Quantification of ultrasound-induced chain scission in Pd II-phosphine coordination polymers

Article abstract of DOI:10.1002/chem.200600120

A kinetically inert, reversible coordination polymer (3) was obtained through complexation of dicyclohexylphosphine telechelic poly(tetrahydrofuran) with palladium(II) dichloride. This coordination polymer is unreactive towards palladium(II) dichloride bis(1-diphenylphosphino)dodecane (4), because ligand dissociation in the coordination polymer is slow. However, upon ultrasonication of solutions of 3 in toluene in the presence of 4, formation of palladium(II) heterocomplexes was observed with ³¹P NMR spectroscopy. Heterocomplex formation, the consumption of 4, and changes in molecular weight were used to quantify the scission process. In the presence of 60 equivalents of the alkyldiphenylphosphine stopper complex, the reduc-tion in molecular weight was strongly enhanced ; over a period of eight hours the weight-averaged molecular weight was reduced from 1.1×10⁵ to 2.3 × 10⁴ g mol^-1 while 47% of the palladium(II) complexes in the coordination polymer had been converted into heterocomplexes. These results show that the system of 3 in combination with scavenger 4 is a suitable system to study the efficiency of ultrasound-induced chain scission of coordination polymers.

Full text of DOI:10.1002/chem.200600120

DOI: 10.1002/chem.200600120
Quantification of Ultrasound-Induced Chain Scission in PdII–Phosphine
Coordination Polymers
Jos M. J. Paulusse, Jeroen P. J. Huijbers, and Rint P. Sijbesma*[a]
Abstract: A kinetically inert, reversible
coordination polymer (3) was obtained
through complexation of dicyclohexyl-
phosphine telechelic poly(tetrahydro-
furan) with palladium(ii) dichloride.
This coordination polymer is unreac-
tive towards palladium(ii) dichloride
bis(1-diphenylphosphino)dodecane (4),
because ligand dissociation in the coor-
dination polymer is slow. However,
upon ultrasonication of solutions of 3
in toluene in the presence of 4, forma-
tion of palladium(ii) heterocomplexes
was observed with 31P NMR spectro-
scopy. Heterocomplex formation, the
consumption of 4, and changes in mo-
lecular weight were used to quantify
the scission process. In the presence of
60 equivalents of the alkyldiphenyl-
phosphine stopper complex, the reduc-
tion in molecular weight was strongly
enhanced; over a period of eight hours
the weight-averaged molecular weight
was reduced from 1.1”105 to 2.3”
104 gmolꢀ1 while 47% of the palladi-
A
polymer had been converted into het-
erocomplexes. These results show that
the system of 3 in combination with
scavenger 4 is a suitable system to
study the efficiency of ultrasound-in-
duced chain scission of coordination
polymers.
Keywords: coordination polymers ·
heterocomplexes
supramolecular
ultrasound
·
phosphanes
chemistry
·
·
Introduction
bonds in the polymer chain allows for preferential scission
at these bonds over other, stronger bonds.[15,16] Ultrasonica-
tion of reversible coordination polymers enables the specific
rupture of a coordinative bond.[17] Our ultimate goal is to
develop mechanochemical synthetic procedures in coordina-
tion chemistry and to enable the mechanical formation of
highly active catalytic species.
