Spectroscopic and reactivity differences in metal complexes derived from sulfur containing Triphos homologs

Article abstract of DOI:10.1039/c7dt01459g

Herein, we report a simplified method for the synthesis of Triphos homologs H₃CC(CH₂X)_n(CH₂Y)_3-n (X = SPh, Y = PPh₂, n = 0-3). The multidentate compounds were tested for their potential to coordinate metals such as Ni, Fe, and Mo under the same experimental conditions. Cyclic voltammetry, spectroelectrochemical IR investigations as well as DFT calculations were used to examine the electronic alterations in a series of [{H₃CC(CH₂X)_n(CH₂Y)_3-n}Mo(CO)₃] complexes and to evaluate their potential to open coordination sites or to release CO upon oxidation or in the presence of different solvents. In addition, we demonstrate that the catalytic hydrosilylation of N,N-dimethylbenzamide to N,N-dimethylbenzylamine is influenced by the applied tripodal ligand. Our investigations show the high potential of such manipulations to selectively alter the dynamics of the binding properties of Triphos-metal complexes and their reactivity.

Full text of DOI:10.1039/c7dt01459g

Dalton
Transactions
View Article Online
View Journal
PAPER
Spectroscopic and reactivity differences in metal
complexes derived from sulfur containing Triphos
homologs†
Cite this: DOI: 10.1039/c7dt01459g
A. Petuker,‡a P. Gerschel,‡a S. Piontek,‡a N. Ritterskamp,a F. Wittkamp,a L. Iffland,a
a
R. Miller,a M. van Gastelb and U.-P. Apfel
*
Herein, we report a simplified method for the synthesis of Triphos homologs H3CC(CH2X)n(CH2Y)3−n (X =
SPh, Y = PPh2, n = 0–3). The multidentate compounds were tested for their potential to coordinate
metals such as Ni, Fe, and Mo under the same experimental conditions. Cyclic voltammetry, spectroelec-
trochemical IR investigations as well as DFT calculations were used to examine the electronic alterations
in a series of [{H3CC(CH2X)n(CH2Y)3−n}Mo(CO)3] complexes and to evaluate their potential to open
coordination sites or to release CO upon oxidation or in the presence of different solvents. In addition, we
demonstrate that the catalytic hydrosilylation of N,N-dimethylbenzamide to N,N-dimethylbenzylamine is
influenced by the applied tripodal ligand. Our investigations show the high potential of such manipula-
tions to selectively alter the dynamics of the binding properties of Triphos-metal complexes and their
reactivity.
Received 23rd April 2017,
Accepted 10th July 2017
DOI: 10.1039/c7dt01459g
rsc.li/dalton
metalloenzymes allowing for a controlled opening and closing
of reactive sites with minimal energetic demands. Therefore,
Introduction
Tripodal compounds such as CH3C(CH2PPh2)3 (Triphos, 1,1,1- the modulation of the metal–ligand binding strength seems to
tris(diphenylphosphinomethyl)ethane) and CH3Si(CH2PPh2)3 be a key feature in the tuning of the catalytic properties. A
(TriphosSi, tris(diphenylphosphinomethyl)methylsilane) and comparable effect on catalysis was recently shown for
their respective metal complexes are receiving increasing atten- [(Triphos)FeCl2] and [(TriphosSi)FeCl2].24 The modulation of
tion due to their impressive catalytic properties.1,2 The the binding properties due to C/Si exchange in the backbone
ruthenium,3–12 rhodium,13–17 molybdenum18,19 and cobalt20,21 of the tripodal ligand revealed a significant effect on the metal
complexes of these ligands have been particularly well studied complexes to support either Sonogashira cross coupling or
and were recently shown to be powerful e.g. in the hydrogen- hydrosilylation reactions. Triphos was shown to have a higher
ation of CO2 and lactams as well as in the activation of N2.10,22 tendency to adopt a κ3-binding mode as compared to its
Notably, the potential to switch between κ2- or κ3-binding TriphosSi counterpart.24 This finding demonstrated the impor-
modes was highlighted as one possible explanation for their tance of investigating the Triphos-ligand flexibility within cata-
remarkable performance.16 Along this line, a flexible, solvent- lysis. However, although Triphos is a common ligand frame-
and temperature-triggered binding of Triphos as well as work in catalytic applications, attempts to alter the catalyst’s
TriphosSi to metals was observed in their Co- and Fe-com- properties, by partially exchanging the phosphine donor
plexes.23,24 As such, these tripodal ligands reveal properties of groups with other substituents, remain limited. Prominent
examples for such a change in reactivity are [(Triphos)Mo
(dppm)]-type complexes (dppm
= bis(diphenylphosphino)
methane) with different substituents on phosphorus that were
reported to allow for an activation of N2.18,22,25 By modulation
of the substitution patterns at the phosphorus donor atoms, a
significant alteration of the N–N stretch vibrations was
observed, suggesting a substituent-dependent N2 activation.
