Gyroscope-like molecules consisting of PdX2/PtX2 rotators within three-spoke dibridgehead diphosphine stators: Syntheses, substitution reactions, structures, and dynamic properties

Article abstract of DOI:10.1002/chem.201304419

Threefold intramolecular ring-closing metatheses of trans-[MCl ₂(P{(CH₂)_mCH=CH₂}₃) ₂] are effected with Grubbs' catalyst. Following hydrogenation catalyzed by [RhCl(PPh₃)₃], the title complexes trans-[MCl₂(P((CH₂)_n)₃P)] (n=2m+2; M/n=Pt/14, 4 c; Pt/16, 4 d; Pt/18, 4 e; Pd/14, 5 c; Pd/18, 5 e) and sometimes isomers partly derived from intraligand metathesis, trans-[MCl ₂{P(CH₂)_n(CH₂)_n}P(CH ₂)_n)] (4c-e, 5e), are isolated. These react with LiBr, NaI, and KCN to give the corresponding MBr₂, MI₂, and M(CN)₂ species (58-99 %). ¹³C NMR data show that the MX₂ moieties rapidly rotate within the diphosphine cage on the NMR timescale, even at -120 °C. The reaction of 4 c and KSCN gives separable Pt(NCS)₂ and Pt(NCS)(SCN) adducts (13 c, 28 %; 14 c, 20 %), and those of 4 c,e and Ph₂Zn give PtPh₂ species (15 c, 61 %; 15 e, 90 %). ¹³C NMR spectra of 13 c-15 c show two sets of CH₂ signals (ca. 2:1 intensity ratios), indicating that MX₂ rotation is no longer rapid. Reactions of 4 c or 4c and excess NaC≡CH afford the free diphosphines P{(CH₂)₁₄}₃P (91 %) and (CH ₂)₁₄P(CH₂)₁₄P(CH₂) ₁₄ (90 %). The latter has been crystallographically characterized as a bis(BH₃) adduct. The crystal structures of eight complexes with P(CH₂)₁₄P linkages (PtCl₂, PtBr₂, PtI₂, Pt(NCS)₂, PtPh₂, PdCl₂, PdBr₂, PdI₂) and 15 e have been determined, and intramolecular distances analyzed with respect to MX₂ rotation. The conformations of the (CH₂)₁₄ moieties and features of the crystal lattices are also discussed. A fascinating set of complexes with trans-R₃P-MX₂-PR₃ linkages are described that mimic the connectivity, symmetry, and rotational dynamics of common toy gyroscopes (see figure).

Full text of DOI:10.1002/chem.201304419

DOI: 10.1002/chem.201304419
Full Paper
&
Olefin Metathesis
Gyroscope-Like Molecules Consisting of PdX2/PtX2 Rotators within
Three-Spoke Dibridgehead Diphosphine Stators: Syntheses,
Substitution Reactions, Structures, and Dynamic Properties
Agnieszka J. Nawara-Hultzsch,[a] Michael Stollenz,[b] Michał Barbasiewicz,[a]
Sławomir Szafert,[c] Tadeusz Lis,[c] Frank Hampel,[a] Nattamai Bhuvanesh,[b] and
John A. Gladysz*[a, b]
Abstract: Threefold intramolecular ring-closing metatheses
of trans-[MCl2(P{(CH2)mCH=CH2}3)2] are effected with Grubbs’
catalyst. Following hydrogenation catalyzed by [RhCl(PPh3)3],
the title complexes trans-[MCl2(P((CH2)n)3P)] (n=2m+2;
M/n=Pt/14, 4c; Pt/16, 4d; Pt/18, 4e; Pd/14, 5c; Pd/18, 5e)
and sometimes isomers partly derived from intraligand
metathesis, trans-[MCl2{P(CH2)n(CH2)n}P(CH2)n)] (4’c–e, 5’e),
are isolated. These react with LiBr, NaI, and KCN to give the
corresponding MBr2, MI2, and M(CN)2 species (58–99%).
13C NMR data show that the MX2 moieties rapidly rotate
within the diphosphine cage on the NMR timescale, even at
ꢀ1208C. The reaction of 4c and KSCN gives separable
Pt(NCS)2 and Pt(NCS)(SCN) adducts (13c, 28%; 14c, 20%),
and those of 4c,e and Ph2Zn give PtPh2 species (15c, 61%;
15e, 90%). 13C NMR spectra of 13c–15c show two sets of
CH2 signals (ca. 2:1 intensity ratios), indicating that MX2 rota-
tion is no longer rapid. Reactions of 4c or 4’c and excess
NaCꢁCH afford the free diphosphines P{(CH2)14}3P (91%) and
(CH2)14P(CH2)14P(CH2)14 (90%). The latter has been crystallo-
graphically characterized as a bis(BH3) adduct. The crystal
structures of eight complexes with P(CH2)14P linkages (PtCl2,
PtBr2, PtI2, Pt(NCS)2, PtPh2, PdCl2, PdBr2, PdI2) and 15e have
been determined, and intramolecular distances analyzed
with respect to MX2 rotation. The conformations of the
(CH2)14 moieties and features of the crystal lattices are also
discussed.
Introduction
limit of gyroscope miniaturization. Many other approaches to
microscopic gyroscopes are also receiving attention,[5] driven
by a variety of technological applications, the most ubiquitous
being mobile-phone display screens.
Macroscopic gyroscopes can assume a number of physical
forms,[1] which include biological assemblies, such as the hal-
teres of flying insects.[2] All serve as devices for sensing and/or
maintaining the orientation of an object. The classic mechani-
cal gyroscope, encountered by many as a childhood toy, con-
sists of an axle, a “flywheel” that can rotate, and a protective
housing. When the object is displaced from its axis of rotation,
there is a restoring force arising from the conservation of an-
gular momentum.[3] The underlying physics also holds at the
molecular level,[4] allowing chemists to pursue the ultimate
Molecular gyroscopes constitute one subset of molecular
rotors,[4] which are composed of “rotators” and “stators”. Since
motion associated with a component of any object is a func-
tion of the frame of reference, the stator is taken as the sub-
unit with the greater moment of inertia. In a series of commu-
nications[6–11] we have described the syntheses of gyroscope-
like molecules in which a metal-based rotator, MLy, is encased
in a macrobicyclic dibridgehead diphosphine stator, such that
a P-M-P axle or axis results. Full papers that detail related di-
phosphite- and dipyridine-based systems are also avail-
able.[12,13]
[a] Dr. A. J. Nawara-Hultzsch, Dr. M. Barbasiewicz, Dr. F. Hampel,
Prof. Dr. J. A. Gladysz
Institute of Organic Chemistry and
As shown in Scheme 1, the diphosphine and diphosphite
systems have been accessed by surprisingly effective threefold
intramolecular ring-closing metatheses of trans-monophospho-
rus ligands, followed by hydrogenation (I!II). These sequen-
ces can be conducted in trigonal-bipyramidal, square-planar,
and octahedral coordination spheres (III–V). Many elegant
studies of purely organic or organosilicon models for molecular
gyroscopes, some with similar topologies, have been reported
by the Garcia-Garibay[14,15] and Setaka groups.[16,17] The most
relevant assemblies are described in the discussion section.
Interdisciplinary Center for Molecular Materials
Friedrich-Alexander-Universitꢀt Erlangen-Nꢁrnberg
Henkestrasse 42, 91054 Erlangen (Germany)
[b] Dr. M. Stollenz, Dr. N. Bhuvanesh, Prof. Dr. J. A. Gladysz
Department of Chemistry, Texas A&M University
PO Box 30012, College Station, Texas 77842-3012 (USA)
E-mail: gladysz@mail.chem.tamu.edu
[c] Dr. S. Szafert, Dr. T. Lis
Department of Chemistry, University of Wroclaw
F. Joliot-Curie 14, 50-383 Wroclaw (Poland)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304419.
Chem. Eur. J. 2014, 20, 4617 – 4637
4617
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Results
Gyroscope-like complexes from dichloride complexes
Platinum and palladium dichloride complexes of the formula
trans-[MCl2(P{(CH2)mCH=CH2}3)2] (m=4–8) were prepared by
routine procedures described previously.[19] [Note that the no-
tation a–e in the compounds numbers given hereafter, refer to
m=4–8, respectively.] The complexes with six methylene
groups in each phosphine substituent, trans-1c (M=Pt) and
trans-2c (M=Pd), were used for lead experiments throughout
this study. As shown in Scheme 2, solutions of compounds
trans-1c and trans-2c in CH2Cl2 (0.0035m) were slowly treated
with solutions of Grubbs’ catalyst. For all metatheses, catalyst
loadings (here 3 and 14 mol% (two charges)) and reaction
times were initially optimized by assaying aliquots by NMR
spectroscopy.[20] When the =CH2 1H NMR signals of the reac-
tants were no longer detectable, the mixtures were filtered
through alumina.[21] The filtrates were treated with hydrogen
(5 bar) in the presence of Wilkinson’s catalyst (13–14 mol%).
Chromatographic workups gave the target gyroscope-like mol-
ecules derived from threefold intramolecular and interligand
metatheses, trans-4c and trans-5c, in 43 and 36% yields, re-
spectively.
Scheme 1. Syntheses of gyroscope-like complexes based upon trans-span-
ning diphosphines with (CH2)n linkers.
In this full paper, we document our initial efforts with gyro-
scope-like complexes based upon square-planar platinum and
palladium cores. The effectiveness of intramolecular ring-clos-
ing metathesis is probed as a function of macrocycle size. The
feasibility of substitution reactions within the gyroscope cages
is then demonstrated. The sizes of the ligands are varied to
probe rotational dynamics within the cage, which are com-
pared to analogues with different coordination geometries. A
number of crystal structures are determined that provide addi-
tional insight, and define constraints on rotation in the solid
state. A small portion of this work has been communicated,[7]
and additional details have been described elsewhere.[18]
Complexes trans-4c and trans-5c, which feature 17-mem-
bered macrocycles with 14 methylene groups, were initially
obtained as pale yellow waxy oils that solidified with time.
Crystals of each could be grown, and the structures were con-
firmed crystallographically as described below. These and all
other gyroscope-like complexes below were stable in air for
extended periods of time. They were characterized by NMR
(1H, 13C, 31P) and IR spectroscopy, and in most cases by micro-
analyses, DSC, and mass spectrometry (Experimental Section).
Scheme 2. Metathesis/hydrogenation sequences involving complexes of the formulae trans-[MX2(P{(CH2)mCH=CH2}3)2]. [a] Larger scale reaction (see text).
[b] Yields from different experiments.
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4618
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
complex trans-2e, the corresponding complexes trans-5e and
trans-5’e were obtained in lower yields (15 and 3%, respective-
ly).
Table 1. Selected 31P{1H} and 13C{1H} NMR data (CDCl3).
31P{1H} d [ppm]
[1JPPt, Hz][a]
13C{1H} d [ppm]
PCH2CH2
[c]
PCH2
[1JPC, Hz][b]
PCH2CH2CH2
[3JPC, Hz][b]
Comparable sequences were conducted with a platinum
complex with seven methylene groups in each phosphine sub-
stituent, trans-1d. However, the yields of monoplatinum com-
plexes were lower, and the composition varied in different
runs. In one case, only the gyroscope-like complex trans-4d
was isolated (4%). In another, the isomer trans-4’d dominated
(4%). Reactions were also conducted with palladium dichloride
complexes with four or five methylene groups in each phos-
phine substituent (trans-2a,b). As shown in Scheme 2
(bottom), insoluble powders or gums that were presumed to
be oligomers or polymers were obtained. There was no evi-
dence for significant quantities of monopalladium products,
even when Grubbs’ second-generation catalyst was employed.
The exact conditions utilized are detailed elsewhere.[18]
trans-1c
trans-1d
trans-1e
trans-4c
trans-4d
trans-4e
trans-5c
trans-5e
trans-4’c
5.2 [2381]
5.2 [2380]
5.1 [2381]
7.1 [2398]
6.0 [2398]
6.5 [2389]
13.3
21.5 [16.2]
20.4 [16.0]
20.4 [16.1]
21.3 [16.3]
20.8 [16.2]
20.7 [16.2]
22.4 [12.2]
21.7 [13.1]
23.1 [16.3]
18.9 [15.9][d]
22.8 [16.1][f]
19.2 [16.0][d]
22.6 [16.5][f]
19.3 [16.0][d]
23.2[g]
24.2
23.6
23.6
23.1
23.2
23.4
23.5
23.9
21.7[e]
31.5 [6.2]
31.1 [6.5]
31.1 [6.3]
30.0 [6.8]
31.4 [6.7]
30.7 [6.6]
30.2 [6.7]
30.8 [6.6]
31.3 [7.3]
29.0 [6.2][d]
31.6 [7.2]
29.9 [6.3][d]
31.4 [6.8]
30.6 [6.2][d]
31.5 [7.0]
30.6 [6.1][d]
29.8 [7.0]
30.5 [6.9]
30.0 [7.0]
30.3 [7.0]
10.1
5.6 [2374]
trans-4’d
trans-4’e
trans-5’e
5.5 [2374]
5.4 [2374]
11.0
22.6[e]
22.8[e]
25.1[e]
20.5 [12.8][d]
24.9 [17.2]
24.7 [17.3]
25.7 [14.3]
21.6 [17.3]
trans-11 c
trans-11 e
trans-12c
trans-15e
9.2 [2166]
6.8 [2152]
17.4
23.7
24.1
24.0
27.8
Gyroscope-like complexes from dibromide complexes
3.2 [2807]
Metathesis/hydrogenation sequences were also conducted
with analogous palladium dibromide complexes.[19] With trans-
3a,b, which feature four or five methylene groups in each
phosphine substituent, oligomers or polymers were again ob-
tained (Scheme 2, bottom).
[a] Singlets; the platinum complexes exhibit 195Pt satellites (d). [b] Virtual
triplets for which the iJPC values represent the apparent coupling between
adjacent peaks. [c] Singlets. [d] Downfield/upfield signals about 1:2.
[e] The PCH2CH2 signal for the phosphacycle cannot be definitively as-
signed from present data. [f] One line of this triplet is obscured. [g] Due
to overlapping signals, this coupling can only be bounded as between 12
and 15 Hz.
The results with trans-3c are depicted in Scheme 3. Metathe-
sis was slower than with dichloride analog trans-5c, and
20 mol% of Grubbs’ catalyst was employed. Subsequent hy-
Key NMR data are summarized in Table 1. All complexes ex-
hibited a single 31P{1H} NMR signal. The platinum adducts
showed 195Pt satellites, the magnitudes of which were consis-
tent with trans-P-Pt-P linkages.[22] The 13C NMR spectra exhibit-
ed seven (m+1 or n/2) methylene signals. COSY experiments
1
(1H,1H and H,13C{1H}) with trans-4c allowed the PCH2, PCH2CH2
and PCH2CH2CH2 13C and H NMR signals to be assigned.[20] Par-
1
allel assignments were made for the other complexes. The 13C
PCH2 and PCH2CH2CH2 signals were virtual triplets with J values
of 17.3–12.2 Hz and 7.0–6.2 Hz, respectively.[23] In contrast, the
PCH2CH2 signals were singlets.
Similar sequences were conducted with complexes with two
additional methylene groups in each phosphine substituent
(trans-1e and trans-2e). Now, as shown in Scheme 2, apprecia-
ble amounts of two types of monometallic metathesis prod-
ucts were obtained. In the case of platinum complex trans-1e,
the gyroscope-like complex trans-4e with 21-membered mac-
rocycles (eighteen methylene groups) was isolated in 28%
yield. The structure followed from its NMR properties, as well
as the crystal structure of a derivative below. However, a more
polar substance, trans-4’e, was also isolated in 23% yield. It ex-
hibited two sets of nine (n/2) 13C NMR signals in an approxi-
mate 2:1 intensity ratio. The mass spectrum showed a molecu-
lar ion identical to that of trans-4e. Accordingly, it was as-
signed an isomeric structure derived from a combination of
inter- and intraligand metathesis, as shown in Scheme 2. This
assignment was supported by a crystal structure derived from
a homologue, as described below. In the case of palladium
Scheme 3. Metathesis/hydrogenation and substitution sequences leading to
the gyroscope-like palladium dibromide complex trans-8c.
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4619
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
drogenation gave a mixture of three monoplatinum com-
plexes, as assayed by 31P NMR spectroscopy, in about 19%
yield (ca. 50:40:10).[20] These could not be separated chromato-
graphically. Pale yellow plates were obtained from CH3OH/THF,
which a crystal structure and independent synthesis below
showed to be the target gyroscope-like dibromide complex
trans-8c. This could be assigned to the 31P NMR signal of inter-
mediate intensity (10.3 ppm). The minor signal had the same
chemical shift as the dichloride analogue trans-5c (13.3 ppm).
The mass spectrum of the mixture showed an ion with the
mass of the mixed monochloride monobromide complex
trans-6c. Accordingly, this was assigned to the major signal
(12.2 ppm). Apparently, the chloride ligands in Grubbs’ and/or
Wilkinson’s catalyst are able to exchange with the bromide li-
gands of palladium.
These assignments were verified by monitoring an inde-
pendent reaction of the dichloride complex trans-5c with LiBr.
As shown in sequence B in Scheme 3, one equivalent was ini-
tially added, thereby generating the mixture of three com-
plexes obtained by metathesis. Then an additional 1.7 equiva-
lents were added, cleanly generating the dibromide complex
trans-8c. Workup of a similar reaction (below) gave trans-8c in
99% yield. Additional substitution processes are detailed in the
section on subsitution reactions and low-temperature NMR
data below.
Scheme 4. Substitutions of chloride ligands by bromide, iodide, and cyanide
ligands.
spectively, as deep yellow powders in 70–99% yields. Similar
reactions with NaI (4.0 equiv) in THF/acetone (homogeneous
conditions) gave the corresponding orange diiodide complexes
trans-9c and trans-10c,e, respectively, in 58–92% yields. NMR
monitoring showed that the palladium complexes underwent
complete substitution within 10–20 min, whereas the platinum
complexes required hours.
