Coordination chemistry of SCS Pd(II) pincer systems

Article abstract of DOI:10.1002/1099-0682(200012)2000:12<2533::AID-EJIC2533>3.0.CO;2-E

We have studied the coordination of substituted pyridines, and phosphorus- and sulfur-containing ligands to an SCS Pd(II) pincer system. These ligands coordinate to Pd(II) (trans to the cyclopalladated awl group) by quantitative substitution of the labile acetonitrile ligand in complex 1. Competition experiments showed that both electronic and steric effects influence the strength of coordination to the Pd(II) pincer of the substituted pyridines. A quantitative analysis of the substituent effect was achieved by a Hammett correlation. Phosphorus-containing ligands also coordinate to this SCS Pd(II) motif, as evidenced by NMR spectroscopy and single-crystal X-ray diffraction studies. They are much stronger ligands than the pyridines. The coordination strength of the thioureas falls in between those of the pyridines and phosphanes/phosphites. Our results lead therefore to the following order of ligand strength towards Pd(II) in SCS Pd(II) pincers: PRa > P(OR)₃ > N,N'-disubstituted thiourea > (substituted) pyridines > MeCN.

Full text of DOI:10.1002/1099-0682(200012)2000:12<2533::AID-EJIC2533>3.0.CO;2-E

FULL PAPER
Coordination Chemistry of SCS PdII Pincer Systems
Henk-Jan van Manen,[a] Kazuaki Nakashima,[a,b] Seiji Shinkai,[b] Huub Kooijman,[c]
Anthony L. Spek,[c][‡] Frank C. J. M. van Veggel,*[a] and David N. Reinhoudt*[a]
Keywords: Coordination chemistry / N ligands / P ligands / Palladium / Tridentate ligands
We have studied the coordination of substituted pyridines,
phorus-containing ligands also coordinate to this SCS PdII
and phosphorus- and sulfur-containing ligands to an SCS PdII motif, as evidenced by NMR spectroscopy and single-crystal
pincer system. These ligands coordinate to PdII (trans to the
cyclopalladated aryl group) by quantitative substitution of
X-ray diffraction studies. They are much stronger ligands
than the pyridines. The coordination strength of the thio-
the labile acetonitrile ligand in complex 1. Competition ex- ureas falls in between those of the pyridines and phos-
periments showed that both electronic and steric effects in- phanes/phosphites. Our results lead therefore to the follow-
fluence the strength of coordination to the PdII pincer of the
substituted pyridines. A quantitative analysis of the substitu-
ing order of ligand strength towards PdII in SCS PdII pincers:
PR3 Ͼ P(OR)3 Ͼ N,NЈ-disubstituted thiourea Ͼ (substituted)
ent effect was achieved by a Hammett correlation. Phos- pyridines Ͼ MeCN.
Introduction
meta-Xylene derivatives in which the two methylene
groups carry suitable donor atoms (e.g. N, P, or S), so-called
pincer ligands,[1] are used for the complexation of a wide
range of transition metals.[2] Cyclometallation gives rise to
a variety of systems displaying rich coordination chemistry.
A number of stable PdII complexes containing SCS
pincer systems have been reported to date,[3] with recent
examples mainly related to supramolecular chemistry.[4] We
have been exploiting the SCS PdII pincer motif in the non-
covalent synthesis of nanosize metallodendrimers.[5] Repres-
entative dendritic building blocks used in our investigations
are depicted in Scheme 1.
Controlled metallodendrimer assembly up to generation
5 has been achieved by divergently growing from the core
G0.[6] In these metallodendritic structures, the coordination
of nitrile ligands is used as the assembling principle. Large
self-assembled hyperbranched organopalladium spheres
(100Ϫ400 nm) have also been synthesized divergently by an
uncontrolled one-pot strategy.[7]
By employing ligands of different coordination strengths,
Scheme 1. Representative building blocks that have been employed
it is also possible to synthesize metallodendrimers in a con-
to assemble metallodendrimers, see refs.[6Ϫ8]
vergent manner.[8] This growth strategy commences by coor-
dinating the strongest ligands first (at the periphery). Sub-
sequent growth proceeds inwards using building blocks of
decreasing coordination strength. This order is imperative
in preventing ligand scrambling. We have demonstrated this
type of growth in the synthesis of a third generation metal-
lodendrimer containing cyano- and pyridine-based building
blocks. In order to broaden the scope of this metallodend-
rimer synthetic method, other ligands with different bind-
ing strengths are needed. Therefore, we have studied the
relative coordination strengths of a series of ligands towards
[a]
Laboratory of Supramolecular Chemistry and Technology,
MESAϩResearch Institute, University of Twente,
P. O. Box 217, 7500 AE, Enschede, The Netherlands
Fax: (internat.) ϩ 31-53/4894645
E-mail: smct@ct.utwente.nl
[b]
Chemotransfiguration Project, Japan Science and Technology
Corporation,
2432 Aikawa, Kurume, Fukuoka 839-0861, Japan
[c]
Bijvoet Center for Biomolecular Research, Department of
Crystal and Structural Chemistry, Utrecht University,
Padualaan 8, 3584 CH Utrecht, The Netherlands
[‡]
To whom correspondence pertaining to crystallographic studies
should be addressed (E-mail: a.l.spek@chem.uu.nl)
Eur. J. Inorg. Chem. 2000, 2533Ϫ2540
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1434Ϫ1948/00/1212Ϫ2533 $ 17.50ϩ.50/0
2533

F. C J. M. van Veggel, D. N. Reinhoudt et al.
FULL PAPER
the SCS PdII pincer system. We report here our results for at δ ϭ 6.80 (in 5) or δ ϭ 6.92 (in 6) are absent from the 1H
the parent PdII pincer. Moreover, a dipincer has been syn- NMR spectra, indicating complete cyclopalladation. In the
thesized which could be used as a new rigid core for metal- 13C NMR spectra of the acetonitrile complexes 1 and 3, the
lodendrimer synthesis.
signal of the carbon atom directly linked to PdII is too weak
to be observed (probably due to long relaxation times),
whereas in the chloro complex 2 it is present at δ ϭ 150.9
(30 ppm downfield compared to ligand 5). In the positive
FAB-MS spectrum of 2, a signal corresponding to [M Ϫ Cl
ϩ H]ϩ (m/z ϭ 499.6) is present, whereas in the negative
FAB-MS spectrum the chloride ion remains coordinated to
PdII, resulting in a signal at m/z ϭ 533.7 ([M]ϩ). Loss of
the acetonitrile ligand is observed in both the positive and
negative FAB-MS spectra of 1, resulting in signals corres-
ponding to [M Ϫ MeCN Ϫ BF4]ϩ (m/z ϭ 499.0) and [M
Ϫ MeCN ϩ BF4]Ϫ (m/z ϭ 671.6) respectively, while both
acetonitriles are lost in the positive FAB-MS spectrum of 3
(m/z ϭ 1001.1 for [M Ϫ 2MeCN Ϫ 2BF4 ϩ H ]ϩ).
