H, d, J = 7.6, C(5)H); 6.97 (1H, d, J = 7.6, C(3)H);
7
.11 (1H, d, J = 7.6, C(2)terminal-H); 7.21 (1H, d, J = 7.6, C(5)terminal-H); 7.45 (1H, t, J = 7.6, C(4)H); 7.56
(
1H, dt, J = 1.8, 7.6, C(4)terminal-H); 8.49-8.55 (1H, m, C(6)terminal-H).
6
,6'-Dimethyl-2,2'-dipyridylmethane, Me
2
-DPM (3). The reagent ratio and yield of the condensation
1
product are indicated in Table 1. Bp 117-121°C at 0.7 Torr. H NMR (25°C, CDCl
3
), δ, ppm: 2.51 (6H, s, CH );
3
4
.26 (2H, s, CH
2
); 6.95 (4H, d, J = 7.6, C(3)H, C(5)H); 7.43 (2H, t, J = 7.6, C(4)H). High-resolution mass
spectrum, m/z: M = 198.11611. C13
H
14
N . Calculated 198.11570
2
2
,6-Bis(2-pyridylmethyl)pyridine, Tripyridinedimethane, TPDM (4). The reagent ratio and yield of
1
the condensation product are indicated in Table 1. Bp 165-168°C at 0.1 Torr. H NMR (25°C, CDCl
3
), δ, ppm:
4
.31 (4H, s, CH ); 7.03 (2H, d, J = 7.7, C(3)central-H, C(5)central-H); 7.10 (2H, dd, J = 7.6, 4.9,
2
C(5)terminal-H); 7.21 (2H, d, J = 7.7, C(3)terminal-H); 7.48 (1H, t, J = 7.7, C(4)central-H); 7.55 (2H, dt,
J = 1.8, 7.7, C(4)terminal-H); 8.49-8.54 (2H, m, C(6)terminal-H). High-resolution mass spectrum, m/z:
found 261.12730. C17
H
15
N . Calculated 261.12660.
3
2
-(6-Methyl-2-pyridylmethyl)-6-(2-pyridylmethyl)pyridine, 6-Methyltripyridinedimethane,
Me-TPDM (5). The reagent ratio and yield of the condensation product are indicated in Table 1. Bp 169-171°C
1
at 0.7 Torr. H NMR (25°C, CDCl
3
), δ, ppm: 2.51 (3H, s, CH
3
); 4.28 (2H, s, CH
2
); 4.31 (2H, s, CH ); 6.95 (2H,
2
d, J = 7.7, C(3)H, C(5)H); 7.00 (1H, d, J = 7.7, C(3)central-H); 7.02 (1H, d, J = 7.7, C(5)central-H); 7.09 (1H,
dd, J = 7.7, 4.8, C(5)terminal-H); 7.22 (1H, dd, J = 7.7, 0.9, C(3)terminal-H); 7.42 (1H, t, J = 7.7, C(4)H); 7.46
(
1H, t, J = 7.7, C(4)central-H); 7.55 (1H, dt, J = 1.8, 7.7, C(4)terminal-H); 8.48-8.54 (1H, m, C(6)terminal-H).
High-resolution mass spectrum, m/z: found 275.13685. C18
H
17
N . Calculated 275.142247.
3
2
,6-Bis(6-methyl-2-pyridylmethyl)pyridine, 6,6'-Dimethyltripyridinedimethane, Me
-TPDM (6).
2
The reagent ratio and yield of the condensation product are indicated in Table 1. Bp 173-178°C at 0.15 Torr.
1
H NMR (25°C, CDCl
3
), δ, ppm: 2.51 (6H, s, CH
3
); 4.29 (4H, s, CH ); 6.95 (2H, d, J = 7.5, C(5)terminal-H);
2
6
.96 (2H, d, J = 7.5, C(3)terminal-H); 6.99 (2H, d, J = 7.2, C(3)central-H, C(5)central-H); 7.42 (2H, t, J = 7.5,
C(4)terminal-H); 7.45 (1H, t, J = 7.2, C(4)central-H). High-resolution mass spectrum, m/z: found 289.15664.