Preparative reactions with platinum complex trans-1c
Subsequent to the chemistry described above, the platinum di-
chloride complex trans-4c became the lead compound for
a number of investigations, including studies of a dibridgehead
diphosphine obtained below. As such, the sequence with
trans-1c in Scheme 2 has been conducted many dozens of
times by many researchers, with the yield of trans-4c typically
slightly greater than 40% on scales up to one gram.
Rotators featuring metal–carbon bonds were also sought.
Thus, trans-4c,e and trans-5c were analogously treated with
KCN (2.0–4.0 equiv) in CH2Cl2/CH3OH (homogeneous condi-
tions). Workups gave the dicyanide complexes trans-11 c,e and
trans-12c, respectively, as white powders in 86–91% yields.
The 13C NMR spectra showed diagnostic signals for the cyanide
ligands (124.3–129.5 ppm), which were coupled to phosphorus
2
In larger scale reactions, somewhat lower yields of trans-4c
have usually been obtained. A procedure involving five parallel
1.8–2.2 g scale reactions (each 1.0–1.2 L total solvent) and
a combined chromatographic workup is given in the Experi-
mental Section. Noteworthy is that, in addition to the 32%
yield of trans-4c isolated, 5% of a new monoplatinum product,
trans-4’c, was obtained (Scheme 2). This species was not de-
tected in the small-scale reactions communicated earlier,[7] al-
though close examinations of NMR spectra of some samples
prior to chromatography revealed observable quantities. The
structural assignment was supported by a crystal structure of
the BH3 adduct of the free diphosphine ligand, obtained as de-
scribed below.
(t, JCP =17.2–14.1 Hz). Strong IR nCN bands were also apparent
(2123–2142 cmꢀ1).
All of the complexes in Scheme 4 showed only a single set
of n/2 methylene 13C NMR signals. In this context, consider the
idealized conformational energy minimum VI in Scheme 5,
which features a mirror plane. This would give rise to two sets
of n/2 methylene 13C NMR signals with a 2:1 intensity ratio.
1
Thus, H and 13C{1H} NMR spectra of trans-4c, trans-5c, trans-
[24]
10c, and trans-12c were recorded in CDFCl2 at low tempera-
tures. However, even at ꢀ1208C, no decoalescence—only
signal broadening—was observed. Hence, the MCl2, MBr2, MI2,
and M(CN)2 moieties appear to have very low rotational barri-
ers, rendering the three methylene bridges equivalent on the
NMR timescale. The carbon–phosphorus coupling seen in the
cyanide ligand 13C NMR signals rules out alternative dynamic
processes involving dissociation of the cyanide or diphosphine
ligands.
Substitution reactions and low-temperature NMR data
The scope of substitution reactions that can be carried out
within the diphosphine cages of the gyroscope-like complexes
was probed. As shown in Scheme 4, trans-4c and trans-5c,e
were combined with LiBr (2.3–4.0 equiv) in THF at room tem-
perature (homogeneous conditions). Workups gave the corre-
sponding dibromide complexes trans-7c and trans-8c,e, re-
Complexes with more longitudinally extended rotors were
sought. Thus, trans-4c and the thiocyanide salt KSCN
(4.1 equiv) were combined in THF/acetone (homogeneous con-
ditions). As shown in Scheme 6 (top), a chromatographic
workup gave two white crystalline products, trans-13c and
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4620
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
hibited two IR nNCS bands, one close to that of trans-13c
(2107 cmꢀ1), and the other at a lower frequency (2005 cmꢀ1)—
a shift typical of a thiocyanide ligand.[26] Hence, trans-14c was
assigned as an unsymmetrical NCS/SCN linkage isomer.[27] In
a series of palladium thiocyanide/isothiocyanide adducts with
diphosphine ligands Ph2P(CH2)m’PPh2 (m’=2, 3, 4), either the
SCN/SCN, NCS/SCN, or NCS/NCS isomers can dominate, de-
pending upon chelate ring size.[28] Importantly, the 13C NMR
spectra of trans-13c and trans-14c exhibited 12 and 13 meth-
ylene carbon signals, respectively, showing PtX2 rotation to be
slow on the NMR timescale at room temperature.[29]
Finally, trans-4c,e were combined with the carbon nucleo-
phile Ph2Zn (3.0 equiv) in THF. After 20 h, NMR spectra showed
that substitution was ꢂ95% complete. As depicted in
Scheme 6 (bottom), workups gave the corresponding diphenyl
complexes trans-15c and trans-15e as white air stable pow-
ders in 61–90% yields. The 13C NMR spectrum of trans-15c
showed 14 methylene carbon signals, or two sets of seven
(n/2) with an approximate 2:1 intensity ratio. However, that of
trans-15e showed only nine signals. This indicates slow and
fast PtPh2 rotation, respectively, on the NMR timescale at room
temperature. Interestingly, the 1H and 13C NMR spectra of
trans-15c exhibited two ortho- and two meta-phenyl CH sig-
nals. This further requires restricted rotation about the PtꢀCipso
axis as depicted in Scheme 6. No coalescence or line broaden-
ing was observed when spectra were recorded at 808C in
Scheme 5. Effect of rotator symmetry upon numbers and types of energy
minima and maxima in gyroscope-like complexes.
1
CDCl3. The H and 13C NMR spectra of trans-15e did not exhibit
additional phenyl signals at room temperature.
Crystal structures of gyroscope-like complexes
As noted above, single crystals of many complexes could be
obtained. The crystal structures of six platinum and palladium
dichloride, dibromide, and diiodide complexes with 14 carbon
methylene segments, trans-4c, trans-7c, trans-9c, trans-5c,
trans-8c, and trans-10c, were determined as summarized in
the Experimental Section and Table S1 in the Supporting Infor-
mation. With trans-5c and trans-7c, four to five carbon atoms
were disordered, but the major and minor conformations
could be resolved. Unit cell parameters and selected inter-
atomic distances and bond angles are provided in Table 2.
Views of each structure are given in Figures 1–6.
The crystal structures of two additional platinum complexes
with 14 carbon methylene segments were determined, the bis-
(isothiocyanide) complex trans-13c and a pentane monosol-
vate of the diphenyl complex, trans-15c·n-C5H12. With the
former, which exhibited a C2 axis perpendicular to the coordi-
nation plane, four carbon atoms were disordered and refined
with 50:50 occupancy ratios. The NꢀPtꢀN and N=C=S linkages
were nearly linear (178.8(2)8 and 178.6(5)8), but consistent with
literature structures[28] the PtꢀN=C linkage was more bent
(169.6(5)8). The crystal structure of the diphenyl complex with
18 carbon methylene segments, trans-15e, could also be de-
termined. Data for these compounds are incorporated into
Table 2, and the molecular structures are depicted in Fig-
ures 7–9.
Scheme 6. Additional chloride ligand substitution reactions.
trans-14c, in 28 and 20% yields, respectively. When a similar
reaction was monitored by 31P NMR spectroscopy, the initial
13c/14c ratio was about 50:50, and shifted with time in favor
of trans-14c.[25]
Complex trans-13c exhibited
a single IR nNCS band
(2111 cmꢀ1), and a crystal structure (below) showed two nitro-
gen bound or isothiocyanide ligands. Complex trans-14c ex-
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4621
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 2. Lattice parameters and key interatomic distances [ꢁ] and angles [8] for gyroscope-like complexes.
trans-4c
trans-7c
trans-9c
trans-5c
trans-8c
trans-10c
trans-13c
trans-15c
·n-C5H12
trans-15e
¯
¯
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
a [8]
P21/n
P21/n
Pbca
P21/n
P21/n
Pbca
C2/c
P1
P1
16.264(3)
13.698(3)
19.998(4)
90
91.29(3)
90
16.2375(4)
13.9462(4)
20.2226(5)
90
90.31(3)
90
20.702(4)
13.778(3)
33.197(6)
90
90
90
16.2233(2)
13.7836(3)
20.1744(4)
90
91.207(1)
90
16.318(7)
13.796(6)
20.043(9)
90
90.52(4)
90
20.737(4)
13.785(3)
33.259(5)
90
90
90
18.225(4)
15.265(4)
18.179(4)
90
104.27(3)
90
11.9863(2)
16.4670(3)
16.9113(2)
67.387(1)
72.547(1)
79.382(1)
2930.44(8)
2
9.4061(2)
13.1694(2)
28.3030(5)
79.746(1)
80.949(1)
71.449(1)
3251.61(10)
2
b [8]
g [8]
V [ꢁ3]
Z
4454.1(16)
4
4579.4(2)
4
9469(3)
8
4510.31(14)
4
4512(3)
4
9507(3)
8
4901(2)
4
MꢀP1
MꢀP2
MꢀX1
MꢀX2
P1···P2
X-M-X
P-M-P
2.3051(8)
2.3082(8)
2.3165(10)
2.3074(10)
4.611
2.3100(13)
2.3169(13)
2.4281(8)
2.4313(8)
4.625
2.3207(17)
2.3278(17)
2.6192(6)
2.6045(7)
4.646
2.3199 (6)
2.3129(6)
2.2976(6)
2.3080(6)
4.630
2.317(4)
2.319(4)
2.416(2)
2.384(2)
4.633
2.3398(9)
2.3470(9)
2.6183(5)
2.5988(5)
4.685
2.3189(15)
2.3189(15)
1.961(4)
1.961(4)
4.634
2.2849(7)
2.2845(7)
2.079(3)
2.080(3)
4.568
2.2838(11)
2.2761(11)
2.096(4)
2.092(4)
4.559
178.10(3)
176.75(3)
177.13(2)
177.10(5)
173.508(18)
176.64(6)
177.49(3)
176.15(2)
176.98(9)
175.74(14)
172.891(11)
176.55(3)
178.8(2)
175.00(6)
175.87(10)
176.74(2)
179.13(16)
177.74(4)
Figure 1. Thermal ellipsoid plots of trans-4c (50% probability level): side
view (top); view along P2-Pt-P1 axis (bottom).
Figure 2. Thermal ellipsoid plots of trans-7c (50% probability level): side
view (top); view along P2-Pt-P1 axis (bottom). Five carbon atoms are disor-
dered; the dominant conformation is depicted.
Other metrical parameters, of interest in the context of anal-
yses presented in the discussion section, were also tabulated.
Torsion angles associated with the methylene segments are
presented in Tables 3 and 4 in a format that facilitates identifi-
cation of “conserved conformations”. Table 5 provides other
quantities associated with individual molecules or the crystal
lattice.
Demetalation reactions
In the course of exploring the above substitution chemistry,
unexpected demetalations that liberated the diphosphine li-
gands were discovered. Although these represent the start of
major ongoing investigations involving the demetalation
mechanism and diphosphine properties,[30,31] certain results
bear upon key synthetic or structural issues above. Hence, se-
lected observations are presented here.
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4622
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 5. Thermal ellipsoid plots of trans-8c (50% probability level): side
view (top); view along P-Pd-P axis (bottom).
Figure 3. Thermal ellipsoid plots of trans-9c (50% probability level): side
view (top); view along P2-Pt-P1 axis (bottom).
Figure 4. Thermal ellipsoid plots of trans-5c (50% probability level): side
view (top); view along P2-Pd-P1 axis (bottom). Four carbon atoms are disor-
dered; the dominant conformation is depicted.
Figure 6. Thermal ellipsoid plots of trans-10c (50% probability level): side
view (top); view along P2-Pd-P1 axis (bottom).
Small-scale reactions of trans-4c and NaCꢁCH (2–4 equiv;
commercial xylene slurry or powder) in THF were monitored
by 31P NMR spectroscopy. The reactants were gradually con-
sumed over the course of 24–48 h. As shown in Scheme 7, the
data suggested the formation of both mono- (trans-16c;
9.83 ppm, 1JPPt =2429 Hz) and disubstituted (trans-17c;
6.46 ppm, 1JPPt =2421 Hz) products. In an effort to promote
complete conversion to trans-17c, a larger excess of NaCꢁCH
(30 equiv) was employed. However, now a third species, with
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4623
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 7. Molecular structure of trans-13c with atoms represented as iso-
tropic spheres: side view (top); view along P-Pt-P axis (bottom).
Figure 8. Thermal ellipsoid plots of trans-15c·n-C5H12 with solvent molecule
omitted: side view (top); view along P2-Pt-P1 axis (bottom).
a signal that was not coupled to platinum, was clearly ob-
served (18, ꢀ31.9 ppm). After 140 h, trans-16c and trans-17c
could no longer be detected. Workup of a comparable prepa-
rative reaction gave 18 as a moderately air-sensitive white
powder in 91% yield. Based upon the mass spectrum, micro-
analysis, and chemistry described elsewhere,[30,31] this was as-
signed as the dibridgehead diphosphine corresponding to the
stator of trans-4c.
The isomeric complex trans-4’c and NaCꢁCH (5–50 equiv)
were also combined in THF at room temperature. After 9 d,
workup gave the isomeric diphosphine (CH2)14P(CH2)14P(CH2)14
(19) as a white powder in 90% yield. Alternatively, the crude
Table 3. Torsion angles [8] for gyroscope-like complexes with P(CH2)14P linkages for trans-4c, trans-7c, trans-9c, and trans-5c.[a]
trans-4c
trans-7c
trans-9c
trans-5c
P1-C1-C2-C3
C1-C2-C3-C4
C2-C3-C4-C5
C3-C4-C5-C6
178.2(2)/177.8(2)/ꢀ167.0(2) ꢀ168.8(4)/173.8(5)/177.8(4)
168.0(5)/174.2(5)/179.4(5)
ꢀ63.2(7)/56.3(8)/170.3(6)
ꢀ61.8(8)/62.4(8)/71.6(8)
ꢀ179.5(6)/178.2(6)/174.6(7)
172.4(2)/ꢀ176.44(17)/167.42(18)
ꢀ55.9(4)/54.7(4)/176.3(3) 175.6(5)/ꢀ63.8(8)/54.0(7)
163.8(3), ꢀ155.9(6)[b]/55.9(3)/ꢀ178.8(2)
61.7(4),ꢀ69.7(10)[b]/57.5(3)/61.3(3)
174.7(2),ꢀ176.8(7)[b]/ꢀ178.8(2)/60.3(3)
ꢀ53.8(4)/60.9(4)/ꢀ179.0(3) ꢀ178.3(5)/ꢀ49.5(10)/59.4(7)
ꢀ58.5(4)/178.1(3)/56.2(4)
ꢀ174.0(3)/71.2(4)/55.9(4)
178.1(3)/72.1(4)/63.9(4)
58.3(7)/ꢀ77.4(12),ꢀ57.8(12)[b]
/
179.0(5)
C4-C5-C6-C7
C5-C6-C7-C8
C6-C7-C8-C9
C7-C8-C9-C10
C8-C9-C10-C11
53.6(8)/ꢀ176.2(9),180(11)[b]
/
179.3(6)/60.8(8)/63.1(10)
172.6(6)/58.6(8)/60.8(10)
174.8(6)/173.9(6)/172.0(6)
56.8(9)/176.4(6)/60.1(9)
63.0(8)/ꢀ177.3(6)/53.6(10)
66.8(4),171.4(9)[b]/92.0(3),51.7(7)[b]
179.6(2)
/
71.6(7)
63.9(7)/ꢀ170.7(12),62(2)[b]
/
61.6(4),79.2(6)[b]/ꢀ66.7(4), 72.0(12)[b]
/
72.4(7)
ꢀ177.1(2)
ꢀ176.9(3)/178.7(3)/177.4(3) ꢀ179.6(5)/ꢀ172.1(10),174.8(11)[b]
/
179.2(2)/ꢀ178.5(3),173.5(8)[b]/178.8(2)
64.6(3)/178.2(3), 77.4(13)[b]/72.4(3)
54.3(3)/ꢀ172.2(3),159.1(8)[b]/71.3(3)
57.7(3)/ꢀ59.7(4)/178.3(2)
177.8(5)
ꢀ64.7(4)/ꢀ176.1(3)/62.5(4)
91.7(3)/ꢀ179.6(3)/67.0(4)
61.2(9),78.5(11)[b]/ꢀ67.2(12),
66.9(12)[b]/ꢀ177.4(5)
68.6(9),168.5(15)[b]/93.6(7),
51.6(9)[b]/ꢀ179.4(5)
C9-C10-C11-C12 ꢀ179.2(2)/60.0(4)/174.8(3)
173.3(6),ꢀ175.5(13)[b]/ꢀ179.1(5)/ ꢀ178.8(6)/ꢀ179.7(6)/59.9(9)
61.6(7)
C10-C11-C12-C13
C11-C12-C13-C14
57.0(4)/59.8(4)/60.9(4)
55.1(4)/ꢀ178.8(3)/165.1(3) 161.8(6),ꢀ157.7(9)[b]/56.1(6)/
ꢀ67.7(17),61.3(9)[b]/57.9(6)/63.3(7)
63.8(8)/ꢀ62.9(8)/ꢀ176.4(7) ꢀ178.5(2)/ꢀ52.9(4)/60.3(3)
56.6(8)/ꢀ62.8(8)/176.9(6)
175.4(2)/ꢀ58.5(3)/55.7(3)
ꢀ177.8(5)
C12-C13-C14-P2 ꢀ175.9(2)/168.0(2)/172.2(2)
171.5(5)/ꢀ177.1(4)/164.0(4)
171.4(5)/165.5(5)/ꢀ163.1(5) ꢀ167.94(17)/178.57(18)/178.27(18)
[a] The torsion angles for each of the three P(CH2)14P segments are separated by slashes. [b] One or more carbon atoms associated with this torsion angle
are disordered. The first value represents that derived from the carbon atoms with the greater occupancy factors (dominant conformation). The second
value represents that from the carbon atoms with the lower occupancy factors.
