A serendipitous C-C bond formation reaction between michael donors and diiminoquinoid ring assisted by quaternary ammonium fluoride

Article abstract of DOI:10.1021/ol9025843

An efficient C-C bond formation reaction assisted by a fluoride ion has been Identified for N,N'-diphenyl-1,4-phenylenediimine with the compounds having an active methylene group. The reaction follows the typical Michael addition fashion and proceeds to c

Full text of DOI:10.1021/ol9025843

ORGANIC
LETTERS
2009
Vol. 11, No. 24
5586-5589
A Serendipitous C-C Bond Formation
Reaction between Michael Donors and
Diiminoquinoid Ring Assisted by
Quaternary Ammonium Fluoride
Vijaykumar V. Paike, R. Balakumar, Hsin-Yu Chen, Hong-Pin Shih, and
Chien-Chung Han*
Department of Chemistry, National Tsing Hua UniVersity, Hsinchu, Taiwan, ROC
cchan@mx.nthu.edu.tw
Received July 16, 2009
ABSTRACT
An efficient C-C bond formation reaction assisted by a fluoride ion has been identified for N,N′-diphenyl-1,4-phenylenediimine with the
compounds having an active methylene group. The reaction follows the typical Michael addition fashion and proceeds to completion within
1 h at 70 °C in acetonitrile in the presence of tetrabutylammonium fluoride (TBAF), while the same reaction failed to proceed in the absence
of a fluoride anion. The new finding reported herein also offers an unprecedented method for a direct functionalization of polyaniline backbone
with versatile functional alkyl groups.
The Michael addition reaction is one of the most useful
carbon-carbon bond forming reactions, which are of general
synthetic interest in organic synthesis. Enolates or stabilized
carbanions could be regarded as proper nucleophiles for a
conjugate addition reaction to the R,ꢀ-unsaturated carbonyl
moieties.1,2 Various types of catalysts such as metal com-
plexes,3 chiral ionic liquid,4 phase transfer catalyst,5 orga-
nocatalyst,6 L-proline salt,7 and TBAF are used to facilitate
the reaction.8 Similar types of reactions for the R,ꢀ-
unsaturated imino compounds or diiminoquionoid rings have,
however, rarely been reported.9 Polyaniline (Pan) represents
(6) (a) Xu, D. Q.; Yue, H. D.; Luo, S. P.; Xia, Ai.-B.; Zhang, S.; Xu,
Z. Y. Org. Biomol. Chem., 2008, 6, 2054–2057. (b) Wang, J.; Li, H.; Zu,
L.; Jiang, W.; Xie, H.; Duan, W.; Wang, W. J. Am. Chem. Soc. 2006, 128,
12652–12653. (c) Yoon, T. P.; Jacobsen, E. N. Angew. Chem., Int. Ed.
2005, 44, 466–468. (d) Wang, C.; Zhou, Z.; Tang, C. Org. Lett. 2008, 10,
1707–1710.
(1) (a) Jacobsen, E. N.; Pfaltz, A.; Yamamoto, H.; Eds. ComprehensiVe
Asymmetric Catalysis, 1st ed.; Springer: Berlin, 1999. (b) Tsogoeva, S. B.
Eur. J. Org. Chem. 2007, 1701–1716
.
(2) (a) Yang, X.; Zhou, X.; Lin, L.; Chang, L.; Liu, X.; Feng, X. Angew.
Chem., Int. Ed. 2008, 47, 1–4. (b) Pansare, S. V.; Pandya, K. J. Am. Chem.
Soc. 2006, 128, 9624–9625. (c) Corey, E. J.; Zhang, F. Y. Org. Lett. 2000,
2, 4257–4259. (d) Nugent, B. M.; Yoder, R. A.; Johnston, J. N. J. Am.
(7) (a) Kotrusz, P.; Toma, S. ArkiVoc 2006, (v), 100–109. (b) Yamaguchi,
M.; Shiraishi, T.; Hirama, M. J. Org. Chem. 1996, 61, 3520–3530.
