Electropolymerization of Pd(II) complexes containing phosphinoterthiophene ligands

Article abstract of DOI:10.1021/ja016465m

A series of Pd complexes of 3′-diphenylphosphino-2,2′:5′2″-terthiophene (1a, dppterth) in which the metal is coordinated in three different modes have been prepared and electropolymerized, resulting in the formation of conductive thin films. In [Pd_2<

Full text of DOI:10.1021/ja016465m

J. Am. Chem. Soc. 2001, 123, 9963-9973
9963
Electropolymerization of Pd(II) Complexes Containing
Phosphinoterthiophene Ligands
Olivier Clot, Michael O. Wolf,* and Brian O. Patrick
Contribution from the Department of Chemistry, The UniVersity of British Columbia,
VancouVer, British Columbia V6T 1Z1, Canada
ReceiVed June 20, 2001
Abstract: A series of Pd complexes of 3′-diphenylphosphino-2,2′:5′2′′-terthiophene (1a, dppterth) in which
the metal is coordinated in three different modes have been prepared and electropolymerized, resulting in the
formation of conductive thin films. In [Pd2(µ-Cl2)(dppterth-P,C3)2] (3a) the metal is P,C-coordinated, in [PdCl2-
(dppterth-P)2] (4a) the coordination is monodentate via the phosphine, and in [Pd(dppterth-P,C3)(dppterth-
P,S1)][PF6] (5a) both P,C- and P,S-coordination modes are found. In 5a, the coordinated thiophene is hemilabile
and may be displaced by reaction with more strongly coordinating ligands such as isocyanides. To probe the
effect of blocking the R-position of the terthienyl moiety with methyl groups, 3′-diphenylphosphino-5-methyl-
2,2′:5′2′′-terthiophene (1b, Me-dppterth) and 3′-diphenylphosphino-5,5′′-dimethyl-2,2′:5′2′′-terthiophene (1c,
Me2-dppterth) were prepared, and the corresponding series of Pd complexes was synthesized. One of these
complexes, [Pd(Me2-dppterth-P,C3)(Me2-dppterth-P,S1)][PF6] (5c), has been crystallographically characterized.
The electropolymerized films prepared from 5a react with isonitriles, and shifts in the absorption spectra of
the electropolymerized materials are observed upon reaction. A Pd complex has also been prepared from
5-diphenylphosphino-2,2′:5′2′′-terthiophene (2, 5dppterth), and this complex has been electropolymerized. All
the electropolymerized thin films have been characterized using EDX analysis, which demonstrates good
correspondence with the elemental analysis of the respective monomers, and the maximum conductivities of
the films are near 10-4 S cm-1. Comparing the electropolymerization behavior of the complexes, along with
their electrochemical and spectroscopic data, allows conclusions to be drawn regarding the involvement of
π-delocalization and the metal group in the conductivity of the materials.
Introduction
coupling results when a saturated linker is present. Functional
materials based on conjugated frameworks in which the reactiv-
ity of the metal center may be tuned by using the redox
characteristics of the backbone have also been explored.8,9
Materials with π-conjugated backbones are being intensively
studied because of their interesting electronic and optical
properties, which include high conductivities, electrolumines-
cence, and photochromism.1 π-Conjugated materials which
incorporate transition metals are of significance, because the
metal has the capacity to influence the overall properties of these
materials, possibly giving rise to unprecedented or enhanced
properties.2,3 A wide range of possible applications is being
explored for these materials, including applications as photo-
refractive materials,4 as chemosensory materials,5,6 and in
electroluminescent devices.7 In all of these applications, the
nature of the electronic interaction between the metal and the
conjugated backbone is of central importance, and this is
influenced by the bonding between the metal and the backbone.
The metal may be strongly electronically coupled by direct
coordination to (or insertion in) the backbone, while weaker
Our own efforts have focused on metal-containing hybrid
materials in which the π-backbone is an oligo- or polythio-
phene,10-12 principally because thiophene-based materials are
readily synthetically accessible, relatively stable, and may be
polymerized via electrochemical routes. A structural motif which
has been less explored involves the metal center linking
conjugated chains leading to cross-linked networks; in this case,
the metal may be involved in charge transport through the bulk
material. Weder and co-workers recently demonstrated such a
hybrid material in which Pt centers bridge poly(p-phenylene
ethynylene) chains, resulting in changes in the photophysical
behavior of the polymer.13 In a recent Communication, we
reported the first example of a polythiophene derivative in which
conjugated chains are linked by Pd bridges in which the metal
is σ-bonded to a C of the backbone.14 Here, we further explore
* To whom correspondence should be addressed. E-mail: mwolf@
chem.ubc.ca. Fax: 604-822-2847.
(1) Skotheim, T. A.; Elsenbaumer, R. L.; Reynolds, J. R. Handbook of
Conducting Polymers, 2nd ed.; Marcel Dekker: New York, 1998.
(2) Kingsborough, R. P.; Swager, T. M. Prog. Inorg. Chem. 1999, 48,
123-231.
(3) Wolf, M. O. AdV. Mater. 2001, 13, 545-553.
(4) Peng, Z.; Yu, L. J. Am. Chem. Soc. 1996, 118, 3777-3778.
(5) Zhu, S. S.; Carroll, P. J.; Swager, T. M. J. Am. Chem. Soc. 1996,
118, 8713-8714.
(6) Zhu, S. S.; Swager, T. M. J. Am. Chem. Soc. 1997, 119, 12568-
12577.
(8) Weinberger, D. A.; Higgins, T. B.; Mirkin, C. A.; Liable-Sands, L.
M.; Rheingold, A. L. Angew. Chem., Int. Ed. Engl. 1999, 38, 2565-2568.
(9) Weinberger, D. A.; Higgins, T. B.; Mirkin, C. A.; Stern, C. L.; Liable-
Sands, L. M.; Rheingold, A. L. J. Am. Chem. Soc. 2001, 123, 2503-2516.
(10) Zhu, Y.; Millet, D. B.; Wolf, M. O.; Rettig, S. J. Organometallics
1999, 18, 1930-1938.
(11) Zhu, Y.; Wolf, M. O. Chem. Mater. 1999, 11, 2995-3001.
(12) Clot, O.; Wolf, M. O.; Yap, G. P. A.; Patrick, B. O. J. Chem. Soc.,
Dalton Trans. 2000, 2729-2737.
(13) Huber, C.; Bangerter, F.; Caseri, W. R.; Weder, C. J. Am. Chem.
Soc. 2001, 123, 3857-3863.
(7) Wong, C. T.; Chan, W. K. AdV. Mater. 1999, 11, 455-459.
10.1021/ja016465m CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/25/2001

9964 J. Am. Chem. Soc., Vol. 123, No. 41, 2001
Clot et al.
Scheme 1
is most likely cis to the σ-bonded carbon, as is typically the
case in related compounds.15
the materials formed by electropolymerization of Pd(II) com-
plexes containing phosphinoterthiophene ligands exhibiting
several different bonding modes. This gives insight into
structural control and electronic interactions in such cross-linked
materials.
Results
Syntheses and Structures of Ligands and Complexes. The
phosphinoterthiophene ligand 3′-diphenylphosphino-2,2′:5′2′′-
terthiophene (1a, dppterth)12 has been previously reported. The
related mono- and dimethyl terminated derivatives of dppterth
(1b and 1c) have been prepared in an attempt to address
questions related to the electropolymerization of metal com-
plexes containing these ligands (vide infra). Compounds 1b (Me-
dppterth) and 1c (Me2-dppterth) were synthesized by the reaction
of 3′-bromo-5-methyl-2,2′:5′2′′-terthiophene (11) and 3′-bromo-
5,5′-dimethyl-2,2′:5′2′′-terthiophene (12), respectively, with
n-butyllithium at -15 °C followed by reaction with PPh2Cl at
room temperature (Scheme 1). Compounds 1b and 1c were
isolated as bright yellow powders, which oxidized slowly in
air. The structurally allied compound 5-diphenylphosphino-2,2′:
5′2′′-terthiophene (2, 5dppterth) which is expected to bind in a
monodentate fashion via the terminal phosphine has been
prepared using a similar procedure.
The products obtained from the reactions of 1a-c with PdCl2
are dependent on the amount of metal halide used in the reaction.
Complexes 3a-c were synthesized via the reactions of 1a-c
in ethanol/acetonitrile solution with a small excess of PdCl2 in
water (Scheme 2). Upon addition of the phosphine to the
solution of PdCl2, the products 3a-c immediately precipitated
and were collected by filtration as yellow-orange, air-stable
powders.