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4624
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 10. Thermal ellipsoid plot of 19·2BH3 (50% probability level). Key
crystallographic distances (ꢁ) and bond angles (8): P1···P2 19.406(9), P1ꢀB1
1.903(5), P2ꢀB2 1.916(5), P1ꢀC1 1.823(4), P1ꢀC15 1.811(4), P1ꢀC28 1.806(4),
P2ꢀC14 1.797(5), P2ꢀC29 1.848(6), P2ꢀC42 1.802(5), C1-P1-B1 113.4(2), C15-
P1-B1 113.4(2), C28-P1-B1 114.6(2), C14-P2-B2 113.8(2), C29-P2-B2 111.4(3),
C42-P2-B2 113.9(3), C1-P1-C15 102.4(2), C15-P1-C28 107.3(2), C28-P1-C1
104.7(2), C14-P2-C29 103.5(2), C29-P2-C42 110.1(3), C42-P2-C14 103.4(2).
low overall yields by means of ring-closing metathesis/hydro-
genation sequences involving trans-phosphine ligands of the
formula P((CH2)mCH=CH2)3, in which m is greater than or equal
to six. Systems with 17-membered macrocycles (trans-4c,
trans-5c) and 14 methylene groups are most efficiently ob-
tained, and appreciable quantities of 21-membered analogues
can also be isolated.
Reactions with dibromo analogues (Scheme 3) are slower,
and give byproducts arising from catalyst-derived chloride/bro-
mide exchange. Exchange was not a problem in a similar se-
quence involving an octahedral rhenium bromide complex.[10]
However, PtO2 was used as the hydrogenation catalyst, and
the yield was lower than with the rhenium chloride analogue.
These data suggest that smaller ligands are advantageous for
threefold interligand alkene metathesis. Indeed, when one of
the chloride ligands in trans-1c is replaced by a pentafluoro-
phenyl ligand, only oligomers are obtained.[32]
Figure 9. Thermal ellipsoid plots of trans-15e: side view (top); view along
P2-Pt-P1 axis (bottom).
product was treated with BH3·S(CH3)2. This gave the previously
reported[30] bis(borane) adduct 19·(BH3)2 as a white solid in
78% overall yield. Single crystals were obtained, and the crys-
tal structure determined (Experimental Section). The molecular
structure and key metrical parameters are given in Figure 10.
This unequivocally establishes the nature of the ligand in
trans-4’c and related complexes above.
Ring-closing metatheses of pentafluorophenyl-substituted
platinum chloride complexes with trans-diphenylphosphine li-
gands of the formula Ph2P(CH2)mCH=CH2 have also been inves-
tigated.[33] In these cases, metathesis/hydrogenation sequences
efficiently yield 13 to 23-membered macrocycles with 10–20
methylene groups (m=4, 67%; 5, 82%; 6, 69%; 8, 53%; 9,
43%). The failure to detect monometallic products from the
substrates with m=4 and 5 in Scheme 2 may in part reflect
Discussion
Scope of syntheses
Scheme 2 shows that dichloroplatinum- and dichloropalladi-
um-based gyroscope-like complexes can be accessed in fair to
Table 4. Torsion angles [8] for gyroscope-like complexes with P(CH2)14P linkages for trans-8c, trans-10c, trans-13c, and trans-15c·n-C5H12.[a]
trans-8c
trans-10c
trans-13c
trans-15c·n-C5H12
P1-C1-C2-C3
C1-C2-C3-C4
C2-C3-C4-C5
C3-C4-C5-C6
C4-C5-C6-C7
C5-C6-C7-C8
C6-C7-C8-C9
C7-C8-C9-C10
C8-C9-C10-C11
C9-C10-C11-C12
C10-C11-C12-C13
C11-C12-C13-C14
C12-C13-C14-P2
ꢀ162.7(8)/ꢀ180.0(10)/172.9(9)
173.1(11)/ꢀ61.6(17)/55.1(16)
179.3(11)/ꢀ52(2)/56.3(17)
56.9(18)/ꢀ58(2)/174.7(12)
52.4(19)/ꢀ174.9(14)/73.4(17)
68.9(17)/174.9(16)/68.6(17)
176.8(12)/ꢀ178.1(15)/ꢀ177.6(11)
63.3(18)/ꢀ57(2)/ꢀ171.5(11)
66(2)/90.3(16)/ꢀ177.7(11)
174.1(14)/ꢀ179.5(12)/57.8(15)
62.7(17)/52.6(16)/63.5(15)
161.7(12)/57.6(16)/178.4(10)
169.3(10)/ꢀ175.5(9)/162.7(9)
167.4(2)/174.0(2)/179.6(2)
ꢀ64.0(4)/56.3(4)/171.0(3)
ꢀ62.9(4)/62.2(4)/71.2(4)
ꢀ179.6(3)/177.9(3)/173.7(3)
179.5(3)/61.7(4)/63.9(5)
172.5(3)/58.8(4)/60.9(5)
174.8(3)/174.2(3)/170.4(3)
56.2(4)/176.4(3)/58.9(4)
64.4(4)/ꢀ177.8(3)/54.1(4)
ꢀ179.1(3)/ꢀ180.0(3)/60.0(4)
175.3(4)/ꢀ176.2(8)/157.9(5)
56.4(7)/ꢀ164.1(13)/ꢀ76.6(8)
58.9(7)/ꢀ88.4(19)/ꢀ172.9(9)
171.2(5)/ꢀ143(3)/ꢀ169.8(9)
63.8(8)/ꢀ129(2)/ꢀ70.5(12)
61.2(9)/ꢀ96(3)[b]/ꢀ174.8(7)
172.4(6)/146(3)[b]/172.4(6)
ꢀ174.8(7)/ꢀ96(3)[b]/61.2(9)
ꢀ70.5(12)/ꢀ129(2)[b]/63.8(8)
ꢀ169.8(9)/ꢀ143(3)[b]/171.2(5)
ꢀ161.7(3)/ꢀ177.4(2)/169.4(2)
168.1(4)/ꢀ58.1(4)/ꢀ179.9(3)
ꢀ69.8(5)/ꢀ65.9(4)/177.3(3)
ꢀ80.2(5)/ꢀ175.4(3)/180.0(3)
178.3(3)/ꢀ64.7(4)/59.8(5)
ꢀ179.5(3)/ꢀ62.9(5)/58.6(5)
ꢀ173.0(3)/ꢀ176.5(3)/178.2(3)
ꢀ56.7(5)/ꢀ179.3(3)/59.6(4)
ꢀ64.1(5)/177.8(3)/59.9(4)
ꢀ175.2(3)/ꢀ71.8(4)/178.6(3)
ꢀ65.5(4)/ꢀ70.4(4)/177.7(3)
ꢀ59.1(4)/172.9(3)/ꢀ179.4(3)
63.5(4)/ꢀ62.1(4)/ꢀ176.8(3) ꢀ172.9(9)/ꢀ88.4(19)[b]/58.9(7)
56.4(4)/ꢀ65.2(4)/177.6(3)
170.7(2)/167.2(2)/ꢀ163.6(2)
ꢀ76.6(8)/ꢀ164.1(13)[b]/56.4(7)
157.9(5)/ꢀ176.2(18)[b]/175.3(4) ꢀ175.9(2)/ꢀ162.3(2)/169.6(2)
[a] The torsion angles for each of the three P(CH2)14P segments are separated by slashes. [b] One or more carbon atoms associated with this torsion angle
are disordered with 50:50 occupancy factors. The torsion angle given is for the chain with the better geometrical parameters.
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4625
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 5. Intramolecular and intermolecular distances involving rotator and stator atoms in gyroscope-like complexes [ꢁ], and angles defined by gyroscope
axes [8].
trans-4c
trans-7c
trans-9c
trans-5c
trans-8c
trans-10c
trans-13c
trans-15c
·n-C5H12
trans-15e
MꢀX[a]
2.312
4.062
7.448
6.746
6.385
5.734
5.457
6.350
3.757
5.748
4.619
2.919
79.42
79.74
2.430
4.280
7.448
6.798
6.461[h]
5.566[h]
6.388
5.735
3.866
5.748
4.800
3.100
78.07
78.14
2.612
4.592
7.295[f]
6.589
5.630
5.490
5.629
6.676
3.790
5.595
5.878
2.303
4.053
7.438
6.762
5.741
6.383
6.394[h]
5.518[h]
3.818
5.738
4.685
2.985
79.16
79.00
2.400
4.250
7.453
6.736
5.691
6.441
6.342
5.434
3.734
5.753
4.714
3.014
79.87
78.84
2.609
4.589
7.293
6.529
6.682
5.635
5.642
6.713
3.935
5.593
5.869
4.708[b]
6.508
7.38[g]
7.38[g]
6.954
6.001
6.001
6.954
4.301
5.68[g]
5.020(8)
5.883[c]
7.083
7.790
7.760
5.774
5.206
5.348
5.759
3.506
6.090
6.153
5.873[c]
7.073
10.373
10.486
5.915
6.162
7.928
7.793
4.215
8.786
5.732
radius of rotator[d]
MꢀCa[e]
MꢀC’a[e]
[e]
MꢀCb
[e]
MꢀC’b
MꢀCc[e]
MꢀC’c[e]
[i]
MꢀCdistal-vdW
[j]
MꢀCdistal’-vdW
[k]
MꢀCneighbor
[l]
MꢀCneighbor-vdW
4.178
[n]
4.169
[n]
3.320(8)
[o]
4.453
[o]
4.032
[o]
a [P-M-P + P-M-P][m]
a [P-MX2-P][m]
[n]
[n]
[o]
[o]
[o]
[a] The crystallographic distances (Table 2) are averaged. [b] Distance from platinum to the non-bonded sulfur atom. [c] Distance from platinum to the
para-hydrogen atom of the phenyl ring. [d] MꢀX plus the van der Waals radius of the terminal atom (halide, sulfur, or hydrogen). [e] Distance from the
metal to the two remote carbon atoms of the macrocycles that are closest to the plane of the rotator (MꢀCdistal). The indices a, b, and c refer to the three
different macrocycles as illustrated in Figure 11. [f] In this complex, a carbon atom adjacent to this atom was the farthest from the metal (7.388 ꢁ).
[g] These distal carbon atoms are disordered with a 50:50 occupancy ratio. The average distance to the metal is given. [h] These distal carbon atoms are
disordered with 53:47 or 82:18 occupancy ratios. The distance corresponds to that to the remote carbon with the greater occupancy factor. [i] The shortest
of the previous six distances, minus the van der Waals radius of carbon (1.70 ꢁ). [j] The longest of the previous six distances, minus the van der Waals
radius of carbon. [k] Distance from the metal to the nearest carbon atom of a neighboring molecule. [l] Previous entry minus the van der Waals radius of
carbon. [m] Planes used for angle calculations between sets of molecules with parallel P-M-P axes; see also the text and Figure 14. [n] This lattice exhibits
four sets of molecules with parallel P-M-P axes. [o] All of the P-M-P axes in the lattice are parallel.
Scheme 7. Isolations of diphosphines by demetalations of platinum complexes.
a greater likelihood for intermolecular metathesis with multiple
shorter chain (CH2)mCH=CH2 substituents.
spect to the CO ligands, as shown in X in Scheme 8. This di-
rects the (CH2)mCH=CH2 groups that must metathesize to
afford III into the same interstices. Once one linkage is in
place, subsequent intramolecular and interligand metathesis
becomes more probable.
In this context, it should be noted that the trigonal bipyra-
midal iron tricarbonyl analogs III (Scheme 1) can be accessed
in good overall yields from precursors trans-Fe(CO)3(P-
((CH2)mCH=CH2)3)2 with m=4 (50–64%), 5 (63%), 6 (64%), and
8 (74%).[6] The superior yields may in part be due to a preorga-
nization effect. Specifically, both the Fe(CO)3 fragment and
phosphine ligands have a C3 symmetry axis. In the precursors,
each phosphine ligand would logically adopt a conformation
in which the (CH2)mCH=CH2 substituents are staggered with re-
In contrast, consider the limiting conformations XI and XII
depicted in Scheme 8 for the dichloroplatinum and dichloro-
palladium precursors. The latter directs three phosphine sub-
stituents on each side of the metal square plane, while mini-
mizing P-M-P torsional interactions, and may be the more
stable. In any case, after an initial interligand macrocyclization
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4626
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
unambiguously attained in other cases.[35] Also, there remain
possibilities for optimizing the preceding chemistry. For exam-
ple, Fogg has emphasized that oligomers sometimes dominate
kinetically in ring-closing metatheses, but can convert to mon-
ocyclic products given sufficient time, catalyst, and ethylene
co-product.[36]
Some conceptually related synthetic efforts from other re-
search groups merit note. Setaka and co-workers have shown
that threefold intramolecular olefin metatheses can be effected
with several aromatic systems that bear para-substituents of
the formula Si((CH2)mCH=CH2)3.[16c,e] A representative sequence
leading to the gyroscope-like bis(silane) 21, which features a p-
phenylene rotator and 20-membered macrocycles, is shown in
Scheme 9 (top). Related compounds with silicon containing
spokes have been accessed by other routes.[16a,b]
Scheme 8. Some conformational considerations in ring-closing metatheses
leading to gyroscope-like complexes.
of XI or XII, an equilibrium mixture of conformers XIIIa and
XIIIb should result. While the former is positioned to lead to
the gyroscope-like product XVI, the latter can undergo either
1) a productive macrocyclization to give XIV, which can with
appropriate torsional motion be transformed to XVI, or 2) a
dead-end macrocyclization to give XV, which due to mis-
matched connectivity can only lead to oligomer or polymer.
The latter has no counterpart in a trigonal-bipyramidal geome-
try.
Scheme 9. Gyroscope-like organic and organosilicon molecules and related
species synthesized by the Garcia-Garibay and Setaka groups.
Except in the case of trans-1e, the yields of products derived
from a combination of inter- and intraligand metathesis (trans-
4’c,d,e, trans-5’e; Scheme 2) are much lower than those of the
gyroscope-like complexes. They may be derived from XIII in
Scheme 8, or by initial intraligand metathesis. Analogous prod-
ucts have not yet been detected with any trigonal-bipyramidal
substrate. However, they can be observed with various octahe-
dral substrates,[34] and one has been crystallographically char-
acterized. No attempt has been made to establish thermody-
namic control in any of these reactions, although this has been
The Garcia-Garibay group has reported that the gyroscope-
like molecule 23, which features 28-membered macrocycles,
can be obtained in 27% yield by a threefold intramolecular
Williamson ether synthesis using the tris(phenol) 22 (Scheme 9,
middle).[15a,b] Interestingly, reaction of the hexa(alkene) 24 with
Grubbs’ catalyst afforded only the isomer 26 after hydrogena-
tion, which is derived from a combination of inter- and intra-
endgroup metathesis. This is opposite to the dominant cycliza-
tion mode found with all coordination geometries in
Scheme 1. Homologues with shorter and longer methylene
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4627
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
chains gave analogous results. An intermolecular Williamson
ether synthesis with the hexa(phenol) 25 and an a,w-dibro-
mide also afforded 26 (Scheme 9, bottom). However, reactions
of 25 with other types of a,w-electrophiles gave low yields of
compounds with topologies analogous to 23.[15b]
substituents closely approximate an idealized staggered con-
formation, with torsion angles of 53.48 to 57.148 (vs. 608). The
corresponding torsion angles in the other complexes are much
smaller. The remaining two macrocycles in trans-15e exhibit
spiral-staircase-like conformations that span two gauche posi-
tions. One features a much higher proportion of gauche seg-
ments than the ring analyzed above (8 of 17), apropos to the
chiral and helical nature of such constructs. This underscores
the conformational variability of the stators, especially as the
lengths of the methylene chains are increased.
Structural data: bond lengths, bond angles, and torsion
angles
Several platinum and palladium dichloride, dibromide, and
diiodide complexes of the formula trans-[MX2(PR3)2] have been
crystallographically characterized.[37] The six new dihalide com-
plexes presented above (Figure 1–Figure 6) are unexceptional
from the standpoint of bond lengths about the metals. As
summarized in Table 2, the X-M-X angles in the diiodide com-
plexes trans-9c and trans-10c exhibit the greatest deviations
from 1808 (172.891(11) and 173.508(18)8), perhaps reflecting re-
pulsive interactions of this larger ligand with the macrocycles;
values of >179.58 are common in acyclic analogues.[37d,f] The
smallest P-M-P angle is found in the palladium dibromide com-
plex trans-8c (175.74(14)8). Among the three non-halide com-
plexes, the X-M-X and P-M-P angles in the diphenyl adduct
trans-15c are the smallest (175.87(10) and 176.74(2)8).
Cage dimensions and dynamic properties
The ultimate objective of this research is to prepare transition-
metal-based gyroscope-like rotors in which the rotators can
turn within cage like stators with energy barriers that are as
low as possible. Towards this end, the interior space available
within the stator as compared to the sizes of the rotor ligands
is important. Figure 11 provides a starting point for a simple,
but admittedly approximate, analysis that uses the preceding
crystallographic data.
The torsion angle data in Table 3 reveal a number of rela-
tionships for the eight complexes with P(CH2)14P linkages. In all
adducts, the terminal P1-C1-C2-C3 and C12-C13-C14-P2 seg-
ments adopt anti conformations, with torsion angles ranging
from ꢃ157.9(5) to ꢃ180.0(10)8. Similarly, the middle C6-C7-C8-
C9 segments are also anti, with torsion angles ranging from
ꢃ179.6(5) to ꢃ170.4(3) or ꢃ146(3)8 (disordered chain). These
units can be viewed as the “top”, “bottom”, and “sides” of the
stators, and crystallize as linearly as possible. The intermediate
segments feature a mixture of anti and gauche conformations,
with the latter providing the necessary “curvature”.
Of the 24 macrocycles in these eight complexes, 17 exhibit
seven anti-PCH2CH2CH2 or -CH2CH2CH2CH2 segments and six
gauche segments (totals reflect the dominant conformations of
the three disordered chains). Four possess six anti and seven
gauche segments, and one has eight and five. Both trans-13c
and trans-15c exhibit one macrocycle with nine anti and four
gauche segments, and the distinctive motif in the latter is ana-
lyzed in the following paragraph. The most striking homology
involves the platinum and palladium diiodide complexes trans-
9c and trans-10c, for which all rings exhibit identical anti/
gauche sequences (seven and six total), and nearly identical
torsion angles. Less extended homologous sequences in the
other complexes are readily apparent in Figure 1–Figure 8.