(8) (a) Liu, L.; Xu, B.; Hammond, G. B. Org. Lett. 2008, 10, 3887–
3890. (b) Sharma, G. V. M.; Reddy, V. G.; Chander, A. S.; Reddy, K. R.
Tetrahedron Asymmetry 2002, 13, 21–24. (c) Park, D. Y.; Gowrisankar,
S.; Kim, J. N. Tetrahedron Lett. 2006, 47, 6641–6645. (d) Ruano, J. L. G.;
Topp, M.; Lopez-Cantarero, J.; Aleman, J.; Remuinan, M. J.; Cid, M. B.
Org. Lett. 2005, 7, 4407–4410.
Chem. Soc. 2004, 126, 3418–3419
.
(3) (a) Agostinho, M.; Kobayashi, S. J. Am. Chem. Soc. 2008, 130, 2430–
2431. (b) Kurz, M. E.; Chen, T. R. J. Org. Chem. 1978, 43, 239–242. (c)
Zhou, H.; Peng, D.; Qin, B.; Hou, Z.; Liu, X.; Feng, X. J. Org. Chem.
2007, 72, 10302–10304. (d) Trost, B.; Lupton, D. W. Org. Lett. 2007, 9,
2023–2026
.
(9) (a) Han, C. C.; Balakumar, R.; Thirumalai, D.; Chung, M. T. Org.
Biomol. Chem. 2006, 4, 3511–3516. (b) Han, C. C.; Hong, S. P.; Yang,
K. F.; Bai, M. Y.; Lu, C. H.; Huang, C. S. Macromolecules 2001, 34, 587–
591. (c) Han, C. C.; Lu, C. H.; Hong, S. P.; Yang, K. F. Macromolecules
2003, 36, 7908–7915. (d) Han, C. C.; Hseih, W. D.; Yeh, J. Y.; Hong,
S. P. Chem. Mater. 1999, 11, 480–486. (e) Han, C. C.; Jeng, R. C. Chem.
Commun. 1997, 553–554.
(4) Li, P.; Wang, L.; Wang, M.; Zhang, Y. Eur. J. Org. Chem. 2008,
1157–1160
.
(5) (a) Gomez-Bengoa, E.; Linden, A.; Lo´pez, R.; Mu´igica-Mendiola,
I.; Oiarbide, M.; Palomo, C. J. Am. Chem. Soc. 2008, 130, 7955–7966. (b)
Francesco, F.; Valentina, S.; Daniel, P.; Raquel, P. H.; Luca, B.; Alfredo,
R. Angew. Chem., Int. Ed. 2005, 44, 7975–7978
.
10.1021/ol9025843  2009 American Chemical Society
Published on Web 11/24/2009

a very interesting class of conducting polymers, which is
renown for its fairly large number of redox states. The
backbone structure of polyaniline, as illustrated in Figure 1,
Scheme 1. Fluorination of Emeraldine Base (EB) and
Pernigraniline Base (PB) Forms of Polyaniline (Pan) via the
CRS Reaction
Figure 1. Representative structure of polyaniline.
can be regarded as consisting of various ratios of the reduced
repeat units A (containing the diaminobenzenoid rings) and
the oxidized repeat units B (containing the diiminoquinoid
rings). Like other classes of conducting polymers, the
application of polyaniline has been greatly hindered by its
poor solution-processability.
Although the solubility of substituted-polyanilines (S-Pans)
prepared from the alkyl- and alkoxy-substituted aniline
monomers are higher, their conductivities are ∼3-5 orders
of magnitude lower than the unsubstituted Pan. Such
conductivity reductions are believed to be caused by the
increased extent of the nonconjugated defect backbone
structures that was induced to form during the growth of
the polymer chain due to the competing electronic directing
placement effect of the substituent’s. Such a problem has
been alleviated by utilizing the concurrent reduction and
substitution (CRS) reaction method to introduce the sub-
stituent group after the formation of a polyaniline backbone.