Reversible coordination polymers are dynamic supramolec-
ular polymers[1,2] with coordinative bonds in their main
chain. Much attention in this relatively new research area is
directed towards the characterization of the dynamic proper-
ties of these polymers, such as exchange kinetics,[3] ring–
chain equilibria,[4–6] solvent interactions,[6,7] and new metal–
ligand combinations.[8] At the same time, coordination poly-
mers are already beginning to find applications as smart ma-
terials, for instance, as stimuli-responsive[9] and photoactive
polymers.[10,11] Our interest in this field concerns the direct
manipulation of the coordination sphere of transition-metal
complexes by mechanical forces. Ultrasound is known as
one of the most efficient techniques to break polymers in
solution,[12,13] but high molecular weights are required to
transduce the mechanical forces.[14] Employingweaker
Recently, we reported on the use of ultrasound to reversi-
bly break coordination polymer 1, which is based on diphe-
nylphosphine telechelic poly(tetrahydrofuran) and palladi-
um(ii) dichloride.[17] The weight-averaged molecular weight
(Mw) of the coordination polymer was reversibly reduced
from 1.7”105 to 1.0”105 gmolꢀ1 upon irradiation with ultra-
sound for one hour. Upon equilibration, the original molec-
ular-weight distribution was fully restored within 24 h. The
reversibility of the ultrasonic scission process indicated that
only coordinative bonds and no covalent bonds were
[a] J. M. J. Paulusse,+ J. P. J. Huijbers,+ Dr. R. P. Sijbesma
Laboratory of Macromolecular and Organic Chemistry
Eindhoven University of Technology
P.O. Box 513, 5600 MB Eindhoven (The Netherlands)
Fax : (+40)245-1036
E-mail: R.P.Sijbesma@tue.nl
[+] These authors contributed equally to this work.
Supportinginformation for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
4928
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4928 – 4934

FULL PAPER
broken. Hence, the procedure constitutes a novel method to
create coordinatively unsaturated metal ions.
2a). The maximum degree of supramolecular polymerization
(DPmax) is approximately 16 units, which corresponds to ap-
proximately 6% impurity in the material; most likely, mono-
functionalized polymeric ligands or ligands with one oxi-
dized phosphine end group act as a stopper.
For the application of this method in coordination chemis-
try and catalysis, the scission process needs to be highly effi-
cient, and a method to quantify scission efficiency is desira-
ble. However, quantification of scission in polymer 1 is com-
plicated by relatively rapid equilibration and the formation
of cyclic oligomers.
In this paper, we present the quantification of ultrasound-
induced chain scission in kinetically stable coordination poly-
mer 3, which is composed of dicyclohexylphosphine tele-
chelic poly(tetrahydrofuran) and palladium(ii) dichloride.
The use of an alkyldiphenylphosphine palladium(ii) com-
plex, which suppresses cyclization by scavenging reactive
phosphine chain ends, is described, and the rate of ultrasonic
chain scission is analyzed by usingtwo independent meth-
ods: size-exclusion chromatography (SEC) and 31P NMR
spectroscopy.
Design of the stopper complex: For the scavenging of reac-
tive phosphine chain ends created by ultrasonic scission, a
compound is required with high reactivity towards alkyldicy-
clohexylphosphine ligands and coordinatively unsaturated
palladium, whereas the scavenger compound itself, and any
of its reaction products, should not display reactivity to-
wards the coordination polymer. These requirements sug-
gested the use of palladium(ii) dichloride with less-nucleo-
philic bis(alkyldiphenylphosphine) ligands, for example,
complex 4, as a candidate for scavenging chain ends. It was
anticipated that the high nucleophilicity of the alkyldicyclo-
hexylphosphine ligands would result in complete displace-
ment of one alkyldiphenylphosphine ligand (Scheme 2),
while the liberated alkyldiphenylphosphine ligand would be
unreactive towards the coordination polymer, but would
react with coordinatively unsaturated Pd chain ends.
The validity of this assumption was first studied by mixing
1-(dicyclohexylphosphino)dodecane 5 with four equivalents
of complex 4 in [D8]toluene (Figure 1). This immediately led
to quantitative formation of heterocomplex 6, as is evident
from the 31P NMR spectrum depicted in Figure 1. After
mixing, the signal of free ligand 5 had disappeared and a
new peak, originating from free 1-(diphenylphosphino)dode-
cane ligand 7, was observed at d=ꢀ16.0 ppm. Formation of
heterocomplex 6 was confirmed by the appearance of two
doublets at d=27.5 and 13.0 ppm (JP, P =527 Hz) in the
31P NMR spectrum. The large coupling constant implies that
the heterocomplex adopts a trans configuration in tol-
uene.[18]
Results
Synthesis of a kinetically stable coordination polymer: Re-
cently we reported on the synthesis of several palladium(ii)
dichloride bis-phosphine complexes.[5] From studies concern-
ing their ligand-exchange kinetics, it was concluded that al-
kyldicyclohexylphosphine ligands form extremely stable
complexes with palladium(ii) dichloride with equilibration
times of several months. These ligands would be suitable for
the synthesis of a coordination polymer that is kinetically
more stable than polymer 1.