The diphenyl complex trans-15e (Figure 9) features 21-mem-
bered macrocycles with 18 methylene groups, each of different
conformational types. Interestingly, trans-15e and the lower
homologue trans-15c (Figure 8) both have one ring with two
extended series of anti segments (P1 to C6A or C8A and C9A
or C11A to P2), connected by a “curved” gauche/gauche/anti/
gauche/gauche sequence. This motif is not found in any of the
other complexes. The conformation about the phosphorus–
phosphorus vector (CH2ꢀP···PꢀCH2) in trans-15e is also note-
worthy. As can be seen in Figure 9 (bottom), the phosphorus
Figure 11. Calculations involving the spatial void between the rotator and
the carbon spokes, illustrated using trans-4c as a model; a, b, c=corre-
sponding macrocycles, vdW=van der Waals.
The first step is to calculate the radius of the rotator. For the
dihalide complexes, this entails summing the average MꢀX
bond lengths and the van der Waals radius of X.[38] For the bis-
(isothiocyanide) complex trans-13c, this entails summing the
platinum–sulfur distance and the van der Waals radius of
sulfur. For the diphenyl complexes trans-15c,e, the distance
from platinum to the para-hydrogen atom is calculated, and
the van der Waals radius of hydrogen is added. Values are pre-
sented in Table 5.
The horizontal interior space of the stator cage is then ana-
lyzed. First the distances between the metal atom and the
remote carbon atoms of each macrocycle are calculated (C7
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4628
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
and C8 for P(CH2)14P linkages). These are normally the two
closest to the plane that would be defined by MX2 rotation. As
depicted in Figure 11 and summarized in Table 5, the distances
vary, and the shortest among the six possibilities is taken. The
van der Waals radius of an sp3-hybridized carbon atom is then
subtracted, giving a “horizontal clearance” or minimum “bridge
height”. The resulting distances (MꢀCdistal-vdW) are listed in
Table 5. However, it should be emphasized that an equilibrium
ensemble of macrocycle conformations would be expected in
solution, and any impression of a rigid steric barrier is unin-
tended.
(P1 to C8A; C11A to P2) is more than twice as great (8.786 ꢁ).
This suggests that as the macrocycles become larger, the
radius of the rotator might better be compared to the average
clearance of the three macrocycles.
Only non-hydrogen atoms of the stator have been consid-
ered in this admittedly qualitative analysis. As a partial justifica-
tion, many conformational processes readily take place in or-
ganic molecules despite van der Waals interactions involving
hydrogen atoms and/or difficulties executing them with space-
filling models. Nonetheless, space-filling representations that
include hydrogen atoms are provided in Figure 12 for the plat-
The minimum “bridge heights” for the six dihalide com-
plexes range from 3.734 to 3.935 ꢁ, and the radii of the rota-
tors range from 4.053 to 4.592 ꢁ. In all cases the latter are
somewhat greater. The radius of the rotators in the dicyanide
complexes can be estimated as 4.68–4.69 ꢁ from other crystal-
lographically characterized platinum and palladium bis(phos-
phine) adducts.[39] Nonetheless, the 13C{1H} NMR data for all of
these complexes show that rotation is rapid on the NMR time-
scale, even at low temperatures. Presumably, the passage of
the MX2 ligands through the macrocycles is correlated with
conformational changes that enhance the horizontal clearance.
As an upper bound on clearance, the longest of the six metal–
carbon distances noted above is taken, and the van der Waals
radius of a carbon atom subtracted (MꢀCdistal’-vdW). These values,
5.593 to 5.753 ꢁ (Table 5), easily exceed the radii of the rota-
tors.
Figure 12. Space filling representations of trans-4c, trans-7c and trans-9c.
inum dichloride, dibromide, and diiodide complexes trans-4c,
trans-7c, and trans-9c. Visually the void spaces are modest, es-
pecially as compared to the larger iodide ligands in trans-9c.
This underscores the necessity of coupling conformational pro-
cesses involving the macrocycles to MX2 rotation. Some space
filling representations of trans-15c and trans-15e are provided
in Figure 13. The last view highlights the thin space through
which a phenyl ring must pass.
A related issue involves the “floor-to-ceiling” or vertical inte-
rior space of the stator cage. Although the PꢀMꢀP distances
present a possible starting point for analysis, we prefer to
focus on the closest CH2ꢀP···PꢀCH2 spacings (i.e., those associ-
ated with the smallest torsion angles, which are typically 34.88
to 52.28). Consider the CH2ꢀP···PꢀCH2 linkage in the platinum
dichloride complex trans-4c that has a torsion angle of 37.98.
This corresponds to a carbon–carbon distance of 6.22 ꢁ. When
twice the van der Waals radius of carbon is subtracted, a rough
floor-to-ceiling or vertical clearance of 2.82 ꢁ is obtained. Im-
portantly, this is less than the van der Waals diameter of
a chlorine atom (2ꢂ1.75=3.50 ꢁ).[38] The fit becomes tighter
for a larger iodide ligand (2ꢂ1.98=3.96 ꢁ), but is comparable
for a cyanide ligand, for which the “width” can be approximat-
ed as twice the van der Waals radii of sp-hybridized nitrogen
and carbon atoms (2ꢂ1.55=3.10 ꢁ, 2ꢂ1.78=3.56 ꢁ).
Figure 13. Space filling representations of trans-15c·n-C5H12 (left, with sol-
vent molecule omitted) and trans-15e (middle) along the P-Pt-P axes, and of
trans-15e (right) along the C-Pt-C axis.
The new MCl2, MBr2, MI2, and M(CN)2 complexes appear to
have considerably lower rotational barriers than their Fe(CO)3
and [Fe(CO)2(NO)]+ analogues (DH° 9.5 kcalmolꢀ1, DS°
ꢀ6.5 eu for the cation).[6] In another study involving the same
With the bis(isothiocyanide) complex trans-13c, the radius
of the rotator is substantially greater than the horizontal clear-
ance (6.508 vs. 4.301 ꢁ; Table 5). Thus, MX2 rotation is effective-
ly braked, and two sets of 13C NMR signals are observed for the
methylene carbon atoms. The same holds for the diphenyl
complex trans-15c (7.083 vs. 3.506 ꢁ), although the clearance
afforded by the macrocycle with the extended anti segments
(vide supra; P1 to C6A and C9A to P2) is much greater
(6.090 ꢁ). With the diphenyl complex trans-15e, which features
21-membered macrocycles, the horizontal clearance (4.215 ꢁ)
is set by one of the “spiral staircase” methylene chain motifs
noted above. This would also seemingly brake PtPh2 rotation,
but rotation is fast for this complex. Importantly, the clearance
afforded by the macrocycle with the extended anti segments
diphosphine stator,
a trivalent trigonal rhodium rotator
(Rh(CO)2I) was shown to have a distinctly higher rotational bar-
rier than a divalent linear rhodium rotator (Rh(CO)I). This can
be rationalized in the context of Scheme 5. A 3608 rotation of
a trigonal rotator would require three IX transition states with
threefold eclipsing interactions. All three macrocycles must
extend to accommodate the three eclipsed ligands. On the
other hand, a 3608 rotation of a divalent linear rotator would
require six VII transition states with onefold eclipsing interac-
tions. Only one macrocycle must be sufficiently extended. It is
furthermore well established that in purely organic molecules,
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4629
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
sixfold rotational barriers are generally much lower than three-
fold rotational barriers.[40]
constants (Z=4, V=4454.1(16)–4510.31(14) and 4579.4(2)–
4512(3) ꢁ3). The diiodide complexes trans-9c and trans-10c
also crystallize in the same space group (Pbca) with very similar
lattice constants (Z=8, V=9507(3)–9469(3) ꢁ3). In all cases, the
orientations of the P-M-P axes are of interest.
One must be mindful that mechanisms other than ML2 rota-
tion may render the methylene bridges equivalent on NMR
timescales. NMR evidence against ligand dissociation was pre-
sented in the Results section. However, cis/trans isomerization
about the metal also presents possibilities. Consider the
isomer cis-4c shown in Scheme 10, which can be accessed by
In the dichloride and dibromide complexes, there are two
“non-parallel” sets of molecules with parallel P-M-P axes. A
packing diagram illustrating this feature is given in Figure 14.
The angle between the sets of axes can be quantified in sever-
Scheme 10. Attempted equilibration of cis-4c and trans-4c.
Figure 14. Packing diagram of trans-5c illustrating the two “non-parallel”
sets of molecules with parallel P-M-P axes.
a threefold intramolecular ring-closing metathesis of cis-1c.
One macrocycle lies roughly in the platinum square plane, and
the others above and below. In principle, these may exchange
by a “jump rope” mechanism. However, cis-4c and trans-4c do
not readily interconvert, either in solution or the solid state.
al ways. One approach is to define a least-squares plane con-
sisting of the six atoms of two parallel P-M-P axes of one set,
and the analogous plane of the other set (this avoids dealing
with the imprecise planes defined by three nearly collinear
points). Alternatively, the least-square planes defined by the
five-atom coordination planes (P-MX2-P) may be employed. As
summarized in Table 5, these values are in good agreement,
and range from 78.07 to 79.878.
In the diiodide complexes, which have larger unit cells, there
are four “non-parallel” sets of molecules with parallel P-M-P
axes. Least-squares planes may be analogously defined, but
the multiple angles generated are not tabulated. These diverse
lattice geometries provide thought-provoking conceptual rela-
tionships to various macroscopic devices that are based upon
multiple gyroscopes,[41] some of which feature non-parallel
axes.
Structural data: lattice properties
Another question is whether the MX2 moieties in the dihalide
complexes might be able to rotate in the crystal lattice without
interference from nearby molecules. Accordingly, for each com-
plex the atom from a neighboring molecule closest to the
metal was identified. These proved to be carbon atoms associ-
ated with macrocycles that intercalate between those of the
reference molecule. The distances are summarized in Table 5
(MꢀCneighbor, 4.619 to 5.878 ꢁ), and increase monotonically with
halide ligand size. The van der Waals radius of carbon was sub-
tracted to give the corresponding clearances (MꢀCneighborꢀvdW
,
2.919 to 4.178 ꢁ). These were always much less than the radii
of the respective rotors (4.053–4.592 ꢁ). Hence, rotation can be
considered blocked in the crystal lattice. The same holds for
the other complexes in Table 5.
In the other crystallographically characterized complexes,
trans-13c (space group C2/c, Z=4) and trans-15c,e (both
¯
space group P1, Z=4), all P-Pt-P axes are parallel. However,
In this context, note that the conformational reorganization
of the macrocycles necessary to sterically accommodate the
passage of the rotator should require significantly more energy
in the solid state than in solution. This might be facilitated by
a phase transition corresponding to a disordering of the meth-
ylene chains. DSC data (Experimental Section) do reveal five
complexes for which a melting endotherm is accompanied by
a second lower temperature endotherm. However, crystal
structures determined above and below the transition temper-
ature did not reveal any new disorder, and these phenomena
remain under investigation.
most square-planar complexes crystallize with “non-parallel”
sets of molecules with parallel P-M-P axes. Interestingly, trigo-
nal bipyramidal gyroscope-like complexes tend to crystallize
with parallel M-P-M axes. Also, in many cases the rotator can in
principle rotate in the solid state without van der Waals inter-
actions with
a neighboring molecule. Examples include
[Fe(CO)3(P{(CH2)14}3P)],[6] analogues with different diphosphine
ligands or osmium, and carbonyl ligand substitution pro-
ducts.[34b,42]
Towards molecular gyroscopes
As can be seen in Table 2, the dichloride complexes trans-4c
and trans-5c and dibromide complexes trans-7c and trans-8c
crystallize in identical space groups (P21/n) with similar lattice
Ideally, one would like to engineer stators that enclose their ro-
tators in frictionless environments, both in solution and the
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4630
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
crystal lattice. The preceding structural analyses reveal some of
the strengths and weaknesses of the square-planar systems re-
ported herein. The “horizontal clearance” can be optimized by
varying the sizes of the ligands on the rotator and the lengths
of the methylene chains in the stator. The replacement of
some methylene groups by more rigid units may also provide
benefits.
a means of orienting the rotators or driving unidirectional rota-
tion, as in compasses and gyroscopes.
Experimental Section
General: Except for hydrogenations, reactions were conducted
under dry inert atmospheres. Chemicals were treated as follows:
THF, benzene, toluene, hexanes, and pentane, distilled from Na/
benzophenone or Na; acetone, distilled from CaCl2; CH2Cl2 and
CDCl3, distilled from CaH2; CH2Cl2 (for workups) and CH3OH, simple
distillation; NaCꢁCH (Acros, 18 wt% slurry in xylene), used as re-
ceived or isolated by filtration, washed with toluene under argon,
and dried in vacuo (18 h); Grubbs’ catalyst ([(Cy3P)2Cl2Ru(=CHPh)];
Aldrich), [RhCl(PPh3)3] (97%, Alfa Aesar or Strem), LiBr (Fluka, 99%),
NaI (Rienst, 99%), KCN, KSCN (2ꢂFluka, 98%), Ph2Zn (Acros, 98%),
BH3·S(CH3)2 (2.0m in THF, Acros), used as received.
However, there is much less flexibility with respect to the
top/bottom or “vertical clearance”. Towards this end, we re-
cently prepared diarsine analogues of the iron diphosphine
systems III in Scheme 1. Since iron–arsenic bonds are about
4% longer than iron–phosphorus bonds, the length of the axis
is increased. However, additional progress will require spacer
groups. One possibility would be an axis composed of
a PAuCꢁCM(L2)CꢁCAuP linkage.
From the standpoint of minimizing intermolecular interac-
tions in the solid state, equatorial “belts” as found in most toy
gyroscopes may prove useful. One approach would entail
switching to alkyne metathesis for macrocyclization, the feasi-
bility of which has been established for trans-phosphine li-
gands of the type Ph2P(CH2)mCꢁCCH3.[43] The CꢁC linkages
could then be used for binding additional metals and possibly
attendant bridging groups.
NMR spectra were recorded on 500–300 MHz spectrometers and
1
referenced as follows: H: residual internal CHCl3 (d, 7.24 ppm); 13C:
internal CDCl3 (d, 77.0 ppm); 31P: external H3PO4 (d=0.00 ppm). IR
spectra were recorded using ATR accessories. DSC and TGA data
were treated by standard methods.[45] Instrument models and relat-
ed details are provided in the Supporting Information of earlier
papers.[7,30,31]
trans-[PtCl2(P{(CH2)14}3P)] (trans-4c) and trans-[PtCl2({P(CH2)14}-
(CH2)14P{(CH2)14})] (trans-4’c)
Regardless of the MX2 rotational barriers associated with the
preceding complexes, the MX2 moieties of individual molecules
would exhibit both clockwise and counterclockwise (i.e., Brow-
nian) rotation. However, in order to attain gyroscopic proper-
ties, unidirectional rotation is required. One strategy for con-
trolling the direction of a rotator involves introducing a dipole
moment, such that it can be oriented by a static applied elec-
tric field or driven by a rotating electric field.[4,44] Both substitu-
tion reactions of the types in Schemes 3 and 4, and ring-clos-
ing metatheses of less symmetrical substrates—such as rhodi-
um carbonyl chloride analogues of the preceding dichloride
complexes (see IV, Scheme 1)[9]—should provide practical
routes to such species.
Method A: A two-necked flask was charged with trans-1c[19]
(1.000 g, 1.005 mmol) and CH2Cl2 (285 mL; the resulting solution
was 0.0035m in trans-1c), and fitted with a condenser with a bub-
bler. The solution was cooled to 08C. A solution of Grubbs’ catalyst
(0.0251 g, 0.0305 mmol, 3.0 mol%) in CH2Cl2 (50 mL) was added
dropwise over 100 min.[46] The solution was refluxed (16 h), cooled,
and passed through a pad of neutral alumina (3ꢂ10 cm, rinsed
with CH2Cl2). The filtrate was concentrated to about 10 mL. A Fisch-
er–Porter bottle was charged with the filtrate and [RhCl(PPh3)3]
(0.1202 g, 0.1299 mmol), and flushed and pressurized with H2
(5 bar). The solution was stirred at 458C. After 4 d, the solvent was
removed by rotary evaporation. The residue was passed through
a pad of neutral alumina (2.5ꢂ7.0 cm, rinsed with CH2Cl2). The sol-
vent was removed from the filtrate by rotary evaporation. The resi-
due was subjected to chromatography on a silica column (3.5ꢂ
14 cm, 1:2 v/v CH2Cl2/hexanes). The solvent was removed from the
product fraction by rotary evaporation to give trans-4c as a pale
yellow waxy oil, which solidified after 15 min (0.398 g, 0.434 mmol,
43%).
Conclusion
We have demonstrated that it is possible to 1) enclose rotators
of the formula PdCl2 and PtCl2 in trans-spanning diphosphine
ligands using threefold intramolecular alkene metatheses,
2) elaborate the caged rotators through substitution reactions,
and 3) extrude the rotators from the diphosphines. All of these
steps pose fascinating mechanistic questions, which remain
under investigation. Consistent with certain design elements,
the MCl2, MBr2, MI2, and M(CN)2 complexes appear to have con-
siderably lower rotational barriers than their Fe(CO)3 and
[Fe(CO)2(NO)]+ analogues.[6] Attempts to quantify these remain
in progress. Regardless, the barriers are increased when the
sizes of the ligands are increased to isothiocyanide and phenyl,
and can be modulated by varying the lengths of the methyl-
ene bridges. Current efforts are directed at the synthesis of an-
alogues with 1) more spacious stators, and 2) unsymmetrically
substituted rotators (MXX’). The dipole moments of the latter
should be responsive to external electric fields, providing
M.p. (capillary) 116–1178C; DSC (Ti/Te/Tp/Tf): 77.8/114.7/118.0/
126.38C (endotherm); TGA: onset of mass loss, 2848C; elemental
analysis calcd (%) for C42H84Cl2P2Pt: C 55.01, H 9.23; found: C 54.81,
1
H 9.22; H NMR (CDCl3):[47] d=1H 1.95–1.80 (brm, 12H; PCH2), 1.80–
1.60 (brm, 12H; PCH2CH2), 1.50–1.38 (brm, 12H; PCH2CH2CH2),
1.38–1.20 ppm (brm, remaining CH2); 13C{1H} NMR: d=30.0 (vir-
tual t,[23] 3JCP =6.8 Hz, PCH2CH2CH2), 28.1 (s, CH2), 27.1 (s, 2CH2), 26.8
(s, CH2), 23.1 (s, PCH2CH2), 21.3 ppm (virtual t,[23] 1JCP =16.3 Hz,
PCH2); 31P{1H} NMR: d=7.1 ppm (s, 1JPPt =2398 Hz).[48] IR (powder
~
film): n=2922 (s), 2851 (s), 1461 (m), 1411 (s), 1351 (s), 1261 (s),
1106 (s), 1005 (s), 820 (s), 795 (m), 719 cmꢀ1 (m); MS:[49a] m/z (%):
917 (100) [M]+, 882 (50) [MꢀCl]+.