Results and Discussion
Synthesis and Characterization of Model SCS PdII Pincer
Complexes
The SCS PdII complexes 1 and 2 were prepared in two
and three steps, respectively (Scheme 2). After alkylation of
the previously reported[7b] SCS pincer ligand 4 with n-butyl
bromide (77% yield), cyclopalladation of the resulting li-
gand 5 was achieved by reaction with [Pd(CH3CN)4][BF4]2.
X-ray Structure of 1
Besides the analytical evidence discussed above, struc-
tural proof of the acetonitrile complex 1 is provided by an
X-ray analysis. A drawing of the acetonitrile complex 1 is
shown in Figure 1. The following bond lengths for the
atoms bonded to PdII have been obtained: PdϪS(1)
˚
˚
˚
2.2992(8) A, PdϪS(2) 2.2961(9) A, PdϪN 2.110(2) A,
˚
PdϪC(1) 1.975(2) A.
Scheme 2. Synthesis of SCS PdII pincer complexes; a) nBuBr,
tBuOK, CH3CN
The cationic acetonitrile complex 1 (containing the non-
Ϫ
coordinating BF4 counter anion) was obtained in 90%
yield by precipitation with Et2O from CH3CN solution
after completion of the cyclopalladation. The neutral
chloro complex 2 was isolated (in 94% yield) by stirring a
solution of 1 in CH2Cl2/CH3CN with saturated NaCl, fol-
lowed by column chromatography.
Similarly, dipincer 3 was prepared in two steps from
1,2,4,5-tetrakis(bromomethyl)benzene by first allowing the
benzyl bromides to react with tert-butylthiophenol using
K2CO3 as base and 18-crown-6 as catalyst (81% yield), fol-
lowed by cyclopalladation of the resulting dipincer ligand 6
using [Pd(CH3CN)4][BF4]2 in CH3CN. The bis(acetonitrile)
complex was purified by precipitation from CH3CN solu-
tion with Et2O (60% yield). Complexes 1, 2 and 3 were isol-
ated as air- and moisture-stable yellow solids. All spectro-
scopic and analytical data are consistent with cyclopallad-
Figure 1. Structure of the acetonitrile complex 1
The Pd atom adopts a slightly distorted square-planar
geometry [S(1)ϪPdϪS(2) 171.11(2)° and NϪPdϪC(1)
178.26(11)°] and the two five-membered chelate rings pos-
sess somewhat distorted envelope conformations in which
the two phenyl rings (oriented anti with respect to the
square plane) are in axial positions [PdϪS(1)ϪC(4)
100.51(9)° and PdϪS(2)ϪC(12) 103.17(9)°]. It should be
noted that complex 1 is chiral in the solid state (prepared
as a racemate) due to the lack of symmetry elements.
1
ation. Compared to the free ligands 5 and 6, the H NMR
resonances for the diastereotopic CH2S protons in the pal-
ladium complexes are shifted downfield by ca. 0.5 ppm and
are broadened due to slow conformational interconversion
of the PdII-containing five-membered rings. At lower tem-
peratures the coupling of the two diastereotopic CH2S pro-
tons results in the broad signal splitting into a pair of doub-
Substitution of the Labile MeCN Ligand by Pyridine
Substitution of the labile acetonitrile ligand of complex
lets (vide infra). Further, the signals for the ortho-protons 1 by stronger ligands such as pyridine is facile. Upon treat-
2534
Eur. J. Inorg. Chem. 2000, 2533Ϫ2540

Coordination Chemistry of SCS PdII Pincer Systems
FULL PAPER
ment of 1 with 1 equiv. of pyridine (in CDCl3), fast and
quantitative substitution was observed. In the 1H NMR
spectrum of the pyridine adduct, the resonance of the pyr-
idine α-protons is shifted upfield from δ ϭ 8.60 (free pyrid-
ine) to δ ϭ 8.15 (coordinated pyridine). This is different
from the pyridine complexes of SCS PdII pincers reported
by Loeb and co-workers[9] for which a downfield shift of the
pyridine α-protons was observed upon coordination. This
difference must be due to the presence of the phenyl rings
attached to the sulfur atoms in our system. The ring current
of these aromatic rings will have a shielding effect on the
pyridine protons (mostly the α-protons). The α-proton sig-
nal is also considerably broadened, indicating hindered ro-
tation of the pyridine ligands with respect to the plane of
coordination. The CH2S proton signal shifts downfield by
0.15 ppm, and the aromatic proton signal by 0.05 ppm,
upon substitution of MeCN by pyridine. The greater coor-
dination strength of the pyridine ligand as compared to ace-
tonitrile is also evident from the positive FAB-MS spec-
trum, in which a small, but significant peak corresponding
to the molecular ion [M Ϫ BF4]ϩ is observed at m/z ϭ
578.4, in addition to a large signal at m/z ϭ 499.2 from [M
Ϫ pyr Ϫ BF4]ϩ.
Upon adding excess pyridine, two separate NMR signals
for free and bound pyridine are observed. This observation
can be explained by either slow exchange of pyridine on the
NMR timescale or a kinetically stable pyridine complex.
