Binding Properties of Two Novel Phosphane-Modified Tetrapodants and Their Bimetallic Transition Metal Complexes

Article abstract of DOI:10.1002/(SICI)1099-0682(199812)1998:12<1975::AID-EJIC1975>3.0.CO;2-K

Two novel phosphane ligands 3 and 4 based on the rigid diphenylglycoluril molecule have been synthesized and characterised. Binding studies with 3 and 4, using 1,3-dihydroxybenzene derivatives reveal that ligands 3 and 4 behave similarly to clip molecule 5, which has the same binding site as ligands 3 and 4. The size of the flexible spacers in the ligands has been varied and the effect of this variation on the association constant of resorcinol denvatives has been determined. These cavity-containing ligands are able to coordinate two transition metal centres, leading to bimetallic macrocycles. The metallamacrocycles formed from 4 containing platinum or rhodium bind the guest, olivetol (5-pentylbenzene-1,3-diol), almost four times as strongly as the free tetrapodant 4. Complexes of 4 having palladium centres display similar or reduced binding affinities for resorcinol derivatives, when compared to free 4. Metal complexes of ligand 3 do not form host-guest complexes, probably because of a too small a ring-size of the metallamacrocycle.

Full text of DOI:10.1002/(SICI)1099-0682(199812)1998:12<1975::AID-EJIC1975>3.0.CO;2-K

FULL PAPER
Binding Properties of Two Novel Phosphane-Modified Tetrapodants and Their
Bimetallic Transition Metal Complexes
Georg C. Dola, Sander Gaemersa, Marko Hietikkoa, Paul C. J. Kamera, Piet W. N. M. van Leeuwen*b,
and Roeland J. M. Noltea
Institute of Molecular Chemistry, University of Amsterdama,
Nieuwe Achtergracht 166, NL-1018 WV Amsterdam, The Netherlands
Fax: (internat.) ϩ 31-20/5256456
E-mail: tijdink@sci.kun.nl
Department of Organic Chemistry, University of Nijmegenb,
Toernooiveld, NL-6525 ED Nijmegen, The Netherlands
Fax: (internat.) ϩ31-24/3652929
E-mail: pwnm@anorg.chem.uva.nl
Received June 18, 1998
Keywords: Supramolecular chemistry / Metallamacrocycle / Hydrogen bonding / Phosphane
Two novel phosphane ligands 3 and 4 based on the rigid
diphenylglycoluril molecule have been synthesized and
characterised. Binding studies with 3 and 4, using 1,3-
dihydroxybenzene derivatives reveal that ligands 3 and 4
behave similarly to clip molecule 5, which has the same
binding site as ligands 3 and 4. The size of the flexible
spacers in the ligands has been varied and the effect of this
variation on the association constant of resorcinol derivatives
has been determined. These cavity-containing ligands are
able to coordinate two transition metal centres, leading to
bimetallic macrocycles. The metallamacrocycles formed from
4 containing platinum or rhodium bind the guest, olivetol (5-
pentylbenzene-1,3-diol), almost four times as strongly as the
free tetrapodant 4. Complexes of 4 having palladium centres
display similar or reduced binding affinities for resorcinol
derivatives, when compared to free 4. Metal complexes of
ligand 3 do not form host-guest complexes, probably because
of a too small a ring-size of the metallamacrocycle.
Introduction
bond strength between host and guest was investigated.
These results are of interest for future applications of metal
The use of enzymes as catalysts for organic reactions is a complexes of 3 and 4 as catalysts for e.g. the hydrofor-
major topic in modern chemistry. Enzyme-catalysed reac- mylation reaction or cross-coupling reactions.
tions can reach incredibly high rates and selectivities. Man-
made catalysts do not reach that level of supremacy yet,
but recently some elegant examples of synthetic enzymes
(synzymes) have been published. These synzymes consist of
Results and Discussion
Ligand Synthesis
the building block diphenylglycoluril (ϭ DPGU),[1Ϫ5]
a
cyclodextrin[6Ϫ8] or a cyclophane[9] and a catalytically ac-
tive centre. Synzymes do not match the natural enzymes in
reactivity, but they are interesting molecules to study the
fundamental processes taking place in enzyme catalysis. In
this way processes like substrate coordination and sub-
strate-metal interactions can be studied in a less compli-
cated environment. The knowledge obtained can then be
used in the development of shape-selective cata-
lysts.[3Ϫ5,10,11] Non-covalent binding interactions like hydro-
gen bonding, π-π stacking, electrostatic, and van der Waals
interactions induce substrate discrimination in the processes
mentioned above.
The tetrapodant starting compounds 1 and 2 were pre-
pared according to literature procedures[12] and sub-
sequently converted into 3 and 4 after reaction with four
equivalents of LiPPh2 (see Scheme 1). The two tetradentate
ligands 3 and 4 were obtained in 80% and 67% yield,
respectively. Also, we tried to synthesize compound 4 from
2 and four equivalents of HPPh2 in the presence of 10
equivalents of tBuOK. This reaction afforded only small
amounts of the desired product. Under these conditions,
the main reaction that occurred was cleavage of the ether
In this paper two tetrapodant host molecules 3 and 4 moieties.
based on the rigid building block DPGU are described. The
The two ligands (3 and 4) are moderately air-sensitive,
binding affinities of these host molecules towards 5-substi- but they are stable under an inert atmosphere. When the
tuted resorcinol (1,3-dihydroxybenzene) derivatives will be products were not carefully worked up or stored, 31P-NMR
presented. Several transition metal complexes of 3 and 4 spectra showed signals in the phosphane oxide region (δ ϭ
were prepared and the effect of metal complexation on the ϩ 30). These oxide contaminations could be removed by
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ1948/98/1212Ϫ1975 $ 17.50ϩ.50/0
1975

P. W. N. M. van Leeuwen et al.
FULL PAPER
reduction with an excess of HSiCl3 in the presence of 3
equivalents of NEt3 per HSiCl3 in refluxing benzene.[13]
A
phosphane-free
precursor
compound
[e.g.
(COD)PdCl2] could also be used, provided that a small
amount of PPh3 was added to the reaction mixture. When
the syntheses were performed without PPh3, large amounts
of insoluble compounds were formed in addition to the de-
sired product. The replacement of COD or CH3CN by a
phosphane ligand is fast compared to a phosphane-by-
phosphane replacement. Before intramolecular ring closure
can occur, intermolecular complexation takes place. The
polymeric complexes precipitate and cannot equilibrate to
form the desired bimetallic complexes. These reactions were
also performed at lower concentrations (3 ϫ 10Ϫ4 ), but
always some insoluble product was formed. A suspension
of precipitated (polymeric) complex could be treated with a
small amount of PPh3 to redissolve the polymeric products.
The PPh3 that remained after the reaction could be re-
moved with the same ease as weakly bound ligands such
as COD and CH3CN. A simple precipitation of the metal
complex after the addition of pentane, hexane or diethyl
ether, followed by filtration or decantation of the suspen-
sion was sufficient to remove PPh3.
Scheme 1. Synthesis of ligands 3 and 4
There are two exceptions to the general procedure de-
scribed above: the formation of L[Pd(BF4)2]2 and
L[Pd(TCNE)]2 (L ϭ 3 or 4, TCNEϭ tetracyanoethylene).
For the preparation of the aforementioned ionic palladium
compounds, CH3CN was used as a cosolvent. The products
formed still contained two molecules of CH3CN, which in-
creased the solubility of the products. Hence, polymeric
Metal Complex Synthesis
The starting compounds in the metal complexation reac- structures did not precipitate and the desired products were
tions and the yields are summarised in Table 1. Most of the obtained. For the formation of the L[Pd(TCNE)]2 com-
complexes of 3 and 4 were synthesized from starting metal plexes the reaction of Pd(DBA)2 (DBAϭ dibenzylideneace-
complexes containing triphenylphosphane. After addition tone) with either 3 or 4 was chosen. This resulted in the
of the tetradentate ligands 3 or 4 to these PPh3-containing initial formation of L[Pd(DBA)]2 complexes. The replace-
compounds, the bimetallic complexes of 3 or 4 were ment of the second DBA molecule is known to be less
formed. The driving force of this reaction is the replacement favourable,[14] but this could be achieved by the addition
of a triarylphosphane ligand by an alkyldiphenylphos- of TCNE, an extremely electron-poor alkene. Complexes
phane, which is a stronger donor ligand. The reaction is possessing other electron-poor alkenes like maleic anhy-
also entropy-driven; a monodentate ligand is always re- dride, dimethyl fumarate or fumaronitrile could not be iso-
placed by a bidentate ligand having similar electronic and lated. These alkenes are not as tightly bound to the Pd cen-
steric properties.
tre as TCNE and therefore afford less stable compounds.