Method B: Five two-necked round-bottomed flasks were charged
with trans-1c (i–v: 2ꢂ1.903 g, 2ꢂ1.912 mmol; 2ꢂ2.141 g, 2ꢂ
2.151 mmol; 1.843 g, 1.852 mmol), CH2Cl2 (i–v: 2ꢂ1000 mL, 2ꢂ
1200 mL, 1000 mL) and cooled to 08C. Solutions of Grubbs’ catalyst
(i–v: 0.051 g, 0.062 mmol, 3 mol%; 0.056 g, 0.068 mmol, 4 mol%;
2ꢂ0.107 g, 2ꢂ0.130 mmol, 2ꢂ6 mol%; 0.092 g, 0.112 mmol,
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4631
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
6 mol%;) in CH2Cl2 (i–v: 250 mL, 200 mL, 3ꢂ30 mL) were added
dropwise over 90–15 min, respectively. The cooling baths were re-
moved, and the flasks were fitted with condensers with bubblers.
The mixtures were refluxed (4–8 d). Runs i, ii, and iv were charged
with a second portion of Grubbs’ catalyst (i: 0.050 g, 0.061 mmol,
3 mol%, ii: 0.051 g, 0.062 mmol, 3 mol%, and iv: 0.109 g,
0.132 mmol, 6 mol%) and refluxed (1–6 d additional). The mixtures
were cooled to room temperature, exposed to air (3–7 d), and
passed through pads of neutral alumina (4 cm) that were rinsed
with CH2Cl2. The filtrates were concentrated to about 30 mL. Five
Fischer–Porter bottles were charged with [RhCl(PPh3)3] (i: 0.088 g,
0.095 mmol, 5 mol%; ii: 0.089 g, 0.096 mmol, 5 mol%; iii: 0.120 g,
0.130 mmol, 6 mol%; iv: 0.119 g, 0.129 mmol, 6 mol%; v: 0.104 g,
0.112 mmol, 6 mol%) and the concentrates, and then flushed and
pressurized with H2 (5 bar). The mixtures were stirred at 408C (4–
6 d), and the solvents were removed by rotary evaporation. The
residues were combined and placed at the top of a silica column
(10 cm) that was eluted with hexanes/CH2Cl2 (1:0 to 0:1 v/v). The
solvents were removed from the product containing fractions by
rotary evaporation and oil pump vacuum to give trans-4c (2.895 g,
3.157 mmol, 32%; eluting first) as a yellowish solid and trans-4’c
(0.498 g, 0.543 mmol, 5%) as a yellowish viscous oil.
(m), 752 (m), 718 cmꢀ1 (m); MS:[49a] m/z (%): 1000 (55) [M]+, 964
(50) [MꢀCl]+, 928 (100) [Mꢀ2Cl]+.
Method B: Complex trans-1d (0.609 g, 0.564 mmol), CH2Cl2
(230 mL; the resulting solution was 0.0025m in trans-1d), and
Grubbs’ catalyst (0.0144 g, 0.0175 mmol, 3.1 mol%) in CH2Cl2
(11 mL) were combined as in procedure A. The solution was re-
fluxed (18 h). The mixture was cooled and passed through a pad of
neutral alumina (2.5ꢂ5 cm, rinsed with CH2Cl2). The filtrate was
concentrated to about 7 mL. A Fischer–Porter bottle was charged
with the filtrate and [RhCl(PPh3)3] (0.0754 g, 0.0815 mmol), and
flushed and pressurized with H2 (5 bar). The solution was stirred at
458C. After 4 d, the solvent was removed by rotary evaporation.
The residue was passed through a pad of neutral alumina (2.5ꢂ
5 cm, rinsed with CH2Cl2). The solvent was removed from the fil-
trate by rotary evaporation. The residue was subjected to chroma-
tography on a silica column (2ꢂ24 cm, 1:6 v/v CH2Cl2/hexanes).
The solvent was removed from the product fraction by rotary
evaporation to give trans-4’d as a pale yellow waxy oil, which sol-
idified after 24 h (0.0208 g, 0.0208 mmol, 4%).
M.p. (capillary) 59–608C; DSC (Ti/Te/Tp/Tc/Tf): 36.2/64.4/73.4/78.9/
103.48C (endotherm); 190.1/221.9/260.0/279.9/289.88C (exotherm);
TGA: onset of mass loss, 2828C; elemental analysis calcd (%) for
C48H96Cl2P2Pt: C 57.58, H 9.66; found: C 57.27, H 9.66; 1H NMR
(CDCl3): d=2.19–2.01 (brm, 4H; CH2), 1.79–1.67 (brm, 8H; CH2),
1.66–1.55 (brm, 8H; CH2), 1.46–1.36 (brm, 16H; CH2), 1.41–
Data for trans-4’c: Elemental analysis calcd (%) for C42H84P2PtCl2: C
55.01, H 9.23; found: C 55.13, H 9.45; 1H NMR (CDCl3): d=2.01–
1.95 (m, 4H; CH2), 1.88–1.56 (m, 16H; CH2), 1.48–1.14 ppm (m,
64H; CH2); 13C{1H} NMR: d= 31.3 (virtual t,
JCP =7.3 Hz,
1.22 ppm (brm, 60H; CH2); 13C{1H} NMR: d=31.6 (virtual t,[23] JCP
7.2 Hz, PCH2CH2CH2), 29.9 (virtual t,[23]
CP =6.3 Hz, 2PCH2CH2CH2),
=
PCH2CH2CH2), 29.0 (virtual t, JCP =6.2 Hz, 2PCH2CH2CH2), 28.1 (s,
CH2), 27.2 (s, CH2), 26.87 (s, CH2), 26.84 (s, 2CH2), 26.77 (s, 2CH2),
26.65 (s, 2CH2), 26.2 (s, 2CH2), 25.4 (s, CH2), 24.9 (s, CH2), 23.1 (vir-
J
29.7 (s, CH2), 28.9 (s, CH2), 28.2 (s, CH2), 27.83 (s, 2CH2), 27.76 (s,
CH2), 27.6 (s, 2CH2), 27.5 (s, 2CH2), 27.4 (s, 2CH2), 27.0 (s, 2CH2), 26.8
(s, CH2), 26.3 (s, CH2), 25.0 (s, CH2), 22.8 (center of obscured vir-
tual t, JCP =16.3 Hz, PCH2), 21.7 (s, PCH2CH2), 18.9 ppm (virtual t,
1
CP =15.9 Hz, 2PCH2); 31P{1H} NMR: d=5.6 ppm (s, JPPt
=
tual t,[23] 1JCP =16.1 Hz, PCH2), 22.6 (s, PCH2CH2), 19.2 ppm (vir-
J
2374 Hz);[48] MS:[49b] m/z (%): 881.5 (2) [MꢀCl]+, 651.6 (3)
1
tual t,[23]
J
CP =16.0 Hz, 2PCH2); 31P{1H} NMR: d=5.5 ppm (s, JPPt
=
[MꢀPtCl2]+, 326.3 (100) [MꢀPtCl2 +H]2+
.
2374 Hz);[48] IR (powder film): n=2921 (s), 2852 (s), 1459 (m), 1414
(w), 1351 (w), 1202 (w), 1050 (m), 795 (m), 725 (m), 718 cmꢀ1 (m);
MS:[49a] m/z (%): 1000 (45) [M]+, 965 (50) [MꢀCl]+.
~
trans-[PtCl2(P{(CH2)16}3P)] (trans-4d) and trans-[PtCl2({P(CH2)16}-
(CH2)16P{(CH2)16})] (trans-4’d)
Method A: A two-necked flask was charged with trans–1d[19]
(0.522 g, 0.484 mmol) and CH2Cl2 (250 mL; the resulting solution
was 0.0019m in trans-1d), and fitted with a condenser with a bub-
bler. The solution was cooled to 08C. A solution of Grubbs’ catalyst
(0.0141 g, 0.0171 mmol, 3.5 mol%) in CH2Cl2 (15 mL) was added
dropwise over 1 h.[44] The solution was refluxed (17 h). The mixture
was cooled and passed through a pad of neutral alumina (1.5ꢂ
7.0 cm, rinsed with CH2Cl2). The filtrate was concentrated to about
5 mL. A Fischer–Porter bottle was charged with the filtrate and
[RhCl(PPh3)3] (0.0652 g, 0.0705 mmol), and flushed and pressurized
with H2 (5 bar). The solution was stirred at 508C. After 4 d, the sol-
vent was removed by rotary evaporation. The residue was passed
through a pad of neutral alumina (1.5ꢂ15 cm, rinsed with CH2Cl2).
The solvent was removed from the filtrate by rotary evaporation.
The residue was subjected to chromatography on a silica column
(3ꢂ24 cm, 1:6 v/v CH2Cl2/hexanes). The solvent was removed from
the product fraction by rotary evaporation to give trans-4d as
a pale yellow waxy oil, which solidified after 24 h (0.0172 g,
0.0172 mmol, 4%).
trans-[PtCl2(P{(CH2)18}3P)] (trans-4e) and trans-[PtCl2({P(CH2)18}-
(CH2)18P{(CH2)18})] (trans-4’e): The complex trans-1e[19] (0.750 g,
0.645 mmol), CH2Cl2 (650 mL; the resulting solution was 0.0010m
in trans-1e), and
a solution of Grubbs’ catalyst (0.0181 g,
0.0220 mmol, 3.0 mol%) in CH2Cl2 (25 mL) were combined as in
method A for trans-4d. The solution was refluxed (20 h). The mix-
ture was cooled and passed through a pad of neutral alumina
(2.5ꢂ5 cm, rinsed with CH2Cl2). The filtrate was concentrated to
about 5 mL. A Fischer–Porter bottle was charged with the filtrate
and [RhCl(PPh3)3] (0.0904 g, 0.0977 mmol), and flushed and pressur-
ized with H2 (5 bar). The solution was stirred at 508C. After 4 d, the
solvent was removed by rotary evaporation. The residue was
passed through a pad of neutral alumina (2ꢂ5 cm, rinsed with
CH2Cl2). The solvent was removed from the filtrate by rotary evapo-
ration. The residue was subjected to chromatography on a silica
column (3ꢂ22 cm, 1:8 v/v CH2Cl2/hexanes). The solvent was re-
moved from the product fractions by rotary evaporation to give
trans-4e as a pale yellow waxy oil, which solidified after 24 h
(0.1991 g, 0.1834 mmol, 28%, Rf =0.68), and trans-4’e as a pale
yellow waxy oil, which solidified after 72 h (0.1612 g, 0.1485 mmol,
23%, Rf =0.57).
Elemental analysis calcd (%) for C48H96Cl2P2Pt: C 57.58, H 9.66;
found: C 58.20, H 10.05;[50] 1H NMR (CDCl3):[51] d=1.90–1.70 (brm,
12H; PCH2), 1.70–1.52 (brm, 12H; PCH2CH2), 1.48–1.36 (brm, 12H;
PCH2CH2CH2), 1.36–1.20 ppm (brm, remaining CH2); 13C{1H} NMR:
d=31.4 (virtual t,[23] 3JCP =6.7 Hz, PCH2CH2CH2), 28.2 (s, CH2), 27.9 (s,
CH2), 27.8 (s, CH2), 27.7 (s, CH2), 27.4 (s, CH2), 23.2 (s, PCH2CH2),
20.8 ppm (virtual t,[23] 1JCP =16.2 Hz, PCH2); 31P{1H} NMR: d=
Data for trans-4e: M.p. (capillary) 48–508C; DSC (Ti/Te/Tp/Tc/Tf): 35.6/
49.1/52.3/55.4/68.88C (endotherm); TGA: onset of mass loss,
3308C; elemental analysis calcd (%) for C54H108Cl2P2Pt: C 59.76, H
10.03; found: C 59.54, H 9.74; 1H NMR (CDCl3):[51] d=1.83–1.80
(brm, 12H; PCH2), 1.60–1.57 (brm, 12H; PCH2CH2), 1.42–1.38 (brm,
12H; PCH2CH2CH2), 1.38–1.27 ppm (brm, remaining CH2); 13C{1H}
NMR: d=30.7 (virtual t,[23] 3JCP =6.6 Hz, PCH2CH2CH2), 28.5 (s, CH2),
6.0 ppm (s, 1JPPt =2398 Hz);[48] IR (powder film): n=2922 (s), 2853
~
(s), 1459 (m), 1413 (m), 1351 (w), 1100 (w), 1054 (w), 1019 (w), 791
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4632
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
28.3 (s, CH2), 28.2 (s, 2CH2), 27.9 (s, CH2), 27.6 (s, CH2), 23.4 (s,
PCH2CH2), 20.7 ppm (virtual t,[23] 1JCP =16.2 Hz, PCH2); 31P{1H} NMR:
lyst (0.0202 g, 0.0245 mmol, 2.9 mol%) in CH2Cl2 (45 mL) was
added dropwise over 50 min.[46] The solution was refluxed (16 h;
31P{1H} NMR (aliquot): d=13.5 (s), 13.1 (s), 12.3 (s), 12.1 (s), 12.1–
11.9 ppm (broad oligomer signals)). The mixture was passed
through a pad of neutral alumina (3ꢂ5 cm, rinsed with CH2Cl2).
The filtrate was concentrated to about 5 mL. A Fischer–Porter
bottle was charged with the filtrate and [RhCl(PPh3)3] (0.1101 g,
0.1190 mmol), and flushed and pressurized with H2 (5 bar). The so-
lution was stirred at 408C. After 4 d, the solvent was removed by
rotary evaporation. The residue was passed through a pad of neu-
tral alumina (2ꢂ5 cm, rinsed with 1:1 v/v CH2Cl2/hexanes). The sol-
vent was removed from the filtrate by rotary evaporation. The resi-
due was subjected to chromatography on a silica column (5.5ꢂ
25 cm, 1:6 v/v CH2Cl2/hexanes). The solvent was removed from the
product fractions by rotary evaporation to give trans-5e as
a yellow waxy oil, which solidified after 4 d (0.1223 g, 0.1227 mmol,
15%, Rf =0.43), and trans-5’e as a yellow waxy oil, which solidified
after 16 d (0.0225 g, 0.0226 mmol, 3%, Rf =0.36).
6.5 ppm (s, 1JPPt =2389 Hz);[48] IR (powder film): n=2922 (s), 2853
~
(s), 1459 (m), 1417 (w), 1351 (w), 1124 (w), 1083 (w), 1061 (w), 791
(m), 718 cmꢀ1 (m); MS:[49a] m/z (%): 1084 (40) [M]+, 1048 (45)
[MꢀCl]+.
Data for trans-4’e: DSC (Ti/Te/Tp/Tc/Tf): 36.5/60.8/68.1/73.7/93.58C
(endotherm); 214.2/230.4/247.3/262.3/294.48C (exotherm); TGA:
onset of mass loss, 2918C; elemental analysis calcd (%) for
1
C54H108Cl2P2Pt: C 59.76, H 10.03; found: C 59.30, H 10.04; H NMR
(CDCl3): d=2.05–1.85 (brm, 4H; CH2), 1.85–1.70 (brm, 4H; CH2),
1.70–1.52 (brm, 10H; CH2), 1.48–1.30 (brm, 20H; CH2), 1.35–
1.15 ppm (brm, 70H; CH2); 13C{1H} NMR: d=31.4 (virtual t,[23] JCP
=
6.8 Hz, PCH2CH2CH2), 30.6 (virtual t,[23]
JCP =6.2 Hz, 2PCH2CH2CH2),
29.2 (s, CH2), 28.9 (s, CH2), 28.4 (s, 2CH2), 28.3 (s, 2CH2), 28.2 (s, CH2),
28.1 (s, 2CH2), 28.0 (s, 2CH2), 27.6 (s, 2CH2), 27.5 (s, CH2), 27.4 (s,
CH2), 27.3 (s, 2CH2), 26.6 (s, CH2), 24.7 (s, CH2), 22.8 (s, PCH2CH2),
22.6 (partially obscured virtual t,[23]
JCP =16.5 Hz, PCH2), 19.3 ppm
(virtual t,[23]
J
CP =16.0 Hz, 2PCH2); 31P{1H} NMR: d=5.4 ppm (s,
Data for trans-5e: M.p. (capillary) 56–578C; DSC (Ti/Te/Tp/Tf): 34.4/
53.2/58.8/70.28C (endotherm); TGA: onset of mass loss, 2908C; ele-
mental analysis calcd (%) for C54H108Cl2P2Pd: C 65.07, H 10.92;
1JPPt =2374 Hz);[48] IR (powder film): n=2922 (s), 2853 (s), 1459 (m),
1413 (w), 1124 (w), 1058 (w), 1023 (w), 795 (m), 718 cmꢀ1 (m);
MS:[49a] m/z (%): 1084 (40) [M]+, 1047 (45) [MꢀCl]+.
