Convenient one-pot synthesis of 4-(dialkylaminobenzyl)triphenylphosphonium salts - Application to the synthesis of a fluorescent probe for multiphotonic imaging of biological membranes

Article abstract of DOI:10.1055/s-2003-40521

An efficient one-pot preparation of (dialkylaminobenzyl)triphenylphosphonium salts directly from dialkylbenzenamine is described. The one-pot procedure proceeds quickly in high yields and allows the synthesis of useful reagents for Wittig reaction. Their

Full text of DOI:10.1055/s-2003-40521

PAPER
1541
Convenient One-Pot Synthesis of 4-(Dialkylaminobenzyl)triphenylphosphoni-
um Salts – Application to the Synthesis of a Fluorescent Probe for Multiphoto-
nic Imaging of Biological Membranes
A
minobenzylph
a
osphonium
u
Salts
–
a
Flu
r
orescen
e
t
Probe for
n
Imaging of
B
t
iological
Me
P
mbranes orrès, Bharath Kumar Goud Bhatthula, Mireille Blanchard-Desce*
Synthèse et ElectroSynthèse Organiques (CNRS, UMR 6510) Université de Rennes 1, Campus de Beaulieu, Bât. 10A, 35042 Rennes
Cedex, France
Fax +33(2)23236277; E-mail: Mireille.Blanchard-Desce@univ-rennes1.fr
Received 18 April 2003; revised 13 May 2003
(3 weeks) and we found it was not easily extended to dif-
Abstract: An efficient one-pot preparation of (dialkylaminoben-
ferent types of benzenamines. For instance, the reaction
did not proceed efficiently in the case of phosphonium
salts 6–8.
zyl)triphenylphosphonium salts directly from dialkylbenzenamine
is described. The one-pot procedure proceeds quickly in high yields
and allows the synthesis of useful reagents for Wittig reaction. Their
potential is illustrated by the synthesis of a bolaamphiphilic fluoro-
phore of interest for biological membrane imaging.
In order to overcome these limitations, we have developed
a quicker and more reliable protocol that can be used to
synthesize a number of aminobenzylphosphonium salts
(Scheme 1).
Key words: phosphorus, amines, Wittig reactions, alkenation,
fluorescence, conjugation
Wittig-type olefinations1 are of a wide interest in organic
synthesis due to their versatility, flexibility and efficacy in
carbon–carbon double bond creation. Implementation of
the Wittig reaction beforehand requires the preparation of
phosphonium salts, whose synthesis is in most cases re-
quired. While working on the synthesis of dipolar and
quadrupolar conjugated molecules for applications in the
field of molecular nonlinear optics, we required some 4-
(dialkylaminobenzyl)triphenylphosphonium
halides.
These salts have recently been used for the synthesis of
molecules with nonlinear optical properties2,3 and in the
preparation of biologically active molecules.4 We de-
scribe here an easy one-pot procedure to produce quickly
such phosphonium salts directly from the dialkylben-
zenamine starting material.
Scheme 1
The optimized experimental protocol consists of dissolv-
ing a stoichiometric mixture of amine, triphenylphos-
phine, sodium iodide and solid paraformaldehyde in an
HOAc–H2O–toluene mixture. The heterogeneous reac-
tion mixture is refluxed for several hours until completion
of the reaction (monitored by 1H NMR). After easy work-
up, the phosphonium salts are obtained as stable salts that
can be stored for a long time.
The synthesis of benzylic type phosphonium salts fre-
quently involves the reaction of triphenyphosphine with
benzylic halides. In most cases, this approach requires the
synthesis of reactive halides from the corresponding ben-
zylic alcohols.5 In the case of the dialkylaminobenzyl de-
rivatives, we found it suitable to operate in one-pot
conditions because of the inherent reactivity of such ha-
lide derivatives. By generating the halide reagent in the
presence of the triphenylphosphine, tedious work-up and
isolation of sensitive halide derivatives could be omitted.
We thus developed a one-pot procedure, which allows
their rapid preparation directly from dialkylbenzenamine
(Scheme 1). This procedure was inspired by the route pro-
posed by Bredereck and co-workers for the preparation of
dimethylaminobenzyltriphenylphosphonium iodide.6 The
experimental protocol involves prolonged reaction time
This protocol has been tested with various (aliphatic, cy-
clic) dialkylbenzenamines (Table 1). The variety of start-
ing materials available opens the route to a broad scope of
phosphonium reagents. We observe from Table 1 that the
procedure allows easy and efficient preparation. The reac-
tion time depends on the reactivity of the starting amine.