. Calculated 289.15790.
C
H N
19 19 3
4
14

2
,6-Bis(4-tert-butyl-2-pyridylmethyl)pyridine,
4,4'-Bis(tert-butyl)tripyridinedimethane,
(
tert-Bu)
2
-TPDM (7). The reagent ratio and yield of the condensation product are indicated in Table 1.
1
Bp 195-200°C at 0.1 Torr. H NMR (25°C, CDCl
3
), δ, ppm: 1.23 (18H, s, CH
3
); 4.30 (4H, s, CH ); 7.02 (2H, d,
2
J = 7.7, C(3)central-H, C(5)central-H); 7.09 (2H, dd, J = 1.8, 5.4, C(5)terminal-H); 7.22-7.25 (2H, m,
C(3)terminal-H); 7.47 (1H, t, J = 7.7, C(4)central-H), 8.39-8.43 (2H, m, C(6)terminal-H). High-resolution mass
spectrum, m/z: found 373.25083. C25
H
31
3
N . Calculated 373.25180.
Synthesis of 2,2-Bis(2-pyridyl)propane, 2,2'-Dipyridine-2,2-propane, DPP,
C
13
H
14
N
2
by
Methylation of Dipyridylmethane, DPM. A flame-dried 300 ml Schlenk flask connected to a vacuum-argon
line and equipped with a stirring bar was filled with purified argon. In the flask 50 ml of glyme was added
through a Teflon cannula. Then DPM (5.8 g, 34 mmol) was introduced with a syringe. The flask was placed
then into an ice bath and 10 M butyllithium solution in hexane (3.5 ml, 35 mmol) was added dropwise with
stirring. After the addition of butyllithium was complete, the flask with the dark red solution was left for 15 min.
Then methyl iodide (2.25 ml, 4.83 g, 34 mmol) was added dropwise with syringe with intense stirring. With the
last drops of methyl iodide the reaction mixture became almost colorless. These treatments of the solution with
butyllithium and then with methyl iodide have been repeated one time more. To the resulting yellow solution
with some precipitated lithium iodide, potassium hydroxide (4.3 g, 66 mmol) dissolved in 3 ml of water was
added with intense stirring. Then the liquid phase was separated by filtration from the inorganic salts
precipitated and dried over solid potassium hydroxide, and the solvent was distilled off. The oily residue was
1
then distilled under vacuum to give pure DPP. Yield 6.7 g (90 %); bp 90-95°C at 0.7 Torr. H NMR (25°C,
CDCl
3
), δ, ppm: 1.78 (6H, s, CH ); 7.03-7.08 (2H, m, C(5)H); 7.13-7.18 (2H, m, C(3)H); 7.52-7.57 (2H, m,
3
C(4)H); 8.51-8.55 (2H, m, C(6)H). High-resolution mass spectrum, m/z: found 198.11647. C13
H
14
N .
2
Calculated 198.11570.
CONCLUSION
We conclude that a Chichibabin-like condensation (Eqs. (1), (2)) is a convenient synthetic route to
dipyridinemethanes and tripyridinemethanes driven by a large (not less than 10 pK units) gain in the acidity of
a
its participants. This statement is the basis for understanding its limitations when planning the synthesis of
alkyl-substituted pyridines and, in particular, polypyridine- polyalkanes according to these schemes. In turn,
polypyridinepolyalkanes can be easily cleaved at the bridging carbon–pyridine carbon bond with carbanionic
nucleophiles even if only a small gain in acidity can be reached when going from reactants to the cleavage
products.
Synthetic application of the method described here for the synthesis of macrocyclic
polypyridinepolyalkanes (pyridinophanes or calixpyridines) is now under way.
A.N.V. is grateful to Prof. Kenneth G. Caulton for a post-doctoral Fellowship at Indiana University.
S.V.B. thanks BRHE REC-007 for support of his participation in this work. This work was made possible in
part due to funding from the Russian Foundation for Basic Research (grant No 01-03-32692a).
REFERENCES
1
2
3
4
.
.
.
.
J. Canty, A. Dedieu, H. Jin, A. Milet, and M. K. Richmond, Organometallics, 15, 2845 (1996).
D. D. Wick and K. I. Goldberg, J. Am. Chem. Soc., 119, 10235 (1997).
S. A. O'Reilly, P.S. White, and J. L. Templeton, J. Am. Chem. Soc., 118, 5684 (1996).
C. M. Wang, J. W. Ziller, and T. C. Flood, J. Am. Chem. Soc., 117, 1647 (1995).
4
15

5
.
G. Wilkinson (editor), Comprehensive Coordination Chemistry. The Synthesis, Reactions, Properties,
and Applications of Coordination Compounds, Pergamon Press, Oxford; New York; Toronto, 1987,
Vol. 2, 73.
6
7
8
9
.
.
.
.
L. F. Szczepura, L. M. Witham, and K. J. Takeuchi, Coord. Chem. Rev., 174, 5 (1998).
J. Canty, N. J. Minchin, B. W. Skelton, and A. H. White, J. Chem. Soc., Dalton Trans., 2201 (1986).
J. Manzur, Transit. Met. Chem., 11, 220 (1986).
A. Meister, N. Takano, T. Chuard, M. Graf, K. Bernauer, H. Stoeckli-Evans, and G. Suess-Fink,
Z. Anorg. Allgem. Chem., 621, 117 (1995).
1
1
1
0.
1.
2.
M. T. Garland and D. Grandjean, Acta Crystallogr., C43, 643 (1987).
A. M. Garcia and J. Manzur, Acta Crystallogr., C50, 1882 (1994).
E. Spodine, J. Manzur, M. T. Garland, J. P. Fackler, Jr., R. J. Staples, and B. Trzcinska-Bancroft, Inorg.
Chim. Acta, 203, 73 (1993).
1
1
1
1
1
1
1
2
2
2
2
3.
4.
5.
6.
7.
8.
9.
0.
1.
2.
3.
H. Vorbrueggen and M. Maas, Heterocycles, 27, 2659 (1988).
C. Osuch and R. Levine, J. Am. Chem. Soc., 78, 1723 (1956).
F. P. Schmitz, H. Hilgers, and B. Gemmel, Makromol. Chem., 191, 1033 (1990).
E. Buncel and B. Menon, J. Am. Chem. Soc., 99, 4457 (1977).
D. Algrim, J. E. Bares, J. C. Branca, and F. G. Bordwell, J. Org. Chem., 43, 5024 (1978).
R. R. Fraser, T. S. Mansour, and S. Savard, J. Org. Chem., 50, 3232 (1985).
F. G. Bordwell, J. E. Bartmess, and J. A. Hautala, J. Org. Chem. 43, 3095 (1978).
E. Pasquinet, P. Rocca, F. Marsais, A. Goddard, and G. Queguiner, Tetrahedron, 54, 8771 (1998).
R. G. Parr and W. Yang, Density-functional Theory of Atoms and Molecules, Oxford, 1989.
J. P. Perdew, K. Burke, and M. Ernzerhof, Phys. Rev. Lett., 77, 3865 (1996).
Yu. A. Ustynyuk, L.Yu. Ustynyuk, D.N. Laikov, and V.V. Lunin, J. Organomet. Chem., 597, 182
(
2000).
2
2
2
2
4.
5.
6.
7.
W. J. Stevens, H. Bash, and M. Krauss, J. Chem. Phys., 81, 6026 (1984).
W. J. Stevens, H. Bash, M. Krauss, and P. Jasien, Can. J. Chem., 70, 612 (1992).
T. R. Cundari and W. J. Stevens, J. Chem. Phys., 98, 5555 (1993).
P. J. Chivers and T. A. Crabb, Tetrahedron, 26, 3369 (1970).
4
16