~
1
found: C 65.00, H 11.03; H NMR (d (CDCl3):[51] d=1.82–1.75 (brm,
trans-[PdCl2(P{(CH2)14}3P)] (trans-5c): The complex trans-2c[19]
(0.450 g, 0.496 mmol), CH2Cl2 (140 mL; the resulting solution was
0.0035m in trans-2c), and a solution of Grubbs’ catalyst (0.0301 g,
0.0366 mmol, 7.4 mol%) in CH2Cl2 (40 mL) were combined as in
method A for trans-4d. The solution was refluxed (12 h) and
cooled to 08C. A second charge of Grubbs’ catalyst (0.0251 g,
0.0305 mmol, 6.1 mol%) in CH2Cl2 (35 mL) was added dropwise
over 1 h.[46] The solution was refluxed (15 h; 31P{1H} NMR (aliquot):
d=16.4 (s), 15.0 (s), 13.6 (s) 13.0–11.9 ppm (broad oligomer sig-
nals)). The mixture was passed through a pad of neutral alumina
(3ꢂ10 cm, rinsed with CH2Cl2). The filtrate was concentrated to
about 5 mL. A Fischer–Porter bottle was charged with the filtrate
and [RhCl(PPh3)3] (0.0650 g, 0.0703 mmol), and flushed and pressur-
ized with H2 (5 bar). The solution was stirred at 558C. After 4 d, the
solvent was removed by rotary evaporation. The residue was
passed through two pads of neutral alumina (2ꢂ10 cm and 2ꢂ
5 cm, rinsed with 1:1 v/v CH2Cl2/hexanes). The solvent was re-
moved from the filtrate by rotary evaporation. The residue was
subjected to chromatography on a neutral alumina column (3.5ꢂ
26 cm, 1:9 v/v CH2Cl2/hexanes). The solvent was removed from the
product fraction by rotary evaporation to give trans-5c as a yellow
waxy oil, which solidified after 18 h (0.150 g, 0.181 mmol, 36%).
12H; PCH2), 1.63–1.52 (brm, 12H; PCH2CH2), 1.46–1.38 (brm, 12H;
PCH2CH2CH2), 1.38–1.23 ppm (brm, remaining CH2); 13C{1H} NMR:
d=30.8 (virtual t,[23] 3JCP =6.6 Hz, PCH2CH2CH2), 28.5 (s, CH2), 28.4 (s,
CH2), 28.3 (s, CH2), 28.2 (s, CH2), 27.7 (s, CH2), 27.1 (s, CH2), 23.9 (s,
PCH2CH2), 21.7 ppm (virtual t,[23] 1JCP =13.1 Hz, PCH2); 31P{1H} NMR:
~
d=10.1 ppm (s); IR (powder film): n=2921 (s), 2851 (s), 1461 (m),
1442 (m), 1350 (m), 1262 (m), 1098 (m), 1057 (s), 1039 (m), 1022
(m), 775 (m), 718 cmꢀ1 (m); MS:[49a] m/z (%): 997 (70) [M]+, 960 (80)
[MꢀCl]+, 925 (40) [Mꢀ2Cl]+.
Data for trans-5’e: DSC (Ti/Te/Tp/Tf): 38.2/55.9/67.0/75.48C (endo-
therm); TGA: onset of mass loss, 2938C; elemental analysis calcd
1
(%) for C54H108Cl2P2Pd: C 65.07, H 10.92; found: C 65.14, H 10.96; H
NMR (CDCl3): d=2.03–1.91 (brm, 4H; CH2), 1.78–1.68 (brm, 4H;
CH2), 1.68–1.52 (brm, 10H; CH2), 1.48–1.34 (brm, 20H; CH2), 1.34–
1.18 ppm (brm, 70H; CH2); 13C{1H} NMR: d=31.5 (virtual t,[23] JCP
=
7.0 Hz, PCH2CH2CH2), 30.6 (virtual t,[23]
JCP =6.1 Hz, 2PCH2CH2CH2),
29.2 (s, CH2), 28.9 (s, CH2), 28.4 (s, 2CH2), 28.3 (s, 2CH2), 28.2 (s, CH2),
28.1 (s, 2CH2), 28.0 (s, 2CH2), 27.6 (s, 2CH2), 27.45 (s, CH2), 27.42 (s,
CH2), 27.3 (s, 2CH2), 26.6 (s, CH2), 25.1 (s, PCH2CH2), 23.2 (s,
2PCH2CH2), 20.5 ppm (virtual t,[23]
J
CP =12.8 Hz, 2PCH2); 31P{1H}
~
NMR:d=11.0 ppm (s); IR (powder film): n=2921 (s), 2852 (s), 1461
(m), 1412 (w), 1059 (w), 795 (m), 749 (m), 713 cmꢀ1 (m); MS:[49a] m/z
(%): 997 (70) [M]+, 960 (80) [MꢀCl]+, 925 (40) [Mꢀ2Cl]+.
DSC (Ti/Te/Tp/Tf): 73.9/99.8/111.1/117.88C (endotherm); TGA: onset
of mass loss, 3038C; elemental analysis calcd (%) for C42H84Cl2P2Pd:
trans-[PdBr2(P{(CH2)14}3P)] (trans-8c)
1
C 60.90, H 10.22; found: C 60.90, H 10.33; H NMR (CDCl3):[51] d=
Method A: The complex trans-3c[19] (0.162 g, 0.163 mmol), CH2Cl2
(65 mL; the resulting solution was 0.0025m in trans-3c), and a solu-
tion of Grubbs’ catalyst (0.0140 g, 0.0170 mmol, 10 mol%) in
CH2Cl2 (5 mL) were combined as in method A for trans-4d (95 min
addition time). The solution was refluxed (15 h) and cooled to 08C.
A second charge of Grubbs’ catalyst (0.0141 g, 0. 0171 mmol,
10 mol%) in CH2Cl2 (5 mL) was added dropwise over 30 min.[46] The
solution was refluxed (24 h; 31P{1H} NMR (aliquot): d=15.51 (s),
14.27 (s), 12.82 (s), 12.40–9.50 ppm (broad oligomer signals)). The
mixture was cooled and passed through a pad of neutral alumina
(1.5ꢂ10 cm, rinsed with CH2Cl2). The filtrate was concentrated to
about 3 mL. A Fischer–Porter bottle was charged with the filtrate
and [RhCl(PPh3)3] (0.030 g, 0.032 mmol), and flushed and pressur-
ized with H2 (5 bar). The solution was stirred at 558C. After 92 h,
the solvent was removed by rotary evaporation. The residue was
passed through a pad of neutral alumina (2ꢂ7 cm, rinsed with 1:1
v/v CH2Cl2/hexanes). The solvent was removed from the filtrate by
1.76–1.73 (brm, 12H; PCH2), 1.73–1.56 (brm, 12H; PCH2CH2), 1.52–
1.41 (apparent t, 12H; PCH2CH2CH2), 1.41–1.14 ppm (brm, remain-
ing CH2); 13C{1H} NMR: d=30.2 (virtual t,[23] 3JCP =6.7 Hz,
PCH2CH2CH2), 28.1 (s, CH2), 27.2 (s, CH2), 27.1 (s, CH2), 26.9 (s, CH2),
23.5 (s, PCH2CH2), 22.4 ppm (virtual t,[23] 1JCP =12.2 Hz, PCH2); 31P{1H}
~
NMR: d=13.3 ppm (s); IR (powder film): n=2921 (s), 2850 (s), 1459
(m), 1409 (s), 1262 (m), 1094 (m), 1052 (m), 1015 (w), 791 (m),
714 cmꢀ1 (m); MS:[49a] m/z (%): 827 (100) [M]+, 791 (90) [MꢀCl]+,
755 (85) [Mꢀ2Cl]+.
trans-[PdCl2(P{(CH2)18}3P)] (trans-5e) and trans-[PdCl2({P(CH2)18}-
(CH2)18P{(CH2)18})] (trans-5’e): The complex trans-2e[19] (0.900 g,
0.837 mmol), CH2Cl2 (590 mL; the resulting solution was 0.0014m
in trans-2e), and
a solution of Grubbs’ catalyst (0.0201 g,
0.0244 mmol, 2.9 mol%) in CH2Cl2 (45 mL) were combined as in
method A for trans-4d (95 min addition time). The solution was re-
fluxed (20 h) and cooled to 08C. A second charge of Grubbs’ cata-
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4633
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
rotary evaporation. The residue was subjected to chromatography
on a neutral alumina column (1.5ꢂ15 cm, 1:1 v/v CH2Cl2/hexanes).
The solvent was removed from the product fraction by rotary
evaporation to give a yellow oil (mixture of trans-5c, trans-6c
(Scheme 3), and trans-8c) that solidified after several days
(0.0290 g, 0.0316 mmol,[52] 19%). 31P{1H} NMR (CDCl3): d=13.3 (s,
10%, 5c), 12.2 (s, 50%, 6c), 10.6 ppm (s, 40%, 8c); MS:[49a] m/z (%):
916 (5) [M]+, 872 (55) [PdBrCl(P{(CH2)14}3P)]+, 835 (60) [PdBr(P-
{(CH2)14}3P)]+, 791 (100) [PdCl(P{(CH2)14}3P)]+, 755 (100) [Pd(P-
{(CH2)14}3P)]+.
1.62–1.59 (brm, 12H; PCH2CH2), 1.44–1.39 (brm, 12H;
PCH2CH2CH2), 1.38–1.29 ppm (brm, remaining CH2); 13C{1H} NMR:
29.9 (virtual t,[23] 3JCP =6.8 Hz, PCH2CH2CH2), 28.1 (s, CH2), 27.2 (s,
CH2), 27.1 (s, CH2), 26.8 (s, CH2), 23.2 (s, PCH2CH2), 22.7 ppm (vir-
1
tual t,[23] 1JCP =16.7 Hz, PCH2); 31P{1H} NMR: 2.8 ppm (s, JPPt
=
2344 Hz);[48] IR (powder film): n=2921 (s), 2852 (s), 1455 (m), 1412
(m), 1403 (m), 1351 (m), 1262 (m), 1092 (m), 1065 (m), 1052 (m),
1015 (m), 793 (m), 716 cmꢀ1 (m); MS:[49a] m/z (%): 1006 (10) [M]+,
926 (55) [MꢀBr]+, 845 (100) [Mꢀ2Br]+.
~
trans-[PtI2(P{(CH2)14}3P)] (trans-9c): Complex trans-4c (0.0159 g,
0.0173 mmol), NaI (0.0104 g, 0.0694 mmol) and THF/acetone
(0.6 mL, 1:1 v/v) were combined as in the procedure for trans-7c.
After 24 h, an identical workup gave trans-9c as an orange powder
(0.0110 g, 0.0100 mmol, 58%).
Method B: A 5 mm NMR tube was charged with trans-5c (0.0112 g,
0.0135 mmol) and THF (0.6 mL). Another NMR tube was charged
with LiBr (0.0027 g, 0.031 mmol). The solution in the first tube was
cannulated in stages into the second. After 30 min, a 31P{1H} NMR
spectrum showed >99% conversion. The solvent was removed by
oil pump vacuum. The residue was extracted with benzene. The
solvent was removed from the extract by freeze pump drying to
give trans-8c as a yellow powder (0.0123 g, 0.0134 mmol, 99%).
M.p. (capillary) 112–1148C; DSC (Ti/Te/Tp/Tf): 77.2/106.0/114.0/
123.18C (endotherm); TGA: onset of mass loss, 2808C; elemental
analysis calcd (%) for C42H84Br2P2Pd: C 54.99, H 9.23; found: C
54.17, H 8.99;[50] 1H NMR (CDCl3):[51] d=1.91–1.88 (brm, 12H; PCH2),
1.75–1.65 (brm, 12H; PCH2CH2), 1.65–1.45 (brm, 12H;
PCH2CH2CH2), 1.37–1.11 ppm (brm, remaining CH2); 13C{1H} NMR:
d=30.1 (virtual t,[23] 3JCP =6.8 Hz, PCH2CH2CH2), 28.1 (s, CH2), 27.2 (s,
CH2), 27.1 (s, CH2), 26.8 (s, CH2), 24.1 (virtual t,[23] 1JCP =13.7 Hz,
M.p. (capillary) 122–1238C; DSC (Ti/Te/Tp/Tc/Tf): 98.7/117.5/120.2/
121.6/121.68C (endotherm), 124.7/125.5/128.6/130.4/133.68C (en-
dotherm); TGA: onset of mass loss, 2868C; elemental analysis calcd
(%) for C42H84I2P2Pt: C 45.86, H 7.70; found: C 45.76, H 7.63; 1H
NMR (CDCl3):[51] d=2.21–2.17 (brm, 12H; PCH2), 1.54–1.41 (brm,
12H; PCH2CH2), 1.43–1.38 (brm, 12H; PCH2CH2CH2), 1.38–1.13 ppm
3
(brm, remaining CH2); 13C{1H} NMR: d=29.8 (virtual t,[23] JCP
=
6.5 Hz, PCH2CH2CH2), 28.1 (s, CH2), 27.2 (s, CH2), 27.1 (s, CH2), 26.7 (s,
CH2), 26.3 (virtual t,[23] 1JCP =17.5 Hz, PCH2), 23.5 ppm (s, PCH2CH2);
31P{1H} NMR: d=ꢀ3.6 ppm (s, 1JPPt =2283 Hz);[48] IR (powder film):
~
n=2921 (s), 2852 (s), 1461 (m), 1441 (m), 1204 (m), 1158 (m), 1015
(m), 965 (m), 778 (m), 722 (m), 677 cmꢀ1 (m); MS:[49a] m/z (%): 972
31
1
~
PCH2), 23.7 (s, CH2); P{ H} 10.6 ppm (s); IR (powder film): n=2921
(s), 2852 (s), 1636 (m), 1457 (m), 1410 (w), 1262 (m), 1104 (m), 1015
(m), 800 (m), 715 cmꢀ1 (m); MS:[49a] m/z (%): 916 (40) [M]+, 837
(100) [MꢀBr]+, 755 (85) [Mꢀ2Br]+.
(100) [MꢀI]+.
trans-[PdBr2(P{(CH2)18}3P)] (trans-8e): Complex trans-5c (0.0173 g,
0.0174 mmol). THF (0.6 mL) and LiBr (0.0046 g, 0.053 mmol) were
combined as in procedure B for trans-8c. A similar workup (extract
filtered through a glass fiber filter) gave trans-8e as a yellow
powder (0.0132 g, 0.0122 mmol, 70%).
trans-[PdI2(P{(CH2)14}3P)] (trans-10c): A 5 mm NMR tube was se-
quentially charged with trans-5c (0.0164 g, 0.0198 mmol), NaI
(0.0119 g, 0.0793 mmol) and THF/acetone (0.7 mL; 1:1 v/v). After
10 min, a 31P{1H} NMR spectrum showed >99% conversion. The
solvent was removed by oil pump vacuum. The residue was ex-
tracted with benzene. The solvent was removed from the extract
by freeze pump drying to give trans-10c as an orange powder
(0.0184 g, 0.0182 mmol, 92%).
DSC (Ti/Te/Tp/Tf): 37.1/39.9/51.1/56.38C (endotherm); TGA: onset of
mass loss, 2338C; elemental analysis calcd (%) for C54H108Br2P2Pd: C
59.74, H 10.03; found: C 59.19, H 9.29;[50] 1H NMR (CDCl3):[51] d=
1.99–1.90 (brm, 12H; PCH2), 1.60–1.48 (brm, 12H; PCH2CH2), 1.48–
1.35 (brm, 12H; PCH2CH2CH2), 1.35–1.15 ppm (brm, remaining
CH2); 13C{1H} NMR: d=30.7 (virtual t,[23] 3JCP =6.5 Hz, PCH2CH2CH2),
28.5 (s, CH2), 28.3 (s, CH2), 28.2 (s, 2CH2), 27.9 (s, CH2), 27.6 (s, CH2),
24.0 (s, PCH2CH2), 23.5 ppm (virtual t,[23] 1JCP =13.6 Hz, PCH2); 31P{1H}
M.p. (capillary) 124–1258C; DSC (Ti/Te/Tp/Tf): 79.1/105.4/115.3/
116.38C (endotherm), 116.2/119.8/127.9/131.58C (endotherm); TGA:
onset of mass loss, 2268C; elemental analysis calcd (%) for
C42H84I2P2Pd: C 49.88, H 8.37; found: C 50.23, H 8.55; 1H NMR
(CDCl3):[51] d=2.24–2.12 (brm, 12H; PCH2), 1.60–1.48 (brm, 12H;
PCH2CH2), 1.48–1.38 (brm, 12H; PCH2CH2CH2), 1.38–1.20 ppm (brm,
remaining CH2); 13C{1H} NMR: d=30.0 (virtual t,[23] 3JCP =6.8 Hz,
PCH2CH2CH2), 28.2 (partially obscured virtual t,[23] 1JCP =13.7 Hz,
PCH2), 28.1 (s, CH2), 27.2 (s, CH2), 27.1 (s, CH2), 26.8 (s, CH2),
24.1 ppm (s, PCH2); 31P{1H} NMR: d=4.4 ppm (s); IR (powder film):
~
NMR: d=7.9 ppm (s); IR (powder film): n=2921 (s), 2851 (s), 1460
(m), 1416 (m), 1350 (m), 1262 (m), 1097 (m), 1058 (s), 1036 (m), 795
(m), 719 cmꢀ1 (m).
trans-[PdI2(P{(CH2)18}3P)] (trans-10e): Complex trans-5c (0.0159 g,
0.0160 mmol), NaI (0.0096 g, 0.064 mmol) and THF/acetone
(0.6 mL; 1:1 v/v) were combined as in the procedure for trans-7c.
After 10 min, an identical workup gave trans-10e as an orange
powder (0.0154 g, 0.0131 mmol, 82%).
~
n=2921 (s), 2852 (s), 1461 (m), 1441 (m), 1414 (m), 1351 (m), 1158
(m), 1088 (s), 1013 (m), 775 (m), 721 (m), 710 cmꢀ1 (m); MS:[49a] m/z
(%): 883 (100) [MꢀI]+, 756 (40) [Mꢀ2I]+.