We note that refluxing in toluene allows shorter reaction
time and/or higher reaction yield than chloroform (see en-
tries 1–9 in Table 1). We also observe that increasing the
aliphatic chains requires shorter reaction time, in relation
to the increased electron-donating character of the
dialkylamino substituent. This facilitates the electrophilic
substitution on the phenyl ring (compounds 1 versus 3,4
in Table 1). This effect is even more pronounced when us-
ing the julolidine cyclic starting amine, which requires the
Synthesis 2003, No. 10, Print: 21 07 2003.
Art Id.1437-210X,E;2003,0,10,1541,1544,ftx,en;P03303SS.pdf.
© Georg Thieme Verlag Stuttgart · New York

1542
L. Porrès et al.
PAPER
Table 1 Preparation of Phosphoniums Salts from Dialkylbenzenamines
Product
R1
R2
Time
3 weeks
62 h
Reflux (Y/N)
Solvent
CHCl3
toluene
CHCl3
toluene
CHCl3
toluene
CHCl3
CHCl3
toluene
CHCl3
CHCl3
CHCl3
toluene
toluene
Yield (%)
1
Me
Me
N
Y
N
Y
N
Y
N
Y
Y
Y
N
Y
Y
Y
866
1
Me
Me
78
2
Bu
Bu
8 weeks
8 h
83
2
Bu
Bu
84
3
Hex
Hex
3 weeks
7 h
85
3
Hex
Hex
86
4
Oct
Oct
3 weeks
24 h
85
4
Oct
Oct
78
5
Julolidine
Et
Julolidine
CH2CH2OH
CH2CH2OH
CH2CH2OH
CH2CH2OH
H
5 h
92
6
39 h
69
7a
7a
7a
8
CH2CH2OH
CH2CH2OH
CH2CH2OH
H
3 weeks
50 h
no reaction
45
56
36
8 h
30 h
NMR: Bruker ARX 200 (1H: 200.13 MHz, 13C: 50.32 MHz) or AC
300 (31P: 121.50 MHz); 1H chemical shifts ( ) are given in ppm in
CDCl3 or DMSO-d6 solutions, respectively, relative to TMS or to
the solvent residual signal at = 2.50; J values are in Hz; 13C chem-
ical shifts are given relative to the solvent residual signal at = 77.0
for CDCl3 solutions or at = 39.5 for DMSO-d6; 31P chemical shifts
are referenced to external 85% aqueous H3PO4. HRMS measure-
ments were performed at the Centre Regional de Mesures Physiques
de l’Ouest, using a Micromass MS/MS ZABSpec TOF instrument
with EBE TOF geometry; LSIMS (Liquid Secondary Ion Mass
Spectrometry) at 8 kV with Cs+ in m-nitrobenzyl alcohol (mNBA).
Elemental analyses were performed at I.C.S.N–C.N.R.S. (Gif-sur-
Yvette, France).
shortest reaction time (compound 5). In the absence of
alkyl chains (compound 8) much longer reaction time is
needed and a lower yield is obtained.
This protocol can be used for the preparation of phospho-
nium salts bearing primary alcohol end groups. The phos-
phonium salts 6 and 7a are of interest for the synthesis of
push-pull dipolar chromophores that can be used for the
elaboration of electro-optical polymeric materials.7 We il-
lustrate the use of phosphonium salt 7a for the synthesis
of a bolaamphiphilic fluorophore 10 (Scheme 2) opti-
mized for multiphotonic imaging of lipidic bilayer or cel-
lular membranes.8
Phosphonium Salts; General Method
In summary, we have developed a very simple and effi-
cient synthesis of dialkylaminobenzylphosphonium salts,
useful not only in the field of material science but also for
the synthesis of molecules for biological applications.
To a solution of N,N-dialkylbenzenamine (22 mmol), Ph3P (5.77 g,
22 mmol) and paraformaldehyde (0.66 g, 7.33 mmol) in CHCl3 or
toluene (30 mL) were added NaI (3.30 g, 22 mmol), H2O (1.56 mL)
and HOAc (4.44 mL). The mixture was refluxed for 5–50 h (see
Table 1). After addition of H2O (40 mL), extraction with CH2Cl2,
washing of the combined organic layers with NaHCO3, then with
+PPh3I-
RO
RO
N
7a: R = H
7b: R = THP
i
KOtBu
CH2Cl2
OHC
CHO
RO
RO
OR
OR
N
N
9: R = THP
10: R = H
ii
Scheme 2 (i) DHP (2.2 equiv), PPTS (0.2 equiv), MeCN, reflux, 40 h, 96%; (ii) aq HCl (10%), CH2Cl2–EtOH (7:3), r.t., 90 h, 52%.