Table 1. Starting compounds, products, yields and spectroscopic data of the bimetallic macrocycles of 3 and 4
Entry
Starting compound
Product
Yield [%]
δ(31P) [ppm] (1JM-P [Hz])
ν(CX)[f]
[ ]
Ϫ1
1
(PPh3)2PdCl2
3(PdCl2)2
58
75
84
85
93
79
85
84
79
68
96
96
89
85
12.5[a]
14.9[a]
15.4[b]
21.9[b]
38.4[c]
42.8[c]
24.8[b]
12.8[b]
1710[g]
1710[g]
1710[g]
1710[g]
1648[g]
1670[g]
2
CODPdCl2/PPh3
(PPh3)2PdMeCl
CODPdMeCl/PPh3
(CH3CN)4Pd(BF4)2
(CH3CN)4Pd(BF4)2
Pd(DBA)2 ϩ TCNE
Pd(DBA)2 ϩ TCNE
(PPh3)2PtCl2
4(PdCl2)2
3
3(PdMeCl)2
4(PdMeCl)2
3[Pd(BF4)2]2
4[Pd(BF4)2]2
3[Pd(TCNE)]2
4[Pd(TCNE)]2
3(PtCl2)2
4
5
6
7
1701[g], 2220[h]
1707[g], 2213[h]
1707[g]
8
9
0.4[b] (3550[d]
)
)
10
11
12
13
14
(PPh3)2PtCl2
4(PtCl2)2
4.8[b] (3610[d]
1708[g]
(PPh3)2RhClCO
(PPh3)2RhClCO
(PPh3)2IrClCO
3(RhClCO)2
4(RhClCO)2
3(IrClCO)2
4(IrClCO)2
18.6[b] (124[e])
22.0[b] (123[e])
14.8[a]
1710[g], 1972[i]
1708[g], 1970[i]
1710[g], 1959[j]
1711[g], 1958[j]
(PPh3)2IrClCO
17.7[a]
[d] 1
[e] 1
[a] Measured in C6D6. Ϫ [b] Measured in CDCl3. Ϫ [c] Measured in CD3CN. Ϫ
[g] ν(CO). Ϫ [h] ν(CC). Ϫ [i] ν[(CO)Rh]. Ϫ [j] ν[(CO)Ir].
J
(cis). Ϫ
J
(trans). Ϫ [f] Measured in KBr.
Rh-P
Pt-P
Ϫ
1976
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985

Phosphane-Modified Tetrapodants and Their Bimetallic Transition Metal Complexes
FULL PAPER
Identification
Scheme 2. Schematic representation of the synthesis of the metal
complexes (used starting compounds: (PPh3)2PdCl2;
(COD)PdCl2; (PPh3)2PdMeCl; (COD)PdMeCl; i)
Pd(DBA)2, ii) TCNE; (CH3CN)4Pd(BF4)2; (PPh3)2-
PtCl2; (PPh3)2RhCl(CO); [PPh3)2IrCl(CO)]
1
The H-NMR spectra of the metal complexes of 3 and 4
have characteristic signals, viz. a doublet, at δ ϭ 5.5Ϫ5.7.
This signal is due to one of the inequivalent (CO)NCHHЈ
protons. The coupling constant of these two protons is in
the range of J ϭ 15Ϫ16 Hz.[4,15Ϫ20] The other doublet is
generally found at δ ഠ 3.8 depending on the solvent. When
a large deviation from this ppm value is found, the three-
dimensional structure of the building block is distorted.
1
Another characteristic signal in the H-NMR spectrum is
the singlet arising from the hydroquinone walls at δ ϭ 6.5
± 0.1. This signal was used in the determination of the host-
guest association constants (see below). The IR spectra of
DPGU derivatives always show a strong absorption of the
urea carbonyl groups [ν(CO) at 1705 ± 5 cmϪ1 (KBr disk)].
This carbonyl stretching frequency can be used to observe
interactions involving the carbonyl group. For example, hy-
drogen bonding between the carbonyl group and resorcinol
derivatives induces a shift of Ϫ25 ± 5 cmϪ1, as has been
[16,21,22]
described in the literature
.
1
The H- and 13C-NMR spectra of 3 and 4 indicated that
the concave part of the host, i.e. where the binding of sub-
strates occurs, is similar to that of the rigid clip molecule 5
(see Figure 2). The coupling constant of the doublet at δ ഠ
5.5 and the chemical shift of the hydroquinone parts were
in good agreement with literature values for these type of
compounds[4,15,16,18Ϫ20]. In the 31P{1H}-NMR spectrum
singlets at δ ϭ Ϫ21.1 and Ϫ21.7 were found for 3 and 4,
respectively, which are values similar to those of other alkyl-
diphenylphosphanes. The IR spectra of 3 and 4 showed a
carbonyl resonance at the expected frequency around 1710
cmϪ1. The elementary analyses of ligands 3 and 4 showed is known that the value of ∆δcompl after the formation of a
that two molecules of water are present in the host in the cis-PdCl2 complex is 43Ϫ47 ppm, whereas 32Ϫ36 ppm is
solid state. This could be confirmed by IR and NMR spec- found when trans complexes are formed.[24] In our case the
troscopy, which both revealed small amounts of water. values are ∆δcompl ϭ 34.2 for 3(PdCl2)2 and 36 ppm for
These water molecules did not seem to disturb any of our 4(PdCl2)2, which indicates that bimetallic trans-coordinated
further studies and therefore no attempts were made to re- complexes have been formed.
move them.
In this series of complexes, 3(PtCl2)2 and 4(PtCl2)2 are
The various transition metal complexes of ligands 3 and the only cis-coordinated compounds we were able to isolate.
4 have 31P-NMR resonances (see Table 1) at higher ppm The 31P-NMR spectra of these complexes showed coupling
1
values compared to the free tetrapodants. The difference constants around JPt-P ϭ 3500 Hz (Table 1, entries 9 and
between the chemical shift of the free ligand and that of 10), which indicates the formation of cis complexes.[23]
its transition metal complex (∆δcompl) is a suitable tool to These coupling constants were also observed in the 195Pt-
determine what type of complex has been formed. The 31P- NMR spectrum of 4(PtCl2)2, which afforded a triplet at δ ϭ
NMR values of these complexes are similar to phos- Ϫ4390. In addition to the magnitude of the coupling con-
phanemetal complexes known from literature,[23][24] show- stant (1JPt-P), the 195Pt chemical shift itself is indicative of
ing that ligands 3 and 4 are suitable ligands for coordi- a cis-coordinated complex. The typical chemical shift value
nation chemistry studies. A comparison of the change in of a trans complex is found at δ ϭ Ϫ3950.[25] For electronic
chemical shift on coordination (∆δcompl) with values found reasons, i.e. the larger trans influence of phosphanes com-
in literature leads to the following conclusions about the pared to chloride, PtCl2 complexes of phosphane ligands
immediate surroundings of the transition metal centres. All with small cone angles prefer cis coordination.
complexes except L[Pd(TCNE)]2 have a square planar con-
The zero-valent palladium complexes (Table 1, entries 7
formation. Complexes containing the PdMeCl fragment and 8) are stabilised by the electron-poor alkene TCNE,
have two equal, trans-coordinated phosphane moieties, which reduces the electron density on the Pd0 centre. The
which can be derived from the singlet in the 31P{1H}-NMR coordination mode around the metal centre is probably
spectrum. Complexes L(PdCl2)2 displayed a singlet in the pseudo-tetrahedral, as has been found for several crystal
31P{1H}-NMR spectrum around δ ϭ 13. From literature it structures of related Pd0 complexes.[26] The IR spectra of
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985
1977

P. W. N. M. van Leeuwen et al.
FULL PAPER
3[Pd(TCNE)]2 and 4[Pd(TCNE)]2 showed sharp bands of frequency 3 cmϪ1 below that of the original Vaska complex
the CN vibration of the TCNE molecule. The values were (1962 cmϪ1). In these iridium complexes the band of the
found at 2213 cmϪ1 and 2220 cmϪ1, respectively, which are urea carbonyl group was found at 1710 ± 2 cmϪ1 (in KBr).
in good agreement with values reported in the literature for
We attempted to isolate the cationic complexes
TCNE metal complexes.[26] The urea carbonyl groups in the 3[Rh(CO)ϩ]2 and 4[Rh(CO)ϩ]2, by the reaction of AgBF4
ligands gave rise to strong and sharp bands at 1707 cmϪ1 with the two L[RhCl(CO)]2 complexes. Turbid solutions
and 1710 cmϪ1 in 3[Pd(TCNE)]2 and 4[Pd(TCNE)]2, containing a fine precipitate were formed after the addition
respectively.
of AgBF4, but the precipitate could not be removed; centri-
The two ionic palladium(II) complexes of 3 and 4 are fugation and subsequent filtration with Celite did not result
square-planar complexes. The phosphane moieties in these in the removal of AgCl. IR measurements on the crude
complexes are coordinated in a trans fashion as can be seen complexes showed that the ν(CO)Rh had shifted to a higher
from the phosphorus chemical shift values[27] (δ ϭ 38.4 and frequency (1990 cmϪ1). The peaks of the urea carbonyl
42.8 for 3[Pd(BF4)2]2 and 4[Pd(BF4)2]2, respectively). In the group shifted approximately 30 cmϪ1 to a lower frequency
solid state the two PdII tetrafluoroborate complexes still in both 3[Rh(CO)ϩ]2 and 4[Rh(CO)ϩ]2. These frequency
contain one molecule of CH3CN per Pd atom, as could be shifts for both the urea carbonyl and the coordinated CO
concluded from the elementary analyses of 3[Pd(BF4)2]2 molecule strongly indicate the formation of a cationic RhI
and 4[Pd(BF4)2]2. However, their IR spectra did not show complex in which an interaction exists between the cationic
a band at 2100Ϫ2000 cmϪ1, the typical region of a coordi- Rh centre and the urea carbonyl group. The positive shift
nated CH3CN molecule. The IR spectra did show strong of ν(CO)Rh indicates a reduced π back donation of the elec-
carbonyl bands at 1648 cmϪ1 for 3[Pd(BF4)2]2 and at 1670 tron-poor cationic rhodium(I) centre to the coordinated CO
cmϪ1 for 4[Pd(BF4)2]2. The shifts of 60 cmϪ1 and 40 cmϪ1 molecule. The negative shift of the urea carbonyl band indi-
compared to the L(PdCl2)2 complexes, indicate that there is cates that the carbonyl group coordinates to the cationic
an interaction between the carbonyl group of the ligand and rhodium(I) centre. This electron donation was also found
the cationic palladium centre. Electron donation of the urea in the aforementioned cationic palladium(II) complexes,
carbonyl group to the PdII centre will result in a reduced which showed a larger shift of ∆ν ϭ Ϫ60 cmϪ1 and ∆ν ϭ
bond strength, leading to an absorption at a lower fre- Ϫ40 cmϪ1 for 3[Pd(BF4)2]2 and 4[Pd(BF4)2]2, respectively.