M.p. (capillary) 26–308C; DSC (Ti/Te/Tp/Tc/Tf): 29.8/29.8/39.3/44.9/
48.68C (endotherm); TGA: onset of mass loss, 1848C; elemental
analysis calcd (%) for C54H108I2P2Pd: C 53.52, H 8.53; found: C 54.64,
H 9.24;[50] 1H NMR (CDCl3):[51] d=2.30–2.20 (brm, 12H; PCH2), 1.56–
1.48 (brm, 12H; PCH2CH2), 1.48–1.36 (brm, 12H; PCH2CH2CH2),
1.36–1.25 ppm (brm, remaining CH2); 13C{1H} NMR: d=31.0 (vir-
trans-[PtBr2(P{(CH2)14}3P)] (trans-7c): A 5 mm NMR tube was se-
quentially charged with trans-4c (0.0181 g, 0.0197 mmol), LiBr
(0.0068 g, 0.079 mmol), and THF (0.6 mL). After 27 h, a 31P{1H} NMR
spectrum showed >99% conversion. The solvent was removed by
oil pump vacuum, and benzene was added. The suspension was
filtered through a glass fiber filter. The solvent was removed from
the filtrate by freeze pump drying to give trans-7c as a yellow
powder (0.0148 g, 0.0147 mmol, 75%).
tual t,[23] 3JCP =6.6 Hz, PCH2CH2CH2), 28.9 (s, CH2), 28.7 (s, CH2), 28.6
1
(s, 2CH2), 28.35 (virtual t,[23] partially obscured by other signal, JCP
=
17.1 Hz, PCH2), 28.29 (s, CH2), 28.0 (s, CH2), 24.8 ppm (s, PCH2CH2);
31P{1H} NMR: d=1.5 ppm (s); MS:[49a] m/z (%): 1051 (80) [MꢀI]+.
M.p. (capillary) 119–1218C; DSC (Ti/Te/Tp/Tc/Tf): 88.1/121.4/124.0/
125.9/140.48C (endotherm); TGA: onset of mass loss, 2838C; ele-
mental analysis calcd (%) for C42H84Br2P2Pt: C 50.15, H 8.42; found:
trans-[Pt(CN)2(P{(CH2)14}3P)] (trans-11 c): A 5 mL vial was sequen-
tially charged with trans-4c (0.0231 g, 0.0252 mmol), KCN
(0.0033 g, 0.051 mmol), and CH2Cl2/CH3OH (2.0 mL, 1:2 v/v). After
16 h, a 31P{1H} NMR spectrum showed>99% conversion. The sol-
1
C 50.20, H 8.63; H NMR (CDCl3):[51] d=1.94–1.89 (brm, 12H; PCH2),
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4634
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
vent was removed by oil pump vacuum, and benzene was added.
The suspension was filtered through a glass fiber filter. The solvent
was removed from the filtrate by freeze pump drying to give
trans-11 c as a white powder (0.0205 g, 0.0228 mmol, 90%). M.p.
(capillary) 171–1728C; DSC (Ti/Te/Tp/Tc/Tf): 111.1/125.3/128.1/133.4/
133.98C (endotherm), 151.0/153.4/159.1/163.8/164.38C (endo-
therm), 164.4/170.0/174.3/176.5/180.18C (endotherm); TGA: onset
of mass loss, 2448C; elemental analysis calcd (%) for C44H84N2P2Pt:
C 58.84, H 9.43, N 3.12; found: C 58.61, H 9.53, N 3.02; 1H NMR
(CDCl3):[51] d=2.30–2.03 (brm, 12H; PCH2), 1.62–1.58 (brm, 12H;
PCH2CH2), 1.52–1.46 (brm, 12H; PCH2CH2CH2), 1.42–1.22 ppm (brm,
CH2Cl2 (2:1 v/v) and subjected to chromatography on a silica
column (2ꢂ30 cm, 1:1 v/v CH2Cl2/hexanes). The solvents were re-
moved from the product fractions by rotary evaporation to give
trans-14c (0.0140 g, 0.0145 mmol, 20%; Rf =0.57) and trans-13c
(0.0190 g, 0.0197 mmol, 28%; Rf =0.40) as white crystalline solids.
Data for trans-13c: M.p. (capillary) 210–2128C; DSC (Ti/Te/Tp/Tc/Tf):
179.5/189.6/199.0/205.6/202.68C (endotherm), 205.3/210.1/213.8/
215.6/219.78C (endotherm); TGA: onset of mass loss, 2708C; ele-
mental analysis calcd (%) for C44H84N2P2PtS2: C 54.92, H 8.80, N
2.91, S 6.66; found: C 55.04, H 8.84, N 2.68, S 6.21; 1H NMR
(CDCl3):[51] d=1.95–1.86 (brm, 12H; PCH2), 1.68–1.46 (brm, 24H;
PCH2CH2), 1.43–1.25 ppm (brm, remaining CH2); 13C{1H}[29] NMR: d=
2
remaining CH2); 13C{1H} NMR: 124.3 (t, JCP =17.2 Hz, CN), 29.8 (vir-
tual t,[23] 3JCP =7.0 Hz, PCH2CH2CH2), 27.9 (s, CH2), 27.2 (s, CH2), 27.0
(s, CH2), 26.7 (s, CH2), 24.9 (virtual t,[23] 1JCP =17.2 Hz, PCH2),
30.2 (virtual t,[23] CP =6.8 Hz, 2PCH2CH2CH2), 29.4 (virtual t,[23] JCP
J =
6.5 Hz, PCH2CH2CH2), 28.4 (s, CH2), 27.9 (s, CH2), 27.8 (s, CH2), 27.3 (s,
1
23.7 ppm (s, PCH2CH2); 31P{1H} NMR: d=9.2 ppm (s, JPPt
=
CH2), 26.8 (s, CH2), 26.7 (s, 2CH2), 26.6 (s, CH2), 23.7 (s, CH2), 22.4 (s,
2166 Hz);[48] IR (powder film): n=2919 (s), 2850 (s), 2124 (m, nCN),
1465 (m), 1447 (m), 1414 (m), 1262 (w), 1106 (m), 1004 (m), 801
(m), 721 (m), 708 (m); MS:[49a] m/z (%): 898 (70) [M]+, 683 (80) [OP-
1
CH2), 20.0 ppm (s, CH2); 31P{1H} NMR: d=11.3 ppm (s, JPPt
=
~
2258 Hz);[48] IR (powder film): n=2925 (s), 2852 (s), 2111 (s, nNCS),
1461 (m), 1441 (m), 1414 (m), 1260 (w), 1094 (m), 1042 (m), 1015
(w), 853 (m), 797 (m), 722 (m); MS:[49a] m/z (%): 962 (100) [M]+, 904
(35) [MꢀSCN]+.
~
{(CH2)14}3PO]+,
[P{(CH2)14}3P]+.
667
(100)
[P{(CH2)14}3PO]+,
651
(20)
trans-[Pd(CN)2(P{(CH2)14}3P)] (trans-12c). Complex trans-5c
(0.0605 g, 0.0730 mmol), KCN (0.0190 g, 0.292 mmol), and CH2Cl2/
CH3OH (3.0 mL, 1:2 v/v) were combined as in the procedure for
trans-11 c. After 10 min, an identical workup gave trans-12c as
a white powder (0.0509 g, 0.0629 mmol, 86%).
Data for trans-14c: M.p. (capillary) 211–2138C; DSC (Ti/Te/Tp/Tc/Tf):
182.8/210.6/217.6/222.2/229.08C (endotherm); TGA: onset of mass
loss, 2838C; elemental analysis calcd (%) for C44H84N2P2PtS2: C
54.92, H 8.80, N 2.91, S 6.66; found: C 54.95, H 8.59, N 2.75, S 6.20;
1H NMR (CDCl3): d=2.17–1.97 (brm, 6H; CH2), 1.86–1.69 (brm,
14H; CH2), 1.69–1.60 (brm, 2H; CH2), 1.59–1.47 (brm, 20H; CH2),
1.47–1.39 (brm, 8H; CH2), 1.39–1.17 ppm (brm, 34H; CH2);
M.p. (capillary) 183–1878C; DSC (Ti/Te/Tp/Tc/Tf): 39.1/56.1/64.0/68.3/
77.08C (endotherm), 196.6/197.9/211.2/217.4./219.78C (endotherm);
TGA: onset of mass loss, 2608C; elemental analysis calcd (%) for
C44H84N2P2Pd: C 65.28, H 10.46, N 3.46; found: C 66.65, H 10.26, N
3.04;[50] 1H NMR (CDCl3):[51] d=1.96–1.94 (brm, 12H; PCH2), 1.58–
1.53 (brm, 12H; PCH2CH2), 1.46–1.41 (brm, 12H; PCH2CH2CH2),
1.35–1.15 ppm (brm, remaining CH2); 13C{1H} NMR: d=129.5 (t,
2JCP =14.1 Hz, CN), 30.0 (virtual t,[23] 3JCP =7.0 Hz, PCH2CH2CH2), 27.9
(s, CH2), 27.2 (s, CH2), 26.9 (s, CH2), 26.8 (s, CH2), 25.7 (virtual t,[23]
1JCP =14.3 Hz, PCH2), 24.0 ppm (s, PCH2CH2); 31P{1H} NMR: d=
13C{1H}[29] NMR: d=30.8 (virtual t,[23]
(virtual t,[23]
CP =6.5 Hz, 2PCH2CH2CH2), 28.4 (s, 2CH2), 28.0 (s, CH2),
JCP =7.0 Hz, PCH2CH2CH2), 29.5
J
27.7 (s, 2CH2), 27.2 (s, CH2), 26.83 (s, 2CH2), 26.78 (s, CH2), 26.6 (s,
CH2), 25.7 (s, CH2), 25.1 (s, CH2), 22.2 (s, CH2), 20.9 ppm (virtual t,[23]
1JCP =16.2 Hz, PCH2); 31P{1H} NMR: d=13.4 ppm (s, JPPt
=
1
2224 Hz);[48] IR (powder film): n=2921 (s), 2848 (s), 2107 (s, nNCS),
2005 (m, nSCN), 1457 (m), 1441 (m), 1260 (m), 1260 (w), 1092 (s),
1021 (s), 795 (s), 717 cmꢀ1 (m); MS:[49a] m/z (%): 962 (95) [M]+, 904
(100) [MꢀSCN]+.
~
~
17.4 ppm (s); IR (powder film): n=2919 (s), 2850 (s), 2142 (s, nCN),
1465 (m), 1414 (m), 1106 (m), 1025 (w), 801 (m), 721 (m), 708 ppm
(m); MS:[49a] m/z (%): 809 (50) [M]+, 683 (100) [OP{(CH2)14}3PO]+, 667
(55) [P{(CH2)14}3PO]+, 651 (25) [P{(CH2)14}3P]+.
trans-[PtPh2(P{(CH2)14}3P)] (trans-15c).
A
Schlenk flask was
charged with trans-4c (0.1004 g, 0.1095 mmol) and Ph2Zn
(0.0718 g, 0.327 mmol), and THF (3.5 mL) was added with stirring.
After 20 h, a few drops of CH3OH were added. The sample was ex-
posed to air. After 1 h, the solvent was removed by trap to trap
distillation, and benzene was added. The suspension was filtered
through a pipette packed with silica gel. The pipette was rinsed
with benzene. The solvent was removed from the filtrate by freeze
pump drying to give trans-15c as a white powder (0.0663 g,
0.0663 mmol, 61%).
trans-[Pt(CN)2(P{(CH2)18}3P)] (trans-11 e): Complex trans-4e
(0.0301 g, 0.0277 mmol), KCN (0.0038 g, 0.0584 mmol) and CH2Cl2/
CH3OH (1:2 v/v) were combined as in the procedure for trans-11 c.
After 18 h, an identical workup gave trans-11 e as a white powder
(0.0269 g, 0.0252 mmol, 91%).
DSC (Ti/Te/Tp/Tc/Tf): 109.2/151.0/168.1/178.3/173.28C (endotherm);
TGA: onset of mass loss, 2868C; elemental analysis calcd (%) for
C56H108N2P2Pt: C 63.07, H 10.21, N 2.63; found: C 64.18, H 10.45, N
2.25;[50] 1H NMR ( CDCl3):[51] d=2.59–1.92 (brm, 12H; PCH2), 1.68–
1.52 (brm, 12H; PCH2CH2), 1.52–1.36 (brm, 12H; PCH2CH2CH2),
1.36–1.16 ppm (brm, remaining CH2); 13C{1H} NMR: d=124.7 (vir-
tual t,[23] 2JCP =17.2 Hz, CN), 30.5 (virtual t,[23] 3JCP =6.9 Hz,
PCH2CH2CH2), 28.5 (s, CH2), 28.4 (s, CH2), 28.3 (s, CH2), 28.1 (s, CH2),
M.p. (capillary) 178–1798C (slight discolorization, 1018C); DSC (Ti/
Te/Tp/Tc/Tf): 68.3/78.5/89.2/102.3/110.08C (endotherm), 173.4/186.4/
188.9/190.8/196.68C (endotherm); TGA: onset of mass loss, 3128C;
elemental analysis calcd (%) for C54H94P2Pt: C 64.84, H 9.47; found:
1
1
C 65.37, H 9.58; H NMR (CDCl3):[53] d=7.49 (d, JHH =7.2 Hz, 2H; o-
1
1
Ph), 7.38 (d, JHH =7.0 Hz, 2H; o’-Ph), 7.02 (apparent t, JHH =7.2 Hz,
27.8 (s, CH2), 27.5 (s, CH2), 24.7 (virtual t,[23] 1JCP =17.3 Hz, PCH2),
2H; m-Ph), 6.95 (apparent t, 1JHH =7.2 Hz, 2H; m’-Ph), 6.79 (appa-
1
24.1 ppm (s, PCH2CH2); 31P{1H} NMR: d=6.8 ppm (s, JPPt
=
1
rent t, JHH =7.2 Hz, 2H; p-Ph), 1.94–1.74 (brm, 6H; CH2), 1.70–1.55
2152 Hz);[48] IR (powder film): n=2918 (s), 2849 (s), 2123 (w, nCN),
1471 (m), 1417 (w), 1108 (w), 1124 (w), 799 (w), 749 (m), 722 (m);
MS:[49a] m/z (%): 1066 (100) [M]+.
~
(brm, 6H; CH2), 1.55–1.30 (brm, 50H; CH2), 1.30–1.12 ppm (brm,
2
22H; CH2); 13C{1H} NMR: d=162.0 (s, JCPt =10.8 Hz, i-Ph), 141.4 (s,
4
o-Ph), 139.2 (s, o’-Ph), 126.7 (s, JCPt =21.2 Hz, m-Ph),[48] 126.0 (s, m’-
trans-[Pt(NCS)2(P{(CH2)14}3P)] (trans-13c) and trans-[Pt(NCS)-
Ph), 120.7 (s, p-Ph), 30.6 (virtual t,[23]
J
CP =6.9 Hz, CH2), 29.0 (s, 2CH2),
(SCN)(P{(CH2)14}3P)] (trans-14c):
A
5 mL vial was sequentially
28.8 (s, 2CH2), 28.7 (partially obscured virtual t,[23]
JCP =6.6 Hz,
charged with trans-4c (0.0656 g, 0.0715 mmol), KSCN (0.0286 g,
0.294 mmol), and THF/acetone (3.0 mL, 1:1 v/v). After 5 h, the mix-
ture was filtered. The solvent was removed from the filtrate by
rotary evaporation. The sample was partially dissolved in THF/
PCH2CH2CH2), 28.3 (s, CH2), 27.6 (s, CH2), 27.1 (s, CH2), 26.8 (s, CH2),
26.6 (s, 2CH2), 26.4 (s, 2CH2), 26.0 (s, CH2), 25.1 (s, CH2), 22.3 (vir-
tual t,[23]
J
CP =15.7 Hz, PCH2), 21.4 ppm (virtual t,[23]
J
CP =16.8 Hz,
PCH2); 31P{1H} NMR: d=3.0 ppm (s, 1JPPt =2837 Hz);[48] IR (powder
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4635
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
~
film): n=3019 (w), 2922 (m), 1174 (s), 1440 (w), 1366 (s), 1216 (s),
18 wt% in xylene, 0.157 mmol), and THF (1 mL) with stirring. After
72 h, the solvent was removed by oil pump vacuum. The residue
was placed on a plug of silica gel in a Pasteur pipette and rinsed
with benzene. The solvent was removed from the filtrate, and THF
(1 mL) and BH3·S(CH3)2 (0.10 mL, 2.0m in THF, 0.20 mmol) were
added. After 15 min, the solvent was removed by oil pump
vacuum. The residue was placed on a plug of silica gel in a Pasteur
pipette and rinsed with hexanes/CH2Cl2 (2:1 v/v). The solvent was
removed by oil pump vacuum to give 19·2BH3 (0.017 g,
0.025 mmol, 78%) as a white solid. The NMR data closely corre-
sponded to those obtained with an independently prepared
sample that was additionally characterized by microanalysis and
mass spectrometry.[30]
1029 (w), 1019 (w), 799 cmꢀ1 (w); MS:[49a] m/z (%): 1000 (10) [M]+,
923 (100) [MꢀPh]+.
trans-[PtPh2(P{(CH2)18}3P)]
(0.0431 g, 0.0397 mmol), Ph2Zn (0.0261 g, 0.1189 mmol), and THF
(4 mL), and CH3OH (0.1 mL) were combined as in the procedure for
trans-15c. An identical workup gave a white powder. Crystallization
from CH2Cl2/pentane gave trans-15e as colorless needles (0.0417 g,
0.0357 mmol, 90%).