For example, highly conductive and processable function-
alized polyanilines with various alkylthio and dialkylamino
groups have been prepared via the CRS method by reacting
the preformed Pan of a desirable redox state with nucleo-
philes, like thiols and amines.9 The reactions were believed
to occur at the diiminoquinoid sites and follow a typical
Michael addition fashion, which converted the unsubstituted
diiminoquinoid ring into the substitituted diaminobenzenoid
rings. The efforts in extending the same reaction to the
oxygen and carbon-based nucleophiles have however proved
to be futile, up to now.
Recently, such CRS reaction was found to be a useful
synthetic method for making highly conductive and electroactive
fluorinated-polyanilines (F-Pans) having a controlled extent of
fluorination (ranging from 25 to 130 mol %),10 by reacting Pan
with a fluoride anion, as illustrated in Scheme 1. Most
interestingly, the fluoriation reaction did not go in the typical
aprotic solvent media (i.e., NMP, DMF, CH3CN), but it worked
effectively (though slowly) in some selected protic solvent
media (i.e., MeOH and EtOH). The plausible mechanism was
proposed as the one illustrated in Scheme 2.
Scheme 2. Plausible Mechanism for the Fluorination of Pan
cloud system. The equilbrium concentration of the flu-
orinated intermediate 1 may not be high, but the basicity
of the intermediate is high enough to seize the proton from
the employed protic solvents to form the protonated and
fluorinated intermediate 2 (Scheme 2), which would
ultimately result in the fluorinated-polyaniline leucoem-
eraldine base as driven by the extra energy gain from the
resonance energy of the newly formed aromatic ring. To
further evaluate the fundamental properties of the F-Pans
and understand the reaction mechanism of such fluorina-
tion, similar reactions between the fluoride anion and the
model compound N,N′-diphenyl-1,4-phenylenediimine (PDI)
was investigated. Surprisingly, there was no reaction in
the aprotic solvents (e.g., CH3CN, CH2Cl2, DMSO, DMF,
NMP, and THF) as well as in the protic solvents (e.g.,
MeOH and EtOH), even with excess amounts of fluoride
anion. We thought that maybe in this much less conjugated
PDI system, the much reduced polarizability of the
π-electron cloud system is no longer sufficient to suppport
the fluoride to make an effective nucleophilic attack at
the diiminoquinoid rings of PDI. Thus, an the alternative
way to facilitate the above reaction is to activate the PDI
counterpart through the protonation of the imino sites by
a much more efficient proton source than ROH.
We rationalized that although the nucleophilicity of the
fluoride anion has been weakened in these protic solvents,
it is still sufficient to make an effective nucleophilic attack
at the diiminoquinoid rings of Pan, as facilitated by the
high polarizability of the highly conjugated π-electron
Therefore, one drop of acetic acid was added to provide
such an essential ingredient, but the reaction failed, which
(10) Han, C. C.; Chen, H. Y. Macromolecules 2007, 40, 8969–8973.
Org. Lett., Vol. 11, No. 24, 2009
5587

we rationalized that the presence of the free acid may
significantly further reduce the nucleophilicity of the fluoride
and thus result in no reaction. Apparently, what we actually
needed is a latent proton source, which should not be a free
proton but could serve as an effective proton source as
needed by the reaction intermediate. For this purpose we
chose nitromethane, because it is a polar aprotic solvent and
contains a proton source which is more acidic than MeOH
but not as free as acetic acid.
as with a lesser amount of TBAF (0.5-4 equivalents), giving
quite similar yields (entries 10-13, Table 1). The significance
of this reaction was shown by the fact that when the CRS
reaction was tried between PDI and n-BuLi, there was no
sign for the formation of expected butyl-substituted PDA,
even when CH3NO2 was used as the solvent medium. Other
bases (e.g., KOH, NaOH, NEt3, and DBU) in CH3NO2 have
also failed to promote the reaction. The tetrabutylammonium
salts with a counteranion of stronger nucleophilicity, like
chloride and bromide, also failed.