Therefore, dicyclohexylphosphine telechelic poly(tetrahy-
drofuran) was synthesized by cationic polymerization of tet-
rahydrofuran, and was terminated with lithium dicyclohexyl-
phosphine (LiPCy2). This yielded polymeric ligand 2 with a
number-average molecular weight (Mn) of 6.7”103 gmolꢀ1
,
To evaluate the reactivity of stopper complex 4 towards
the coordination polymer, a solution of 3 in toluene
(1.5 mm) containingcomplex 4 (1.7 equiv) was stirred with-
out sonication at 208C. The reaction was followed by moni-
toringthe reduction of the peak of 4 in the SEC trace (see
the SupportingInformation). Within the first hour approxi-
mately 1.2% heterocomplex formed, whereas after five
hours only 2.2% heterocomplex was observed. This time
course suggests that the initial exchange is due to the pres-
ence of impurities, such as bridged palladium(ii) dichloride
complexes and free dicyclohex-
1
obtained from H NMR analyses, and a polydispersity index
(PDI) of 1.11. Coordination polymer 3 was obtained by stir-
ringa solution of ligand 2 in toluene (20.0 gLꢀ1, 2.91 mm)
with an excess of palladium(ii) dichloride for seven days
(Scheme 1). SEC analysis of 3 showed the presence of high-
molecular-weight material (Mtop =1.1”105 gmolꢀ1 (Mtop
=
peak molecular weight), Mw =6.1”104 gmolꢀ1, based on pol-
ystyrene (PS) standards), accompanied by a peak at 1.0”
104 gmolꢀ1 that represents the cyclic monomer (see Figure
ylphosphine ligands, and that
the true exchange rate between
coordination polymer
3 and
stopper complex 4 is very low.
This hypothesis was verified by
mixinga solution of the
chloro-bridged
coordination
polymer 8 with stopper com-
plex 4. After 30 min, approxi-
mately 15% exchange had
taken place, leadingto the for-
Scheme 1. Synthesis of a,w-bis(dicyclohexylphosphino)poly(tetrahydrofuran) (2) and its palladium(ii) chloride
coordination polymer 3. Tf=trifluoromethanesulfonyl.
Chem. Eur. J. 2006, 12, 4928 – 4934
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
4929

R. P. Sijbesma et al.
probe in the presence of 0, 1.7, 10, and 60 equivalents of
complex 4. Samples for 31P NMR and SEC analyses were
taken at regular intervals (see Figure 2a). In the absence of
Scheme 2. General scheme for the interception of dicyclohexylphosphine
chain ends with a low-molecular-weight complex containing less-nucleo-
philic diphenylphosphine ligands.
Figure 2. SEC traces measured before and after a) 3 h of sonication of 3
in the presence of 0, 1.7, and 60 equivalents of 4 and b) before and after
8 h of sonication of 3 in the presence of 60 equivalents of 4, with assign-
ment.
mation of heterocomplexes (see the SupportingInforma-
tion).
Quantification of scission by monitoring the reduction of
molecular weight: Solutions of 3 in toluene (1.5 mm mono-
mer) were sonicated for three hours by usinga sonication
4, a decrease in the weight-averaged molecular weight from
1.3”105 to 7.6”104 gmolꢀ1 was observed after three hours,
while a greater decrease in Mw
was observed in the presence
of 4, to 4.5”104 (1.7 equiv),
4.4”104 (10 equiv), and 4.1”
104 gmolꢀ1 (60 equiv). After
eight hours of sonication, in
the presence of 60 equivalents
of 4, Mw was further decreased
to 2.3”104 gmolꢀ1
.