quency. This frequency shift is larger in 3[Pd(BF4)2]2 than
that in 4[Pd(BF4)2]2, probably because the shorter spacer
groups in the latter complex enforce a shorter metal-to-car-
bonyl distance. The interaction between the carbonyl group
and the metal centre could not be detected in solution. The
Variable-Temperature NMR Experiments
To investigate a possible interaction between the phos-
cationic complexes dissolve in polar, coordinating solvents phane end groups of 4 and guests we recorded the 31P-
only (e.g. CH3CN or DMSO), which replace the coordi- NMR spectra of 4 and five host-guest complexes of 4 with
nated carbonyl groups.
resorcinol derivatives. Figure 1 shows the temperature de-
The iridium and rhodium complexes in Table 1 are simi- pendence of the 31P{1H} line width of these complexes. At
lar in structure to the original Vaska complex[28] that con- 333 K all the 31P{1H}-NMR signals were visible as sharp
tains two triphenylphosphane ligands coordinated in a trans singlets having a line width of ∆ν1/2 ϭ 5Ϫ7 Hz. These are
fashion to an iridium centre. The rhodium complex showed the normal values in 31P{1H}-NMR spectra. When 4 was
1
a doublet in the 31P{1H}-NMR spectrum with JRh-P ϭ 124 measured without additives, this sharp singlet remained in
Hz, indicative of trans coordination.[29] We also determined the temperature range of T ϭ 223Ϫ333 K. The metal com-
the 103Rh chemical shift of 4[RhCl(CO)]2, which was found plex 4[RhCl(CO)]2 also gave rise to a sharp signal in this
at δ ϭ Ϫ419, close to that of reference compound temperature range. Experiments using a mixture of 4 and a
(PPh3)2RhCl(CO).[25] The IR spectra of 3[RhCl(CO)]2 and resorcinol derivative , however, showed a gradual increase
4[RhCl(CO)]2 showed bands for the urea carbonyl groups in the ∆ν1/2 of the 31P signal on lowering of the temperature.
(ν˜ ϭ 1710 and 1708 cmϪ1) and for the carbon monoxide The ∆ν1/2 passed through a maximum value at Tmax, fol-
ligand [ν(CO)Rh]. The latter absorptions were found at 1970 lowed by a decrease in ∆ν1/2 on further lowering of the tem-
and 1972 cmϪ1, respectively, which are slightly lower values perature. The chemical shift of the 31P signal also shifted
than those found in (PPh3)2RhCl(CO). Finally, we deter- on variation of the temperature. At high temperatures the
mined the molecular weight of complex 4[RhCl(CO)]2 in a signal was found at δ ϭ Ϫ20.5, whereas at low tempera-
dichloromethane solution (by vapour pressure osmometry) tures, the signal shifted to δ ϭ Ϫ24. The substrate appar-
and found a value of 1850 ± 100 (calcd. 1919). The molecu- ently has an influence on both the value of ∆ν1/2 and the
lar weight determination of 3[RhCl(CO)]2 proved to be dif-
T
max at which the maximum broadening is found. Tmax and
ficult due to its poor solubility.
∆ν1/2 were also affected by the amount of substrate present.
The 31P{1H}-NMR spectra of the iridium complexes of When a mixture of one equivalent of 4 and 6.5 equivalents
3 and 4 showed a singlet for both compounds (see Table 1, of octyl 3,5-dihydroxybenzoate was measured, the value of
entries 13 and 14), proving the formation of a trans com- Tmax was found at 313 K having a ∆ν1/2 of 88 Hz. When
plex. The peaks of the CO ligand [ν(CO)Ir] in the infra red the host-guest ratio was lowered to 1:2, the value of Tmax
spectra of 3[IrCl(CO)]2 and 4[IrCl(CO)]2 were found at a was found at a lower temperature (283 K) and the maxi-
1978
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985

Phosphane-Modified Tetrapodants and Their Bimetallic Transition Metal Complexes
FULL PAPER
mum broadening was now 52 Hz. The size of the substitu- Table 2. Association constants of tetrapodant 4 with several resor-
cinol derivatives
ent on resorcinol also had an effect on the value of ∆ν1/2. 5-
Chlororesorcinol, for example, had a smaller effect on ∆ν1/2
Guest
K [Ϫ1 [a]
]
CIS [ppm]
than the aforementioned octyl ester of 3,5-dihydroxyben-
zoic acid. These measurements indicate that the flexible
spacer groups, attached to the concave part of tetrapodant
4, have an interaction with an incoming substrate. In a ref-
erence experiment without addition of a substrate no
change in line width was observed. This indicates that the 5-(4-chlorophenyl)resorcinol
presence of a substrate is required to induce this line broad-
ening. The nature of the interaction between 4 and the sub-
olivetol
1676 (61)
Ϫ0.45
Ϫ0.43
Ϫ0.34
Ϫ0.37
Ϫ0.44
Ϫ0.44
Ϫ0.48
Ϫ0.45
5-chlororesorcinol
5-bromoresorcinol
5-iodoresorcinol
5-phenylresorcinol
4704 (407)
5360[b] (164)
3952 (272)
6537[b] (257)
6228 (132)
6437 (556)
7406 (431)
5-(4-methylphenyl)resorcinol
octyl 3,5-dihydroxybenzoate
strates, however, cannot be determined from these experi-
ments.
[a]
Duplo experiments, obtained from NMR titrations using
0.25Ϫ0.75 ϫ 10Ϫ3  solutions of host and 2.5Ϫ7.5 ϫ 10Ϫ3  solu-
tions of guest compound in CDCl3, errors are given in parentheses,
[b]
Figure 1. Plot of the line width (∆ν1/2) of the 31P-NMR signal of
4 versus temperature (T) in the presence of various resorcinol deri-
vatives (a: [4]/[octyl 3,5-dihydroxybenzoate] ϭ 1:6.5; b: [4]/[octyl
3,5-dihydroxybenzoate] ϭ 1:2; c: [4]/[olivetol] ϭ 1:7.4; d: [4]/[5-
chlororesorcinol] ϭ 1:6.5; e: [4]/[5-(p-tol)resorcinol] ϭ 1:4; f: 4 wi-
thout additives) (conditions: solvent CDCl3, [4] ϭ 8 ϫ 10Ϫ3 )
T ϭ 293 K. Ϫ
Single experiment; association constants were
calculated using the Granot[42] procedure.
ation constant of 4 and a guest, e.g. K is 1676 Ϫ1 for oli-
vetol (5-pentylbenzene-1,3-diol). Resorcinol derivatives
with an aromatic substituent at the 5-position also give rise
to increased association constants (K > 6200 Ϫ1). An ad-
ditional π-π stacking interaction between the phenyl groups
of the diphenylphosphane moieties and the additional aro-
matic ring of the substrates may be responsible for this in-
crease in binding affinity. These numbers (except those
found with aryl-substituted resorcinol derivatives) are in
reasonably good agreement with the values recently pub-
lished by Reek and co-workers.[16] They used compound 5
in their studies of the binding properties of molecular clips
based on DPGU and showed that electron-donating groups
on the resorcinol guests lead to reduced association con-
stants. Electron-withdrawing groups on the guest increase
the acidity of the hydroxy groups and the π-π stacking in-
teractions, which both result in an increase in association
constant. Reek and co-workers tested a number of resorci-
nol derivatives in their study and found a linear Hammet
relationship between the free energy of association (∆Gass
)
and the σmЈ values of the substituents on the 5-position
of resorcinol.
Association Constant Determinations
In order to determine the association constants of host-
A plot of the ∆Gass values calculated from the data col-
guest complexes between 1Ϫ4 and guests the 1H-NMR lected in Table 2 versus σm, did not show a linear relation-
shifts of the aromatic-wall protons of the tetrapodants were ship. The only difference between our tetrapodants and 5
monitored. These signals shift to lower ppm values on as- is that the latter does not contain flexible spacer groups
sociation with an aromatic guest molecule and a chemically (including the PPh2 moieties). These spacer groups are
induced shift (CIS) is observed. Dihydroxybenzenes bind in probably responsible for the non-linear relationship be-
the cavity of the host molecule via hydrogen bonds between tween σm and ∆Gass. They can interact with an incoming
the hydroxy groups of the guest and the urea carbonyl substrate, as shown by the lineshape analyses of the
groups of the host molecule[15,16,22]. A second interaction is 31P{1H}-NMR spectra (vide supra), and therefore influence
π-π stacking between the aromatic walls of the host and the the association constant. In the case of 3 and 4 the steric
aromatic ring of the guest molecule. Thirdly, electrostatic bulk of the end group will also be of great importance and
repulsion might occur, which diminishes the total bond contribute to the magnitude of the association constant.
strength between the host and its guest.
The two resorcinol derivatives substituted with alkyl chains
Resorcinol derivatives containing electron-withdrawing (olivetol and octyl 3,5-dihydroxybenzoate), for example,
groups at the 5-position bind relatively strongly to 4 (see give rise to relatively low association constants compared
Table 2). Titration experiments using 5-chlororesorcinol to the values that would be expected on the basis of their
and 4 resulted in an association constant of K ϭ 4704 Ϫ1
Electron-donating groups on the resorcinol ring, like alkyl
.
σm values.