Hence, it is evident that it is the fluoride anion of TBAF
mediating the substitution reaction to yield substituted
PDA’s. Based on all the above results, we proposed a
plausible mechanism for the nitromethylation of PDI in
Scheme 3.
Interestingly, when the reaction was conducted in ni-
tromethane in the presence of 4 equiv of TBAF at ambient
temperature, all starting material PDI was consumed within
15 h. During the reaction course, the color of the reaction
solution changed gradually from orange to pale yellow,
indicating the gradual disappearance of the starting material
PDI. At higher reaction temperature (70 °C), the same
reaction proceeded much faster and finished within 45 min
1
(Table 1). The H NMR spectrum of the obtained product
Scheme 3. Plausible Mechanism for the Nitromethylation of PDI
Table 1. Nitromethylation of PDI Assisted by Fluoride Anion
entrya
TBAF (equiv)
solvent
yieldc (%)
1
2
3
4
5
6
7
8
4
4
4
4
4
4
4
0
4
4
2
1
MeOH
N.R.
N.R.
N.R.
N.R.
N.R.
N.R.
71
N.R.
73
72
CH3CN
CH2Cl2
DMSO
DMF
THF
CH3NO2
CH3NO2
CH3NO2
CH3CNb
CH3CNb
CH3CNb
CH3CNb
When the fluoride encounters the nitromethane molecule,
small equilibrium amounts of HF and nitromethane anion
may form as driven by the unusually strong bonding energy
of H-F (i.e., 136 kcal/mol)11 and the relatively low pKa value
of nitromethane (10.2).11
9
10
11
12
13
70
68
65
0.5
a Reaction conditions: entries 1-8: at rt for 15 h; entries 9-13: at 70
°C for 45 min. b Containing 1 vol % of CH3NO2. c Yield of isolated product
after column chromatography. N.R. - no reaction.
The released HF is acidic enough (unlike MeOH) to
protonate the imino N of PDI and renders the diiminoquinoid
ring 3 reactive enough to be nucleophilically attacked by
the nitromethane anion. The resultant nitromethylated inter-
mediate 4 then could undergo either a 1,3-H shift or a
deprotonation and successively protonation process with the
aid of fluoride as a base to yield the final nitromethylated-
PDA (NM-PDA). The fluoride anion released from the above
reaction cycle and then could induce another cycle of
reactions, which accounts for the catalytic nature of this
reaction. The same mechanism can also explain the inef-
fectiveness of tetrabutylammonium chloride or tetrabuty-
lammonium bromide, because the initial deprotonation of
the nitromethane by Cl- or Br- are much unlikely due to
the much weaker bonding strength of HCl (103 kcal/mol)11
and HBr (88 kcal/mol)11 than HF.
indeed showed the typical spectrum for the monosubstituted
N,N′-diphenyl-1,4-phenylene-diamine (mono-PDA) with the
substituent appearing at the center ring.9a However, both of
1
its H and 13C NMR spectra did not show the expected
1H-19F splitting pattern, instead, the 1H NMR of the isolated
compound showed the presence of an additional singlet peak
at ∼δ 5.5, which did not undergo H-D exchange as D2O
was added. The peak integration indicated that the new
singlet accounts for methylene protons, which was further
confirmed by 13C DEPT and 2D NMR (e.g., HMBC, HSQC)
study. The HRMS data indicated that it is a PDA being
monosubstituted by a moiety having a mass of 60, possibly
corresponding to a -CH2NO2 group instead of a fluorine.
The control experiment indicated that no reaction occurred
between PDI and nitromethane in the absence of TBAF. It
was clear that TBAF is mediating the reaction. Most
interestingly, we found that the same reaction can proceed
under similar conditions even with nitromethane as low as
1 vol % in other solvent mediums such as CH3CN, as well
Interestingly, we found that this reaction could serve as
an efficient way to introduce a C-C bond, if it worked with
an appropriate compound having a suitable latent proton
source. We have systematically screened the scope of the
reaction for a variety of diketones, keto esters, dicyano and
(11) Wade, L. G., Jr. Organic Chemistry, 5th ed.; Prentice Hall: New
York, 2003.