The structures of 3b,c were assigned by comparison of their
NMR spectra with those of 3a, whose solid-state structure was
reported previously.14 Because all three ligands 1a-c are closely
related, we assign the geometry of 3b,c to dimetallic complexes
in which the Pd centers are in a square planar environment with
bridging chlorides and the phosphinoterthiophene ligands bound
via the phosphine and C3 of the terthienyl group. If an excess
of 1a dissolved in ethanol is added to PdCl2 in water/ethanol
solution cooled to 0 °C, cyclopalladation does not take place,
and the resulting products are exclusively 4a and unreacted
PdCl2, with no 3a detected by NMR.
Complex 3a was shown previously to react with phosphines
and isocyanides to give mononuclear complexes via cleavage
of the dichloro bridge.14 Similarly, the dichloro bridge in 3b
can also be cleaved, and the reaction with tolyl isocyanide yields
complex 7b in which the isocyanide binds in a terminal fashion
to Pd (νNC ) 2193 cm-1 in chloroform). In 7b, the isocyanide
When PdCl2 in water is added to an excess of the appropriate
phosphinoterthiophene ligand in ethanol at 50 °C, complexes
4a-c and 6 are obtained in good yield (Scheme 2). These
complexes are not soluble in the solvent mixture and rapidly
precipitate out of the solution as yellow, air-stable solids during
the reaction. The 31P NMR spectra of all four complexes contain
only one singlet at ∼δP 10, indicating that the phosphines are
equivalent, and elemental analyses are consistent with the
general formula [PdCl2(phosphinoterthiophene)2]. The far-IR
spectrum of complex 4a contains a strong band at 357 cm-1
corresponding to the Pd-Cl stretch, indicating that the chlorines
are trans and resulting in an all-trans square planar geometry
common to many Pd bis(phosphine) complexes.16,17 The
spectroscopic data for compounds 4b,c and 6 indicate that these
have analogous structures to 4a.
Chloride abstraction from bis(phosphine) complexes 4a-c
using 2 equiv of AgBF4, followed by metathesis with NH4PF6,
results in formation of monocationic complexes 5a-c in good
yield (Scheme 2). All three complexes were isolated as orange,
air-stable powders. Although attempts to grow crystals of 5a
were unsuccessful, small platelet-like crystals of 5c were grown
from a chloroform/diethyl ether mixture, and the solid-state
structure was established by X-ray crystallography. The structure
is shown in Figure 1, and selected bond lengths and angles are
collected in Table 1. The two Me2-dppterth ligands bind
differently to the Pd; one ligand coordinates via a P,C mode
resulting in a six-membered ring with a P-Pd-C bite angle of
80.4(3)°. The second ligand coordinates in a P,S mode with
the sulfur coordinated in an η1 fashion via the terminal thiophene
ring, resulting in a six-membered ring with a P-Pd-S bite angle
of 84.96(9)°. The complex is distorted from planarity toward a
tetragonal geometry (S-Pd-C ) 168.4(3)° and P-Pd-P )
163.56(9)°).
A number of species have been reported containing a bond
or interaction between palladium and the sulfur of a thiophene
ring;18-26 however, only two of these have been structurally
characterized.25,26 In [Pd(SDPDTP)]Cl (SDPDTP ) 5,20-
(15) Hayashi, Y.; Isobe, K.; Nakamura, Y.; Okeya, S. J. Organomet.
Chem. 1986, 310, 127-134.
(16) ComprehensiVe Organometallic Chemistry: the Synthesis, Reactions
and Structures of Organometallic Compounds; Wilkinson, G., Ed.; Perga-
mon: Oxford, 1982; Vol. 6.
(17) ComprehensiVe Organometallic Chemistry II: A ReView of the
Literature 1982-1994; Abel, E. W., Stone, G. A., Wilkinson, G., Eds.;
Pergamon: Oxford, 1995; Vol. 9.
(14) Clot, O.; Wolf, M. O.; Patrick, B. O. J. Am. Chem. Soc. 2000, 122,
10456-10457.

Electropolymerization of Pd(II) Complexes
J. Am. Chem. Soc., Vol. 123, No. 41, 2001 9965
Scheme 2
Table 1. Selected Bond Lengths and Angles for 5c
Bond Length, Å
Pd(1)-S(1)
Pd(1)-P(1)
S(1)-C(2)
S(1)-C(5)
C(2)-C(3)
C(3)-C(4)
C(4)-C(5)
2.427(3)
Pd(1)-C(18)
Pd(1)-P(2)
S(4)-C(16)
S(4)-C(19)
C(16)-C(17)
C(17)-C(18)
C(18)-C(19)
2.01(1)
2.307(3)
1.73(1)
1.76(1)
1.37(1)
1.41(1)
1.39(1)
2.352(2)
1.74(1)
1.745(9)
1.34(2)
1.41(2)
1.34(1)
Bond Angle, deg
S(1)-Pd(1)-C(18) 168.4(3)
P(1)-Pd(1)-P(2)
163.56(9)
108.2(1)
80.4(3)
103.9(3)
121.0(3)
116.0(3)
92.5(5)
S(1)-Pd(1)-P(1)
P(1)-Pd(1)-C(18)
Pd(1)-P(1)-C(7)
84.96(9) S(1)-Pd(1)-P(2)
88.1(3)
112.5(3)
Pd(1)-P(1)-C(29) 113.2(3)
Pd(1)-P(1)-C(35) 115.0(3)
P(2)-Pd(1)-C(18)
Pd(1)-P(2)-C(21)
Pd(1)-P(2)-C(41)
Pd(1)-P(2)-C(47)
C(16)-S(4)-C(19)
S(4)-C(16)-C(17)
S(4)-C(19)-C(18)
C(2)-S(1)-C(5)
S(1)-C(2)-C(3)
S(1)-C(5)-C(4)
C(2)-C(3)-C(4)
C(3)-C(4)-C(5)
92.8(5)
109.3(8)
108.3(8)
114(1)
109.7(8)
110.5(8)
Figure 1. X-ray crystal structure of 5c. The hydrogen atoms, phenyl
groups, and PF6 anion are omitted for clarity. Thermal ellipsoids are
drawn at 30% probability.
C(16)-C(17)-C(18) 116(1)
115(1)
C(17)-C(18)-C(19) 111(1)
diphenyl-10,15-ditolyl-21-thiaporphyrin),26 the Pd-S bond length
(2.208(5) Å) is shorter than the corresponding bond in 5c and
shorter than the sum of the covalent radii of Pd and S (2.35
Å).27 On the other hand, in [Pd(µ-C3H5)(L)][CF3SO3] (L )
2,5,8-trithia[9](2,5)thiophenophane),25 the Pd-S (thiophene)
distance (2.786(4) Å) is much longer than both the Pd-S bond
in 5c and the sum of the covalent radii; however, because the
Pd-S distance is shorter than the sum of the van der Waals
radii (3.4 Å),28 an apical Pd-S bond was postulated.25 Many
of the metal-thiophene sulfur bonds reported in the literature
are near or slightly longer than the sum of the covalent
radii,12,25,29-40 and this is also the case for 5c where the Pd-S
(18) Russell, M. J. H.; White, C.; Yates, A.; Maitlis, P. M. J. Chem.
Soc., Dalton Trans. 1978, 857-861.
(19) Larock, R. C.; Leach, D. R.; Bjorge, S. M. Tetrahedon Lett. 1982,
23, 715-718.
(20) Athar, M.; Ahmed, N. Indian J. Chem., Sect. A 1983, 22A, 1055-
1057.
(21) Joginadham, C.; Bhagwat, U. A.; Mukhedkar, V. A.; Mukhedkar,
A. J. Indian J. Chem., Sect. A 1987, 26A, 341-344.

9966 J. Am. Chem. Soc., Vol. 123, No. 41, 2001
Clot et al.
distance is 2.427(3) Å. The thiophene ring plane has a tilt angle
of 141° with respect to the Pd-S bond, which is within the
range typically observed for η1-S thiophene bound to various
metals.12,41-43 The rather large value of the tilt angle suggests
that there is significant back-bonding to the thiophene.44
The bond lengths and angles of the sulfur-coordinated
thiophene in 5c differ from the analogous values in thiophene.45
The coordinated ring is distorted in a fashion similar to that of
the coordinated thiophenes in ruthenium complexes in which
dppterth is chelated in a P,S fashion with longer CsS and
shorter CdC bonds than in thiophene.12 Similar distortions have
also been observed in other metal-thiophene complexes and
are believed to be due to a slight loss of aromaticity in the ring
upon coordination.46 In compound 5c, it is also possible to
compare within the same complex the relative distortions
imposed on the terthienyl moieties by the P,C- and P,S-
coordination modes. Most of the bond lengths and angles within
the two terthienyl moieties are comparable, except for those in
the terminal ring that is coordinated to the metal. The deforma-
tions in the C-coordinated thiophene are larger than those in
the S-coordinated ring, indicating that coordination of thiophene
to palladium via the sulfur is less disruptive to the aromaticity
than the formation of a metalscarbon σ-bond between Pd and
thiophene.