Exchange of pyridine was indeed evidenced by treatment of
the pyridine complex with an excess (5 equiv.) of [D5]pyrid-
ine. The signal at δ ϭ 8.15 disappeared and a new
peak appeared at δ ϭ 8.6, indicating free pyridine. This
kinetic lability is consistent with other pyridineϪPdII
complexes.[10]
1
Figure 2. H NMR spectra (298 K) of the titration of [D6]DMSO
to a CDCl3 solution of the pyridine complex of the SCS PdII pincer
system containing an excess of 1 equiv. pyridine; the designations
u and c correspond to the pyridine α-proton signals of uncoordin-
ated and coordinated pyridine, respectively
Substituent Effects on Pyridine Complex Stability
Within a series of substituted pyridines, we anticipated
that subtle differences in nitrogen electron density would
control the coordination strength towards PdII. When two
different pyridines (each 1 equiv.) were mixed with 1 equiv.
of the acetonitrile complex 1 in CDCl3 (Scheme 3), the α-
proton signals of the two pyridines in the 1H NMR spectra
were used as a probe to determine the extent of coordina-
tion of each pyridine (e.g. see Figure 3).
The pyridine exchange rate is dependent on the solvent.
In CDCl3 two separate signals for bound and free pyridine
were observed up to 50 °C. However, when acetonitrile
complex 1 was treated with 2 equiv. of pyridine in
[D6]DMSO, a coordinating solvent, only one broad signal
1
for the pyridine α-protons was observed in the H NMR
spectrum, indicating fast exchange on the NMR timescale.
When [D6]DMSO was titrated into a CDCl3 solution of
the pyridine complex containing 1 equiv. of excess pyridine
(Figure 2), the rate of pyridine exchange varied with the
concentration of [D6]DMSO.
It is known that the kinetics of substitution in d8 square-
planar complexes depends on two factors, one related to
direct ligand attack (associative-dissociative mechanism)
and the other to an associative solvolysis pathway, in which
the leaving group is first displaced by a solvent molecule
and which is subsequently substituted by the incoming li-
gand (dissociative-associative mechanism).[11] The observed
Scheme 3. Competition experiments between substituted pyridines
Table 1 shows that electron-donating pyridines bind more
pyridine exchange rate can be rationalized in this way. With strongly than pyridine, whereas electron-withdrawing pyrid-
increasing [D6]DMSO concentration, the pyridine exchange ines are weaker ligands than pyridine. The stability of these
increasingly follows the solvolysis pathway. In CDCl3 this pyridine complexes follows the Hammett equation using the
pathway is not possible and only slow direct pyridine at- σϩ constant of Brown and Okamoto.[13] This modified sub-
tack occurs.
stituent constant applies to reactions in which strong ac-
Eur. J. Inorg. Chem. 2000, 2533Ϫ2540
2535

F. C J. M. van Veggel, D. N. Reinhoudt et al.
FULL PAPER
would be expected that 2-picoline would bind stronger than
3-picoline. However, as the reverse is found experimentally,
it must be concluded that steric hindrance between the 2-
methyl group and the aryl groups of the PdII pincer system
overrules electronic effects in this case.
Coordination of Triphenylphosphane to PdII
The coordination of phosphorus-containing ligands to
SCS PdII pincers has, to the best of our knowledge, not
been investigated. Only phosphanes coordinated to trans-
ition metal pincer systems of the PCP[14] and NCN[15] type,
and a phosphane coordinated to an SCS PtII pincer[3] have
been reported. Upon addition of 1 equiv. of PPh3 to 1, sev-
eral distinct spectroscopic features evidenced phosphane co-
ordination to PdII. Firstly, in the 1H NMR spectrum a
broad peak corresponding to the CH2S protons is present.
The signal is sharper and is shifted downfield (Ϯ 0.25 ppm)
when compared to the MeCN complex. Furthermore, the
signal from the aromatic protons is a doublet (J ϭ 2.9 Hz)
due to proton-phosphorus coupling. In the 31P NMR spec-
trum, the phosphorus signal shifts from δ ϭ Ϫ5.6 (free
PPh3) to δ ϭ 13.6 upon coordination. These combined re-
sults prove the coordination of PPh3 trans to the cyclo-
metallated aryl group. Opening of the SϪPd chelate ring,
as suggested by Shaw et al.,[16] does not occur.
Figure 3. Part of the 1H NMR spectrum (CDCl3, 298 K) for a com-
petition experiment between pyridine and 4-methoxypyridine. The
designated signals correspond to the pyridine α-proton signals of:
a) free pyridine, b) free 4-methoxypyridine, c) bound pyridine,
d) bound 4-methoxypyridine
ceptor character develops at the aromatic ring, e.g. solvo-
lysis of tertiary halides in which ionization is rate-determin-
ing. Using the selectivities reported in Table 1 (only the data
of the 4-substituted pyridines are analyzed), a Hammett
plot as shown in Figure 4 is obtained with a ρϩ value of
Ϫ1.0.
Table 1. Results of coordination competition experiments between
substituted pyridines
Selectivity[a]
X-ray Structure of 3·(PPh3)2
X
Y
% bound pyrϪX/pyrϪY
The dipincer complex 3, which is more soluble in organic
solvents than reported dipincers[17] due to its tert-butyl
groups, was mixed with 2 equiv. of PPh3. Crystals were
grown by vapor diffusion of diethyl ether into a CHCl3 so-
lution of the bis(PPh3) complex (Figure 5).
4-H
4-H
4-H
4-H
4-H
4-H
4-H
2-Me
2-Me
4-NH2
4-MeO
4-tBu
4-Me
17:83
32:68
40:60
45:55
44:56
62:38
64:36
43:57
57:43
4.88
2.13
1.50
1.22
1.27
0.61
0.56
1.33
0.75
4-Ph
4-Ac
4-CO2Me
4-Me
3-Me
[a]
Ratio of bound pyridineϪY over bound pyridineϪX.
Figure 4. Hammett plot of the stability of 4-substituted pyridine
complexes with the SCS PdII pincer system against the
BrownϪOkamoto σϩ substituent constant
Usually in systems obeying the BrownϪOkamoto equa-
tion, the values of ρϩ are much more negative. However,
in those cases a full positive charge develops in the rate-
determining step instead of the partial positive charge de-
veloping on nitrogen in pyridineϪPdII coordination.