5588
Org. Lett., Vol. 11, No. 24, 2009

cyano esters having an active methylene group and we found
that the reaction worked very well and delivered desired
compounds in good to excellent isolated yield, as summarized
in Table 2. From Table 2, it was also observed that the CRS
group of the PDA backbone system, as driven by the energy
gains from their enlarged conjugation extent and the ad-
ditional H-bonding (or Coulombic) interaction.
Similar enolization has also occurred to the acetylaceto-
nated-PDA (entry 4, Table 2), as evidenced by the missing
benzylic proton and the formation of two different methyl
groups. Interestingly, in the case of 2-nitropropane it was
found that the original nitro group at the benzylic position
seems to undergo further elimination to yield a diamino-
substituted R-methylstyrene. In the case of nitroethane, the
product 6b is stable in its solid form at low temperature, but
in solution state it gradually lost its HNO2 group, forming a
diamino-substituted styrene that readily underwent further
polymerization, possibly being catalyzed by the released
HNO2 acid. Regarding these low pKa Michael donor systems,
the same reaction mechanism as illustrated in Scheme 3
might be applied.
Table 2. Reaction of PDI with Various Michael Donors
Assisted by TBAF
Most interestingly, these novel Michael addition reactions
have also been successfully expanded to the polymer system,
providing the first feasible synthetic method for a direct
addition of a C-based substituent to the benzene rings of
Pan and rendering the possible syntheses of various novel
functionalized polyanilines (Scheme 4) for the first time.
Scheme 4. Preparation of Novel Functionalized Alkylated Pans
a All reactions were carried out at 70 °C in CH3CN containing 1 vol %
of Michael donor and 4 equiv of TBAF. b Yield of isolated product after
column chromatography.
Further studies on these new Pans are under progress and
their application on dye-sensitized solar cell is currently
explored in our Lab.
reaction between the PDI and Michael donors can tolerate
both the electron-donating (entries 1-3, Table 2) as well as
the electron-withdrawing groups (entries 4-8, Table 2) that
appeared on the Michael donors and give excellent isolated
yield (68-80%) in a short span of reaction time (40-55
In conclusion, we have demonstrated a fluoride-assisted
serendipitous C-C bond formation reaction between diimi-
noquionoid ring and a Michael donor in CH3CN. The
reaction proceeds well at 70 °C within 1 h to give moderate
yields (70-80%) and can tolerate both the electron-donating
and electron-withdrawing substituents on the substrate.
Control experiments demonstrated that other stronger nu-
cleophilic halides or stronger bases however failed to promote
the same reaction. A plausible mechanism is also proposed
to account for all the observations.
1
min) at 70 °C in acetonitrile. Interestingly, the H NMR
spectra (in DMSO-d6) of those compounds having a CN
substituent (entries 5-8, Table 2) at the benzylic position
however showed no peaks corresponding to the benzylic
C-H moiety. In addition, their 1H NMR spectra all showed
two different NH peaks, with one peak at ∼δ5.9-6.6
accounting for two protons and another peak at ∼δ6.5-7.9
for one proton (as confirmed by running H-D exchange
experiments). Furthermore, their 13C NMR spectra and their
corresponding 2D NMR (e.g., HMBC, HSQC) studies
confirmed that they all have a quarternary, but not a tertiary
benzylic carbon.
Acknowledgment. We thank the National Science Council
of ROC for supporting this work.
Supporting Information Available: Typical experimental
procedure with the spectral data of all compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
We rationalized that these compounds might have under-
gone further enolization due to the high acidity of the
resultant benzylic protons and the basic nature of the NH
OL9025843
Org. Lett., Vol. 11, No. 24, 2009
5589