The results obtained for tetrapodants 1؊4 and olivetol
substituents have a slightly negative influence on the associ- (see Table 3), lead to two conclusions: Larger end groups
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985
1979

P. W. N. M. van Leeuwen et al.
FULL PAPER
Figure 2. Molecular clip 5 used in association constant studies by
Table 4. Association constants of bimetallic complexes of 3 and 4
with olivetol
Reek[16] and co-workers
Metallohost
K [Ϫ1 [a]
]
CIS [ppm]
3(PdCl2)2
n.d.[b]
n.d.[b]
Ϫ0.02
n.d.[b]
n.d.[b]
Ϫ0.02
Ϫ0.44
Ϫ0.47
Ϫ0.48
Ϫ0.47
Ϫ0.51
3(PdMeCl)2
3[Pd(TCNE)]2
3(PtCl2)2
562 (367)
n.d.[b]
n.d.[b]
3[RhCl(CO)]2
4(PdCl2)2
164 (73)
2346 (499)
2502 (77.83)
960 (362)
4314[c] (338)
4514[c] (835)
4(PdMeCl)2
4[Pd(TCNE)]2
4(PtCl2)2
4[RhCl(CO)]2
[a]
on the ethylene glycol chains have a negative effect on the
association constant using olivetol as the guest. Replace-
ment of the chlorine groups in 1 by diphenylphosphane
moieties giving 3 causes a decrease of the association con-
stant of 20%. A similar decrease of 20% is found when 2
(chloro compound) and 4 (PPh2 compound) are compared.
This is probably the result of a steric interaction between
the large PPh2 groups and the substrate. Another expla-
nation for this drop in binding affinity might be that one of
the phenyl groups of the PPh2 moieties is clammed between
the aromatic walls of the cavity, which will compete with
binding of the guests. The latter explanation could not be
confirmed by low-temperature NMR because of the low
solubility of 4 at lower temperatures.
Single-fold experiments, obtained from NMR titrations using 5
ϫ 10Ϫ3  solutions of host and 50 ϫ 10Ϫ3  solutions of guest
compound in CDCl3, errors are given in parentheses, T ϭ 293 K.
[b]
Ϫ
n.d. ϭ not determined, NMR signals did not change and
K < 100 Ϫ1. Ϫ Duplo experiment; association constants were
[c]
calculated using the Granot[42] procedure.
“larger” analogues 2 and 4, because less rotational move-
ments are available. Therefore, after binding of a guest, the
∆Sass value will be less for hosts 1 and 3 than that of 2 and
4. A less negative ∆Sass value will result in a more negative
∆Gass value and thus higher association constants for 1
and 3.
Host-Guest Binding Properties of Metal Complexes
As discussed above the tetrapodant ligands 3 and 4 and
appropriate transition metals form basket-shaped bimetallic
complexes, which have different cavity sizes. The binding
properties of transition metal complexes of 3 were found to
be rather poor compared to the free ligand. The complexes
3[RhCl(CO)]2 and 3(PdMeCl)2 have binding constants of
K ϭ 164 Ϫ1 and K ϭ 562 Ϫ1, respectively, for olivetol
(see Table 4). From the small chemically induced shifts
(CIS) values of these host-guest complexes it can be con-
cluded that there is hardly any interaction between host and
guest. The 1H-NMR signals of complexes 3(PtCl2)2,
3(PdCl2)2, and 3[Pd(TCNE)]2 did not change after addition
of 10 equivalents of olivetol. These NMR experiments show
Table 3. Association constants of tetrapodants 1؊4 with olivetol
in CDCl3
Host
K [Ϫ1 [a]
]
CIS [ppm]
1
2
3
4
2769 (148)
2025 (98)
2201 (165)
1676 (61)
Ϫ0.56
Ϫ0.54
Ϫ0.46
Ϫ0.45
[a] Duplo experiments, obtained from NMR titrations using 0.5 ϫ
10Ϫ3  solutions of host and 5 ϫ 10Ϫ3  solutions of guest com-
pound, errors are given in parentheses, T ϭ 293 K; association
constants were calculated using the Granot[42] procedure.
The increase of the number of ethylene glycol units in the that, after coordination of 3 to two transition metal centres,
spacer arms also has a negative effect on the association a basket is formed that is too small for substrate binding.
constant between host and guest, as can be seen when 1 is Coordination of the phosphane moieties to a metal centre
compared with 2 and when 3 is compared with 4. A de- should facilitate the binding of a substrate in the cavity be-
crease of the association constant of approximately 20% is cause a preshaped cavity is formed and less entropy will be
found when the flexible arms are extended with one ethyl- lost during the association process. However, the cavity of
ene glycol unit. This decrease of the association constant is complexes of ligand 3 is not large enough to give access to
probably caused by a difference in entropy loss after the resorcinol derivatives.
formation of a host-guest complex. Binding of a guest mol-
A similar effect has been observed for host 6, which is
ecule will restrict the rotational and translational move- formed by a ring-closure reaction between 1 and benzyla-
ments of the side arms of the host. In particular rotation mine.[12,30,31] A crystal structure of 6 showed that, due to
around the ArOϪCH2 bond will be hampered. This can be the small ring size in 6, the ArOCH2 groups (see Figure 3)
1
derived from the H-NMR spectra of the host in the pres- are directed towards the cavity and block the entrance of
ence of guest molecules. The CH2 groups of the side arms this cavity[20] for entering guests. Although the transition
give rise to two signals, which indicates a restricted rotation. metal complexes of ligand 3 possess a larger cavity than 6,
It should be noted that the effect becomes less for the dis- i.e. instead of one amine two phosphanes and a transition
tant CH2 groups in host 4.
metal are now incorporated in the macrocycle, the cavity is
In the absence of a substrate, the “small” tetrapodants 1 probably still too small to allow a strong interaction of the
and 3 possess less degrees of freedom compared to their host with resorcinol guest molecules.
1980
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985

Phosphane-Modified Tetrapodants and Their Bimetallic Transition Metal Complexes
FULL PAPER
Host 7 and its derivatives do bind dihydroxybenzene de- the transition metal chemical shift of either the rhodium or
rivatives as was shown in previous studies.[3,15,20,30,32] The the platinum complex. This indicates that, although the
cavity of this host is large enough to give access to resorci- metal centre is probably able to come into close proximity
nol derivatives and 2,7-dihydroxynaphthalene. The metal of the binding site, it does not have an interaction with
complexes of ligand 4 have an even larger macrocyclic cav- the guest.
ity than 7 and will therefore be accessible for organic sub-
Infrared spectroscopy was used to investigate the nature
strates. This is confirmed by the experimental data (see of hydrogen-bond formation between host and guest. The
Table 4).
experiments indicated that additional hydrogen bonds be-
tween the carbon monoxide ligand of the rhodium complex
and the hydroxy groups of the guest could be excluded;
after the addition eight equivalents of olivetol a shift of ϩ4
cmϪ1 was found in the IR spectrum of 4[RhCl(CO)]2. This
is a too small a difference to be the result of hydrogen bond-
ing and additionally, the shift is in the wrong direction. Hy-
drogen bonds to the carbon monoxide ligand would shift
the carbonyl frequency to a lower value, as described by
Kazarian and co-workers for other carbon monoxide con-
taining compounds.[33] Hydrogen bonds to the metal centre
itself would induce a positive shift,[33] but the reported
Figure 3. Basket-shaped host molecules 6 and 7
Compared to free 4, complex 4Pd(TCNE) has a lower shifts are usually larger than 20 cmϪ1
binding affinity to olivetol. All the other complexes of 4 One example of a hydrogen bond between a platinum
.
show an increased affinity to this guest. The increase in as- coordinated chloride and a phenolic hydroxy group has
sociation constant is partly caused by fixation of the flexible been reported in the literature.[34] Evidence for hydrogen
arms of the tetrapodant. The degrees of freedom are re- bonding between a chloride ligand of our metal complex
duced compared to the free ligand, resulting in an increase and a hydroxy group of the substrate, however, could not
in the ground-state free energy. After complexation of a be found.
substrate, a relatively small loss in entropy (∆Sass) will result
in a more negative ∆Gass value, and a higher association group, showed that ligand 4 as well as its metal complexes
constant. bind olivetol via hydrogen bonds. After addition of eighth
Monitoring the carbonyl frequency of the urea carbonyl
Compounds 4(PtCl2)2 and 4[RhCl(CO)]2 form host-guest equivalents of olivetol to several 4M2 complexes the IR
complexes with olivetol which display association constants spectrum showed a shift of ν(CO) of Ϫ30 cmϪ1. Appar-
of 4314 Ϫ1 (Pt) and 4514 Ϫ1 (Rh), respectively. These ently, unlike Naϩ ions the metal centres do not prevent the
association constants are nearly three times as high as the formation of hydrogen bonds between host and guest.
value of the free phosphane tetrapodant 4. Complexes Hosts of type 7 bind two Naϩ ions to their crown ether
4(PdCl2)2 and 4(PdMeCl)2 also show an increase in associ- parts. In these host-guest complexes the sodium ions are
ation constant when compared to free 4 and values of K ϭ responsible for the binding of olivetol, as hydrogen bonds
2346 Ϫ1 and 2505 Ϫ1 were found, respectively.