Figure 2. 31P NMR spectrum of 5b in CDCl3 showing a mixture of
(A) trans- and (B) cis-[Pd(Me-dppterth-P,C3)(Me-dppterth-P,S1)][PF6].
The 31P NMR spectra of 5a-c all contain two pairs of
doublets due to the phosphines, indicating the presence of a
mixture of two products in solution, as well as a septet
corresponding to the PF6 anion (Figure 2). The pair of doublets
with the larger JPP corresponds to the complex in which the
phosphines are trans while the other pair is due to a cis
orientation of the phosphines. The structures of 5a,b were
established to be analogous to that of 5c with one of the ligands
P,S-coordinated via the terminal thiophene bound η1(S) and the
other ligand chelating P,C via the C3 of the terthienyl moiety,
by comparing their NMR spectra with that of 5c.
(22) Jain, C. L.; Mundley, P. N.; Kumar, Y.; Sethi, P. D. J. Indian Chem.
Soc. 1989, 66, 431-434.
(23) Li, C. S.; Jou, D. C.; Cheng, C. H. Organometallics 1993, 12, 3945-
3954.
The cis/trans ratios of 5a-c are solvent dependent. Solutions
of 5c contain ∼30% of the cis compound in CDCl3, ∼50% in
acetone-d6, and near 0% in MeOH-d4 (ratios determined by
NMR). For compounds 5a,b, the cis/trans ratios in solution are
∼6% cis for 5a and ∼34% cis for 5b (in CDCl3, determined
by NMR). Cis-trans isomerization in these complexes may
occur via two different pathways: either via a trigonal inter-
mediate by dissociation of the coordinated thiophene ring or
via a nondissociative pathway, involving a tetragonal intermedi-
ate in which the two dppterth ligands remain chelated to Pd.
Dissociative pathways are generally promoted by the presence
of coordinating solvents such as MeOH. However, no evidence
of coordination of MeOH was detected in the 31P NMR spectrum
of 5c in CDCl3 when a large excess (∼10 000 equiv) of MeOH
was added, and virtually no cis isomer was detected by NMR
in MeOH-d4. Because in the solid state 5c has a pseudotetra-
hedral structure, and complexes such as [NiBr2(PPh2Me)2] are
known to isomerize between square planar and tetrahedral
geometries,47 the nondissociative pathway is more likely to be
involved in the cis-trans isomerization of 5a-c.
(24) Liu, S.; Lucas, C. R.; Newlands, M. J.; Charland, J.-P. Inorg. Chem.
1990, 29, 4380-4385.
(25) Lucas, C. R.; Liu, S.; Newlands, M. J.; Gabe, E. J. Can. J. Chem.
1990, 68, 1357-1363.
(26) Latos-Grazynski, L.; Lisowski, J.; Chmielewski, P.; Grzeszczuk,
M.; Olmstead, M. M.; Balch, A. L. Inorg. Chem. 1994, 33, 192-197.
(27) Dean, J. A. Lange’s Handbook of Chemistry; 15th ed.; McGraw-
Hill: New York, 1999.
(28) Bondi, A. J. Phys. Chem. 1964, 68, 441-451.
(29) Bucknor, S. M.; Draganjac, M.; Rauchfuss, T. B.; Ruffing, C. J.;
Fultz, W. C.; Rheingold, A. L. J. Am. Chem. Soc. 1984, 106, 5379-5381.
(30) Draganjac, M.; Rauchfuss, T. B.; Rheingold, A. L. Proc. Arkansas
Acad. Sci. 1992, 46, 36-38.
(31) Van Stein, G. C.; Van Koten, G.; Blank, F.; Taylor, L. C.; Vrieze,
K.; Spek, A. L.; Duisenberg, A. J. M.; Schreurs, A. M. M.; Kojic-Prodic,
B.; Brevard, C. Inorg. Chim. Acta 1985, 98, 107-120.
(32) Goodrich, J. D.; Nickias, P. N.; Selegue, J. P. Inorg. Chem. 1987,
26, 3424-3426.
(33) Constable, E. C.; Henney, R. P. G.; Tocher, D. A. J. Chem. Soc.,
Chem. Commun. 1989, 913-914.
(34) Latos-Grazynski, L.; Lisowski, J.; Olmstead, M. M.; Balch, A. L.
Inorg. Chem. 1989, 28, 1183-1188.
(35) Benson, J. W.; Angelici, R. J. Organometallics 1992, 11, 922-
927.
(36) Draganjac, M.; Ruffing, C. J.; Rauchfuss, T. B. Organometallics
1985, 4, 1909-1911.
(37) Alvarez, M.; Lugan, N.; Mathieu, R. Inorg. Chem. 1993, 32, 5652-
5657.
In complexes 5a-c, the dppterth ligand that is P,S-
coordinated is hemilabile. For example, when 5a reacts with
tolyl isocyanide in methylene chloride at 25 °C, the resulting
product 8a shows a new IR band at 2194 cm-1, characteristic
of a C-coordinated NtC group. The 31P NMR spectrum of 8a
contains a doublets of doublets (JPP ) 388 Hz) similar to that
obtained for trans-5a. The chemical shifts in trans-5a and 8a
are very close, indicating that the phosphines in the two
complexes are in similar environments. On the basis of the
(38) Latos-Grazynski, L.; Olmstead, M. M.; Balch, A. L. Inorg. Chem.
1989, 28, 4065-4066.
(39) Deeming, A. J.; Shinhmar, M. K.; Arce, A. J.; Sanctis, Y. D. J.
Chem. Soc., Dalton Trans. 1999, 1153-1159.
(40) Hanton, L. R.; Turnbull, J. M.; Richardson, C.; Robinson, W. T.
Chem. Commun. 2000, 2465-2466.
(41) Angelici, R. J. Bull. Soc. Chim. Belg. 1995, 104, 265-282.
(42) Angelici, R. J. NATO Sci. Ser., Ser. 3 1998, 60, 89-127.
(43) Rauchfuss, T. B. Prog. Inorg. Chem. 1991, 39, 259-329.
(44) Harris, S. Polyhedron 1997, 16, 3219-3233.
(45) Bak, B.; Christensen, D.; Hansen-Nygaard, L.; Rastrup-Andersen,
J. J. Mol. Spectrosc. 1961, 7, 58-63.
(46) Choi, M. G.; Angelici, R. J. J. Am. Chem. Soc. 1989, 111, 8753-
8754.
(47) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry; John
Wiley & Sons: New York, 1988.

Electropolymerization of Pd(II) Complexes
J. Am. Chem. Soc., Vol. 123, No. 41, 2001 9967
Table 2. Cyclic Voltammetry Data
spectroscopic data, we conclude that in the reaction of 5a with
Ep,ox,a (0.01 V
Ep,red,a (0.01 V
isocyanide the coordinated thiophene is displaced from the Pd.
complex
1a
1b
1c
2
+1.30
+1.18
+1.05
+0.92
3ab
3b
3c
4a
4b
4c
5a
5b
5c
6
+1.20
-1.31
-1.42
-1.22
-1.24
-1.43
-1.30
-1.32
-1.07
-1.11
+1.10 (sh), +1.13
+0.91,c,d +0.97,c,d +1.35
+1.27, +1.53
+1.34, +1.52
+1.07,c,d +1.15,c,d +1.47
+1.31
+1.18, 1.39
+1.05,c+1.42, +1.73
+1.24, +1.49, +2.03
a Volts vs SCE, Pt working electrode, CH2Cl2 containing 0.1 M [(n-
Bu)4N]PF6, 20 °C. bReference.14 cE1/2. dResolved via semiderivative.49
Although 5a exists as a cis/trans mixture, only one isomer
of 8a is detected by 31P NMR. It is possible that the incoming
isocyanide reacts more readily with the trans isomer of 5a,
yielding only 8a. Another possibility is that the displacement
reaction occurs via a trigonal intermediate, in which the
structural strain observed in the structure of 5c is removed. In
5c, the Pd(II) center lies in an unfavorable geometry deviating
from planarity (Figure 1). Upon coordination of isocyanide,
displacement of the coordinated thiophene causes one of the
dppterth ligands to become monodentate, yielding a complex
containing one chelating and two monodentate ligands. This
should allow the resulting complex to achieve a more favorable
square planar geometry, similar to that of bis(phosphine)
complexes 4a,b, where the phosphines are trans, therefore
yielding 8a as the trans isomer only.
Complexes 5b,c also react in a similar fashion with isocya-
nides to yield products in which the isocyanide is bound via
the terminal carbon. For instance, when 5b was treated with
tolyl isocyanide, the product contains a new band at 2191 cm-1
in the IR region, characteristic of a C-coordinated isocyanide.
Cyclic Voltammetry of Ligands and Complexes. The cyclic
voltammograms (CVs) of compounds 1a-c and 2 contain one
irreversible oxidation wave (Table 2) with peak potentials close
to that measured for terthiophene (Ep ) +1.05 V).48 The peak
oxidation potential of 1a to 1c decreases from +1.30 to +1.05
V, because of the addition of the electron donating methyl
groups. No reduction wave was observed between 0 and -1.5
V for any of the compounds. Compound 2 has the lowest
oxidation potential of all the terthienyl-based compounds
reported here, possibly because of the stabilizing influence of
the phosphine group on the resulting cation.
None of these phosphinoterthiophene compounds electropo-
lymerized by sweeping the potential repeatedly past the oxida-
tion wave, as evidenced by the absence of film growth on the
electrode. For 1b,c and 2, this result is due in part to the lack
of two free R-positions on the oligothiophene necessary for
polymerization. In all the compounds, electropolymerization
may also be prevented by the nucleophilicity of the free
Figure 3. Cyclic voltammetry of 4a in CH2Cl2 containing 0.1 M [(n-
Bu)4N]PF6, scan rate ) 200 mV/s. (a) Multiple scans from 0 to +1.70
V for 4a. Inset: first scan of 4a. (b) Scan of a film of poly-4a on a
gold electrode.
phosphine toward the radical cation formed on the oligoth-
iophene upon oxidation. This is likely the reason 1a does not
electropolymerize, despite two free R-positions and a low
oxidation potential.
The cyclic voltammetry of the Pd complexes was carried out
in CH2Cl2 containing 0.1 M [(n-Bu)4N]PF6, and the data are
summarized in Table 2. The CVs of 3a, 4a, and 5a show an
oxidation wave near +1.3 V versus SCE and, in the case of 4a,
a second oxidation at +1.53 V (Figure 3a, inset). For 5a, two
overlapping reduction features are visible on the reverse scan
(Figure 4a, 1st scan), possibly because of reduction of oligomers
of different lengths formed upon scanning past the oxidation
wave. The CVs of the complexes containing 1b all contain two
oxidation waves (Table 2). For complex 3b, these waves are
(48) Diaz, A. F.; Crowley, J.; Bargon, J.; Gardini, G. P.; Torrance, J. B.
J. Electroanal. Chem. Interfacial Electrochem. 1981, 121, 355-361.

9968 J. Am. Chem. Soc., Vol. 123, No. 41, 2001
Clot et al.
Figure 4. Cyclic voltammetry of 5a and 5b in CH2Cl2 containing 0.1 M [(n-Bu)4N]PF6, scan rate ) 200 mV/s. (a) Multiple scans from +0.20 to
+1.60 V for 5a. (b) Scan of a film of poly-5a on a gold electrode. (c) Multiple scans from +0.25 to +1.50 V for 5b. (d) Scan of a film of poly-5b
on a gold electrode.
potential is scanned past the second oxidation wave (+1.42 V).
A third oxidation wave at higher potential is also consistently
observed in the CVs of all three complexes containing 1c. In
these complexes, the lowest potential waves are due to oxidation
of the Me2-dppterth ligands.
The number of electrons n involved in the oxidation waves
for 4c and 5c was determined by adding an internal standard
(cis-[RuCl2(dmpe)2], dmpe ) bis(dimethylphosphino)ethane) to
the solution. This standard was chosen because the RuII/III redox
couple is removed from the oxidation peaks of the Pd
complexes. The semiderivatives of the voltammograms of these
solutions were fitted according to the procedure of Toman and
Brown,49 and the resulting peak heights are then proportional
to n2. For 4c, the ratio of the peak heights for each of the two
resolved oxidation waves to the Ru wave is 0.80; for 5c, the
ratio of the height of the single oxidation wave to the Ru wave
is 0.75 (after correcting for concentration differences between
the standard and compound). The ratios are both less than unity,
consistent with the larger diffusion coefficient expected for the
smaller Ru complex. Assuming similar diffusion coefficients
for 4c and 5c, the similarity between these two ratios shows
that the overlapping waves in 4c are due to two one-electron
oxidations, and the single wave in 5c is due to a one-electron
oxidation. The semiderivative peak widths at half-height, which
are inversely proportional to n, are also close to those measured
for cis-[RuCl2(dmpe)2], confirming this conclusion.
Figure 5. Cyclic voltammetry of 4c in CH2Cl2 containing 0.1 M [(n-
Bu)4N]PF6, scan rate ) 200 mV/s. (a) Scan from -0.30 to +1.20 V
(solid) and semiderivative (dashed). (b) Scan from -0.30 to +1.60 V.
irreversible, while for 4b the lower potential wave has an
associated reduction feature on the reverse scan at +0.88 V. In
the case of 5b, the lower of the two oxidation waves has an
associated reduction wave which overlaps with a wave that
appears at +0.93 V because of the reduction of longer oligomers
(Figure 4c, 1st scan).
The CVs of compounds 3c, 4c, and 5c all contain several
redox waves (Table 2). For complexes 3c and 4c, the CV
contains two overlapping waves near 1 V (shown for 4c in
Figure 5). The waves are reversible if the potential is kept under
+1.20 V and were resolved by taking the semiderivative of the
CVs, which yield symmetric peaks with a flat baseline.49 The
redox wave for 5c at +1.05 V becomes irreversible when the
Blocking of the R-positions of terthienyl derivatives with
methyl groups results in reversible oxidation waves, because
the resulting radical cations do not polymerize.50,51 In 3c and
4c, the two terthienyl groups are identical, and the observed
splitting in the oxidation waves indicates that there is some
(50) Hill, M. G.; Mann, K. R.; Miller, L. L.; Penneau, J. F. J. Am. Chem.
Soc. 1992, 114, 2728-2730.
(51) Hill, M. G.; Penneau, J. R.; Zinger, B.; Mann, K. R.; Miller, L. L.
Chem. Mater. 1992, 4, 1106-1113.
(49) Toman, J. J.; Brown, S. D. Anal. Chem. 1981, 53, 1497-1504.

Electropolymerization of Pd(II) Complexes
J. Am. Chem. Soc., Vol. 123, No. 41, 2001 9969
Figure 6. Cyclic voltammetry of 6 in CH2Cl2 containing 0.1 M [(n-Bu)4N]PF6, scan rate ) 200 mV/s. (a) Multiple scans from 0 to +1.40 V for
6. Inset: first scan of 6. (b) First scan (dashed) between 0 and +1.60 V and second scan (solid) between 0 and +2.10 V of a film of poly-6 on a
gold electrode.
interaction between the two redox groups. In metal complexes,
observation of such splitting has been interpreted as an indication
of charge delocalization across the bridge on the electrochemical
time scale.52 In the case of 3c and 4c, the splitting is small,
suggesting that the interaction is weak. In 5c, the two terthienyl
groups are not identical, and only one of the two is reversibly
oxidized at +1.05 V, while the second group is irreversibly
oxidized at a higher potential (g1.4 V). From these results, we
can conclude that the interaction between the two terthienyl
groups is largest in 5c (causing the oxidation potential of the
second terthienyl group to be shifted well positive of the first),
as expected on the basis of the bidentate coordination of both
ligands to the Pd in this complex.
trodes.54 Similarly, electropolymerization of 5a (Figure 4a) gave
red films (reduced) with a conductivity of 10-4 S cm-1 when
oxidized. The CV of a film of poly-5a (Figure 4b) is similar to
that observed for poly-3a and poly-4a.
In contrast, none of the Pd complexes containing Me2-
dppterth (3c, 4c, and 5c) could be electropolymerized, as
evidenced by the absence of film formation on the electrode
surface during oxidative cycling. This is not unexpected, because
all four R-positions of the terthienyl group in these complexes
are blocked by methyl groups, which are known to inhibit
thiophene polymerization.50,51
In complexes containing Me-dppterth, only two out of four
terthienyl R-positions are blocked by methyl groups; nonethe-
less, all attempts to electropolymerize compound 3b only
resulted in the formation of thin, insulating films on the surface
of the electrode. For 4b, the first few scans showed some
increase in current; however, no material was observed on the
electrode surface. Variation of the sweep limits and scan rate
also did not result in electropolymerization, and any material
that eventually formed was either insulating or very poorly
conductive. On the other hand, compound 5b did successfully
electropolymerize (Figure 4c) to form orange conductive films,
and a CV of this film is shown in Figure 4d.
The CVs of all the palladium complexes containing 1a-c
show one irreversible reduction wave between -1.0 and -1.5
V versus SCE (Table 2). This wave is assigned to a Pd-based
reduction process, as 1a-c do not exhibit any reduction waves
between 0 and -1.5 V.
Electropolymerization of Pd Complexes. All three Pd
complexes containing dppterth electropolymerize when solutions
of these monomers are scanned repeatedly over their whole
oxidation range. Compound 3a was previously electropolymer-
ized to form a purple film that is conductive (10-3 S cm-1
)
when oxidized.14 Electropolymerization of complex 4a gives a
red film of poly-4a on the electrode surface, with a concomitant
increase in current in the CV (Figure 3a). The CV of the
resulting film shows a broad redox feature between +0.7 and
+1.6 V (Figure 3b), similar to that observed for poly-3a and
for other polythiophene derivatives.14,53 The maximum conduc-
tivity of oxidized poly-4a is 3 × 10-4 S cm-1, determined in
situ by deposition of the film on interdigitated Pt microelec-
The cyclic voltammogram of compound 6 contains three
oxidation waves and two reduction features on the reverse scan
(Figure 6). The lower potential reverse wave is likely due to
the reduction of oligomeric species deposited on the electrode
after oxidation of the complex. Compound 6 was electropoly-
merized in a mixture of CH2Cl2 and ClCH2CH2Cl (1/1 v/v)
because of the relative insolubility of this complex in neat CH2-
Cl2. Films of poly-6 are red when reduced and turn green-black
(52) Barlow, S.; O’Hare, D. Chem. ReV. 1997, 97, 637-669.
(53) Roncali, J. Chem. ReV. 1992, 92, 711-738.
(54) Kittlesen, G. P.; White, H. S.; Wrighton, M. S. J. Am. Chem. Soc.
1984, 106, 7389-7396.

9970 J. Am. Chem. Soc., Vol. 123, No. 41, 2001
Clot et al.
Table 3. EDX Elemental Ratios for the Electropolymerized
Materials and Monomers
Table 4. UV-Visible Spectroscopic Dataa
UV-vis λmax, nm (ꢀ, M-1cm-1
complex
)
elemental ratio
compound
polymere
species
Pd/S
Pd/P
Pd/Cl
1ab
1b
254 (2.08 × 104), 354 (1.77 × 104)
252 (3.42 × 104), 282 (sh) (2.57 × 104),
364 (2.63 × 104)
poly-4aa
4b
1.93 ( 0.16
1.58 ( 0.03
1.45 ( 0.02
1.50 ( 0.08
1.44 ( 0.15
1.51 ( 0.01
1.23 ( 0.08
1.31 ( 0.02
0.66 ( 0.05
0.49 ( 0.02
0.80 ( 0.02
0.65 ( 0.02
0.70 ( 0.05
0.65 ( 0.01
0.55 ( 0.04
0.52 ( 0.01
1.37 ( 0.11
0.64 ( 0.03
1c
2
254 (2.33 × 104), 362 (2.15 × 104)
248 (1.67 × 104), 376 (3.16 × 104)
288 (sh) (3.84 × 104), 388 (6.64 × 104)
292 (sh) (3.07 × 104), 346 (3.19 × 104),
396 (4.21 × 104)
poly-5ac
5ab
3ac
3b
522 (br)
442 (br)
poly-5bc
5bb
poly-6c
6b
0.58 ( 0.01
0.55 ( 0.02
3c
4a
4b
4c
5a
5b
5c
388 (3.25 × 104), 400 (4.59 × 104)
260 (sh) (3.32 × 105), 348 (5.07 × 105)
264 (sh) (6.72 × 104), 348 (1.02 × 105)
272 (sh) (4.86 × 104), 354 (6.50 × 104)
330 (2.43 × 105),d 380 (2.09 × 105)d
338 (2.27 × 104),d 393 (1.04 × 104)d
268 (sh) (6.47 × 104), 338 (5.32 × 104),d
380 (4.62 × 104)d
a Thin film on indium tin oxide (ITO)/glass. bPowder on carbon tape,
carbon coated. Thin film on Au.
c
454 (br)
422
upon oxidation. The CV of the film shows two oxidation features
(Figure 6b). If the potential is scanned only to +1.60 V, a
partially reversible redox wave is observed, but when the
potential is scanned to +2.10 V, a second oxidation feature is
visible and no reduction waves are observed on the reverse scan.
This loss of reversibility indicates that poly-6 is not stable under
strongly oxidizing conditions.
6
7b
258 (sh) (1.33 × 105), 374 (2.94 × 105)
242 (4.64 × 104), 316 (1.28 × 104),
378 (1.17 × 104),d 422 (9.46 × 103)d
252 (sh) (9.48 × 104), 352 (4.46 × 104)
431
8a
a In CH2Cl2. bReference.12 cReference.14 dOverlapping bands.
e Reduced film on an ITO/glass electrode.
All the electropolymerized films were characterized by EDX
analysis and the elemental ratios compared to those for the
corresponding monomers (Table 3). For all the films except
poly-4a, the Pd/S ratios measured in the films match those of
their monomers within experimental error. The deviation
observed for poly-4a suggests that side reactions occur during
electropolymerization of this species, resulting in anomously
high metal content in the film. The Pd/P ratios in the films are
comparable to those measured for the monomers, although some
excess P is present in the electropolymerized films because of
the presence of residual PF6-. Further support for the metal
complexes remaining intact upon polymerization is provided
by the similarity in reactivity between the polymer films and
the monomers (vide infra). Attempts to characterize the elec-
tropolymerized materials by mass spectrometry were inconclu-
sive because of their instability under MALDI-TOF conditions.
Efforts to characterize the Pd monomers using this method, as
well as via other mass spectrometric techniques including FAB
and EI, were also not fruitful; the parent ions were not observed
although fragments due to the ligands were observed in most
cases.
seen for poly-3a. When poly-5a is reacted with a solution of
p-tolylisocyanide, the coordinated thiophene is displaced from
the metal, leaving the rest of the palladium coordination intact.
This results in the addition of one solubilizing group per metal
center causing the observed increase in solubility. Unlike the
product of the reaction of poly-3a with isocyanides, the chains
in poly-5a and poly-5b are not expected to cleave to smaller
fragments upon reaction with isocyanide, because the metal
remains coordinated to the two dppterth ligands via the
phosphine groups, even after displacement of the thiophene.
Reactivity of the Electropolymerized Films with Isocya-
nides. The reactivity of poly-5a and poly-5b films toward
isocyanides parallels that of their respective monomers 5a,b.
Exposure of films of poly-5a,b to solutions of t-BuNC in
hexanes at 25 °C resulted in rapid reaction. The surface
reflectance IR spectrum of a film of poly-5a after exposure
shows a new band at 2219 cm-1 due to the CN stretch of an
isocyanide coordinated via the terminal carbon. The IR spectrum
of 8a contains a CN stretch of comparable energy. Similarly,
when a film of poly-5b is exposed to tert-butyl isocyanide in
hexanes, a new band at 2224 cm-1 appears in the surface
reflectance IR spectrum of the film, comparable to the CN
stretch observed for complex 5b treated with tolyl isocyanide
(2191 cm-1).
In a similar fashion, a film of poly-3a was reacted with
t-BuNC in hexanes, and a new, strong band at 2208 cm-1
appeared in the IR spectrum.14 The film remained insoluble in
hexanes after reaction but became highly soluble in CH2Cl2.
This was attributed to the cleavage of the dichloro bridges in
poly-3a yielding oligomers with solubilizing t-Bu groups added
to their backbones.
UV-Visible Spectroscopy of 1-6. The UV-vis spectral
data for complexes 1-6 are shown in Table 4. The spectra of
1a-c and 2 all have absorption bands with a maximum
absorbance between 354 and 376 nm, similar to λmax for the
π-π* transition of terthiophene (351 nm).55 The spectra of
complexes 3a-c contain several absorption bands in the UV
tailing into the visible region. All the bands are assigned to
ligand-based π-π* transitions, by comparison with the UV-
An increase in the solubility of poly-5a and poly-5b was also
observed after treatment with isocyanides, comparable to that
(55) DiCesare, N.; Belletete, M.; Marrano, C.; Leclerc, M.; Durocher,
G. J. Phys. Chem. A 1999, 103, 795-802.

Electropolymerization of Pd(II) Complexes
J. Am. Chem. Soc., Vol. 123, No. 41, 2001 9971
vis spectra of dppterth ligands 1a-c and Ru complexes
containing dppterth such as [RuCl2(dppterth)2].12 The low energy
bands for 3a-c (388, 396, and 400 nm) are red-shifted from
the corresponding bands in 1a-c and follow the same trend
observed in the series 1a-c.
The spectra of the bis(phosphine) complexes 4a-c are very
similar to each other and contain two bands, corresponding to
those observed for the ligands 1a-c. The higher energy bands
are slightly red-shifted in the complexes with respect to the
corresponding bands in the ligands, while the lower energy
bands in 4a-c are blue-shifted both relative to the ligands 1a-c
and to the dimetallic complexes 3a-c. A similar blue shift with
respect to dppterth was also observed in the ruthenium mono-
carbonyl complexes cis- and trans-[RuCl2(CO)(dppm)(dppterth-
P)].12 In 4a-c, the blue-shift is likely due to a small electron
withdrawing effect by the palladium from the ligands.
materials are expected to consist predominantly of sexithiophene
units linking metal bis(phosphine) groups.
Upon reaction of a poly-3a film with t-BuNC or PhNC, the
absorption band red-shifts to λmax ) 528 nm (t-BuNC) and 542
nm for (PhNC).14 Similarly, the absorption maxima of poly-5a
and poly-5b films treated with t-BuNC shift to λmax ) 462 nm
and λmax ) 434 nm, respectively. These shifts are indicative of
the electronic effect on the backbone caused by the reaction at
the pendant metal, and the differences in λmax observed for
reactions with two similar isonitriles with poly-3a indicate how
sensitive the electronic structure of the conjugated backbone is
to the ligands at the Pd center.
Discussion
The solid-state structures of 5c, 3a,14 cis-[RuCl2(dppm)-
(dppterth-P,S)],12 and [RuCl2(CO)(dppterth)2]12 all demonstrate
that coordination of dppterth via the phosphine alone has little
influence on the structure of the terthienyl moiety relative to
that of terthiophene. When the dppterth ligands are chelated in
a bidentate fashion to the metal, via either the sulfur or C3, the
deformations in bond lengths and angles are greater in the
coordinated thiophene ring. However, only minor structural
deformations are observed in the two remaining rings of the
terthienyl moiety. These observations suggest that significant
electronic effects (resulting in structural deformations) due to
metal centers linked to a polythiophene backbone will occur
only when a thiophene ring along the backbone is directly bound
to the metal via a sulfur or carbon atom. It is important to note,
however, that the extended conjugation that exists along
polythiophene chains means that disruption of a single ring in
the backbone is sufficient to cause extensive perturbations to
the overall electronic properties of the polymer.
Complexes 5a-c show a broad absorption consisting of two
overlapping bands with maxima near 340 and 380 nm for all
three complexes. These bands are also attributed to ligand-based
π-π* transitions and are red-shifted both from the parent
complexes 4a-c and from the dppterth ligands 1a-c. The red
shift is likely due to chelation by the ligands in a bidentate
fashion, imposing a rigid conformation on the terthienyl groups.
The lower energy band in 5a-c is comparable to the lower
band observed both in the dimetallic Pd complexes 3a-c and
in the ruthenium complexes in which dppterth is P,S-coordinated
such as [RuCl2(dppterth-P,S)2] and cis- and trans-[RuCl2(dppm)-
(dppterth-P,S)].12 When complex 3b is reacted with p-CH3C6H4-
NC, the UV-vis spectrum of the resulting mononuclear
complex 7b contains a higher energy band at 316 nm and two
lower energy bands at 378 and 422 nm, all of which are assigned
to π-π* ligand-based transitions.
Thiophene oligomers polymerize mainly by C-C coupling
at the R-(2-) positions.53 Therefore, it is likely that the obtained
films consist primarily of R,R-coupled material, a proposal
which is strongly supported by the lack of polymerization
observed with the complexes containing 1c, in which the methyl
substituents prevent R,R-coupling.
The spectrum of 8a contains two bands, both of them blue-
shifted relative to parent complex 5a. Similar behavior was
observed when compound 5b was treated with tolyl isocyanide.
A corresponding blue shift also occurs when the ruthenium
complexes cis- and trans-[RuCl2(dppm)(dppterth-P,S)] are
treated with CO to give cis- and trans-[RuCl2(CO)(dppm)-
(dppterth-P)]. The blue shift is therefore likely due to removal
of the imposed rigidity in the terthienyl moiety when the ligand
becomes monodentate. The absorption band of 8a is also slightly
blue-shifted compared to dppterth, probably because the pal-
ladium center is somewhat electron withdrawing. In the
spectrum of 6, two bands are observed, nearly equal in energy
to the bands observed for 5dppterth, and are assigned to ligand-
based π-π* transitions. In complex 6, coordination of the
phosphinoterthiophene ligand via the phosphine does not appear
to induce a large perturbation in the electronic structure of the
terthienyl group.
The materials described in this paper may be divided into
two classes on the basis of their structures. In the first class,
comprising poly-5b and poly-6, R,R-coupling results in materials
in which the backbone includes the metal centers. In poly-6, a
linear backbone consisting of phosphine-terminated sexithiophene
units linked via metal-dichloro moieties is expected to result
from R,R-coupling. Poly-5b would form structures in which the
methyl-terminated sexithiophene groups are cross-linked via Pd
coordinated to phosphorus, sulfur, and carbon atoms. The UV-
visible absorption spectra of both materials is similar to that of
sexithiophene, supporting the conclusion that the oligothienyl
groups in these materials are composed of six-ring units.
In these materials, conductivity results from (a) delocalization
of charge along the main chains (including the metal), (b)
π-stacking of the oligothiophene units, or (c) a combination of
both effects. A conducting copolymer composed of quater-
thiophene linked by saturated polyester chains has been reported
by Miller et al.,57 and in this polymer, conductivity only occurs
via π-stacks formed with oxidized quaterthiophenes, because
delocalization of polarons or bipolarons along the backbone is
impossible because of the long saturated linkers between the
conjugated moieties.
UV-Visible Spectroscopy of the Electropolymerized Ma-
terials. The spectrum of neutral poly-4a shows a broad
absorbance at λmax ) 442 nm which shifts to 766 nm upon
oxidation. The spectra of neutral poly-5a and poly-5b contain
broadened bands with λmax at 454 and 422 nm, respectively.
The band maxima for poly-4a and poly-5a are red-shifted from
those of their corresponding monomers; however, they are also
blue-shifted with respect to that of poly-3a (522 nm).14 This is
consistent with these materials being composed of oligomers
with longer conjugation lengths than their corresponding
monomers, but with shorter conjugation than in poly-3a. The
absorption maximum of neutral poly-6 is at 431 nm. In poly-
5b and poly-6, the absorption maxima are comparable to that
of sexithiophene (432 nm),56 which is reasonable because these
(56) Van Pham, C.; Burkhardt, A.; Shabana, R.; Cunningham, D. D.;
Mark, H. B.; Zimmer, H. Phosphorus, Sulfur Silicon Relat. Elem. 1989,
46, 153-168.
(57) Hong, Y.; Miller, L. L. Chem. Mater. 1995, 7, 1999-2000.

9972 J. Am. Chem. Soc., Vol. 123, No. 41, 2001
Clot et al.
The structures of poly-5b and of a hypothetical poly-4b would
be very similar except for the coordination at the palladium
center. In 4b, the oligothienyl units are linked to the metal only
via the phosphine groups. On the other hand, in 5b, the
oligothienyl group is linked to the metal not only via the
phosphine ligands but also via Pd-C and Pd-S bonds. While
complex 5b electropolymerizes, compound 4b only yields
insulating films upon electrochemical oxidation. Because of the
similarities in the structures of the two monomers, this intriguing
difference can best be accounted for by examining the CVs of
4c and 5c (vide supra). These indicate that the degree of
interaction across the metal is larger in 5c, suggesting that
conductivity in poly-5b may have a significant contribution from
delocalization through the metal. In 4b, π-stacking alone with
a diminished contribution from delocalization across the metal
must be insufficient to support the growth of conductive films.
Poly-6 also forms readily by electropolymerization, despite a
similar coordination at the metal to 4b. Here, steric factors also
play a role, perhaps allowing better overlap between the
conjugated oligothienyl segments than in 4b, and allowing
conductivity via π-stacking to occur.
The second class of materials includes poly-3a, poly-4a, and
poly-5a. In these materials, the geometry of the precursor
complexes suggests that highly branched species in which oligo-
or polythiophene chains are cross-linked by Pd groups will form.
Conductivity may arise by delocalization strictly along the
conjugated thiophene backbone, as well as via the pathways
described above for the first class. In poly-3a, conductivity must
result primarily from delocalization along the longer poly-
thiophene chains and chain-chain interactions by π-stacking
rather than through the Pd-dichloro bridges. The fact that 3b
does not electropolymerize, combined with the small degree of
electrochemical interaction across the metal bridge in 3c,
supports this conclusion. The UV-vis absorbance spectrum of
poly-3a, which is similar to that of other polythiophene
derivatives,58 also demonstrates that the Pd centers have a
relatively minor influence on the electronic properties of the
material.
electronic interactions with the polymer backbone may be
needed in order to increase the response obtained upon modify-
ing the ligand environment at the metal.
Experimental Section
General. All the reactions were performed using standard Schlenk
techniques with dry solvents under nitrogen. Tolyl isocyanide,59 3′-
diphenylphosphino-2,2′:5′2′′-terthiophene (dppterth) (1a),12 and [Pd2-
(µ-Cl2)(dppterth-P,C3)2] (3a)14 were all prepared according to literature
procedures. All other reagents were purchased from either Strem
Chemicals or Aldrich and used as received. 1H and 31P NMR
experiments were performed on either a Bruker AC-200E or a Bruker
AV-300 spectrometer, and spectra were referenced to residual solvent
(1H) or external 85% H3PO4 (31P). Electronic absorption spectra were
obtained on a UNICAM UV2 UV-vis spectrometer. Infrared spectra
were obtained on a Bomem MB-series spectrometer on methylene
chloride solutions, potassium bromide pellets, or cesium iodide pellets.
Surface reflectance IR experiments were performed using a Pike
Technologies VeeMax specular reflectance accessory mounted in the
Bomem spectrometer. All the surface IR spectra were taken with an
incident angle of 45°. Electrochemical measurements were conducted
on a Pine AFCBP1 bipotentiostat using a Pt disk working electrode,
Pt coil wire counter electrode, and a silver wire reference electrode.
An internal reference (decamethylferrocene) was added to correct the
measured potentials with respect to saturated calomel electrode (SCE).
The supporting electrolyte was 0.1 M [(n-Bu)4N]PF6, which was purified
by triple recrystallization from ethanol and dried at 90 °C under vacuum
for 3 days. Methylene chloride used in cyclic voltammetry was dried
by refluxing over CaH2. EDX (energy-dispersive X-ray) analysis were
performed on a Kevex Quantum light element X-ray detector equipped
with a Quartz XOne X-ray analyzer. Note that only a representative
synthetic procedure for each group of compounds is given here. Detailed
synthetic procedures and complete characterization for the remaining
complexes are provided in the Supporting Information.
3′-Diphenylphosphino-5-methyl-2,2′:5′2′′-terthiophene (Me-dpp-
terth) (1b). A solution of n-butyllithium in hexanes (6.86 mL, 1.6 M,
11.0 mmol) was added dropwise to a solution of 3′-bromo-5-methyl-
2,2′:5′,2′′-terthiophene (11) (3.40 g, 9.97 mmol) in diethyl ether (120
mL) at -15 °C. The mixture was stirred for 1 h at -15 °C and PPh2Cl
(3.31 g, 15.0 mmol) added dropwise. The reaction was then allowed
to warm to room temperature and stirred for another 30 min, after which
time 1 M HCl was added to quench the reaction. The organic layer
was separated, washed with water, and dried over anhydrous MgSO4,
and the solvent was removed to yield the crude product, which was
purified by chromatography on silica gel with hexanes/methylene
chloride (4/1 v/v). The yellow band was collected and the solvent
removed to leave a yellow oil. Trituration with a minimum amount of
Conclusions
The complexes described in this paper allow the electronic
interactions in π-conjugated materials containing pendant metal
centers to be probed. The electrochemical results for 3c, 4c,
and 5c demonstrate that the largest degree of interaction between
the dimethylterthienyl groups is seen in 5c where the metal is
both P,C- and P,S-coordinated. In the monomers that have free
R-positions, polymerization is observed to occur when the
resulting materials have sufficient conductivity to sustain film
growth. In the case of 5a and 5b, the conductivity appears to
involve a contribution from cross-metal delocalization. In 3b
and 4b, poor delocalization across the metal and poor π-π
overlap between oligothienyl groups prevent electropolymer-
ization because of low conductivity of the resulting films. The
observed conductivity in poly-3a results primarily from charge
delocalization along the extended polythiophene chains and
π-stacking rather than through the metal bridges in this material.
In poly-5a and poly-5b, the backbone acts as a hemilabile ligand
on the Pd center, remaining coordinated even after displacement
of the thienyl moiety. However, even when the material is
reacted with a strong σ-donor such as an isocyanide, only small
changes in the UV-vis absorption of the film are observed.
These results suggest that metal centers that have larger
1
methanol gave 1b as a yellow powder. Yield: 2.2 g (50%). H NMR
(200 MHz, CDCl3): δ 7.34 (m, 10H, Ph), 7.15 (m, 1H, Th), 7.05 (m,
1H, Th), 6.95 (m, 2H, Th), 6.65 (m, 1H, Th), 6.58 (s, 1H, Th), 2.46 (s,
3H, CH3). 31P{1H} NMR (81.015 MHz, CDCl3): δ -24.1 (s). Anal.
C25H19PS3 requires C, 67.26; H, 4.26. Found: C, 67.00; H, 4.24%.
[Pd2(µ-Cl2)(Me-dppterth-P,C3)2] (3b). A warm solution of Me-
dppterth (1b) (452 mg, 1.01 mmol) in an ethanol/acetonitrile mixture
(40/12 mL) was slowly added at 50 °C to PdCl2 (200 mg, 1.13 mmol)
in water (12 mL) and concentrated HCl (0.20 mL). A yellow precipitate
immediately formed, and the mixture was stirred for 2 h at 50 °C and
filtered warm to yield a yellow powder. The solid was washed with
water (2 × 10 mL), ethyl ether (1 × 5 mL), and hexanes (2 × 20 mL)
and dried in air. Recrystallization from a chloroform/hexanes mixture
1
yielded an orange solid. Yield: 475 mg (79%). H NMR (300 MHz,
CDCl3): δ 7.66 (m, 4H, Ph), 7.52-7.24 (m, 14H, Ph and Th), 7.15
(m, 2H, Th), 7.05 (m, 2H, Th), 6.95 (m, 2H, Th), 6.49 (s, 2H, Th),
2.40 and 2.29 (s, 6H, CH3). 31P{1H} NMR (121.5 MHz, CDCl3): δ
19.9 (s), 18.7 (s). Anal. C50H36Cl2P2PdS6 requires: C, 51.12; H, 3.07.
Found: C, 50.83; H, 3.19%.
[PdCl2(dppterth-P)2] (4a). To a solution of dppterth (300 mg, 0.69
mmol) dissolved in an ethanol/acetonitrile mixture (20/4 mL) at 50 °C
(58) McCullough, R. D.; Lowe, R. D.; Jayaraman, M.; Anderson, D. L.
J. Org. Chem. 1993, 58, 904-912.
(59) Krapcho, A. P. J. Org. Chem. 1962, 27, 1089-1090.