In addition to giving rise to
a faster decrease in molecular
weight, the presence of 4 also
led to the formation of differ-
ent products. Upon extensive
sonication for eight hours, the
SEC trace of the solution con-
taining60 equivalents of 4 dis-
played two peaks correspond-
ingto molecular weihgts of
Figure 1. 31P NMR spectrum of the product mixture obtained upon addition of 5 to a solution of 4 in
[D8]toluene.
4930
www.chemeurj.org
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4928 – 4934

PdII–Phosphine Coordination Polymers
FULL PAPER
1.5”104 and 2.6”104 gmolꢀ1 (PS standards, see Figure 2b).
These peaks are absent in the SEC trace of 3 sonicated in
the absence of 4, and are assigned to dodecyldiphenylphos-
phine-stoppered monomers and dimers, respectively. The
molecular weights of these species were used to calibrate
the number and weight-averaged degrees of polymerization
(DPn and DPw, respectively) in the SEC traces of the sonica-
tion experiment in the presence of 60 equivalents of 4 (Fig-
ure 3a). These changes were used to calculate the amount of
complex 4, two new doublets appeared in the 31P NMR
spectrum at d=27.5 and 13.0 ppm (JP, P =527 Hz), which
were assigned to the alkyldicyclohexylphosphine and the al-
kyldiphenylphosphine phosphorus atoms in the heterocom-
plex, respectively, in accordance with the earlier described
ligand-exchange experiment. The fraction of broken com-
plexes of 3 was calculated by determiningthe ratio of the
peak intensities (integral, I) by using Equation (2):
I
PCy2 hetero
Fraction of broken complexes in 3 ¼
I
hetero þ I
PCy2
PCy2 homo
ð2Þ
The extent of ultrasonic scission was determined in an inde-
pendent fashion by monitoringthe decrease in intensity of
the signal of 4 in SEC traces of the same solutions.[20]
Duringsonication, 4 is consumed by ligand exchange and
the intensity of the correspondingpeak decreases. Taking
the stoichiometry of the stopper complex into account, the
amount of ultrasonic scission was calculated with Equa-
tion (3):
½4ꢁt¼0ꢀ½4ꢁt¼x
ð3Þ
Fraction of broken complexes in 3 ¼
½3ꢁt¼0
In Figure 4, ultrasonic scission, as derived from NMR and
SEC analyses, has been plotted against time. For sonication
times below 30 min, NMR data are not available, because
the signal to noise ratio of the peaks of the heterocomplex
was insufficient. Nevertheless, a good agreement between
the two quantification methods is observed. The plots of
Figure 4 show that a higher concentration of stopper com-
plex 4 led to increased heterocomplex formation. In the
Figure 3. a) Weight- and number-averaged degree of polymerization and
b) ligand exchange of 3 duringsonication with 60 equivalents of 4.
chain scission expressed as the fraction of broken complexes
in the coordination polymer [Eq. (1)] by usingthe assump-
tion that no recombination takes place. This analysis indi-
cates that sonication results in 39% scission of coordination
complexes over the course of eight hours, with 25% being
broken after three hours (Figure 3b).
1
1
Fraction of broken complexes in 3 ¼
ꢀ
ð1Þ
DPN,t¼x DPN,t¼0
in which DPN is the number averaged degree of polymeriza-
tion. The SEC trace can only be used to evaluate chain scis-
sion when a high concentration of 4 is present. If scavenging
is inefficient, an unknown fraction of cyclic species with
lower hydrodynamic volumes[19] is formed, makingcalibra-
tion impossible.
Quantification of scission by monitoring ligand exchange
with 31P NMR and SEC analyses: The rate of ultrasound-in-
duced chain scission of polymer 3 was further quantified by
monitoringthe formation of the palladium( ii) heterocom-
plex with 31P NMR spectroscopy. Upon sonicatingthe solu-
tions of 3 (in toluene) containingalkyldiphenylphosphine
Figure 4. Ligand exchange in 3 in the presence of a) 1.7 and b) 60 equiv-
alents of 4.