Given the impressive catalytic properties of Triphos derived
compounds, it is important to develop synthetic pathways for
tripodal compounds comprising altered donor sets that are
aAnorganische Chemie I, Ruhr-Universität Bochum, Universitätsstraße 150,
D-44801 Bochum, Germany. E-mail: ulf.apfel@rub.de
bMax-Planck-Institut für Chemische Energiekonversion, Stiftstraße 34-36,
45470 Mülheim, Germany
†Electronic supplementary information (ESI) available: IR and SEC-IR spectra,
GC-MS analysis and X-ray crystallographic analysis. CCDC 1545160–1545164 and
1545379. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c7dt01459g
‡These authors contributed equally.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
both efficient and affordable. Synthetic pathways towards
Triphos ligands with mixed substitution patterns, including
the ligands presented in this work, have been reported in
Results and discussion
Synthesis of the ligands
literature.22,26–35 The synthetic pathways are, however, long Compounds SSS and PPP were synthesized according to litera-
and cumbersome and therefore limit investigation of such ture procedures by reaction of 1 with either excess amounts of
ligands in catalytic applications. Furthermore, reactivity thiophenol or lithium diphenylphosphide (Scheme 2).42–44
changes via substitution of the donor atoms remain Subsequently, we attempted the synthesis of compounds PSS
elusive.26,29,35–38 In this work, we have focused on a straight- and PPS applying the routes described by Kabachnik45 as well
forward synthetic route for substitution of phosphorus by as Tuczek22 and coworkers. While the reaction of 1 with one or
sulfur in the ligand series [SSS] (1,1,1-tris(phenylthiomethyl) two equivalents HPPh2/KOtBu in DMSO resulted in the selec-
ethane), [PPP] (1,1,1-tris(diphenylphosphinomethyl)ethane), tive formation of compounds 2 and 3, respectively, subsequent
[PSS] (1-(diphenylphosphinomethyl)-1,1-bis(phenylthiomethyl) attempts to convert the remaining chlorides into thiophenol
ethane) and [PPS] (1,1-bis(diphenylphosphinomethyl)-1-bis containing products PSS and PPS failed in our hands and
(phenylthiomethyl)ethane) (Scheme 1). The different donor solely afforded starting material. This observation can be
sets are of interest in redox catalysis due to their different sta- assigned to the cumulated steric shielding of the neopentyl
bilizing properties towards metals.39–41
backbone and the diphenylphosphine moieties. Equally to the
The syntheses of SSS and PPP were previously reported synthesis of 2 and 3, the reaction of compound 1 and stoichio-
from 1,3-dichloro-2-(chloro-methyl)-2-methylpropane (1) via metric amounts of thiophenol, NaH and KI in DMF at 50 °C
nucleophilic substitution with thiols42,43 or phosphines.44 yielded the one- or twofold substituted compounds 4 or 5 as
Synthetic access towards the mixed-donor compounds PSS and the main products in 42 and 55% yield, respectively. The reac-
PPS is not as straightforward.