Figure 5. Structure of 3·(PPh3)2; a) side view, b) top view
In the bis(triphenylphosphane) complex, the Pd atoms
are in
a
distorted square-planar environment:
The effect of steric hindrance was investigated in the S(1)ϪPd(1)ϪS(2) 167.15(3)° and P(1)ϪPd(1)ϪC(1)
picoline series. On the basis of electronic effects alone, it 178.53(9)°. The following bond lengths for the atoms bon-
2536
Eur. J. Inorg. Chem. 2000, 2533Ϫ2540

Coordination Chemistry of SCS PdII Pincer Systems
FULL PAPER
ded to PdII were obtained: PdϪS(1) 2.3075(10) A, PdϪS(2)
As can be seen from Table 2, the coalescence temperature
˚
˚
˚
˚
2.3057(9) A, PdϪP 2.3785(9) A, PdϪC(1) 2.032(3) A. The decreases in the order acetonitrile Ͼ pyridine Ͼ chloride
PdIIϪC bond is substantially longer than in complex 1 Ͼ triphenylphosphane. For the neutral ligands this order
˚
(2.03 vs. 1.98 A). The complex has Ci symmetry (inversion corresponds to the lone pair electron donation (the nucleo-
point at the center of the bispalladated ring) in which the philicity) towards PdII. In the case of PPh3, no unambigu-
two tert-butyl phenyl groups attached to the metal-bound ous coalescence temperature could be observed. NMR ex-
sulfur atoms of the same SCS PdII pincer moiety are ori- periments at low temperatures reveal multiple signals for
ented towards the same side of the square plane.
the CH2S protons for all the complexes mentioned above.
On one hand, freezing out the ring puckering results in loss
of the apparent C2v symmetry of the complex that exists at
Stability of Coordinated PPh3 Complexes
In transition metal inorganic complexes, PPh3 is in gen- higher temperatures in solution. On the other hand, in the
eral a stronger ligand than pyridine. This is expressed in PPh3 complex one very broad signal is observed, ranging
their respective nucleophilicity parameters, for example, from δ ϭ 4.3 to δ ϭ 5.2 at Ϫ10 °C. Splitting of the signal
which differ substantially (n0Pt ϭ 3.19 for pyridine and 8.93 into two doublets is not observed, suggesting that besides
1
for PPh3, respectively).[11] In the H NMR spectrum of a lowering the rate of ring interconversion, rotation of the
mixture of 1 equiv. of PPh3 and the pyridine complex of bulky PPh3 ligand around the PdIIϪP bond is also re-
compound 1 (Scheme 2: X ϭ pyridine), quantitative substi- stricted, leading to different conformational isomers. More-
tution of pyridine by PPh3 is shown. The spectrum is a su- over a long range coupling to 31P cannot be excluded a
perposition of the spectra of free pyridine and the PPh3 priori.
complex. Heating the PPh3 complex in refluxing CHCl3
with a 10-fold excess of pyridine did not result in any de-
tectable exchange.
Table 2. Coalescence temperatures and free energies of activation
for ring interconversion of SCS PdII pincer complexes
As mentioned before, ring opening of the SϪPd chelate
by PPh3 has been suggested. Upon adding 5 equiv. of PPh3
to acetonitrile complex 1, no changes were observed in
either the 1H or 31P NMR spectra. The situation was differ-
ent when PPh3 was added to the chloro complex 2. Increas-
ing amounts (0.25, 0.5, 1.0, 1.5, 2.5 and 5.0 equiv.) of PPh3
Ligand X
Coalescence temperature
Free energy of activation
Tc [K]
∆G‡c [kJ/mol] at Tc
MeCN
Pyridine
ClϪ
289 Ϯ 2
286 Ϯ 2
271 Ϯ 2
Ͻ 268[a]
57.5 Ϯ 0.3
56.5 Ϯ 0.3
54.8 Ϯ 0.5
Ϫ
PPh3
1
were added to 2. The H and 31P NMR spectra show that
[a]
an equilibrium is established after each addition because in
every case two signals are observed in the 31P NMR spec-
trum, one corresponding to free PPh3 (δ ϭ Ϫ5.6), and the
other indicating coordination of PPh3 different from that
described for the substitution of acetonitrile above because
the signal of coordinated PPh3 now appears at δ ϭ 21.9
instead of δ ϭ 13.6. In the 1H NMR spectrum a signal
appears that corresponds to ‘‘free’’ CH2S protons. This me-
ans that the SϪPd chelate ring opens upon PPh3 addition. It
also seems that both chelate rings open together, as the
opening of only one would destroy the symmetry of the
system. Substitution of the chloro ligand by PPh3 is also
unlikely, as can be inferred from the 31P NMR spectra (i.e.
no signal observed at δ ϭ 13.6). Therefore it is concluded
that 2 molecules of PPh3 displace the sulfur atoms as li-
gands. This only occurs quantitatively with a large excess
(Ͼ 4Ϫ5 equiv.) of PPh3.
No unambiguous coalescence temperature was observed, see
also text.
From the coalescence temperature the free energy of ac-
tivation for ring puckering can be estimated.[12] The free
energies of activation for ring puckering vary between 55
and 58 kJ/mol in the ligand order chloride Ͻ pyridine Ͻ
acetonitrile. From Table 2 it is evident that higher coales-
cence temperatures correspond to higher free energies of
activation at Tc for ring puckering in this system.
Coordination of Other Phosphorus-Containing Ligands
Besides triphenylphosphane other phosphorus-con-
taining ligands were investigated as ligands for the SCS PdII
pincer system. Coordination of these ligands to PdII was
shown by 1H and 31P NMR spectroscopy and by FAB mass
spectrometry (Table 3).