Chem. Eur. J. 2006, 12, 4928 – 4934
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
4931

R. P. Sijbesma et al.
presence of 1.7 equivalents of 4, approximately 20% of the
complexes in the coordination polymer were converted into
a heterocomplex after three hours, against 35% when
60 equivalents of 4 were present; after eight hours of sonica-
tion, a total of 47% of heterocomplex had formed. Because
the presence of 4 is not expected to influence the rate of
scission, the differences reflect the increased efficiency of
scavenging. Even in the presence of 60 equivalents of 4,
scavenging is not guaranteed to be quantitative. The rate of
heterocomplex formation in the presence of 60 equivalents
of 4 is therefore a lower limit for the true rate of chain scis-
sion.
scission process. In principle, the method based on the re-
duction in DP observed with SEC analyses can be used to
analyze the scission experiment either in the absence or the
presence of an excess of 4, because the coordination poly-
mer is kinetically inert and does not re-equilibrate during
the experiment. However, only in the presence of 4 can the
method be used in a quantitative manner, because in the ab-
sence of stopper an unknown amount of cyclic material with
smaller hydrodynamic volume is formed. Quantification of
the scission efficiency in the presence of 60 equivalents of
stopper shows that the DPn is reduced from 2.7 to 1.3 over
the course of eight hours, implying that during this time
each chain, on average, is broken more than once. It is also
evident that the efficiency of the process is much higher in
the initial stages of sonication, when longer chains are still
present, than duringthe last four hours. This is in line with
the higher susceptibility of longer chains to scission.[12,21,22]
The chain-length dependence of the scission rate means that
in the present system of highly polydisperse polymers it is
difficult to give a number for the overall rate of scission.
The other two methods to quantify scission described
here directly probe the amount of ligand exchange by meas-
uringeither the intensity of the heterocomplex in the
31P NMR spectrum, or the amount of stopper complex con-
sumed accordingto SEC analyses. These data give scission
rates that are in very good
Discussion
In a previous publication on the reversible, ultrasound-in-
duced chain scission of polymer 1, we proposed that the
phosphine-free chain ends initiate a series of ligand-ex-
change reactions, which is terminated by recombination of
chain ends (Scheme 3). Because this process only leads to
redistribution, and not to a reduction in molecular weight,
the stable fragments observed with SEC analyses were pro-
posed to be cyclic species formed by backbiting(step 3 in
Scheme 3).
agreement with each other.
Comparison of these data with
the DP measurements also
gives
a fair agreement, al-
though the amount of chain
scission based on DP gives a
lower estimate for the rate of
scission (39 vs 47% after 8 h,
see Figure 3b). The difference
may be ascribed to a reduction
in the fraction of cyclic species
upon
sonication.
Because
linear products have a larger
hydrodynamic volume than the
correspondingrinsg, this re-
sults in an underestimatation
of the change in DP.
Scheme 3. Proposed mechanism for the reversible ultrasonic scission of coordination polymers in the absence
of a chain-end scavenger.[17]
In the kinetically stable system of coordination polymer 3
in combination with efficient scavenging of chain ends by
complex 4, cyclization and redistribution are suppressed,
and a higher efficiency of chain scission is predicted. Quali-
tative comparison of the SEC traces in Figure 2a shows that
the addition of complex 4 indeed strongly enhances the effi-
ciency of molecular-weight reduction. This confirms the hy-
pothesis that without a stopper, only a fraction of the scis-
sion events results in a reduction of molecular weight.
The system of 3 in combination with scavenger 4 is a suit-
able system to study the efficiency of ultrasound-induced
chain scission of coordination polymers. Three independent
methods have been used to quantify the efficiency of the
Conclusion
In summary, all three methods used to quantify chain scis-
sion have limitations. Although the SEC traces give the
most detailed information on the scission process, using
these traces to extract changes in DP is restricted by the
need for calibration and the presence of a changing fraction
of cyclic species. The other two methods, usingNMR or the
signal of the stopper, are limited in the amount of informa-
tion they give on changes in molecular-weight distribution,
but these methods are convenient and their results are in ex-
cellent agreement with each other.
4932
www.chemeurj.org
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4928 – 4934

PdII–Phosphine Coordination Polymers
FULL PAPER
SEC (PS standards): Mn =15200 gmolꢀ1, Mw =16900 gmolꢀ1, PDI=Mw/
Mn =1.11.