(trans-15e):
Complex
trans-4e
DSC (Ti/Te/Tp/Tc/Tf): 116.8/140.9/144.7/146.3/164.38C (endotherm);
TGA: onset of mass loss, 2348C; elemental analysis calcd (%) for
C66H118P2Pt: C 67.83, H 10.18; found: C 66.66, H 10.43;[50] 1H NMR
(CDCl3):[51] d=7.38 (d, 1JHH =7.2 Hz, 4H; o-Ph), 6.95 (apparent t,
1JHH =7.2 Hz, 4H; m-Ph), 6.77 (apparent t, 1JHH =7.2 Hz, 2H; p-Ph),
1.45–1.42 (brm, 12H; PCH2), 1.36–1.28 (brm, 42H; PCH2CH2), 1.24–
1.05 ppm (brm, 54H; CH2); 13C{1H}[54] NMR: d=140.5 (s, o-Ph), 127.2
(s, m-Ph), 121.1 (s, p-Ph), 30.3 (virtual t,[23] 3JCP =7.0 Hz,
PCH2CH2CH2), 28.7 (s, CH2), 28.7 (s, 2CH2), 28.4 (s, CH2), 28.1 (s, CH2),
28.0 (s, CH2), 27.8 (s, PCH2CH2), 21.6 ppm (virtual t,[23] 1JCP =17.3 Hz,
PCH2); 31P{1H} NMR: d=3.2 ppm (s, 1JPPt =2807 Hz);[48] IR (powder
Crystallography: Details of the crystallographic investigation can
be found in the Supporting Information. CCDC 956025 (trans-4c),
608961 (trans-5c), 956148 (trans-7c), 956019 (trans-8c), 956020
(trans-9c), 956024 (trans-10c), 956021 (trans-13c), 608960 (trans-
15c·n-C5H12), 955782 (trans-15e) and 956018 (15c·2BH3) contain
the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
~
film): n=2922 (s), 2853 (s), 1567 (w), 1459 (m), 1262 (m), 1092 (m),
1023 (m), 799 (s), 734 cmꢀ1 (s); MS:[49a] m/z (%): 1090 (85) [MꢀPh]+.
P{(CH2)14}3P (18). A Schlenk flask was charged with NaCꢁCH
(1.368 g, 28.49 mmol) and THF (30 mL). A solution of trans-4c
(0.523 g, 0.570 mmol) in THF (20 mL) was added to the suspension
with stirring. After 3 d, the mixture was filtered. The filter cake was
washed with THF (3ꢂ5 mL). The solvent was removed from the fil-
trate by oil pump vacuum. Hexanes (50 mL) were added to the
solid residue. The suspension was filtered through celite and the
filter cake was washed with hexanes (3ꢂ5 mL). The solvents were
removed from the filtrate and the solid residue dried (15 h) by oil
pump vacuum to give 18 as a white powder (0.339 g, 0.521 mmol,
91%). M.p. 688C; elemental analysis calcd (%) for C42H84P2: C 77.48,
H 13.00; found: C 77.67, H 13.08; 1H NMR (CDCl3): d=1.37–1.32,
Acknowledgements
We thank the US National Science Foundation (CHE-0719267
and CHE-1153085), the Deutsche Forschungsgemeinschaft
(DFG, GL 300/9-1), and Johnson Matthey PMC (precious metal
loans) for support, and Dr. Tobias Fiedler for some helpful cal-
culations.
Keywords: diphosphine ligands
metathesis · palladium · platinum · substitution reactions
· hydrogenation · olefin
1.31–1.23 ppm (2 br m, 84H; CH2); 13C{1H} NMR: d=31.2 (d, JCP
=
10.3 Hz, CH2), 29.2 (s, CH2), 29.14 (s, CH2), 29.08 (s, CH2), 28.7 (s,
CH2), 26.3 (d, JCP =12.2 Hz, CH2), 25.2 ppm (d, JCP =11.0 Hz, CH2);
31P{1H} NMR: d=ꢀ30.1 ppm (s); MS:[49b]m/z (%): 651.6 (58)
[M+H]+.
[1] See http://www.gyroscopes.org/uses.asp. The authors are unaware of
any books or peer reviewed articles that match the diversity of exam-
ples on the gyroscope.org web site, which is maintained by G. Turner.
[2] R. M. Alexander, Science 2007, 315, 771.
[3] a) E. Butikov, Eur. J. Phys. 2006, 27, 1071; b) D. Kleppner, R. J. Kolenkow,
An Introduction to Mechanics, Cambridge University Press, Cambridge,
2010, Chapters 7.3–7.5.
[4] G. S. Kottas, L. I. Clarke, D. Horinek, J. Michl, Chem. Rev. 2005, 105, 1281.
[5] a) K. Liu, W. Zhang, W. Chen, K. Li, F. Dai, F. Cui, X. Wu, G. Ma, Q. Xiao, J.
Micromech. Microeng. 2009, 19, 113001; b) M. N. Armenise, C. Ciminelli,
F. Dell’Olio, V. M. N. Passaro, Advances in Gyroscope Technologies, Spring-
er, Heidelberg, 2011.
[6] T. Shima, F. Hampel, J. A. Gladysz, Angew. Chem. 2004, 116, 5653;
Angew. Chem. Int. Ed. 2004, 43, 5537.
[7] A. J. Nawara, T. Shima, F. Hampel, J. A. Gladysz, J. Am. Chem. Soc. 2006,
128, 4962.
[8] L. Wang, F. Hampel, J. A. Gladysz, Angew. Chem. 2006, 118, 4479; Angew.
Chem. Int. Ed. 2006, 45, 4372.
(CH2)14P(CH2)14P(CH2)14 (19). A Schlenk flask was charged with
NaCꢁCH (0.399 g, 8.309 mmol) and THF (10 mL). A solution of
trans-4’e (0.153 g, 0.167 mmol) in THF (10 mL) was added to the
suspension with stirring. After 9 d, the mixture was filtered through
celite. The filter cake was washed with THF (6ꢂ5 mL). The solvent
was removed from the filtrate and the residue dried (10 h) by oil
pump vacuum. Hexanes (20 mL) were added. The suspension was
filtered through celite and the filter cake was washed with hexanes
(6ꢂ5 mL). The solvents were removed from the filtrate and the res-
idue dried (24 h) by oil pump vacuum to give 19 as a white
powder (0.098 g, 0.151 mmol, 90%).
M.p. 748C; elemental analysis calcd (%) for C42H84P2: C 77.48, H
13.00; found: C 75.94, H 12.90;[50] 1H NMR (CDCl3): d=1.46–1.41,
1.37–1.31, 1.28–1.26, 1.23 ppm (3 br m, 1 apparent s, 84H; CH2);
13C{1H} NMR: d=31.5 (d, JCP =11.2 Hz, CH2), 29.66 (s, 2CH2), 29.56
[9] L. Wang, T. Shima, F. Hampel, J. A. Gladysz, Chem. Commun. 2006, 4075.
[10] G. D. Hess, F. Hampel, J. A. Gladysz, Organometallics 2007, 26, 5129.
[11] See also a) K. Skopek, M. Barbasiewicz, F. Hampel, J. A. Gladysz, Inorg.
Chem. 2008, 47, 3474; b) J. Han, C. Deng, L. Wang, J. A. Gladysz, Organo-
metallics 2010, 29, 3231.
(s, CH2), 29.42 (s, CH2), 29.16 (d, JCP =9.8 Hz, 2CH2), 27.8 (d, JCP
=
9.8 Hz, CH2), 27.15 (d, JCP =13.1 Hz, 2CH2), 27.08 (s, 2CH2), 26.8 (s,
2CH2), 26.6 (s, 2CH2), 26.19 (s, 2CH2), 26.16 (d, JCP =13.8 Hz, CH2),
24.7 ppm (d, JCP =12.0 Hz, 2CH2); 31P{1H} NMR: d=ꢀ30.9 ppm (s);
MS:[49b] m/z (%): 651.6 (12) [M+H]+, 326.3 (100) [M+2H]2+; HRMS
(ESI+): m/z calcd for C42H85P2 [M+H]+: 651.6127; found: 651.6533.
[12] K. Skopek, J. A. Gladysz, J. Organomet. Chem. 2008, 693, 857.
[13] P. D. Zeits, G. P. Rachiero, F. Hampel, J. H. Reibenspies, J. A. Gladysz, Or-
ganometallics 2012, 31, 2854.
[14] a) T.-A. V. Khuong, J. E. NuÇez, C. E. Godinez, M. A. Garcia-Garibay, Acc.
Chem. Res. 2006, 39, 413; b) C. S. Vogelsberg, M. A. Garcia-Garibay,
Chem. Soc. Rev. 2012, 41, 1892.
H3B(CH2)14P(C14H28)P(CH2)14BH3 (19·2BH3). A Schenk flask was
charged with trans-4’e (0.029 g, 0.032 mmol), NaCꢁCH (0.042 g,
[15] Additional literature subsequent to reference [14a] or [14b]: a) J. E.
NuÇez, A. Natarajan, S. I. Khan, M. A. Garcia-Garibay, Org. Lett. 2007, 9,
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4636
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3559; b) P. Commins, J. E. NuÇez, M. A. Garcia-Garibay, J. Org. Chem.
2011, 76, 8355; c) B. Rodrꢃguez-Molina, S. Pꢄrez-Estrada, M. A. Garcia-
Garibay, J. Am. Chem. Soc. 2013, 135, 10388; d) M. Hughs, M. Jimenez, S.
Khan, M. A. Garcia-Garibay, J. Org. Chem. 2013, 78, 5293.
[16] a) W. Setaka, S. Ohmizu, C. Kabuto, M. Kira, Chem. Lett. 2007, 36, 1076;
b) W. Setaka, S. Ohmizu, M. Kira, Chem. Lett. 2010, 39, 468; c) W. Setaka,
K. Yamaguchi, J. Am. Chem. Soc. 2012, 134, 12458; d) W. Setaka, K. Ya-
maguchi, Proc. Natl. Acad. Sci. USA 2012, 109, 9271; e) W. Setaka, A.
Koyama, K. Yamaguchi, Org. Lett. 2013, 15, 5092.
[17] See also N. S. Khan, J. M. Perez-Aguilar, T. Kaufmann, P. A. Hill, O. Taratu-
la, O.-S. Lee, P. J. Carroll, J. G. Saven, I. J. Dmochowski, J. Org. Chem.
2011, 76, 1418.
[18] A. J. Nawara-Hultzsch, Doctoral Dissertation, Universitꢅt Erlangen-Nꢆrn-
berg, 2008; the crystallographic data for certain complexes have been
further refined since this dissertation, so some values reported herein
differ slightly.
[36] a) J. C. Conrad, M. D. Eelman, J. A. Duarte Silva, S. Monfette, H. H.
Parnas, J. L. Snelgrove, D. E. Fogg, J. Am. Chem. Soc. 2007, 129, 1024;
b) S. Monfette, D. E. Fogg, Chem. Rev. 2009, 109, 3783.
[37] Representative structures: a) trans-[MCl2(P(iPr)(tBu)2)2] (M=Pd, Pt): C. M.
DiMeglio, K. J. Ahmend, L. A. Luck, E. E. Weltin, A. L. Rheingold, C. H.
Bushweller, J. Phys. Chem. 1992, 96, 8765; b) trans-[PdBr2(PCy3)2]: M. L.
Clarke, A. G. Orpen, P. G. Pringle, E. Turley, Dalton Trans. 2003, 4393;
c) trans-[PtBr2(PEt3)2]: G. G. Messmer, E. L. Amma, Inorg. Chem. 1966, 5,
1775; d) trans-[PdI2(PEt3)2]: R. Frꢅnkel, J. Kinczek, H. Poniewar, H. Nçth,
K. Polbrom, W. P. F. Fehlhammer, Inorg. Chim. Acta 2001, 312, 23;
e) trans-[PtBr2(PCy3)2]: T. S. Cameron, H. C. Clark, A. Linden, A. M. Nicho-
las, Polyhedron 1990, 9, 1683; f) trans-[MI2(PCy3)2] (M=Pd, Pt): N. W.
Alcock, P. G. Leviston, J. Chem. Soc. Dalton Trans. 1974, 1834.
[38] A. Bondi, J. Phys. Chem. 1964, 68, 441.
[39] The van der Waals radius of an sp hybridized nitrogen atom, 1.55 ꢁ, is
added to the average metal–nitrogen distances in the following com-
plexes: a) trans-[Pt(CN)2(PCy3)2]: B.-H. Xia, C.-M. Che, D. L. Phillips, K.-H.
Leung, K.-K. Cheung, Inorg. Chem. 2002, 41, 3866; b) trans-[Pd(CN)2{P-
(CH2Ph)3}2]: B. Bendiksen, W. C. Riley, M. W. Babich, J. H. Nelson, R. A. Ja-
cobson, Inorg. Chim. Acta 1982, 57, 29.
[40] a) E. V. Ansyln, D. A. Dougherty, Modern Physical Organic Chemistry, Uni-
versity Science Books, Sausalito, 2006, pp. 97–98; b) E. N. Eliel, S. H.
Wilen, Stereochemistry of Organic Compounds, Wiley, Hoboken; 1994,
pp. 624–629 and 1142–1155.
[41] This voluminous literature is easily accessed via the usual search en-
gines; see for example F. E. Udwadia, B. Han, J. Appl. Mech. 2008, 75,
021011.
[42] D. Skaper, Doctoral Dissertation, Universitꢅt Erlangen-Nꢆrnberg, 2010.
[43] E. B. Bauer, F. Hampel, J. A. Gladysz, Adv. Synth. Catal. 2004, 346, 812.
[44] a) R. D. Horansky, L. I. Clarke, J. C. Price, T.-A. V. Khuong, P. D. Jarowski,
M. A. Garcia-Garibay, Phys. Rev. B 2005, 72, 014302; b) R. D. Horansky,
L. I. Clarke, E. B. Winston, J. C. Price, S. D. Karlen, P. D. Jarowski, R. Santil-
lan, M. A. Garcia-Garibay, Phys. Rev. B 2006, 74, 054306.
[45] H. K. Cammenga, M. Epple, Angew. Chem. 1995, 107, 1284; Angew.
Chem. Int. Ed. Engl. 1995, 34, 1171; the Te values best represent the tem-
perature of the phase transition or endotherm. DSC measurements
were generally not continued above the initial mass loss temperature
(TGA).
[19] A. J. Nawara-Hultzsch, K. Skopek, T. Shima, M. Barbasiewicz, G. D. Hess,
D. Skaper, J. A. Gladysz, Z. Naturforsch. B 2010, 65, 414.
[20] A figure is provided in the Supporting Information.
[21] a) 31P{1H} NMR spectra at this stage showed several peaks.[20] Note that
Z/Z/Z, Z/Z/E, Z/E/E, and E/E/E C=C isomers are possible for the inter-
mediate tris(alkenes)in Scheme 2; b) when the filtrates were taken to
dryness, the yields of crude cyclized products were typically 85–45%.
[22] S. O. Grim, R. L. Keiter, W. McFarlane, Inorg. Chem. 1967, 6, 1133.
[23] W. H. Hersh, J. Chem. Educ. 1997, 74, 1485: the J values given represent
the apparent coupling between adjacent peaks of the triplet.
[24] J. S. Siegel, F. A. L. Anet, J. Org. Chem. 1988, 53, 2629.
[25] A reviewer questioned whether these data were correctly presented.
Given the workup conditions, the isolated yields of trans-13c and trans-
14c (28 and 20%, respectively) should not be regarded as reflecting
that one isomer may be major/minor. With time, the 50:50 mixture ob-
served spectroscopically becomes biased in favor of trans-14c.
[26] a) D. W. Meek, P. E. Nicpon, V. I. Meek, J. Am. Chem. Soc. 1970, 92, 5351;
b) R. R. Cooke, J. L. Burmeister, J. Organomet. Chem. 1973, 63, 471;
c) Y. S. Wong, S. Jacobson, P. C. Chieh, A. Carty, Inorg. Chem. 1974, 13,
284.
[27] This could be confirmed crystallographically, but the bent SCN and ap-
proximately linear NCS ligands were disordered.
[46] The slow addition of the Grubbs’ catalyst at 08C is essential; otherwise
the proportion of oligomeric products increases substantially.
[47] The 1H and 13C NMR assignments were made by 1H,1H COSY and
1H,13C{1H} COSY spectra.
[28] G. J. Palenik, M. Mathew, W. L. Steffen, G. Beran, J. Am. Chem. Soc. 1975,
97, 1059.
[29] As is often the case with such complexes, the 13C NMR spectra did not
exhibit SCN signals; one would have been expected for trans-13c, and
two for trans-14c.
[30] M. Stollenz, M. Barbasiewicz, A. J. Nawara-Hultzsch, T. Fiedler, R. Laddu-
saw, N. Bhuvanesh, J. A. Gladysz, Angew. Chem. 2011, 123, 6777; Angew.
Chem. Int. Ed. 2011, 50, 6647.
[31] M. Stollenz, N. Bhuvanesh, J. H. Reibenspies, J. A. Gladysz, Organometal-
lics 2011, 30, 6510.
[32] E. B. Bauer, T. Shima, unpublished results, Universitꢅt Erlangen-Nꢆrn-
berg.
[33] E. B. Bauer, F. Hampel, J. A. Gladysz, Organometallics 2003, 22, 5567.
[34] a) G. D. Hess, Doctoral Dissertation, Universitꢅt Erlangen-Nꢆrnberg,
2010; b) T. Fiedler, Doctoral Dissertation, Universitꢅt Erlangen-Nꢆrnberg,
2011.
[35] T. Shima, E. B. Bauer, F. Hampel, J. A. Gladysz, Dalton Trans. 2004, 1012.
The product/catalyst system in reference [19] therein rapidly attains
monomer/oligomer equilibrium, which can be reversibly varied as
a function of concentration.
[48] This coupling represents a satellite (d, 195Pt=33.8%) and is not reflect-
ed in the peak multiplicity given.
[49] m/z (relative intensity,%); the most intense peak of isotope envelope is
given; FAB, 3-NBA; ESI(+).
[50] The microanalytical data are not in good agreement with this spectro-
scopically pure sample, but are reported to illustrate the best result ob-
tained. The 1H and 13C{1H} NMR spectra of 19 are provided in the sup-
porting information.
[51] The 1H and 13C NMR assignments were made by analogy to trans-4c.
[52] The total mmol was calculated from the mass using the product ratios
found by 31P{1H} NMR.
1
[53] The phenyl protons were assigned by a H,1H COSY NMR spectrum.
[54] The i-Ph 13C signal was not observed.
Received: November 12, 2013
Published online on March 6, 2014
Chem. Eur. J. 2014, 20, 4617 – 4637
www.chemeurj.org
4637
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim