Synthesis of 4-[N-methyl-4-pyridinio]-phenolate (POMP) and negative solvatochromism of this model molecule in view of nonlinear optical applications

Article abstract of DOI:10.1016/j.tetlet.2005.05.031

4-[N-Methyl-4-pyridinio]-phenolate or [4-(phenyloxido)-N-methylpyridinium, POMP] has been synthesized via Suzuki cross-coupling reaction. POMP undergoes a strong negative solvatochromism (blue shift) as the solvent polarity increases and behaves qualitati

Full text of DOI:10.1016/j.tetlet.2005.05.031

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 4737–4740
Synthesis of 4-[N-methyl-4-pyridinio]-phenolate (POMP)
and negative solvatochromism of this model molecule in view
of nonlinear optical applications
a
´ `
Vincent Diemer, Helene Chaumeil, Albert Defoin,
a,
a
*
b
b
´
Patrice Jacques and Christiane Carre
a
´
Laboratoire de Chimie Organique et Bioorganique CNRS UMR 7015, Ecole Nationale Superieure de Chimie de Mulhouse,
´
Universite de Haute Alsace, 3 rue Alfred Werner, 68093 Mulhouse Cedex, France
´
´
Universite de Haute Alsace, 3 rue Alfred Werner, 68093 Mulhouse Cedex, France
b
´ ´ ´
Departement de Photochimie Generale, CNRS UMR 7525, Ecole Nationale Superieure de Chimie de Mulhouse,
Received 7 April 2005; revised 29 April 2005; accepted 11 May 2005
Available online 31 May 2005
Abstract—4-[N-Methyl-4-pyridinio]-phenolate or [4-(phenyloxido)-N-methylpyridinium, POMP] has been synthesized via Suzuki
cross-coupling reaction. POMP undergoes a strong negative solvatochromism (blue shift) as the solvent polarity increases and
behaves qualitatively as the famous empirical polarity parameter ET(30). However, it should be noted that the previously published
theoretical investigations of POMP revealed unexpected difficulties in the interpretation of its structure. Interestingly, it is confirmed
here in the sense that the experimental transition energy in nonpolar solvents seems by far less important than predicted by different
computational methods. Therefore, POMP is highly prospective for (i) the comprehension of betaines solvatochromism, (ii) the
development of new molecules with large first and second hyperpolarizabilities useful for doping polymers, the aim being to ela-
borate efficient processable materials for quadratic optics.
Ó 2005 Elsevier Ltd. All rights reserved.
1. Introduction
the synthesis of 4-[N-methyl-4-pyridinio]-phenolate
or [4-(phenyloxido)-N-methylpyridinium, POMP], see
Chart 1. This betaine is constituted by two oppositely
charged heteroatomic rings linked directly. It is a model
compound of a series of new zwitterionic molecules,1
which are potentially very promising to dope photo-
polymerizable systems for the elaboration of organic
optical devices offering good quadratic NLO proper-
ties.2,3 To the best of our knowledge, only two deriva-
tives of POMP have been experimentally investigated
until now, compound 2, in view of NLO applications3
and compound 3 as a putative polarity indicator.4
Materials with high nonlinear optical (NLO) activities
are increasingly used as electro-optic switching elements
for telecommunication and optical information proces-
sing. There is, thus, a tremendous need of organic
molecules exhibiting large first and second hyper-
polarizabilities. This is generally achieved by extended
p-conjugated systems, which are mostly very solvent
sensitive, undergoing bathochromic shift as the polarity
increases. As a consequence, depending on the medium
in which they should be embedded, the chromophore
transparency in the visible region—an essential property
for many photonic applications—may be eroded. Fortu-
nately, there are also molecules which undergo negative
solvatochromism, essentially merocyanines or zwitter-
ionic compounds. We report here for the first time
2. Synthesis
POMP was synthesized by Suzuki cross-coupling reac-
tion between the protected bromophenol 7 and the pyr-
idineboronic ester 9 in THF with Pd(PPh3)4 as catalyst
and CsCO3 as base. The O-protection of 4-bromophenol
6 in methylal with P2O5 in CH2Cl2 led to 7 in good
yield.5 The borinic ester 9 was obtained by metal-halide
Keywords: Betaine; Solvatochromism.
*
Corresponding author. Tel.: +33 389336864; fax: +33
(0)389336875; e-mail: helene.chaumeil@uha.fr
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.05.031