The heterocomplex formation reported here is an impor-
tant step toward ultrasound-induced coordination mechano-
chemistry and mechanocatalysis. For application in catalysis,
however, practical applications require an increased scission
rate. In future work, addressingthis issue by usingcoordina-
tion polymers with star and network architectures will be in-
vestigated. The use of ultrasound in the synthesis of new co-
ordination compounds on the other hand, is within reach
when monofunctionalized polymeric phosphine ligands are
used. These ligands will lead to complexes with a single
metal center in the middle of the chain, which will allow
quantitative scission and enable ultrasound-controlled coor-
dination chemistry.
Palladium(ii) dichlorido [a,w-bis(dicyclohexylphosphino)poly(tetrahydro-
furan)] (3): Due to the sensitivity of ligand 2 towards oxygen, a large
amount of the ligand is rapidly complexed by using a soluble palladium
source and the insoluble palladium dichloride was used to obtain perfect
stoichiometry.[5]
Dichloro(1,5-cyclooctadiene)palladium(ii)
(91 mg,
0.32 mmol, 0.75 equiv), palladium(ii) dichloride (27 mg, 0.15 mmol,
0.35 equiv), and 2 (2.69 g, ꢂ0.27 mmol) were dissolved in dry toluene
(135 mL) and this mixture was stirred for 7 d. The mixture was filtered to
remove the excess palladium(ii) dichloride yieldinga clear, yellow solu-
tion. The solvent was not evaporated; the solution (20 gLꢀ1) was immedi-
ately used in sonication experiments. Only a small sample was dried for
analysis purposes. 1H NMR (400 MHz, CDCl3): d=3.58–3.22 (m, n”4H,
CH2O), 1.90–1.19 ppm (m, m”n”4H, CH2CH2O;m”44H, alkyl);
13C NMR (100 MHz, CDCl3): d=70.74 (s, m”n”2C, CH2O), 33.03 (s,
m”4C, alkyl), 29.43 (s, m”8C, alkyl), 28.96 (d, m”8C, alkyl), 27.52–26.3
(m, m”n”2C, CH2CH2O; 4C, alkyl), 22.09 ppm (s, 2C, PCH2); 31P NMR
(162 MHz, CDCl3): d=23.24 ppm (s, m”2P) (n is the degree of polymer-
ization of polymer 2, m is the degree of supramolecular polymerization
of coordination polymer 3).
Experimental Section
Dipalladium(ii) (di-m-chlorido) dichlorido [a,w-bis(dicyclohexylphosphi-
General procedures: 1H NMR (400 MHz), 13C NMR (100 MHz), and
31P NMR (162 MHz) spectra were recorded on a Bruker 400 spectrome-
ter. Chemical shifts are referenced to tetramethylsilane and chloroform
(proton and carbon, respectively) and external 85% phosphoric acid
(phosphorus). Size-exclusion chromatography was performed on a Shi-
madzu LC10-AT instrument, usinga Polymer Laboratories PL Gel 5 mm
mixed-D column (linear range of Mr =200–400000 gmolꢀ1), a Shimadzu
SPD-10AV UV/Vis detector at 254 nm, and chloroform as the eluent at a
flow rate of 1 mLminꢀ1 (208C). Polystyrene standards were used for cali-
bration. Sonication experiments were carried out with a Sonics VCX
500 Watt Ultrasonic Processor purchased from Sonics & Materials Inc. A
13 mm probe was used at a frequency of 20 kHz, at 30% of the maxi-
mum amplitude of 125 mm. Syntheses of the ligands and the complexes
were carried out under a dry argon atmosphere using standard Schlenk
no)poly(tetrahydrofuran)] (8): Coordination polymer
3
(1.23 g,
0.178 mmol) was dissolved in toluene (50 mL) and palladium(ii) dichlor-
ide (51 mg, 0.178 mmol) was added. The mixture was heated to 808C for
24 h. The solvent was removed under vacuum yieldinga dark yellow oil
(1.18 g, 95%).