Synthesis 2003, No. 10, 1541–1544 © Thieme Stuttgart · New York

PAPER
Aminobenzylphosphonium Salts – A Fluorescent Probe for Imaging of Biological Membranes
1543
H2O, and drying (Na2SO4), the solvent was evaporated. The residue
was finally purified by recrystallization from EtOH.
{[4-[Ethyl(2-hydroxyethyl)amino]phenyl]methyl}triphenyl-
phosphonium Iodide (6)
Mp 165–170 °C.
{[4-(Dibutylamino)phenyl]methyl}triphenylphosphonium
Iodide (2)
Mp 206–209 °C.
1H NMR (CDCl3): = 7.78 and 7.61 (m, 15 H), 6.75 (dd, J = 8.9,
2.4, 2 H), 6.46 (d, J = 8.7, 2 H), 4.78 (d, J = 12.8, 2 H), 3.68 (m, 2
H), 3.38 (m, 2 H), 3.31 (t, J = 7.0, 2 H), 2.61 (m, 1 H), 1.04 (t,
J = 7.0, 3 H).
13C NMR (CDCl3): = 147.3 (d, J = 2.6), 134.4 (d, J = 2.3), 133.3
(d, J = 9.6), 131.8 (d, J = 5.1), 129.5 (d, J = 12.3), 115.3 (d,
J = 85.0), 111.4, 110.3 (d, J = 8.7), 58.1, 51.4, 44.4, 29.8 (d,
J = 46.3), 11.2.
1H NMR (CDCl3): = 7.80 and 7.68 (m, 15 H), 6.82 (dd, J = 8.5,
1.8, 2 H), 6.37 (d, J = 8.5, 2 H), 4.93 (d, J = 12.7, 2 H), 3.18 (t,
J = 7.3, 4 H), 1.48 (m, 4 H), 1.28 (m, 4 H), 0.93 (t, J = 7.1, 6 H).
13C NMR (CDCl3): = 148.0 (d, J = 1.9), 134.9 (d, J = 2.8), 134.2
(d, J = 9.6), 132.0 (d, J = 4.9), 130.0 (d, J = 12.3), 117.7 (d,
J = 84.8), 111.5, 110.6 (d, J = 7.9), 50.4, 30.5 (d, J = 46.7), 29.0,
20.4, 13.5.
HRMS: m/z calcd for C29H31NOP (M+·): 440.2143, found:
440.2138.
31P NMR (CDCl3): = 20.9.
({4-[Bis(2-hydroxyethyl)amino]phenyl}methyl)triphenylphos-
phonium Iodide (7a)
Mp 175–180 °C.
Anal. Calcd for C33H39INP (607.56): C, 65.24; H, 6.47; I, 20.89; N,
2.31; P, 5.10. Found: C, 65.01; H, 6.62; I, 20.73; N, 2.32; P, 5.28.
{[4-(Dihexylamino)phenyl]methyl}triphenylphosphonium
Iodide (3)
Mp 147–148 °C.
1H NMR (DMSO-d6): = 7.90 and 7.69 (m, 15 H), 6.71 (dd,
J = 8.6, 1.9, 2 H), 6.50 (d, J = 8.8, 2 H), 4.97 (d, J = 14.5, 2 H), 4.70
(t, J = 5.2, 2 H), 3.45 (t, J = 5.1, 4 H), 3.34 (m, 4 H).
1H NMR (CDCl3): = 7.80 and 7.66 (m, 15 H), 6.80 (dd, J = 8.5,
2.1, 2 H), 6.37 (d, J = 8.5, 2 H), 4.85 (d, J = 12.7, 2 H), 3.17 (t,
J = 7.3, 4 H), 1.50 (m, 4 H), 1.29 (m, 12 H), 0.89 (t, J = 7.1, 6 H).
13C NMR (DMSO-d6): = 147.5 (d, J = 2.6), 134.7 (d, J = 2.5),
133.8 (d, J = 9.7), 131.3 (d, J = 5.2), 129.6 (d, J = 12.3), 118.0 (d,
J = 84.7), 111.9 (d, J = 8.8), 111.0, 57.7, 52.8, 27.4 (d, J = 45.8).