Coalescence Temperatures of Different SCS PdII Pincer
Complexes
The two five-membered chelate rings of the cyclopallad-
As with the substituted pyridines competition experi-
ated pincer system display slow conformational interconver- ments were employed, establishing that the relative coor-
1
sion, which is reflected in the H NMR spectrum by the dination strengths of these ligands increase in the order
broadening of the CH2S proton signal of the pincer com- PCy3 Ͼ PPh3 Ͼ P(OMe3) Ͼ P(OPh)3. The relative differ-
plexes. Different ligands have a different influence on the ences in coordination strengths are quite high, as in all cases
conformational interconversion of the chelate rings. This [PCy3 vs. PPh3, PPh3 vs. P(OMe)3 and P(OMe)3 vs.
phenomenon is reflected in the coalescence temperature Tc P(OPh)3] the selectivity exceeds a 4:1 ratio. Further com-
where the two separate signals of the CH2S protons coalesce petition experiments revealed that all the phosphorus con-
into a broad singlet.
taining ligands coordinate much more strongly than pyrid-
Eur. J. Inorg. Chem. 2000, 2533Ϫ2540
2537

F. C J. M. van Veggel, D. N. Reinhoudt et al.
FULL PAPER
Table 3. Complexes of phosphorus-containing ligands with the SCS PdII pincer system
Complex 1 with
X ϭ
1H NMR Ar H[a]
(ppm)/J (Hz)
31P NMR coord/free ligand
(ppm)
FAB-MS
[M]ϩ found/calcd
P(OMe)3
P(OPh)3
PCy3
6.72/4.8
6.64/5.1
6.66/2.9
122.7/141.0
114.7/127.9
24.0/11.5
623.2/623.1
809.1/809.1
779.5/779.3
[a]
Aromatic protons of cyclopalladated ring.
ine. Only a minor amount of pyridine complex was found sphanes. This was confirmed experimentally by competition
1
in competition experiments with the weakest phosphite li- experiments. The H NMR spectra show that the ratio of
gand [P(OPh3)].
coordinated thiourea to coordinated pyridine is higher than
In the case of PCy3 and P(OPh)3, addition of excess li- 95:5. In the presence of PPh3 the thiourea did not coordin-
1
gand to their respective phosphane or phosphite complexes ate significantly, as was inferred from H and 31P NMR
had no effect on the complex stability, as was the case for spectroscopy. FAB mass spectrometry only showed the
PPh3 (vide supra). However, upon addition of excess PPh3 complex of the SCS PdII pincer system in such mix-
P(OMe)3 to complex 1 [X ϭ P(OMe)3], the SPh ligands tures. This means that the order of ligand strength is PPh3
were substituted by P(OMe)3 to form a complex containing Ͼ thiourea Ͼ pyridine.
three phosphite ligands. This was evidenced by 1H (the
CH2S protons give a sharp signal at δ ϭ 4.2) and 31P NMR
Conclusions
spectroscopy (the two trans phosphite ligands couple to the
cis phosphite ligand and vice versa to give one triplet and
one doublet, J ϭ 73 Hz). In the other cases the presence of
one bulky ligand apparently prevents the coordination of
additional ligands.
The observed order of phosphorus ligand strengths
matches the order of χi values [χi ϭ 0.1 (Cy), 4.3 (Ph), 7.7
(OMe) and 9.7 (OPh)], which is regarded as a parameter
measuring substituent contribution to the σ donor and π
acceptor abilities of phosphorus-containing ligands.[18] On
the other hand, the cone angle θ (the ligand bulkiness) in-
creases in the order P(OMe)3 (107°) Ͻ P(OPh)3 (128°) Ͻ
PPh3 (145°) Ͻ PCy3 (170°).[18] Obviously in our case steric
effects can be ignored, as the ligand strength order is the
same as the electronic parameter χi and not the steric para-
meter θ.
In this paper, a study of the coordination strengths of
various ligands towards a SCS PdII pincer system has been
reported. Several para-substituted pyridines have been co-
ordinated to this pincer system, and the differences in coor-
dination strengths found to arise from electronic and steric
effects. The effect of para substituents was quantitatively
analyzed by correlation with the BrownϪOkamoto σϩ sub-
stituent constants. Moreover, it was demonstrated that
phosphorus- and sulfur-containing ligands are stronger li-
gands than pyridines. These results allow for the convergent
metallodendrimer assembly methodology to be extended.
Investigations concerning these topics are currently un-
derway and the results will be described elsewhere.
Experimental Section
Coordination of Sulfur-Containing Ligands
Finally, we have investigated the coordination of sulfur-
containing ligands to the SCS PdII pincer. Thioureas have
high nucleophilicity values (n0Pt ϭ 7.17 for unsubstituted
thiourea),[11] so they are possible ligands.[19] N-Benzyl-NЈ-
butylthiourea was prepared from butyl isothiocyanate and
benzylamine. Upon addition of this thiourea to the aceton-
General: Melting points were determined with a Reichert melting
point apparatus and are uncorrected. CH2Cl2 and hexane were
freshly distilled from CaCl2. CH3CN (p.a. from Merck) was stored
˚
over molecular sieves (4 A). Other solvents (EtOH, CHCl3, acet-
one) were used as received (p.a. from Merck). All reagents were
purchased from Aldrich and used without further purification. So-
lutions containing phosphanes or phosphites were degassed prior
to use. Ϫ NMR spectra were recorded in CDCl3 (unless stated
otherwise) at 298 K with a Varian Unity 300 locked to the deuter-
ated solvent. Chemical shifts are given relative to tetramethylsilane
(TMS). Ϫ FAB-MS spectra were recorded with a Finnigan MAT
90 spectrometer with m-nitrobenzyl alcohol (NBA) as the matrix.
Ϫ Elemental analyses were performed using a Carlo Erba EA1106.
Ϫ Column chromatography was performed using silica gel (SiO2,
E. Merck, 0.040Ϫ0.063 mm, 230Ϫ240 mesh).
1
itrile complex 1, the H NMR spectrum indicated thiourea
coordination. Firstly, the acetonitrile signal was present at
δ ϭ 1.99, indicating the release of acetonitrile (coordinated
MeCN gives rise to a signal at δ ϭ 2.08). Secondly, signific-
ant changes in the chemical shifts of the thiourea NH pro-
tons and the adjacent CH2 protons were observed. Finally,
the broad signals of the SPh protons of the pincer and es-
sentially all protons of the thiourea indicated restricted
thiourea rotation. Moreover, the FAB-MS spectrum dis-
played an intense signal at m/z ϭ 721.2 corresponding to
the thiourea complex.