There are several possible explanations for the observed groups of the guest could not be detected.[21]
increase in binding affinity of the metal complexes and oli- Summarising, we can state that a different type of bind-
between the carbonyl groups of the host and hydroxy
vetol: (i) An additional interaction between the substrate ing is not responsible for the change in binding affinity with
and the metal centre, e.g. a dipole-dipole interaction, (ii) a the guest. Conformational changes in the metal complexes
weak interaction between the oxygen atom or the aromatic or restricted flexibility of the spacer groups might therefore
ring of the substrate with the metal centre, e.g. coordination be a likely explanation. The difference in flexibility of the
at the axial position of the metal centre, (iii) extra hydrogen- spacer groups might be caused by partial ligand dis-
bonding interactions between the hydroxy groups of oli- sociation in solution. Perhaps the palladiumϪphosphorus
vetol and auxiliary ligands in the complexes, e.g. Cl or CO bond in complex 4[Pd(TCNE)]2 is more labile compared to
at the metal centre, (iv) difference in flexibility of the spacer the metalϪphosphorus bond in the other complexes. This
groups, (v) conformational differences between the metal may have an effect on the flexibility of the crown ether frag-
complexes.
ments, which are not as rigidly oriented as in the other
To investigate an interaction between olivetol and a tran- metal complexes. The crown ether fragments may still have
sition metal centre, we measured the 195Pt and 103Rh chemi- an interaction with the incoming substrate, and a com-
cal shifts of mixtures of 4(PtCl2)2 with olivetol and pound more similar to the parent ligand 4 is formed. The
4[RhCl(CO)]2 with olivetol. Changes in the second and lower association constant can then be explained by the
outer coordination sphere of the metal centre, as well as larger end group on the spacer groups (i.e. PPh2ϪPd in-
steric or electronic variations are known to have a pro- stead of PPh2 only) which will result in a relatively low as-
nounced effect on the chemical shift of the transition me- sociation constant (K ϭ 960 Ϫ1).
tal.[25] Addition of eight equivalents of olivetol to an NMR
sample containing 4(PtCl2)2 or 4[RhCl(CO)]2, did not affect 4[RhCl(CO)]2 may be more hampered in their movements
The crown ether fragments of complexes 4(PtCl2)2 and
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985
1981

P. W. N. M. van Leeuwen et al.
FULL PAPER
[2,1,8-ija]benz[f]azulene-6,13-dione[12] (1), 1,4,8,11-tetrakis[2-(2-
chloroethoxy)ethoxy]-5,7,12,13b,13c,14-hexahydro-13b,13c-
diphenyl-6 H,13H-5a,6a,12a,13a-tetraazabenz[5,6]azuleno[2,1,8-
ija]benz[f]azulene-6,13-dione [12] (2), (PPh3)2R hCl(CO), [37]
and form a preshaped cavity for an incoming guest. Since
the loss in entropy after complexation of a guest (∆Sass) will
be smaller for these complexes compared to the parent li-
gand 4, an increase in binding affinity of a factor of 2.5 is
found. The same positive effect on the association constant
was found for complexes 4(PdCl2)2 and 4(PdMeCl)2, al-
though the increase in binding affinity is only a factor of
1.5 for these complexes. Finally, a change in conformation,
induced by the transition metals, cannot be excluded but
the required structural information about these complexes
is still lacking.
(PPh3)2IrCl(CO),[28]
Pd(DBA)2,[38]
(CH3CN)4Pd(BF4)2,[39]
[40]
(COD)PdCl2
and all the 5-substituted resorcinol derivatives[41]
were prepared according to literature procedures. (PPh3)2PdCl2 and
(PPh3)2PtCl2 were obtained after refluxing PPh3 and PdCl2 or
PtCl2 in CH3CN, followed by evaporation of the solvent.
Ligands
1,4,8,11-Tetrakis[ ( 2-diphenylphosphanyl) ethoxy] -5,7,12,
13b,13c,14hexahydro-13b,13c-diphenyl-6H,13H-5a,6a,12a,13a-tetra-
benz[ 5,6] azuleno[ 2,1,8-ija] benz[ f] azulene-6,13-dione (3): A solu-
tion of 2.5 g of compound 1 (3.1 mmol) in 40 ml of THF and 40
ml of benzene was titrated under N2 at 0°C with 20 ml of a THF
solution of LiPPh2. The LiPPh2 solution was prepared from 2.52 g
of HPPh2 (13.6 mmol) and 5.4 ml of 2.5  nBuLi in hexane (13.5
mmol). When the solution remained orange, 40 ml of degassed
H2O was added. The H2O layer was separated and washed with 20
ml of benzene. The collected organic layers were dried with MgSO4
and concentrated. The product was dissolved in 15 ml of CH2Cl2
and precipitated with 100 ml of diethyl ether. After filtration, the
remaining solvent was removed in vacuo. The product was purified
by column chromatography with silica gel, eluent MeOH/CH2Cl2
(1:99, v/v). Yield 3.47 g (80%) of a white solid, m.p. > 195°C (dec.).
Ϫ NMR (CDCl3): δH ϭ 7.45Ϫ7.29 (m, 40 H, PϪAr), 7.08Ϫ7.01
(m, 10 H, ArH), 6.47 (s, 4 H, HϪAr side wall), 5.45 [d, 4 H,
Conclusions
In this paper we report novel host molecules that possess
a cavity in which substrates can be bound in addition to
two transition metal centres that are potential catalysts for
many reactions. A surprising increase in binding affinity for
olivetol was found upon coordination of ligand 4 to various
transition metals, whereas the transition metal complexes
of ligand 3 are unable to bind substrates.
For host molecules 3 and 4 the association equilibria are
influenced by the following factors: (i) the presence of flex-
ible spacers, (ii) the bulk of the spacer, (iii) the size and σm
value of substituents on the resorcinol derivative, (iv) the
ring size of the (metalla)macrocycle after coordination of
the phosphane moieties to a metal centre and finally, (v)
the metal centre itself. As yet, not all contributing factors
can be understood in full detail. These results are promising
in view of future catalytic applications with this type of
host compounds.
2
(CO)NCHH, J ϭ 15.8 Hz], 4.15 (m, 4 H, OCHH), 3.94 (m, 4 H,
2
OCHH), 3.76 [d, 4 H, (CO)NCHH, J ϭ 15.8 Hz], 2.65Ϫ2.59 (m,
8 H, CH2P); δP ϭ Ϫ21.7 (s); δC ϭ 158.2 (CϭO), 150.5 (ArCϪO),
138.5Ϫ138.2, 134.7, 133.12Ϫ132.8, 128.8Ϫ128.5, 128.3, 113.52,
2
85.3 [NC(Ar)N], 67.59 (d, OϪCH2, JP-C ϭ 24.8 Hz), 37.31
1
(NCH2Ar), 28.61 (d, CH2ϪP, JP-C ϭ 12.8 Hz). Ϫ IR (KBr): ν˜ ϭ
3051 cmϪ1 (w), 2911 (w, OCϪH), 1730 (s), 1710 (s, CO), 1482 (m),
1461 (s), 1433 (m), 1255 (m). Ϫ Mw (CH2Cl2): calcd. 1411; found
1445 ± 100. Ϫ C88H78N4O6P4 · H2O (1414): calcd. C 73.94, H 5.64,
N 3.92; found C 73.92, H, 5.51, N 4.05.
This work was supported by the Netherlands Foundation for
Chemical Research (SON) with financial aid from the Netherlands
Organisation for Scientific Research (NWO).
1,4,8,11-Tetrakis{2-[ 2( diphenylphosphanyl) ethoxy] ethoxy}-5,7,
12,13b,13c,14-hexahydro-13b,13c-diphenyl-6H,13H-5a,6a,12a,13a-
tetrabenz[ 5,6] azuleno[ 2,1,8-ija] benz[ f] azulene-6,13-dione (4): This
compound was prepared in a similar way as compound 3. The fol-
lowing amounts were used: 3.21 g of HPPh2 (17.3 mmol), 6.9 ml
of a 2.5  solution of nBuLi (17.25 mmol) and 4.26 g of compound
2 (4.31 mmol). The eluent used for column chromatography was
MeOH/CH2Cl2 (6:94, v/v). Yield 4.7 g (67%) of a white solid, m.p.
91Ϫ114°C. Ϫ NMR (CDCl3): δH ϭ 7.4Ϫ7.2 (m, 40 H, ArHϪP),
7.03 (m, 10 H, ArH), 6.61 (s, 4 H, HϪAr side wall), 5.47 [d, 4 H,
(CO)NCHH, 2J ϭ 15.9 Hz], 4.0Ϫ3.6 [m, 28 H, (CO)NCHH,
Experimental Section
General: NMR: 1H NMR (300.1 MHz), 31P{1H} NMR (121.5
MHz), 13C{1H} NMR (75.5 MHz), 195Pt NMR (64.3 MHz) and
103Rh NMR (9.48 MHz) were measured with Bruker AMX 300
and DRX 300 machines. The external references were: TMS (1H
NMR and 13C NMR), H3PO4 (31P NMR) and K2PtCl6 195Pt
(
NMR), Ξ(103Rh NMR) ϭ 3.16 MHz at 100 MHz. 195Pt NMR was
measured directly, 103Rh NMR was measured via the 31P-inverse
HMQC technique.[35][36] Ϫ IR spectra were measured with a Nico-
let 510m FT-IR spectrophotometer. Ϫ Melting points were deter-
mined with a Gallenkamp MFB-595 melting-point apparatus, the
values are uncorrected. Ϫ Column chromatography was performed
with silica gel 60, 70Ϫ230 mesh ASTM (Merck). Ϫ Analytical TLC
was performed on TLC aluminium foil, silica gel 60 F254 (Merck).
Ϫ Microanalyses were carried out in our own laboratory with an
Elementar Vario EL apparatus (Foss Electric). Ϫ Mw determi-
nations were performed with a Hewlett-Packard vapour-pressure
osmometer model 301A, with benzil as the reference compound.