13C NMR (CDCl3): = 148.0 (d, J = 2.5), 134.9 (d, J = 2.9), 134.2
(d, J = 9.5), 132.0 (d, J = 5.1), 130.0 (d, J = 12.3), 117.7 (d,
J = 85.0), 111.6, 110.6 (d, J = 8.7), 50.7, 31.5, 30.6 (d, J = 46.3),
26.9, 26.6, 22.5, 13.9.
31P NMR (DMSO-d6): = 22.14.
HRMS: m/z calcd for C29H31NO2P (M+·): 456.2092, found:
456.2084.
31P NMR: (CDCl3): = 20.95.
2,2 ,2 ,2 -{[(1,1 -Biphenyl)-4,4 -diyl-(1E)-2,1-ethenediyl]di-
nitrilo}tetrakisethanol (10)
Anal. Calcd for C37H47INP (663.67): C, 66.96; H, 7.14; N, 2.11.
Found: C, 66.97; H, 7.14; N, 2.21.
To a solution of 7b (1.654 g, 2.2 mmol) and (1,1 -biphenyl)-4,4 -
dicarboxaldehyde (210 mg, 1 mmol) in anhyd CH2Cl2 (15 mL) was
added t-BuOK (370 mg, 3.3 mmol). The mixture was stirred at r.t.
for 5 h. After filtration through Celite, the solvent was evaporated.
The crude compound was isomerized in Et2O–CH2Cl2 (7:2; 9 mL),
with a catalytic amount of I2, at r.t. for 4 h under light exposure (75
W lamp). The organic layer was washed with aq Na2S2O3 and dried
(Na2SO4). The solvent was evaporated and the residue was purified
by column chromatography (CH2Cl2–EtoAC, gradient from 100:0
to 98:2) to yield 9.
{[4-(Dioctylamino)phenyl]methyl}triphenylphosphonium
Iodide (4)
Mp 147–149 °C.
1H NMR (CDCl3): = 7.79 and 7.66 (m, 15 H), 6.80 (dd, J = 8.6,
2.1, 2 H), 6.36 (d, J = 8.6, 2 H), 4.86 (d, J = 12.6, 2 H), 3.17 (t,
J = 7.4, 4 H), 1.49 (m, 4 H), 1.27 (m, 20 H), 0.88 (t, J = 6.4, 6 H).
13C NMR (CDCl3): = 146.6 (d, J = 2.3), 133.7, 133.0 (d, J = 9.9),
130.9 (d, J = 4.3), 128.9 (d, J = 12.4), 116.4 (d, J = 85.0), 110.2,
109.7 (d, J = 9.0), 49.5, 30.4, 29.0 (d, J = 45.3), 28.0, 27.9, 25.7,
21.3, 12.8.
Yield: 1.586 g (80%).
1H NMR (CDCl3): = 7.61 and 7.54 (AA XX , JAX = 8.7, 8 H), 7.41
and 6.75 (AA XX , JAX = 8.9, 8 H), 7.08 (d, J = 15.8, 2 H), 6.92 (d,
J = 16.2, 2 H), 4.61 (m, 4 H), 3.87 (m, 8 H), 3.70–3.47 (m, 16 H),
1.55 (m, 24 H).
31P NMR (121.50 MHz, CDCl3): = 21.04.
Anal. Calcd for C41H55INP (719.78): C, 68.42; H, 7.70; N, 1.94.
Found: C, 68.42; H, 7.71; N, 1.90.
HRMS: m/z calcd for C56H72N2O8 (M+·): 900.5289, found:
900.5289.
Triphenyl[(2,3,6,7-tetrahydro-1H,5H-benzo[ij]quinolizin-9-
yl)methyl]phosphonium Iodide (5)
Mp 185–189 °C.
1H NMR (CDCl3): = 7.80 and 7.63 (m, 15 H), 6.25 (d, J = 2.6, 2
H), 4.65 (d, J = 12.6, 2 H), 3.09 (t, J = 5.5, 4 H), 2.44 (t, J = 6.2, 4
H), 1.83 (m, 4 H).
13C NMR (CDCl3): = 142.7, 135.0 (d, J = 2.9), 134.2 (d, J = 9.9),
130.0 (d, J = 12.3), 129.5 (d, J = 5.3), 121.6, 117.6 (d, J = 85.0),
110.3, 49.6, 31.0 (d, J = 46.4), 27.2, 21.4.