1-(n-Butoxy)-3,5-bis(phenylthiomethyl)benzene (5): To a solution of
3,5-bis(phenylthiomethyl)phenol (4) (1.40 g, 4.1 mmol) in CH3CN
(200 mL) was added tBuOK (0.93 g, 8.3 mmol) and the resulting
mixture was stirred at room temperature for 1 h. A solution of n-
From their n0Pt values we can expect that thioureas are
stronger ligands than pyridine but weaker ligands than pho- butyl bromide (0.44 mL, 4.1 mmol) in CH3CN (20 mL) was added
2538 Eur. J. Inorg. Chem. 2000, 2533Ϫ2540

Coordination Chemistry of SCS PdII Pincer Systems
FULL PAPER
dropwise, and the mixture was stirred overnight at 40 °C. After
evaporation of the solvent in vacuo, the residue was taken up in
CH2Cl2 (100 mL), washed with 1  HCl (100 mL), saturated
NaHCO3 (100 mL) and brine (100 mL). The dried (MgSO4) or-
ganic layer was concentrated to dryness and the crude product was
purified by column chromatography (SiO2; eluent: CH2Cl2/hexane,
1:1, v/v) to afford a colorless oil. Yield: 1.26 g (77%). Ϫ 1H NMR:
δ ϭ 0.94 (t, J ϭ 7.3 Hz, 3 H, CH3), 1.36Ϫ1.51 (m, 2 H, CH2CH3),
1.63Ϫ1.74 (m, 2 H, OCH2CH2), 3.83 (t, J ϭ 6.6 Hz, 2 H, OCH2),
4.01 (s, 4 H, CH2S), 6.68 (s, 2 H, Ar H), 6.80 (s, 1 H, Ar H),
7.12Ϫ7.29 (m, 10 H, SPh). Ϫ 13C NMR: δ ϭ 13.3, 18.7, 30.7, 38.5,
67.1, 113.4, 121.1, 125.8, 128.3, 129.4, 135.8, 138.5, 158.8. Ϫ FAB-
MS; m/z: 395.1 [M ϩ H]ϩ, calcd. for C24H26OS2: 395.1.
C26H28BF4NOPdS2·H2O: calcd. C 48.35, H 4.68, N 2.17, S 9.93;
found C 48.18, H 4.50, N 2.25, S 9.92.
Chloro Complex of Pd Pincer 2: Acetonitrile complex 1 (0.75 g,
1.2 mmol) was dissolved in a mixture of CH2Cl2 (75 mL) and
CH3CN (25 mL). After the addition of brine (100 mL), the mixture
was stirred vigorously overnight. After separation of the layers, the
organic phase was concentrated in vacuo and the residue was puri-
fied by column chromatography (SiO2; eluent: CH2Cl2/Et2O, 95:5,
v/v) to afford a yellow solid. Yield: 0.60 g (94%). Ϫ M.p. 63Ϫ64
1
°C. Ϫ H NMR: δ ϭ 0.94 (t, J ϭ 7.3 Hz, 3 H, CH3), 1.40Ϫ1.48
(m, 2 H, CH2CH3), 1.66Ϫ1.75 (m, 2 H, OCH2CH2), 3.85 (t, J ϭ
6.4 Hz, 2 H, OCH2), 4.53 (br s, 4 H, CH2S), 6.56 (s, 2 H, Ar H),
7.32Ϫ7.37 (m, 6 H, SPh), 7.78Ϫ7.83 (m, 4 H, SPh). Ϫ 13C NMR:
δ ϭ 13.3, 18.7, 30.8, 51.2, 67.3, 108.3, 129.1, 129.2, 130.9, 131.9,
149.6, 150.9, 156.5. Ϫ FAB-MS; m/z: 499.6 [M Ϫ Cl ϩ H]ϩ, calcd.
500.0. Ϫ C24H25ClOPdS2: calcd. C 53.83, H 4.71, S 11.98; found
C 53.90, H 4.70, S 11.88.
Acetonitrile Complex of Pd Pincer 1: Ligand 5 (233 mg, 0.6 mmol)
was dissolved in CH3CN (50 mL) and the solution was placed un-
der argon. Pd[MeCN]4(BF4)2 (262 mg, 0.6 mmol) was added in one
portion, and the mixture was stirred at room temperature for 2 h.
After evaporation of the solvent in vacuo, the yellow paste was
taken up in CH3CN (3 mL) and the product was precipitated by
dropwise addition of diethyl ether. Yield: 334 mg (90%). Ϫ M.p.
68Ϫ70 °C. Ϫ 1H NMR: δ ϭ 0.95 (t, J ϭ 7.3 Hz, 3 H, CH3),
Dipincer Ligand 6: A mixture of 1,2,4,5-tetrakis(bromomethyl)ben-
zene (3.00 g, 6.7 mmol), 4-tert-butylthiophenol (5.00 g, 30.1 mmol),
K2CO3 (9.20 g, 66.6 mmol) and 18-crown-6 (1.80 g, 6.8 mmol) in
1.37Ϫ1.52 (m, 2 H, CH2CH3), 1.67Ϫ1.76 (m, 2 H, OCH2CH2), acetone (150 mL) was heated at reflux overnight under argon. After
2.08 (s, 3 H, CH3CN), 3.88 (t, J ϭ 6.4 Hz, 2 H, OCH2), 4.57 (br s, evaporation of the solvent, the residue was taken up in CH2Cl2
4 H, CH2S), 6.59 (s, 2 H, Ar H), 7.48Ϫ7.50 (m, 6 H, SPh), (250 mL) and washed with brine (250 mL). After drying the organic
7.77Ϫ7.81 (m, 4 H, SPh). Ϫ 13C NMR (CD3CN): δ ϭ 13.6, 19.4,
layer with MgSO4 and evaporation of the solvent in vacuo, the
31.4, 50.3, 68.2, 110.2, 130.6, 131.3, 131.5, 132.1, 151.8, 158.2. Ϫ crude product was recrystallized from EtOH/CHCl3 to afford a