4738
V. Diemer et al. / Tetrahedron Letters 46 (2005) 4737–4740
exchange followed by reaction with tributyl borate and
θ
2
6
3
3' 2'
then with pinacol.6 The protected biphenyl 10 was then
successively acidic O-deprotected and N-methylated
with iodomethane.7 In a final step, the phenol was
deprotonated with tetrabutylammonium hydroxide. A
washing with a mixture of solvents (CH2Cl2/Et2O,
1:1), afforded the pure expected compound POMP
(Scheme 1).8
4
4'
N
O
N
O
1
1'
5
5' 6'
1
POMP
O
O
Ph
N
3. Solvatochromism
Ph
Ph
Ph
Ph
Figure 1 gives the absorption spectrum of POMP in two
polarity representative solvents: water and acetonitrile.
The absorption bands of betaines are generally broad.9
As a consequence, an overlap with the actinic light can
constitute a serious drawback when the molecules are
introduced in a photopolymerizable matrix. Fortu-
nately, it can be seen that POMP has a moderately large
and structureless band. On the other hand, as usual for
zwitterionic structures, the solubility of POMP becomes
weak for low polar solvents. However, this can be cir-
cumvented by appropriate substitutions.1 Note that it
is crucial to control the basicity of solvents. In neat sol-
vents, especially hydroxylic ones, another band located
at ca. 350 nm is observed, which can be safely attributed
to the protonated form of POMP 12. Table 1 gathers the
transition energies of the neutral form in the solvents
spanning the available polarity range. POMP clearly
N
Ph
N
Ph
O
O
K
2
3
ET(30)
3'
2'
2
6
3
3' 2'
I
4
4'
N
O
1'
Na
1
1'
4'
5
5' 6'
5'
6'
5
4
Chart 1.
I
OR
Br
1) BuLi, Et2O
O
O
2) B(OBu)3
B
3) pinacol
66%
N
N
6 R= H
7 R= MOM
8
9
CH2(OCH3)2
P2O5, CH2Cl2
90%
Cs2CO3, Pd(PPh4)4
THF reflux
80%
OH
OR
O
CH3I
acetone
nBu4OH, MeOH
65%
85%
N
12
N
N
I
POMP
10 R= MOM
11 R= H
HCl 1M
94%
Scheme 1. Synthesis of POMP.