tion conditions are key towards the successful synthesis of
Whilst the synthesis of PSS was previously reported via both compounds. Variations from the synthetic protocol,
3-methyl-3-((phenylthio)methyl)oxetane,26 synthesis of PPS e.g. by varying the amount of KI used or changing the reaction
was reported to proceed via (2,2,5-trimethyl-1,3-dioxan-5-yl) temperature, lead to different product distributions and
methyl-4-methylbenzene-sulfonate.33 The synthetic prerequi- commonly resulted in the formation of tri-substituted phos-
sites and functionalizations required to perform the lengthy phines, thioethers or no conversion at all. The subsequent
procedures render a simple alteration of metal complexes with treatment of compounds 4 and 5 with potassium diphenyl-
tripodal ligands difficult. A possible alternative pathway was phosphide afforded the desired compounds PSS and PPS in
reported by Tuczek and coworkers.22 Starting from the trichlo- very good yields (>44% overall yield). In contrast to the litera-
ride 1, a one- or two-fold substitution with HPiPr2·BH3 allowed ture reported lengthy and cumbersome procedures,26,33 the
for the synthesis of the respective mono- and diphosphines. herein described pathway thus allows for a simple and
However, the synthetic route was limited by the steric demand straightforward synthesis of PSS and PPS in only a two-step
of the borane-protected diisopropyl phosphine and the neo- synthesis.
pentyl backbone. Likewise, a controlled substitution reaction
was reported when compound 1 reacted with stoichiometric
amounts of HPPh2/KOtBu in DMSO.36
To develop tripodal compounds comprising mixed
sulfur/phosphorus donor sets, we also started our synthetic
approach from the trichloride 1, a cheap and readily available
starting material. We aimed at finding a simple and straight-
forward synthetic procedure towards all tripodal compounds
SSS, PPP, PSS and PPS for a stepwise, systematic replacement
of the thioether by phosphine groups and investigation
of their binding affinities towards metals under varying
oxidative conditions. In addition we herein show the
different reactive behavior of their iron carbonyl deriva-
tives for the hydrosilylation of N,N-dimethylbenzamide to
N,N-dimethylbenzylamine.
Scheme 2 Reaction scheme towards the syntheses of tripodal com-
pounds with mixed [PS]-donor sets. (a) 3 equiv. NaSPh, 3 equiv. KI, DMF,
60 °C; (b) 1 equiv. KPPh2, DMSO, 60 °C; (c) 2 equiv. NaSPh, 2 equiv. KI,
DMF, 100 °C; (d) 1 equiv. KPPh2, DMSO, 100 °C; (e) 1 equiv. NaSPh, 1
Scheme 1 Tripodal compounds providing [SSS], [PPP], [PSS] and [PPS]
equiv. KI, DMF, 100 °C; (f) 2 equiv. KPPh2, DMSO, 100 °C; (g) 2 equiv.
donor sets.
KPPh2, DMSO, 100 °C; (h) 3 equiv. LiPPh2, THF, −78 °C.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Reactions with metals and characterization
nate the NiII center, revealing Ni–N distances of 1.901(2)/1.905
(2) Å and Ni–P distances of 2.1972(5)/2.1921(4) Å. The P(1)–Ni–
P(2) (91.84(2)°) and N(1)–Ni–N(2) (90.51(6)°) bond angles give
rise to an ideal square planar assembly. In addition, 31P{1H}
NMR spectroscopy revealed a distinct signal for the co-
ordinated phosphine groups at −15.7 ppm as compared to the
phosphorus signals for the free ligand (−27.10 ppm). In con-
trast, no 31P{1H} NMR signal was observed for compound 7.
The electronic differences due to the different ligand field
environments are also visible within the UV/vis spectra of 7
and 8. While 7 possesses two distinct bands at 422 and
728 nm, compound 8 reveals a solitary band at 428 nm
(Fig. S1†). A similar coordination behavior of PPP and PPS is
observed upon reaction with FeCl2 yielding the isostructural
complexes 6 and 9. It is also worth to note that the reactions
of PPS with NiX2 (X = Cl, Br, I) and CoCl2 have been previously
reported.29 These Ni and Co halide complexes adopted a
similar κ2-coordination involving only the phosphorus donors.