FAB-MS; m/z: 499.0 [M Ϫ BF4 Ϫ CH3CN]ϩ, calcd. 499.0. Ϫ white solid. Yield: 4.30 g (81%). Ϫ M.p. 185Ϫ186 °C. Ϫ 1H NMR:
Table 4. Crystallographic data for 1 and 3·(PPh3)2
1
3·(PPh3)2
Empirical formula
Molecular mass
Crystal system
Space group
C26H28NOPdS2·BF4
627.86
C86H90P2Pd2S4·2BF4·2CHCl3
1939.07
triclinic
monoclinic
P21/c (No. 14)
17.9797(16)
8.7315(12)
20.390(5)
90
¯
P1 (No. 2)
˚
a [A]
10.2613(12)
11.807(2)
19.428(3)
97.598(8)
94.227(10)
103.319(10)
2257.1(6)
1.426
˚
b [A]
˚
c [A]
α [°]
β [°]
γ [°]
122.084(12)
90
3
˚
V [A ]
2712.1(9)
1.538
Dcalcd. [g cmϪ3
]
Z
4
1
F (000)
1272
990
µ [mmϪ1] (Mo-Kα)
Crystal colour
Crystal size [mm]
Data collection:
θmin, θmax [°]
0.9
0.8
yellow
pale yellow
0.05 ϫ 0.15 ϫ 0.15
0.15 ϫ 0.25 ϫ 0.40
1.6, 27.5
1.5
0.9, 25.4
X-ray exposure [h]
Data set
4
Ϫ23:23, Ϫ11:9, Ϫ23:26
Ϫ12:12, Ϫ14:13, Ϫ18:23
Total data
21995
15846
8259
0.0601
0.0757
Ϫ
Total unique data
Rint
6192
0.0396
Rσ
0.0370
Absorption corr. range
Refinement:
0.953, 1.027 [MULTI-SCAN][a]
No. of refined parameters
409
415
Final R1[b]
0.0291 [5013 I Ͼ 2σ(I)]
0.0823
0.0442 [6653 I Ͼ 2σ(I)]
0.1171
Final wR2[c]
Goodness of fit
1.075
1.021
wϪ1[d]
σ2(F2) ϩ (0.0327P)2 ϩ 1.81P
Ͻ 0.001, 0.001
Ϫ0.50, 0.64
σ2(F2) ϩ (0.0589P)2
Ͻ 0.001, 0.001
Ϫ0.52, 0.83
(∆/σ)av, (∆/σ)max
Min. and max. residual density [eϪ
Ϫ3
˚
A
]
[a]
[b]
[c]
2
2 2
2 2
[d]
2
Incorporated in PLATON.[22]
Ϫ
R1 ϭ Σ||Fo| Ϫ |Fc||/Σ|Fo|. Ϫ wR2 ϭ [Σ[w(Fo Ϫ Fc ) ]/Σ[w(Fo ) ]]1/2. Ϫ
P ϭ [Max(Fo , 0) ϩ
2
2Fc ]/3.
Eur. J. Inorg. Chem. 2000, 2533Ϫ2540
2539

F. C J. M. van Veggel, D. N. Reinhoudt et al.
FULL PAPER
δ ϭ 1.29 [s, 36 H, C(CH3)3], 4.04 (s, 8 H, CH2S), 6.92 (s, 2 H, Ar
H), 7.19 (d, J ϭ 8.6 Hz, 8 H, SAr), 7.27 (d, J ϭ 8.6 Hz, 8 H, SAr).
Acknowledgments
Financial support of this research by the Council for the Chemical
Sciences of the Netherlands Organization for Scientific Research
(CW-NWO) is gratefully acknowledged.
Ϫ
13C NMR: δ ϭ 31.3, 34.5, 36.7, 130.8, 132.4, 132.8, 134.9, 150.1.
Ϫ
FAB-MS; m/z: 625.2 [M
Ϫ Ϫ
SAr]ϩ, calcd. 625.3.
C50H62S4·0.5H2O: calcd. C 75.04, H 7.93, S 16.03; found C 75.01,
H 7.86, S 15.90.
[1]
These tridentate ligands are usually abbreviated as NCN, PCP
Bis(acetonitrile) Complex of Pd Dipincer 3: Dipincer ligand 6 (213
mg, 0.3 mmol) was dissolved in CH3CN (40 mL) and the solution
was placed under argon. Pd[MeCN]4(BF4)2 (240 mg, 0.6 mmol)
was added in one portion, and the mixture was heated at reflux
overnight. After evaporation of the solvent, the crude product was
taken up in CH3CN (5 mL) and precipitated by dropwise addition
of Et2O. Yield: 203 mg (60%). Ϫ M.p. Ͼ 280 °C (dec.). Ϫ 1H NMR
(CD3CN): δ ϭ 1.33 [s, 36 H, C(CH3)3], 4.61 (br s, 8 H, CH2S), 7.56
(d, J ϭ 8.8 Hz, 8 H, SAr H), 7.74 (d, J ϭ 8.8 Hz, 8 H, SAr H). Ϫ
13C NMR (CD3CN): δ ϭ 30.8, 35.2, 50.0, 127.8, 127.9, 132.0,
and SCS ligands.
[2] [2a]
A. Weisman, M. Gozin, H.-B. Kraatz, D. Milstein, Inorg.
[2b]
Chem. 1996, 35, 1792, and references cited therein. Ϫ
P.
Steenwinkel, R. A. Gossage, G. van Koten, Chem. Eur. J. 1998,
4, 759, and references cited therein.
Only one PtII SCS pincer complex has been reported: G. S.
Hanan, J. E. Kickham, S. J. Loeb, J. Chem. Soc., Chem. Com-
mun. 1991, 893.
[3]
[4]
[5]
[4a] J. E. Kickham, S. J. Loeb, S. L. Murphy, Chem. Eur. J. 1997,
[4b]
3, 1203 Ϫ
J. R. Hall, S. J. Loeb, G. K. H. Shimizu, G. P.
A. Yap, Angew. Chem. Int. Ed. 1998, 37, 121.
[5a]
Recent reviews on metallodendrimers:
G. R. Newkome, E.
145.8, 155.2. Ϫ FAB-MS; m/z: 1001.1 [M Ϫ 2MeCN Ϫ 2BF4
ϩ
[5b]
He, C. N. Moorefield, Chem. Rev. 1999, 99, 1689 Ϫ
M. A.
H]ϩ, calcd. 1001.2. Ϫ C52H65B2F8NOPd2S4·2H2O: calcd. C 49.15,
H 5.47, N 1.10 S 10.09; found C 48.86, H 5.14, N 1.25, S 10.00
(the found values for the elemental composition correspond to the
substitution of one acetonitrile ligand by water, which probably
occurs upon prolonged standing).