Product 9 was dissolved in EtOH–CH2Cl2 (3:7; 10 mL) and aq HCl
(10%; 1 mL) was added. After stirring for 90 h at r.t., the residue
was filtered and washed with H2O and EtOH to yield 10.
Yield: 517 mg (52%); mp >250 °C.
1H NMR (DMSO-d6): = 7.67 and 7.59 (AA XX , JAX = 8.2, 8 H),
7.41 and 6.70 (AA XX , JAX = 8.3, 8 H), 7.16 (d, J = 16.2, 2 H), 6.96
(d, J = 16.2, 2 H), 4.74 (m, 4 H), 3.55 (m, 8 H), 3.31 (m, 8 H).
31P NMR (CDCl3): = 20.60.
HRMS: m/z calcd for C31H31NP (M+·): 448.2194, found: 448.2193.
Acknowledgments
We acknowledge financial support from C.N.R.S. and M.E.N.R.T.
Synthesis 2003, No. 10, 1541–1544 © Thieme Stuttgart · New York

1544
L. Porrès et al.
PAPER
(4) Banner, B. L.; Weber, G.-F. US 5451600, 1995.
References
(5) An alternative proceeds directly from the alcohol by using
the Ph3P·HBr salt, but this procedure requires high thermal
activation and a costly reagent: (a) Zhang, J.-X.; Dubois, P.;
Jérôme, R. Synth. Commun. 1996, 26, 3091. (b) Samyn, C.;
Heylen, M.; Claes, G.; Boutton, C.; Van Beylen, M.;
Persoons, A. Eur. Polym. J. 1998, 34, 1069.
(6) Bredereck, H.; Simchen, G.; Griebenow, W. Chem. Ber.
1973, 106, 3732.
(7) Dalton, L. R.; Jen, A. K.-Y.; Londergan, T.; Carlson, W. B.;
Phelan, G.; Huang, D.; Casmier, D.; Ewy, T.; Buker, N. WO
0208215, 2002.
(8) (a) Ventelon, L.; Charier, S.; Moreaux, L.; Mertz, J.;
Blanchard-Desce, M. Angew. Chem. Int. Ed. 2001, 40,
2098. (b) Blanchard-Desce, M. C. R. Physique 2002, 3, 439.
(c) Porrès, L.; Mongin, O.; Bhatthula, B. K. G.; Blanchard-
Desce, M.; Ventelon, L.; Moreaux, L.; Pons, T.; Mertz, J.
Proc. SPIE-Int. Soc. Opt. Eng. 2002, 4812, 24.
(1) (a) Maryanoff, B. E.; Reitz, A. B. Chem. Rev. 1989, 89, 863.
(b) Nicolaou, K. C.; Harter, M. W.; Gunzner, J. L.; Nadin, A.
Liebigs Ann. Recueil. 1997, 1283.
(2) (a) Londergan, T.; Todorova, G.; Zhu, J.; Huang, D. WO
0214305, 2002. (b) Maertens, C.; Zhang, J.-X.; Dubois, P.;
Jérôme, R. J. Chem. Soc., Perkin Trans. 2 1996, 713.
(c) Marder, S.; Perry, J. US 6267913, 2001. (d) Segura, J.
L.; Gomez, R.; Martin, N.; Guldi, D. M. Org. Lett. 2001, 3,
2645. (e) Rumi, M.; Ehrlich, J. E.; Heikal, A. A.; Perry, J.
W.; Barlow, S.; Hu, Z.; McCord-Maughon, D.; Parker, T. C.;
Röckel, H.; Thayumanavan, S.; Marder, S. R.; Beljonne, D.;
Brédas, J.-L. J. Am. Chem. Soc. 2000, 122, 9500.
(3) (a) Puccetti, G.; Blanchard-Desce, M.; Ledoux, I.; Lehn, J.
M.; Zyss, J. J. Phys. Chem. 1993, 97, 9385. (b) Alain, V.;
Redoglia, S.; Blanchard-Desce, M.; Lebus, S.; Lukaszuk,
K.; Wortmann, R.; Gubler, U.; Bosshard, C.; Gunter, P.
Chem. Phys. 1999, 245, 51. (c) Brunel, J.; Ledoux, I.; Zyss,
J.; Blanchard-Desce, M. Chem. Commun. 2001, 923.
Synthesis 2003, No. 10, 1541–1544 © Thieme Stuttgart · New York