1H NMR (400 MHz, CDCl3): d=3.60–3.34 (m, n”4H, CH2O), 2.00–
1.25 ppm (m, m”n”4H, CH2CH2O; m”44H, alkyl); 13C NMR
(100 MHz, CDCl3): d=70.54 (s, m”n”2C, CH2O), 35.36 (s, m”4C,
alkyl), 29.24 (s, m”8C, alkyl), 28.76 (d, m”8C, alkyl), 26.62 (m, m”n”
2C, CH2CH2O; 4C, alkyl), 22.051 ppm (s, 2C, PCH2); 31P NMR
(162 MHz, CDCl3): d=55.29 ppm (s, m”2P) (n is the degree of polymer-
ization of polymer 2, m is the degree of supramolecular polymerization
of coordination polymer 8).
Sonication experiments: Part of the 20.0 gLꢀ1 solution of 3 in toluene
(2.91 mm) was diluted with toluene to 10.0 gLꢀ1 (1.45 mm) and a water-
cooled glass vessel was filled with 30.0 mL of this diluted solution. The
solution was sonicated while argon was slowly bubbled through it, and
samples for SEC analysis were taken at regular intervals. These samples
were immediately frozen in liquid nitrogen and left to warm to room
temperature only just before SEC measurement. In the case of quantifi-
cation experiments, a specific amount of complex 4 was added (1.7 equiv:
65.5 mg, 7.40”10ꢀ5 mol; 10 equiv: 385 mg, 4.35”10ꢀ4 mol; 60 equiv:
2.31 g, 2.61”10ꢀ3 mol).
techniques. Tetrahydrofuran was distilled from
a sodium–potassium
alloy; diethyl ether was distilled from molecular sieves; dichloromethane
and deuterated chloroform were distilled from P2O5. Diethyl ether, ace-
tonitrile, and n-hexane were degassed before use. n-Butyllithium was ti-
trated before use. n-Butyllithium (1.6m) was purchased from Aldrich;
palladium dichloride (99.9%) and dicyclohexylphosphine (98%) were
purchased from STREM. The syntheses of 1-(diphenylphosphino)dode-
cane, 1-(dicyclohexylphosphino)dodecane, and their palladium(ii) dichlor-
ide complexes have been reported earlier.[5]
a,w-Bis(dicyclohexylphosphino)poly(tetrahydrofuran) (2): Dicyclohexyl-
phosphine (1.04 g, 5.24 mmol, 2.2 equiv) was dissolved in tetrahydrofuran
(30 mL) and the stirred solution was cooled to ꢀ908C. n-Butyllithium in
hexane (1.6m, 3.27 mL, 5.2 mmol, 2.2 equiv) was slowly added to the mix-
ture. After complete addition, the yellow mixture was stirred and left to
warm to room temperature.
Acknowledgements
A
mixture of di-tert-butylpyridine (DTBP; 42.6 mg, 0.22 mmol,
This work was supported by the National Research School Combina-
tion—Catalysis. The authors thank R. Bovee for help with the SEC meas-
urements.
0.09 equiv) and dry tetrahydrofuran (100 mL) was cooled to 08C with ice.
Trifluoromethane sulfonic anhydride (0.67 g, 2.4 mmol) was added to the
mixture. After stirringfor 50 min at 0 8C, a solution of lithium dicyclohex-
ylphosphine in tetrahydrofuran was added to the reaction mixture. This
mixture was stirred for 30 min and left to warm to room temperature.
The solvent was removed under vacuum, resultingin a yellow oil. In
order to remove residual impurities, the oil was dissolved in 150 mL di-
ethyl ether and this mixture was filtered over silica that had been dried
from water and air. At ꢀ188C, 2 crystallized out and precipitated. This
was repeated once to remove all impurities. Finally the solvent was re-
[1] L. Brunsveld, B. J. B. Folmer, E. W. Meijer, R. P. Sijbesma, Chem.
Rev. 2001, 101, 4071–4097.
[2] A. Ciferri, Supramolecular Polymers, 2nd ed., CRC Press, Boca
Raton, FL, 2005.
[3] W. C. Yount, H. Juwarker, S. L. Craig, J. Am. Chem. Soc. 2003, 125,
15302–15303.
moved under vacuum yieldinga white solid (10.9 ,g 68%).
1H NMR
(400 MHz, CDCl3): d=3.55–3.2 (m, n”4H, CH2O), 1.74–1.17 ppm (m,
n”4H, CH2CH2O; 44H, alkyl); 13C NMR (100 MHz, CDCl3): d=70.70
(s, n”2C, CH2O), 33.40 (d, 4C, alkyl), 30.45 (d, 8C, alkyl), 29.09 (d, 8C,
alkyl) 27.44–25.34 (m, n”2C, CH2CH2O; 4C, alkyl), 21.27 ppm (d, 2C,
PCH2); 31P NMR (162 MHz, CDCl3): d=ꢀ3.25 ppm (s, 2P); molecular
weight: by end-group determination (1H NMR): Mn =6700 gmolꢀ1; by
[4] J. M. J. Paulusse, R. P. Sijbesma, Chem. Commun. 2003, 1494–1495.
[5] J. M. J. Paulusse, J. P. J. Huijbers, R. P. Sijbesma, Macromolecules
2005, 38, 6290–6298.
[6] T. Vermonden, J. van der Gucht, P. de Waard, A. T. M. Marcelis,
N. A. M. Besseling, E. J. R. Sudhçlter, G. J. Fleer, M. A. C. Stuart,
Macromolecules 2003, 36, 7035–7044.
Chem. Eur. J. 2006, 12, 4928 – 4934
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
4933

R. P. Sijbesma et al.
[7] T. Vermonden, W. M. de Vos, A. T. M. Marcelis, E. J. R. Sudhçlter,
Eur. J. Inorg. Chem. 2004, 2847–2852.
[16] K. L. Berkowski, S. L. Potisek, C. R. Hickenboth, J. S. Moore, Mac-
romolecules 2005, 38, 8975–8978.
[17] J. M. J. Paulusse, R. P. Sijbesma, Angew. Chem. 2004, 116, 4560–
4562; Angew. Chem. Int. Ed. 2004, 43, 4460–4462.
[8] U. Michelsen, C. A. Hunter, Angew. Chem. 2000, 112, 780–783;
Angew. Chem. Int. Ed. 2000, 39, 764–767.
[18] J. G. Verkade, Coord. Chem. Rev. 1972, 9, 1–106.
[19] J. A. Semlyen, Pure Appl. Chem. 1981, 53, 1797–1804.
[20] UV and refractive index (RI) data can both be used and give com-
parable data. At high stopper concentrations, UV detection suffers
from signal saturation and therefore RI detection has been used in-
stead.
[21] A. Casale, R. S. Porter, Polymer Stress Reactions, Academic Press,
London, 1978.
[22] A. M. Basedow, K. H. Ebert, Makromol. Chem. 1975, 176, 745–757.
[9] J. B. Beck, S. J. Rowan, J. Am. Chem. Soc. 2003, 125, 13922–13923.
[10] R. Dobrawa, M. Lysetska, P. Ballester, M. Grꢁne, F. Wꢁrthner, Mac-
romolecules 2005, 38, 1315–1325.
[11] R. Dobrawa, F. Wꢁrthner, J. Polym. Sci., Part A: Polym. Chem.
2005, 43, 4981–4995.
[12] T. J. Mason, J. P. Lorimer, Applied Sonochemistry, 1st ed., Wiley-
VCH, Weinheim, 2002.
[13] K. S. Suslick, G. J. Price, Annu. Rev. Mater. Sci. 1999, 29, 295–326.
[14] A. M. Basedow, K. H. Ebert, Adv. Polym. Sci. 1977, 22, 83–148.
[15] M. V. Encina, E. Lissi, M. Sarasua, L. Gargallo, D. Radic, J. Polym.
Sci., Polym. Lett. Ed. 1980, 18, 757–760.
Received: January 26, 2006
Published online: April 4, 2006
4934
www.chemeurj.org
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4928 – 4934