Hearshaw, J. R. Moss, Chem. Commun. 1999, 1. Ϫ [5c]C. Gor-
man, Adv. Mater. 1998, 10, 295.
[6]
W. T. S. Huck, F. C. J. M. van Veggel, D. N. Reinhoudt, Angew.
Chem. Int. Ed. Engl. 1996, 35, 1213.
[7] [7a]
W. T. S. Huck, F. C. J. M. van Veggel, B. L. Kropman, D.
H. A. Blank, E. G. Keim, M. M. A. Smithers, D. N. Reinhoudt,
[7b]
J. Am. Chem. Soc. 1995, 117, 8293. Ϫ
W. T. S. Huck, F.
Crystal Structure Determinations of 1 and 3·(PPh3)2: Crystals suit-
able for X-ray structure determination were mounted on a Lindem-
ann-glass capillary and transferred into the cold nitrogen stream of
the diffractometer. Data were collected with a Nonius Kappa CCD
diffractometer on rotating anode (Mo-Kα radiation, graphite mon-
C. J. M. van Veggel, D. N. Reinhoudt, J. Mater. Chem. 1997,
7, 1213.
[8]
W. T. S. Huck, L. J. Prins, R. H. Fokkens, N. M. M. Nibbering,
F. C. J. M. van Veggel, D. N. Reinhoudt, J. Am. Chem. Soc.
1998, 120, 6240.
J. E. Kickham, S. J. Loeb, Inorg. Chem. 1994, 33, 4351.
For example, see: M. Fujita, F. Ibukuro, H. Hagihara, K. Og-
ura, Nature 1994, 367, 720.
M. L. Tobe, J. Burgess, Inorganic Reaction Mechanisms, Long-
man, Essex, 1999.
M. Oki, ‘‘Applications of Dynamic NMR Spectroscopy to Or-
ganic Chemistry’’, Methods in Stereochemical Analysis, vol. 4,
VCH, Weinheim, 1985.
H. C. Brown, Y. Okamoto, J. Am. Chem. Soc. 1958, 80, 4979.
M. Gozin, A. Weisman, Y. Ben-David, D. Milstein, Nature
[9]
˚
ochromator, λ ϭ 0.71073 A, T ϭ 150 K, ϕ and ω scans). Pertinent
[10]
data for the structure determinations are collected in Table 4.
Structures were solved with automated Patterson and subsequent
difference Fourier methods for structure 1 and direct methods for
structure 3·(PPh3)2, using SHELXS86[20] for both structures. Full-
matrix refinement on F2 was performed with SHELXL-97Ϫ2.[21]
The hydrogen atoms of structure 1 were located on a difference
Fourier map and their coordinates were included as parameters in
the refinement; the hydrogen atoms of 3·(PPh3)2 were included in
the refinement on calculated positions riding on their carrier atoms.
The nonhydrogen atoms were refined with anisotropic displace-
ment parameters. The hydrogen atoms were refined with a fixed
isotropic displacement parameter related to the value of the equiva-
lent isotropic displacement parameter of their carrier atoms. The
tBu groups of 3·(PPh3)2 were refined with a disorder model invol-
ving two sites. The BF4Ϫ counteranions and the chloroform solvent
molecules of this compound were also disordered. Since no satis-
factory disorder model could be obtained the contribution of the
disordered region to the structure factors was taken into account
[11]
[12]
[13]
[14]
1993, 364, 699.
[15]
´
R. A. T. M. Abbenhuis, I. Del Rıo, M. M. Bergshoef, J.
Boersma, N. Veldman, A. L. Spek, G. van Koten, Inorg. Chem.
1998, 37, 1749.
[16]
[17]
J. Errington, W. S. McDonald, B. L. Shaw, J. Chem. Soc., Dal-
ton Trans. 1980, 2312.
S. J. Loeb, G. K. H. Shimizu, J. Chem. Soc., Chem. Commun.
1993, 1395.
[18]
[19]
C. A. Tolman, Chem. Rev. 1977, 77, 313.
Isothiocyanates were found to be too weak to displace the ace-
1
tonitrile from complex 1. In the H NMR spectrum (CDCl3),
the diagnostic signal shift from δ ϭ 2.08 (coordinated CH3CN)
to δ ϭ 1.99 (uncoordinated CH3CN) was not observed upon
addition of butyl isothiocyanate to the acetonitrile complex 1.
FT-IR spectroscopy also failed to demonstrate the coordina-
tion of isothiocyanate to PdII.
using the SQUEEZE procedure as incorporated in PLATON.[22]
A
total of 219 eϪ were found in a cavity of 537 A , located at the
inversion center on (0,0,0). Neutral atom scattering factors and an-
omalous dispersion corrections were taken from the International
Tables for Crystallography.[23] Geometrical calculations were per-
formed with PLATON.[22] Crystallographic data (excluding struc-
ture factors) for the structures reported in this paper have been
deposited with the Cambridge Crystallographic Data Centre as
supplementary publication nos. CCDC-141832 [for 3 · (PPh3)2]
and -141833 (for 1). Copies of the data can be obtained free
of charge on application to CCDC, 12 Union Road, Cambridge
3
˚
[20]
[21]
[22]
G. M. Sheldrick, SHELXS86 Program for crystal structure de-
termination, University of Göttingen, Germany, 1986.
G. M. Sheldrick, SHELXL-97Ϫ2 Program for crystal structure
refinement, University of Göttingen, Germany, 1997.
A. L. Spek, PLATON, A multi-purpose crystallographic tool,
Utrecht University, The Netherlands, 2000; Internet: http://
www.cryst.chem.uu.nl/platon/
[23]
A. J. C. Wilson (Ed.), International Tables for Crystallography,
vol. C, Kluwer Academic Publishers, Dordrecht, The
Netherlands, 1992.
CB2 1EZ, UK [Fax: (internat.)
deposit@ccdc.cam.ac.uk].
ϩ 44-1223/336-033; E-mail:
Received March 21, 2000
[I00103]
2540
Eur. J. Inorg. Chem. 2000, 2533Ϫ2540