OCH2], 2.42 (t, 8 H, CH2P, 3J ϭ 8.1 Hz); δP ϭ Ϫ21.1 (s); δC
ϭ
158.0 (C ϭO), 151.2 (ArCϪO), 138.8Ϫ138.4, 134.5, 133.0 (d,
ArCϪP, 2JP-C ϭ 18.8 Hz), 128.8Ϫ128.5, 115.1, 85.41 [NCϪ(Ar)N],
2
70.5, 69.9, (OϪCH2), 68.9 (d, OϪCH2, JP-C ϭ 24.0 Hz), 37.37
1
(NCH2Ar), 29.10 (d, CH2ϪP, JP-C ϭ 12.8 Hz). Ϫ IR (KBr): ν˜ ϭ
3052 cmϪ1 (w), 2865 (w, OCϪH), 1709 (s, CO), 1482 (m), 1460 (s),
1432 (s), 1264 (m). Ϫ Mw (CH2Cl2): calcd. 1588; found 1540 ± 100.
Ϫ C96H94N4O10P4 · 2 H2O (1586): calcd. C 70.92, H 6.20, N 3.45;
found C 70.77, H 5.84, N 3.45.
Chemicals: CH3CN and CH2Cl2 were distilled from CaH2, C6H6
was distilled from Na/benzophenone. CDCl3 used in association
constant determinations was distilled from P2O5 before use.
1,4,8,11-Tetrakis(2-chloroethoxy)-5,7,12,13b,13c,14-hexahydro-
13b,13c-diphenyl-6 H,13H-5a,6a,12a,13a-tetrabenz[5,6]azuleno-
Metal Complexes
3( PdCl2) 2: Compound 3 (20.3 mg, 14.3 µmol) was dissolved in
10 ml of benzene and 20.1 mg (28.6 µmol) of (PPh3)2PdCl2 was
added. The yellow solution was stirred for 2 h and the volume was
1982
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985

Phosphane-Modified Tetrapodants and Their Bimetallic Transition Metal Complexes
FULL PAPER
reduced to approximately 2 ml. The product was precipitated with
10 ml of pentane, the yellow suspension was decanted and the prod-
uct was washed twice with 10 ml of pentane. The product was dried
in vacuo. Yield 14.5 mg (58%) of a light yellow powder, m.p. >
245°C (dec.). Ϫ NMR ([D6]benzene): δH ϭ 8.2Ϫ6.7 (m, 50 H,
ArH), 6.26 (s, 4 H, HϪAr side wall), 5.66 (m, 4 H, ArOCHH),
5.58 [d, 4 H, (CO)NCHH, 2J ϭ 15.9 Hz], 4.33 (m, 4 H, ArOCHH),
3.49 (m, 4 H, CHHP), 3.23 [d, 4 H, (CO)NCHH, 2J ϭ 15.9 Hz],
3.20 (m, 4 H, CHHP); δP ϭ 12.5. Ϫ IR (KBr): ν˜ ϭ 1710 cmϪ1 (s,
δH ϭ 7.8Ϫ7.3 (m, 50 H, ArH), 7.18 (s, 4 H, HϪAr side wall), 5.5
[d, 4 H, (CO)NCHH, 2J ϭ 16.1 Hz], 4.62 (m, 4 H, OCHH), 4.41
(m, 4 H, OCHH), 4.02 [d, 4 H, (CO)NCHH, J ϭ 16.1 Hz], 3.53
(m, 4 H, CHHP), 3.37 (m, 4 H, CHHP); δP ϭ 38.4. Ϫ IR (KBr):
ν˜ ϭ 3444 cmϪ1 (m, OϪH), 3365 (m, OϪH), 3062 (w), 2924 (w),
2
1648 (s, CO), 1586 (m), 1470 (s), 1437 (s), 1400 (s).
Ϫ
C88H78B4F16N4O6P4Pd2 · 2 H2O · 2CH3CN (1796.8): calcd. C
52.87, H 4.25, N 4.02; found C 52.46, H 4.45, N 4.36.
4[ Pd( BF4) 2] 2: This compound was prepared in a similar way as
3[Pd(BF4)2]2 with the following amounts: 13 mg of
(CH3CN)4Pd(BF4)2 (29.3 µmol) and 23.3 mg of compound 4 (14.7
µmol). Yield 32.1 mg (93%) of a white powder, m.p. > 172°C (dec.).
Ϫ NMR (CD3CN): δH ϭ 8.1Ϫ7.0 (m, 50 H, ArH), 6.82 (s, 4 H,
HϪAr side wall), 5.70 [d, 4 H, (CO)NCHH, 2J ϭ 16.0 Hz], 4.7Ϫ3.7
[m, 28 H, (CO)NCHH, CH2O], 3.08 (m, 8 H, CH2P); δP ϭ 42.8.
Ϫ IR (KBr): ν˜ ϭ 3381 cmϪ1 (m, OϪH), 3060 (m, OϪH), 2928 (m),
2883 (w, OCϪH), 1670 (s, CO), 1587 (m), 1478 (s), 1437 (s), 1259
CO), 1480 (s), 1461 (s), 1435 (s), 1351 (w), 1248 (m).
Ϫ
C88H78Cl4N4O6P4Pd2 · 6 H2O (1764.8): calcd. C 56.39, H 4.84, N
2.98; found C 56.54, H 5.43, N 2.33.
4( PdCl2) 2: Compound 4 (22.5 mg, 14.2 µmol) was dissolved in
5 ml of dichloromethane and 8.3 mg of (COD)PdCl2 (28.4 µmol)
was added. A small amount of PPh3 was added to the solution and
a clear yellow solution was obtained. After 2 h of stirring, the vol-
ume was reduced to 2 ml and 10 ml of pentane was added. The
yellow suspension was decanted and the product was washed twice (m). ؊ C96H94B4F16N4O10P4Pd2 · 2 CH3CN (1972.8): calcd. C
with 10 ml of pentane. The product was dried in vacuo. Yield 20.7
mg (75%) of a yellow powder, m.p. > 180°C (dec.). Ϫ NMR
([D6]benzene): δH ϭ 8.0Ϫ6.6 (m, 50 H, ArH), 6.37 (s, 4 H, HϪAr
side wall), 5.93 [d, 4 H, (CO)NCHH, 2J ϭ 15.8 Hz], 4.6Ϫ3.7 (m,
53.87, H 4.52, N 3.77; found C 53.76, H 4.76, N 3.69.
3[ Pd( TCNE) ] 2: Compound 3 (50 mg, 35 µmol) was dissolved
in 3 ml of benzene and a solution of 40 mg of Pd(DBA)2 (70 µmol)
in 10 ml of benzene was added. The orange-brown solution was
stirred for 10 min and 9.1 mg of TCNE (70 µmol) and 2 ml of
benzene were added. After 4 h, the solution was reduced to ca. 5
ml and the product was precipitated with 15 ml of Et2O. The sus-
pension was decanted and the product was washed with 10 ml of
diethyl ether and 10 ml of pentane. The product was dried in vacuo.
Yield 110 mg (85%) of a green solid, m.p. > 211°C (dec.). Ϫ NMR
(CDCl3): δH ϭ 7.6Ϫ7.1 (m, 50 H, ArH), 6.46 (s, 4 H, HϪAr side
wall), 5.59 [d, 4 H, (CO)NCHH, 2J ϭ 15.8 Hz], 4.45 (m, 4 H,
ArOCHH), 4.24 (m, 4 H, ArOCHH), 3.75 [d, 4 H, (CO)NCHH,
2
24 H, OCH2), 3.53 [d, 4 H, (CO)NCHH, J ϭ 15.8 Hz], 3.08 (m,
4 H, CH2CHHP), 2.92 (m, 4 H, CH2CHHP); δP ϭ 14.9. Ϫ IR
(KBr): ν˜ ϭ 1710 cmϪ1 (s, CO), 1480 (m), 1461 (s), 1435 (s), 1248
(m). ؊ C96H94Cl4N4O10P4Pd2 (1940.8): calcd C 59.36, H 4.88, N
2.88; found C 59.08, H 5.35, N 2.70.
3( PdMeCl) 2: Compound 3 (100 mg, 71 µmol) was dissolved in
5 ml of CH2Cl2 and 97 mg (142 µmol) of (PPh3)2PdMeCl, dissolved
in 10 ml of benzene, was added. The colourless solution was stirred
for 2 h and the volume was reduced to approximately 2 ml. The
product was precipitated with 10 ml of pentane, the suspension was
decanted and the product was washed twice with 10 ml of pentane.
The product was dried in vacuo. Yield 103 mg (84%) of a white
powder, m.p. > 176°C (dec.). Ϫ NMR (CDCl3): δH ϭ 7.8Ϫ6.9 (m,
50 H, ArH), 6.59 (s, 4 H, HϪAr side wall), 5.63 [d, 4 H,
(CO)NCHH, 2J ϭ 15.9 Hz], 4.96 (m, 4 H, ArOCHH), 4.15(m, 4
H, ArOCHH), 3.64 [d, 4 H, (CO)NCHH, 2J ϭ 15.9 Hz], 3.22(m,
4 H, CHHP), 2.95(m, 4 H, CHHP), Ϫ0.36(t, 6 H, PdϪCH3,
3JP-H ϭ 6.0 Hz); δP ϭ 15.4. Ϫ IR (KBr): ν˜ ϭ 1710 cmϪ1 (s), 1482
(m), 1461 (s), 1435 (s), 1253 (m). Ϫ Satisfactory analysis could not
be obtained due to hygroscopicity of the compound.
2J ϭ 15.8 Hz], 3.17 (m, 4 H, CHHP), 2.80 (m, 4 H, CHHP); δP
ϭ
24.8. Ϫ IR (KBr): ν˜ ϭ 3057 cmϪ1 (w), 2922 (w), 2220 (m, CN),
1701 (s, CO), 1465 (s), 1435 (s), 1252 (m). Ϫ C100H78N12O6P4Pd2 ·
2 H2O (1878.8): calcd. C 63.22, H 4.25, N 8.85; found C 63.19, H
4.34, N 8.83.