Electropolymerization of Pd(II) Complexes
J. Am. Chem. Soc., Vol. 123, No. 41, 2001 9973
Table 5. Crystallographic Data for 5c
was added dropwise a solution of PdCl2 (50.0 mg, 0.28 mmol) in a
water/HCl mixture (3 mL, with 0.05 mL concentrated HCl added). Upon
addition, a bright lemon-yellow precipitate appeared. After stirring at
50 °C for 1 h, the suspended yellow solid was collected by filtration
and washed with water, ether, and hexanes. Recrystallization from
methylene chloride/hexanes yielded 4a as a yellow-orange powder.
5c‚2/3(CHCl3)
formula
mol wt
C
52.67H41.67Cl2F6P3PdS6
1251.20
temperature, K
crystal system
198(2)
triclinic
1
Yield: 230 mg (79%). H NMR (200 MHz, CDCl3): δ 7.69 (m, 4H,
space group
P1h
o-PC6H5), 7.48-7.24 (m, 7H, m,p-PC6H5 and 3-H), 7.17-7.08 (m, 2H,
5-H and 5′′-H), 7.02 (dd, 1H, JHH ) 3.65 and 1.13 Hz, 3′′-H), 6.93
(dd, 1H, JHH ) 4.91 and 3.65 Hz, 4′′-H), 6.80 (dd, 1H, JHH ) 5.14 and
3.62 Hz, 4-H), 6.65 (s, 1H, 4′-H). 31P{1H} NMR (81.015 MHz,
CDCl3): δ 11.1 (s). Anal. C48H34Cl2P2S6Pd requires: C, 55.31; H, 3.26.
Found: C, 55.57; H, 3.32%.
crystal dimensions (mm3)
0.25 × 0.15 × 0.03
crystal color
red
a, Å
b, Å
c, Å
10.770(1)
13.788(1)
21.289(4)
99.693(5)
94.388(5)
112.405(5)
2846.4(7)
2
R, deg
[Pd(dppterth-P,C3)(dppterth-P,S1)][PF6] (5a). To a suspension of
4a (200 mg, 0.19 mmol) in acetone (40 mL) was slowly added an excess
of AgBF4 (84.0 mg, 0.42 mmol) in acetone (40 mL, sparged with
nitrogen) at 25 °C. After stirring for 18 h under nitrogen at 25 °C, the
mixture was filtered through Celite, and the red filtrate was concentrated
to 1/10 of the initial volume and added slowly to an aqueous solution of
NH4PF6 (620 mg, 3.80 mmol) in water (50 mL). The resulting orange
slurry was stirred at 25 °C for 30 min, the solid collected by filtration
and washed with water (100 mL), diethyl ether (5 mL), and hexanes
(50 mL). Slow recrystallization from chloroform/diethyl ether yielded
complex 5a as red crystals. Yield: 180 mg (84%). 1H NMR (200 MHz,
CDCl3): δ 7.60-7.40 (m, 21H, Ph and Th), 7.27 (m, 1H, Th), 7.19-
7.13 (m, 2H, Th), 7.07-7.00 (m, 3H, Th), 6.96-6.89 (m, 1H, Th),
6.79 (m, 1H, Th), 6.67-6.58 (m, 2H, Th), 6.20 (m, 1H, Th), 6.16-
â, deg
γ, deg
V, Å3
Z
reflns collected/unique
R1a, wR2b [I > 3σ(I)]
R1a, wR2b (all data)
17 750/7695 [Rint ) 0.104]
0.060, 0.072
0.136, 0.163
a R ) ∑|Fo| - |Fc||/∑|Fo|. bRw ) {∑[w(Fo2 - Fc2)2]/∑[w(Fo )2]}.1/2
2
X-ray Crystallographic Analysis. Suitable crystals of 5c were
grown by slow diffusion of diethyl ether into a solution of the compound
in chloroform at ∼15 °C, and data were collected at 198(1) K. Selected
crystallographic data for 5c appear in Table 5. The final unit cell
parameters for 5c were obtained by least-squares on the setting angles
for 5689 reflections with 2θ ) 4.5-46.1°. All data were processed
and corrected for Lorentz and polarization effects and absorption
(semiempirical, based on symmetry analysis of redundant data). The
structure was solved by direct methods60 and expanded using Fourier
techniques.61 Final refinements were carried out using teXsan.62 The
complex 5c crystallizes in space group P1h, with one molecule in the
asymmetric unit. The material crystallizes with CHCl3 in the lattice.
The solvent molecule site is only partially occupied, with a refined
population of approximately 2/3. All non-hydrogen atoms were refined
anisotropically, while all hydrogens were placed in calculated positions.
6.10 (m, 1H, Th). 31P NMR (81.015 MHz, CDCl3): δ 25.12 (d, JPP
)
19.3 Hz, cis P), 15.7 (d, JPP ) 394 Hz, 2P, trans P), 3.8 (d, JPP ) 394
Hz, 2P, trans P), 1.9 (d, JPP ) 19.3 Hz, cis P), -144.3 (sept, JPF ) 713
Hz, 1P, PF6). Anal. C48H33F6P3S6Pd requires: C, 51.69; H, 2.96.
Found: C, 51.19; H, 3.23%.
[PdCl(p-CH3C6H4NC)(Me-dppterth-P,C3)] (7b). To a suspension
of 3b (200 mg, 0.19 mmol) in CH2Cl2 (7 mL) under N2 at 25 °C was
added p-CH3C6H4NC (45.0 mg, 0.38 mmol). After 5 min, the solid
was completely dissolved, and the solution was stirred for an additional
20 min, after which time the mixture was concentrated to 2-3 mL
under vacuum, and hexanes were added to precipitate a yellow powder.
The crude solid was recrystallized from methylene chloride/hexanes
to yield pure 7b. Yield: 202 mg (75%). 1H NMR (200 MHz, CDCl3):
δ 7.73-7.53 (m, 4H, o-PC6H5), 7.50-7.29 (m, m,p-PC6H5), 7.15 (m,
1H, Th), 7.10-6.90 (m, 5H, Th and C6H4NC), 6.72 (m, 2H, C6H4NC),
6.53 (m, 1H, Th), 2.41 (s, 3H, Th-CH3), 2.31 (s, 3H, CH3C6H4NC).
31P{1H} NMR (81.015 MHz, CDCl3): δ 17.5 (s). IR (CH2Cl2): νNtC
) 2193 cm-1. Anal. C33H25ClNPPdS3 requires: C, 56.26; H, 3.55; N,
1.99. Found: C, 56.57; H, 3.78; N, 2.00%.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada for support of this
research and Prof. T. M. Swager (M.I.T.) for a generous gift of
interdigitated Pt microelectrodes.
Supporting Information Available: Complete synthetic
procedures and characterization for 1c, 2, 3c, 4b,c, 5b,c, 6, 9-12
and procedures for electrode preparation and cleaning (pdf).
X-ray crystallographic file (CIF) for 5c. This material is
available free of charge via the Internet at http://pubs.acs.org.
[Pd(p-CH3C6H4NC)(dppterth)2][PF6] (8a). To a solution of 5a
(40.0 mg, 0.04 mmol) in CH2Cl2 (4 mL) under N2 at 25 °C was added
p-CH3C6H4NC (4.19 mg, 0.04 mmol). The solution was stirred for 1.5
h after which time the mixture was concentrated to ∼0.5 mL under
vacuum, and hexanes were added to precipitate a yellow powder. The
crude solid was recrystallized from methylene chloride/hexanes to yield
pure 8a. Yield: 31 mg (68%). 1H NMR (300 MHz, CDCl3): δ 7.60-
7.42 (m, 18H), 7.35-7.30 (m, 2H), 7.22-6.98 (m, 12H), 6.71 (s, 1H),
6.62 (s, 1H), 6.38 (m, 2H), 6.27 (m, 1H), 2.31 (s, 3H, CH3). 31P{1H}
NMR (121.5 MHz, CDCl3): δ 17.5 (d, JPP ) 388 Hz, 1P, trans P), 6.4
(d, JPP ) 388 Hz, 1P, trans P), -144.3 (sept, JPF ) 713 Hz, 1P, PF6).
IR (KBr): νNtC ) 2194 cm-1. Anal. C56H40F6NP3PdS6 requires: C,
54.57; H, 3.25; N, 1.14. Found: C, 54.29; H, 3.29; N, 1.09%.
JA016465M
(60) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,
R. J. Appl. Crystallogr. 1999, 32, 115-119.
(61) Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.; de
Gelder, R.; Israel, R.; Smits, J. M. M. Technical Report of the Crystal-
lography Laboratory; University of Nijmegen: The Netherlands, 1994.
(62) Crystal Structure Analysis Package; The Molecular Structure
Corporation: Woodland, TX, 1985 and 1992.