V. Diemer et al. / Tetrahedron Letters 46 (2005) 4737–4740
4739
80
75
70
POMP being a ‘‘model’’ molecule, has already been
theoretically investigated. Ratner and co-workers10
using the AM1 model Hamiltonian investigated the
key role played by the torsional angle h in the values
of the transition dipole moment and the hyperpolari-
zability. More relevant to the present experimental
investigation is the single configuration interaction
(SCI) study on conjugated betainic chromophores
based on DFT optimized geometries recently published
by Fabian and co-workers.11 The ground and excited
state dipole moments of POMP were calculated to be
lg = 15.2 D and le = 14.2 D, respectively. Upon excita-
tion, the dipole moment decreases therefore, only
slightly: 1D. This is qualitatively in agreement with
the experimental negative solvatochromism reported
presently, but can hardly explain quantitatively the
strong shift observed.
wat r
water
mmethanol
νm == 00..6666 EET((3300)) ++ 3311..22
ax
νmax
kcal/mol
R = 0.9863
R2
ppropann-1-ol
65
w
a
t
e
r
aacetoonitrile
ppropann-2-ol
bbutann-2-ol
aacetone
CCH3CN
60
55
DDMSO
330
380
430
480
5530
λ3 (nm)
40
45
50
55
ET(30)
kcal/mol
60
65
Moreover, the ab initio SCI calculations correctly pre-
dict an intense electronic transition mainly of the
p ! p* character for the nine betaines computed. But,
in the case of the seven compounds for which experi-
mental data are available, the SCI absorption wave
numbers as well as oscillator strengths are greatly over-
estimated. Interestingly, most of the experimentally
known absorption wave numbers are well reproduced
by the energy difference of the frontier KS orbitals
(DFT HOMO-LUMO energy gap). We report here
their data relevant to POMP, which includes also
semi-empirical methods (see Table 3 of the original pa-
per11 for details on the methods used): SCI/6-31+G**
(324 nm); PM3/CI (435 nm); ZINDO/5 (345 nm);
CNDOL/22 (366 nm); DFT (433 nm). In a first ap-
proach, the comparison with the value found in acetone
(k = 479 nm) qualitatively confirms their conclusions.
Even more remembering, that the reported computed
values correspond to the gas phase (i.e., the solvent is
not included in the methods used), it is instructive to
ÔevaluateÕ the corresponding experimental value of
POMP. From Figure 1, it can be estimated that in a
nonpolar solvent like tetramethylsilane (ET(30) ꢀ
30.7 kcal/molꢁ1) POMP would absorb at 550 nm and
in the gas phase (ET(30) ꢀ 27.1 kcal/molꢁ1) at 580 nm.
Both values are now in large disagreement with the
nearest value computed within the DFT frame (433 nm).
Figure 1. Transition energies of POMP versus the empirical parameter
of polarity ET(30). Insert: POMP absorption spectra in moderate
(acetonitrile) and high polarity (water) solvents.
Table 1. Comparison of transition energies in kcal/mol of POMP
(solvents for which solubility is good) with the data ET(30)9
POMP
kcal/mol
ET(30)9
23b k, nm
k, nm
Water
72.6
68.4
64.9
63.7
62.2
60.6
60.5
59.7
394
418
441
457
460
472
473
479
63.1
55.4
50.7
48.4
47.1
45.6
45.1
42.2
Methanol
Propan-1-ol
Propan-2-ol
Butan-2-ol
CH3CN
482
503
512
DMSO
Acetone
undergoes a strong blue shift as the polarity increases
(negative solvatochromism). Moreover, the transition
energy of POMP is linearly correlated with those of
compound ET(30), the well-known empirical parameter
of polarity.9 ET(30) is also a betaine, related to POMP
but with a different location of N+ atom and with five
phenyl substituents (Chart 1). As the slope in Figure 1
is significantly lower than unity (0.66), it can be deduced
that POMP cannot compete with ET(30) as a polarity
indicator despite its strong solvatochromism.
As indicated in Chart 1, the molecular structure of
POMP can be described as a combination of two limit
structures/zwitterionic and quinoid. These two limit
forms differ principally by their electronic distribution
(the total charge of the heterocyclic ring moiety indi-
cates the charge transferred from the donor and accep-
tor fragment of the molecule) andꢁtheir bond lengths
By the way, it can be seen from Table 1 that POMP ab-
sorbs at higher energies than the POMP derivative 2, in
agreement with the well-known t-butyl substituent effect
on UV–visible absorption spectrum. More importantly,
the difference in energy is very large for methanol which
indicates a stronger solvatochromism for POMP com-
pared to 2. This is due to the fact that in the case of 2,
the two t-butyl substituents preclude any hydrogen
bonding from the methanol with the oxygen atom, the
basic center of betaines. Note that this effect is still much
debated in the interpretation of the betaines
solvatochromism.9
0
0
0
(especially the C1 @O or C1 –O and C4–C4 or
0
C4@C4 single or double bonds). Interestingly, in his de-
tailed theoretical investigation of the POMP structure,
Fabian11 was puzzled by the data obtained. Briefly said,
the different pieces of information about the POMP
structure were inconsistent. On one hand, a relatively
0
long interfragmental bond C4–C4 and C–C bonds
showing the expected quinoid bond alternation were
found, the aromaticity criteria confirming this classifica-
tion. On the other hand, these bond distances differed

4740
V. Diemer et al. / Tetrahedron Letters 46 (2005) 4737–4740
Table 2. Comparison of 13C NMR chemical shifts of 4-phenyl
pyridinium 4, sodium phenolate 5, quinone and POMP in CD3OD
tions with the aim to find the adequate compound
among the series of POMP derivatives. By the way, we
hope to shed light on the old but still unsolved debate
concerning the betaine structures and their solvent
dependencies.
4
POMP
5
O
O
C1
49
46.3
144.2
120.7
156.4
117.3
131.1
121.7
177.1
C2–C6
C3–C5
C4
146.7
125.8
157.1
135.0
129.1
130.9
133.3
Acknowledgements
0
C4
C3–C5
0
C2 –C6
115.2
129.9
119.7
167.4
´
The Region Alsace, the program Interreg IIIA Rhenapho-
0
tonics are gratefully acknowledged for their financial
supports. We are grateful to Dr. A. Fort and Dr. A.
Boeglin (IPCMS, CNRS UMR 7504, Strasbourg) for
helpful discussions and suggestions.
0
137.6
188.8
0
C1
not greatly from typical zwitterionic structure 1. Even
more intriguing was the large dipole moment found
(see supra) and its small decrease in the excited state
which implies a negative solvatochromism, as experi-
mentally found in the present work. However, positive
solvatochromism is rather expected from quinoid for-
mula. Anyhow, inspection of the literature shows that
the search for correct description of the betaine struc-
ture (zwitterionic/quinoid) is a tricky task (e.g., the
recent publication of Morley and the references
therein12).
References and notes
´
1. Diemer, V.; Chaumeil, H.; Defoin, A.; Carre, C.; Fort, A.,
in preparation.
2. Soppera, O.; Diemer V.; Bombenger, J.-P.; Feuillade, M.;
Croutxe-Baghorn, C.; Chaumeil, H.; Defoin, A.; Mery, S.;
´
Mager, L.; Fort, A.; Carre, C. Proc. Macro 2004 (World
Polymer Congress)-IUPAC, Paris, 2004 (CD).
3. (a) Fort, A.; Boeglin, A.; Mager, L.; Amiot, C.; Comb-
ellas, C.; Thiebault, A.; Rodriguez, V. Synth. Met. 2001,
124, 209; (b) Boeglin, A.; Fort, A.; Mager, L.; Combellas,
C.; Thiebault, A.; Rodriguez, V. Chem. Phys. 2002, 282,
353; (c) Boeglin, A.; Rodriguez, V.; Combellas, C.;
Diemer, V.; Chaumeil, H.; Defoin, A.; Fort, A. Proc.
SPIE, Linear and Nonlinear Optics of Organic Materials
2004, 5517, 88.
Clearly, POMP appears as a challenge for theoretical
chemistry, which underlines the interest of its present
synthesis. Finally, it is useful to recall that 13C NMR
spectroscopy is a suitable tool for the confirmation of
partial positive or negative charges on certain carbon
atoms. However, in the case of POMP, low solubility se-
verely restricted the number of solvents for which reli-
able data are available. However, 13C NMR data
argue in favour of a zwitterionic structure for POMP
(Table 2). Actually, the data of the pyridine moiety of
POMP are close to those of 4-phenyl pyridinium 4 and
those of the phenyl moiety close to phenolate ion 5.
4. Reichardt, C.; Milart, P.; Schafer, G. Liebigs Ann. Chem.
1990, 441.
5. Fuji, K.; Nakano, S.; Fujita, E. Synthesis 1975, 276.
6. Coudret, C. Synth. Commun. 1996, 26, 3543.
7. Combellas, C.; Subra, S.; Thiebault, A. Tetrahedron Lett.
1992, 33, 4923.
8. Spectroscopic data of POMP: mp(dec) = 120 °C. 1H
NMR (400 MHz, CD3OD, d, ppm): 4.13 (s, 3H, N+–
0
0
0
0
CH3), 6.67 (d, H2 , H6 , 8.8 Hz), 7.76 (d, H3 , H5 , 8.8 Hz),
7.99 (d, H3, H5, 7.0 Hz), 8.34 (d, H2, H6, 7.0 Hz). 13C
0
0
NMR (100.6 MHz, CD3CN, d, ppm): 122.0 (C2 , C6 ), 131
0
0
(C3, C3 , C5, C5 ), 142.93 (C2, C6). Anal. Calcd for
C12H11NO, 2H2O (221.25): C, 65.14; H, 6.83, N, 6.33.
Found: C, 65.5; H, 6.7; N, 6.6.
4. Conclusion
9. Reichardt, C. Solvents and Solvents Effects in Organic
Chemistry; Wiley, VCH: Weinheim, 2003.
10. Albert, I. D. L.; Mark, T. J.; Ratner, M. A. J. Am. Chem.
Soc. 1998, 120, 11174.
11. Fabian, J.; Rosquete, G. A.; Montero-Cabrera, L. A. J.
Mol. Struct. (Theochem) 1999, 469, 163.
12. Morley, J. O.; Parfield, J. J. Chem. Soc., Perkin Trans. 2
2002, 1698.
The optimization of materials for nonlinear optical
devices requires on one side a thorough understanding
of NLO processes involved as a function of electronic
and geometrical molecular structures. On the other
side, a concomitant tailoring of the doped polymers is
necessary in order to be in accordance with industrial
expectations. Works are in progress in these two direc-