4[ Pd( TCNE) ] 2: This compound was prepared in a similar way
as 3[Pd(TCNE)]2 with the following amounts: 40.7 mg of
Pd(DBA)2 (71 µmol), 62.6 mg of compound 4 (39.5 µmol) and 9.2
mg of TCNE (71 µmol). Yield 61 mg (84%) of a green powder,
m.p. > 275°C (dec.). Ϫ NMR (CDCl3): δH ϭ 7.5Ϫ7.35 (m, 40 H,
ArH), 7.12 (m, 10 H, ArH), 6.67 (s, 4 H, HϪAr side wall), 5.65
[d, 4 H, (CO)NCHH, 2J ϭ 15.8 Hz], 4.0Ϫ3.8 [m, 28 H, OCH2,
4( PdMeCl) 2: Compound 4 (30 mg, 18.9 µmol) was dissolved in
5 ml of CH2Cl2, followed by addition of 2 ml CH2Cl2, containing
10 mg of (COD)PdMeCl (37.8 µmol) and a small amount of PPh3.
The colourless solution was stirred for 2 h, the volume was reduced
to 2 ml and 10 ml of pentane was added. The suspension was de-
canted and the product was washed twice with 10 ml of pentane.
The product was dried in vacuo. Yield 30 mg (85%) of a white
powder, m.p. > 154°C (dec.). Ϫ NMR (CDCl3): δH ϭ 7.8Ϫ6.9 (m,
50 H, ArH), 6.63 (s, 4 H, HϪAr side wall), 5.59 [d, 4 H,
2
(CO)NCHH, J ϭ 15.8 Hz], 2.69 (m, 4 H, CH2CHHP), 2.59 (m, 4
H, CH2CHHP); δP ϭ 12.8. Ϫ IR (KBr): ν˜ ϭ 3058 cmϪ1(w), 2871
(w), 2213 (m, CN), 1707 (s, CO), 1458 (s), 1436 (s), 1258 (m). ؊
C108H94N12O10P4Pd2 (2054.8): calcd. C 63.07, H 4.64, N 8.17;
found C 63.05, H 4.74, N 7.94.
3( PtCl2) 2: (PPh3)2PtCl2 (28.8 mg, 36.4 µmol) and 25.7 mg of
compound 3 (18.2 µmol) were dissolved in 5 ml of CH2Cl2 and the
solution was stirred for 2 h. The volume was reduced to ± 2 ml
followed by addition of 15 ml of pentane. The suspension was de-
canted and the white solid was washed three times with 10 ml of
pentane. The product was dried in vacuo. Yield 27.9 mg (79%) of
a white solid, m.p. > 290°C (dec.). Ϫ NMR (CDCl3): δH ϭ 8.0Ϫ7.0
(m, 50 H, ArH), 6.76 (s, 4 H, HϪAr side wall), 5.69 [d, 4 H,
2
(CO)NCHH, J ϭ 15.8 Hz], 4.5Ϫ3.7 (m, 24 H, CH2O), 3.66 [d, 4
2
H, (CO)NCHH, J ϭ 15.8 Hz], 3.07 (m, 4 H, CHHP), 2.92 (m, 4
3
H, CHHP), Ϫ0.05 (t, 6 H, Pd-CH3, JP-H ϭ 6.0 Hz); δP ϭ 21.9. Ϫ
IR (KBr): ν˜ 1710 cmϪ1 (s, CO), 1483 (m), 1460 (s), 1433 (s), 1248
(m). Ϫ C98H100Cl2N4O10P4Pd2 · 4 H2O (1899.8): calcd. C 59.64, H
5.52, N 2.84; found C 59.73, H 5.34, N 2.52.
2
(CO)NCHH, J ϭ 15.8 Hz], 4.99 (m, 4 H, ArOCHH), 4.70 (m, 4
2
3[ Pd( BF4) 2] 2: A solution of 24.8 mg of compound 3 (17.5
µmol) in 2 ml of CH2Cl2 was added to a solution of 15.5 mg of 4 H, CHHP), 2.94(m, 4 H, CHHP); δP ϭ 0.4 (1JPt-P ϭ 3550 Hz).
(CH3CN)4Pd(BF4)2 (35 µmol) in 5 ml of CH3CN. After stirring for
H, ArOCHH), 3.80 [d, 4 H, (CO)NCHH, J ϭ 15.8 Hz], 3.12 (m,
Ϫ IR (KBr): ν˜ ϭ 1707 cmϪ1(s, CO), 1464 (s), 1436 (s), 1247 (m).
3 h, the solution was concentrated to dryness, yielding 35.5 mg Ϫ C88H78Cl4N4O6P4Pt2 (1678): calcd. C 54.38, H 4.05, N 2.88;
(95%) of a yellow powder, m.p. > 175°C (dec.). Ϫ NMR (CD3CN): found C 54.15, H 4.33, N 2.62.
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985
1983

P. W. N. M. van Leeuwen et al.
FULL PAPER
4( PtCl2) 2: (COD)PtCl2 (12.7 mg, 34 µmol) was dissolved in 5
NMR ([D6]benzene): δH ϭ 8.1Ϫ7.9 (dm, 16 H, ArH), 7.2Ϫ6.8 (m,
ml of CH2Cl2 and after addition of 27.0 mg of compound 4 (17 34 H, ArH), 6.34 (s, 4 H, HϪAr side wall), 5.95 [d, 4 H,
µmol), the solution was stirred for 2 h. The volume was reduced to
± 2 ml followed by addition of 15 ml of pentane. The suspension
was decanted and the white solid was washed three times with 10
ml of pentane. The product was dried in vacuo. Yield 24.6 mg
(CO)NCHH, 2J ϭ 15.9 Hz], 4.68 (m, 4 H, OCH2), 4.48 (m, 4 H,
OCH2), 4.31 (m, 4 H, OCH2), 4.13 (m, 12 H, OCH2), 3.54 [d, 4 H,
(CO)NCHH, J ϭ 15.9 Hz], 3.40 (m, 4 H, CH2CHHP), 3.27 (m, 4
H, CH2CHHP); δP ϭ 17.7. Ϫ IR (KBr): ν˜ ϭ 3057 cmϪ1 (w), 2922
(w), 1958 (s, COIr), 1711 (s, CO), 1482 (m), 1458 (s), 1435 (s), 1258
(m). ؊ C98H94Cl2N4O12P4Ir2 · 2 H2O (2097.4): calcd. C 55.13, H
4.63, N 2.62; found C 55.08, H 4.42, N 2.58.
2
(68%) white powder, m.p. > 255°C (dec.). Ϫ NMR (CDCl3): δH
ϭ
7.9Ϫ7.0 (m, 50 H, ArH), 6.73 (s, 4 H, HϪAr side wall), 5.63 [d, 4
H, (CO)NCHH, 2J ϭ 15.8 Hz], 4.2Ϫ3.5 [m, 28 H, OCH2,
(CO)NCHH], 2.66 (m,
4 H, CH2CHHP), 2.53 (m, 4 H,
CH2CHHP); δP ϭ 4.8 (1JPt-P ϭ 3610 Hz); δPt (CD2Cl2) ϭ Ϫ4390
[1]
1
(t, JPt-P ϭ 3610 Hz). Ϫ IR (KBr): ν˜ ϭ 1708 cmϪ1 (s, CO), 1481
M. Bonchio, T. Carofiglio, F. Di Furia, R. Fornassier, J. Org.
Chem. 1995, 60, 5986Ϫ5988.
(m), 1458 (s), 1435 (s), 1256 (m). ؊ C96H94Cl4N4O10P4Pt2 · H2O
(2118): calcd. C 53.94, H 4.53, N 2.62; found C 53.94, H 4.40,
N 2.62.
[2]
J. Rebek Jr., Pure Appl. Chem. 1996, 68, 1261Ϫ1266.
[3]
H. K. A. C. Coolen, J. A. M. Meeuwis, P. W. N. M. van
Leeuwen, R. J. M. Nolte, J. Am. Chem. Soc. 1995, 117,
11906Ϫ11913.
[4]
3[ RhCl( CO) ] 2: To a solution of 100 mg of compound 3 (71
µmol) in 3 ml of CH2Cl2 93 mg of (PPh3)2RhCl(CO) (135 µmol),
dissolved in 6 ml of CH2Cl2 and 14 ml of C6H6, was added and
the solution was stirred for 3 h. The volume was reduced to ca. 5
ml followed by addition of 25 ml of pentane. The suspension was
decanted and the white solid was washed three times with 20 ml of
pentane. The product was dried in vacuo. Yield 119 mg (96%) of a
yellow solid, m.p. > 195°C (dec.). Ϫ NMR (CDCl3): δH ϭ 7.9Ϫ7.0
(m, 50 H, ArH), 6.55 (s, 4 H, HϪAr side wall), 5.57 [d, 4 H,
C. F. Martens, R. J. M. Klein Gebbink, M. C. Feiters, R. J. M.
Nolte, J. Am. Chem. Soc. 1994, 116, 5667Ϫ5670.
[5]
H. K. A. C. Coolen, P. W. N. M. van Leeuwen, R. J. M. Nolte,
Angew. Chem. Int. Ed. Engl. 1992, 31, 905Ϫ907.
[6]
M. T. Reetz, S. R. Waldvogel, Angew. Chem. Int. Ed. Engl. 1997,
36, 865Ϫ867.
[7]
O. S. Tee, B. C. Javed, J. Chem. Soc., Perkin Trans 2 1994,
23Ϫ29.
[8]
H. J. Schneider, F. XIao, J. Chem. Soc., Perkin Trans 2 1992,
387Ϫ391.
[9]
F. Diederich, H.-D. Lutter, J. Am. Chem. Soc. 1989, 111,
2
8438Ϫ8446.
(CO)NCHH, J ϭ 15.6 Hz], 5.1 (m, 4 H, ArOCHH), 4.1 (m, 4 H,
[10]
2
Y. Murakami, J.-I. Kukuchi, Y. Hisaeda, O. Hayashida, Chem.
ArOCHH), 3.64 [d, 4 H, (CO)NCHH, J ϭ 15.6 Hz], 3.0(m, 4 H,
Rev. 1996, 96, 721.
CHHP), 2.7 (m, 4 H); δP ϭ 18.6 (d, 1JRh-P ϭ 124 Hz). Ϫ IR (KBr):
ν˜ ϭ 1972 cmϪ1 (s, CORh), 1710 (s, CO), 1482 (m), 1462 (s), 1435 (s),
[11]
G. Wulff, T. Gross, R. Schönfeld, Angew. Chem. Int. Ed. Engl.
1997, 36, 1962Ϫ1964.
[12]
R. P. Sijbesma, R. J. M. Nolte, Recl. Trav. Chim. Pays-Bas 1993,
1385 (m), 1354 (w), 1307 (w), 1253 (m). ؊ C90H78Cl2N4O8P4Rh2
·
112, 643Ϫ647.
3 H2O (1743): calcd. C 60.11, H 4.71, N 3.12; found C 59.98, H
4.45, N 3.06.
[13]
K. Tani, M. Yabuta, S. Nakamura, T. Yamagata, J. Chem. Soc.,
Dalton Trans. 1993, 2781Ϫ2789.
[14]
C. Amatore, G. Broeker, A. Jutand, F. Khalil, J. Am. Chem.
4[ RhCl( CO) ] 2: This compound was prepared in a similar way
Soc. 1997, 119, 5176Ϫ5185.
as
3 [RhCl(CO)]2 with the following amounts: 230 mg of
[15]
R. P. Sijbesma, R. J. M. Nolte, J. Org. Chem. 1991, 56, 3199.
[16]
(PPh3)2RhCl(CO) (333 µmol) and 240 mg of compound 4 (170
µmol). Yield 285 mg (96%) of a yellow powder, m.p. > 149°C (dec.).
Ϫ NMR (CDCl3): δH ϭ 7.9Ϫ7.0 (m, 50 H, ArH), 6.55 (s, 4 H,
HϪAr side wall), 5.57 [d, 4 H, (CO)NCHH, 2J ϭ 15.8 Hz], 4.5Ϫ3.8
(m, 24 H, OCH2), 3.68 [d, 4 H, (CO)NCHH, 2J ϭ 15.8 Hz], 3.12(m,
4 H, CH2CHHP), 2.91 (m, 4 H, CH2CHHP); δP ϭ 22.0 (d,
1JRh-P ϭ 123 Hz); δRh ϭ Ϫ418. Ϫ IR (KBr): ν˜ ϭ 1970 cmϪ1 (s,
CORh), 1708 (s, CO), 1481 (m), 1458 (s), 1435 (s), 1256 (m) Ϫ Mw
(CH2Cl2): calcd. 1919, found 1850. Ϫ C98H94Cl2N4O12P4Rh2 · 3
H2O (1919): calcd. C 59.61, H 5.10, N 2.84; found C 59.47, H 4.98,
N 2.80.
J. N. H. Reek, A. H. Priem, H. Engelkamp, A. E. Rowan, J.
A. A. W. Elemans, R. J. M. Nolte, J. Am. Chem. Soc. 1997,
119, 9956Ϫ9964.
[17]
J. W. H. Smeets, R. P. Sijbesma, L. van Dalen, A. L. Spek, W.
J. J. Smeets, R. J. M. Nolte, J. Org. Chem. 1989, 54, 3710.
[18]
J. W. H. Smeets, R. P. Sijbesma, F. G. M. Niele, A. L. Spek, W.
J. J. Smeets, R. J. M. Nolte, J. Am. Chem. Soc. 1987, 109, 928.
[19]
H. K. A. C. Coolen, P. W. N. M. van Leeuwen, R. J. M. Nolte,
J. Org. Chem. 1996, 61, 4739.
[20]
C. F. Martens, R. P. Sijbesma, R. J. M. Klein Gebbink, R. J.
M. Nolte, Recl. Trav. Chim. Pays-Bas 1993, 112, 400.
[21]
G. C. Dol, F. Hartl, P. C. J. Kamer, P. W. N. M. van Leeuwen,
R. J. M. Nolte, J. Chem. Soc., Dalton Trans., accepted.
[22]
R. P. Sijbesma, A. P. M. Kentgens, E. T. G. Lutz, J. H. van der
Maas, R. J. M. Nolte, J. Am. Chem. Soc. 1993, 115, 8999Ϫ9005.
3[ IrCl( CO) ] 2: A mixture of 44.4 mg of compound 3 (31.5 µmol)
and 49 mg of (PPh3)2IrCl(CO) (63 µmol) in 10 ml of C6H6 was
stirred overnight. The volume was reduced to ca. 2 ml and 15 ml
of pentane was added. The solvent was decanted and the residue
washed twice with 10 ml of pentane. The product was dried in
vacuo. Yield 54 mg (89%) of a yellow solid, m.p. > 180°C (dec.).
Ϫ NMR ([D6]benzene): δH ϭ 7.96 (m, 16 H, ArH), 7.2Ϫ6.8 (m, 34
H, ArH), 6.34 (s, 4 H, HϪAr side wall), 5.90 [d, 4 H, (CO)NCHH,
2J ϭ 15.8 Hz], 5.8 (m, 4 H, ArOCHH), 4.5 (m, 4 H, ArOCHH),
[23]
C. J. Cobley, P. G. Pringle, Inorg. Chim. Acta 1997, 265,
107Ϫ115.
[24]
S. O. Grim, R. L. Keiter, Inorg. Chem. Acta 1970, 4, 56.
[25]
P. S. Pregosin, Studies in Inorganic Chemistry 13, Elsevier, Am-
sterdam, Oxford, New York, Tokyo, 1991.
[26]
M. Kranenburg, J. P. G. Delis, P. C. J. Kamer, P. W. N. M. van
Leeuwen, K. Vrieze, N. Veldman, A. L. Spek, K. Goubitz, J.
Fraanje, J. Chem. Soc., Dalton Trans. 1997, 1839Ϫ1849.
[27]
S. Murata, Y. Ido, Bull. Chem. Soc. Jpn. 1994, 67,1746Ϫ1748.
[28]
J. P. Collman, C. T. Sears Jr., M. Kubota, Inorg. Synth. 1990,
28, 92Ϫ94.
2
4.1 (m, 4 H, CHHP), 3.43 [d, 4 H, (CO)NCHH, J ϭ 15.8 Hz], 3.3
[29]
T. H. Brown, P. J. Green, J. Am. Chem. Soc. 1970, 92,
(m, 4 H, CHHP); δP ϭ 14.8. Ϫ IR (KBr): ν˜ ϭ 3056 cmϪ1 (w), 2922
(w), 1959 (s, COIr), 1710 (s, CO), 1480 (s), 1462 (s), 1435 (s), 1248
(m). ؊ C90H78Cl2N4O8P4Ir2 · 2 H2O (1921.4): calcd. C 55.18, H
4.22, N 2.86; found C 55.20, H 4.63, N 2.62.
2359Ϫ2362.
[30]
R. P. Sijbesma, A. M. Kentgens, R. J. M. Nolte, J. Org. Chem.
1991, 56, 3122.
[31]
R. P. Sijbesma, W. P. Bosman, R. J. M. Nolte, J. Chem. Soc.,
Chem. Commun. 1991, 885.
[32]
R. P. Sijbesma, R. J. M. Nolte, Top. Curr. Chem. 1995, 175,
4[ IrCl( CO) ] 2: This compound was prepared in a similar way as
3[IrCl(CO)]2 with the following amounts: 49 mg of
(PPh3)2IrCl(CO) (63 µmol) and 50 mg of compound 4 (31.5 µmol).
Yield 56 mg (85%) of a yellow powder, m.p. > 165°C (dec.). Ϫ
26Ϫ56.
[33]
S. G. Kazarian, P. A. Hamley, M. Poliakoff, J. Am. Chem. Soc.
1993, 115, 9069Ϫ9079.
[34]
P. J. Davies, N. Veldman, D. M. Grove, A. L. Spek, B. T. G.
1984
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985

Phosphane-Modified Tetrapodants and Their Bimetallic Transition Metal Complexes
FULL PAPER
[39]
Lutz, G. van Koten, Angew. Chem. Int. Ed. Engl. 1996, 35,
1959Ϫ1961.
B. J. Hathaway, A. E. Underhill, J. Chem. Soc. 1962, 2257, 2444.
J. Chatt, L. M. Vallarino, L. M. Venanzi, J. Chem. Soc. 1957,
3413.
[40]
[41]
[42]
[35]
[36]
C. J. Elsevier, J. M. Ernsting, d. L. W.G.J., J. Chem. Soc., Chem.
Commun. 1989, 585Ϫ586.
G. C. Dol, P. C. J. Kamer, P. W. N. M. van Leeuwen, Eur. J.
Org. Chem. 1998, 359Ϫ364.
A. Bax, R. H. Griffey, B. L. Hawkins, J. Magn. Reson. 1983,
55, 301.
J. A. Osborn, G. Wilkinson, Inorg. Synth. 1990, 28, 79Ϫ80.
M. F. Rettig, P. M. Maitlis, Inorg. Synth. 1977, 17, 134.
J. Granot, J. Magn. Reson. 1983, 55, 216.
[I98190]
[37]
[38]
Eur. J. Inorg. Chem. 1998, 1975Ϫ1985
1985