Structural analysis of complex 9 unequivocally confirms a tetra-
hedrally coordinated FeII center bearing the tripodal ligand
PPS in κ2-coordination mode and two additional chloride ions
(Fig. 1). The Fe–Cl distances of 2.213(2)/2.241(3) Å and Fe–P
distances of 2.432(3)/2.451(3) Å are within the expected range
and comparable to those of 6 (Table 1).46 The ligand PPS
coordinates iron with both P-atoms as donors, leaving the
thioether group uncoordinated. Mössbauer data (δ = 0.59
0.02 mm s−1, ΔEq = 2.23 0.02 mm s−1) are in line with a
high-spin FeII center.24,47 Unlike the reactions of compounds
PPP and PPS, the reaction of ligands SSS and PSS with FeII- or
With the simple synthetic protocol, we were able to efficiently
synthesize the ligands SSS, PPP, PSS and PPS in gram scale for
further coordination studies. We were especially interested in
their comparative coordination behavior towards FeII- and NiII-
salts (Scheme 3). The coordination reactions of Triphos PPP
with FeCl2 and [Ni(CH3CN)6](BF4)2 are literature reported and
selectively afford complexes 6 and 7 in high yields, with the
PPP ligand in a κ2-coordinated manner (Scheme 3).24,46
Likewise, the reaction of ligand PPS and [Ni(CH3CN)6](BF4)2
afforded complex 8 that reveals a κ2-coordinated tripodal
ligand.46 However, in contrast to the previously reported pyra-
midal complex 7, compound 8 adopts a square-planar coordi-
nation (Fig. 1). In both cases the remaining coordination sites
are occupied by CH3CN molecules – two in the case of the
square planar complex 8 and three in the case of the pyramidal
complex 7. As the uncoordinated S/P donor site in PPP and
PPS is electronically isolated from the metal center, in both 7
and 8, it is unlikely that this change in geometry is due to a
change in ligand field strength. This is consistent with the
weakly bound nature of the axial CH3CN molecule in 7. For 8,
two CH3CN molecules as well as κ2-coordinated PPS coordi-
Table 1 Selected bonding angles and lengths of Fe- and Ni-complexes
6–9
Ni
Fe
7
8
6
9
Bond lengths [Å]
M–P(1)
M–P(2)
M–N(1)
M–N(2)
M–N(3)
M–Cl(1)
M–Cl(2)
Scheme 3 Ligand controlled synthesis of iron and nickel complexes
6–10 starting from ligands SSS, PPP, PSS and PPS. Reaction conditions:
(a) FeCl2, THF, 25 °C; (b) [Ni(CH3CN)6](BF4)2, CH3CN, 25 °C.
2.1717(7)
2.1951(7)
1.925(2)
1.928(3)
2.268(3)
—
2.1921(6)
2.1972(6)
1.905(2)
1.901(2)
—
2.404(2)
2.414(2)
—
—
—
2.218(2)
2.253(2)
2.430(3)
2.452(2)
—
—
—
2.241(3)
2.213(3)
—
—
—
Bond angles [°]
P(1)–M–P(2)
P(1)–M–N(1)
P(1)–M–N(2)
P(1)–M–N(3)
P(2)–M–N(1)
P(2)–M–N(2)
P(2)–M–N(3)
N(1)–M–N(2)
N(1)–M–N(3)
N(2)–M–N(3)
P(1)–M–Cl(1)
P(1)–M–Cl(2)
P(2)–M–Cl(1)
P(2)–M–Cl(2)
Cl(1)–M–Cl(2)
93.12(3)
86.85(7)
161.68(9)
102.97(7)
171.84(8)
89.59(7)
98.48(6)
87.93(9)
89.47(9)
94.53(11)
—
91.84(2)
87.64(6)
172.72(8)
—
178.29(8)
89.80(6)
—
90.51(8)
—
—
91.08(2)
—
—
—
—
—
—
—
93.57(3)
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
113.12(3)
104.18(3)
123.69(3)
94.07(2)
124.39(3)
114.2(1)
98.57(9)
120.55(8)
103.37(10)
121.24(11)
Fig. 1 Molecular structures of compounds 8 (left) and 6 (right) with
thermal ellipsoids drawn at the 50% probability level (hydrogen atoms
are omitted for clarity). Color code: C gray, S yellow, P purple, Ni tur-
quoise, N blue, Fe dark blue and Cl green.
—
—
—
—
—
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
NiII-salts did not result in the formation of the desired
complexes. Moreover, the attempted complexation of FeCl2
with both Triphos analogs resulted in the formation of
[Fe4Cl8(THF)6] 10 and is in line with reports on the substitu-
tional lability of diphosphine ligands in tetrahedral FeII-chloro
complexes.48 The results clearly show that the binding affinity
of the ligands can be tuned by the simple alterations shown
herein. The limited coordination chemistry of SSS and PSS
towards NiII- and FeII-salts led us to direct our attention
towards another metal that would enable coordination of SSS,
PPP, PSS and PPS. Mo0-complexes were previously shown to
allow for stable coordination of tripodal ligands comprising
a [SSS], [PSS], [PPS] or [PPP] donor set and thus allow
for comprehensive structural and spectroscopic compari-
son.26,28,33,49,50 While a κ2/κ3-isomerization of the ligands
binding modes of the ligands is feasible by changing the reac-
tion conditions (e.g. temperature, solvent, oxidation state,
counter ion) and would be in line with our observations
on [PPP]Fe- and [PPP]Ni-complexes,24,46 release of CO was
reported for [PPP]M(CO)3 (M = W, Mo) upon oxidation to Mo(III)
and formation of seven-coordinate Mo-complexes with Cl2
or Br2.51 This CO-releasing reaction, which is analogous to the
chemistry of CO-releasing molecules (CORMs),52 thus displays
a second possibility of Mo(CO)3-moieties comprising the SSS,
PPP, PSS or PPS ligands to react on altered reaction conditions.
Reaction of compounds SSS, PPP, PSS and PPS with 1
equiv. Mo(CO)3(toluene) in THF at room temperature afforded
the respective Mo0-complexes [Mo(SSS)(CO)3] 11, [Mo(PSS)
Fig. 2 Molecular structures of compounds 11 (top left), 12 (top right)
and 13 (bottom left) as well as the structural motif of 14 (bottom right,
CO ligands shown in ball and stick representation) with thermal ellip-
soids drawn at the 50% probability level (hydrogen atoms are omitted
for clarity). Color code: C gray, S yellow, P purple, Mo blue and O red.
Table 2 Selected bonding angles and lengths of Mo-complexes 11–13
11
12
13
Bond lengths [Å]
Mo–P(1)
Mo–P(2)
—
—
—
2.532(1)
—
—
2.517(2)
2.539(2)
—
(CO)3] 12, [Mo(PPS)(CO)3] 13 and [Mo(PPP)(CO)3] 14 in good Mo–P(3)
yields (Scheme 4). Structural analysis confirmed a κ3-coordi-
Mo–S(1)
2.564(1)
2.571(2)
2.561(1)
1.973(9)
1.944(6)
1.968(5)
1.143(8)
1.169(7)
1.148(7)
2.5414(9)
2.585(1)
—
1.949(4)
1.967(4)
1.936(4)
1.165(4)
1.159(4)
1.164(4)
2.552(2)
—
—
1.958(6)
1.994(6)
1.992(5)
1.124(7)
1.143(8)
1.155(8)
Mo–S(2)
Mo–S(3)
nation mode for all ligands. Notably, binding of PPS was solely
reported in a κ2-fashion in an analogue compound to 13, Mo–C(1)
Mo–C(2)
Mo–C(3)
C(1)–O(1)
which was obtained by treatment of the respective ligand and
Mo(CO)6.26 This difference clearly shows that even small
changes in the reaction protocol can lead to significant altera- C(2)–O(2)
tions in the product formation. Furthermore, it highlights the
C(3)–O(3)
importance of comparative evaluations under exactly the same
Bond angles [°]
experimental conditions. Analysis of the molecular structures P(1)–Mo–P(2)
—
—
—
76.96(5)
87.56(4)
84.94(5)
—
—
—
—
—
—
84.64(3)
—
83.87(5)
—
—
—
—
—
87.28(5)
—
79.99(5)
85.7(2)
84.7(2)
88.7(2)
reveals an octahedrally coordinated Mo0 center bearing the
P(1)–Mo–P(3)
P(2)–Mo–P(3)
S(1)–Mo–S(2)
κ3-bound tripodal ligands SSS, PPP, PSS or PPS and three
additional CO ligands for all complexes 11–14 (Fig. 2 and S(1)–Mo–S(3)
S(2)–Mo–S(3)
—
Table 2). Compound 11 shows Mo–S distances of 2.564(1)/
P(1)–Mo–S(1)
P(1)–Mo–S(2)
P(2)–Mo–S(1)
C(1)–Mo–C(2)
C(1)–Mo–C(3)
C(2)–Mo–C(3)
80.30(3)
87.02(3)
—
85.4(2)
83.5(2)
87.7(2)
85.8(3)
84.6(3)
85.5(2)
2.571(2)/2.561(1) Å and Mo–C distances of 1.973(9)/1.944(6)/
1.968(5) Å with an average C–O distance of 1.153 Å. Similarly,
the Mo-complexes 12 and 13 with mixed donor sets reveal
Mo–S and Mo–C distances in the same range as observed for 11
(Table 2). While single crystal X-ray diffraction confirmed
the composition and coordination mode of compound 14,
the quality of the obtained crystals did not allow for any dis-
cussion of bond lengths and angles.
Scheme 4 Synthesis of Mo-complexes with κ3-bound ligands.
Reaction conditions: (a) [Mo(CO)3(toluene)], THF, 25 °C.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Table 3 Spectroscopic and electrochemical comparison of complexes 11–14, and symmetry classification of the CO stretching frequencies
according to idealized C3v symmetry
Infrared spectroscopy ν [cm−1
]
31P{1H} NMR δ [ppm]
Mo0/MoI E1/2 [V]
Solida
THF
THF ox.
CH3CN
CH3CN ox.
CDCl3
CH3CN
11
12
13
—
1925(a1)
—
1821(e)
1791(e)
—
—
—
—
1919
—
—
1796
—
2037
—
—
—
—
—
−0.130
1919
1898
1828
1780
—
1939
1919
1837
1780
—
1939
1852
1834
—
—
—
1919
1903
—
1785
1969
1938
1894
1835
—
2019
1938
1853
1833
—
—
—
18.5
18.1
−0.072
—
—
1926(a1)
1837(e)
1788(e)
—
1925(a1)
—
1829(e)
—
—
1931
1830
1801
—
1930
1850
1836
1817
1807
—
1931b
1830b
1801b
1990
1933
—
0.068
—
—
—
n.d.
—
14
—
2021
1938
1850
—
18.1
0.134
1925(a1)
1831(e)
—
1938
1844
—
1929
1838
—
a The symmetry of the vibration frequencies was assigned according to an idealized C3v symmetry. b The intensity of this band decreased, com-
pared to the spectrum for non-oxidized complexes.
Notably, the stronger π-acceptor nature of the phosphine
respective to the thioether donor influences the trans located
CO ligands. Due to a weaker π-backbonding in this position,
Mo–C distances are increased by 0.03 Å compared to the other
CO ligands. In addition, 31P{1H} NMR spectroscopy showed
one signal for the coordinated phosphine groups at ∼18 ppm
for all complexes 12–14, indicating electronic equivalence for
all Mo-bound phosphorus atoms (Table 3). However, compared
to the free ligand (−27.10 ppm) the phosphorus atoms are
Fig. 3 A: IR spectra of compounds 11–14 in the carbonyl region. B:
strongly deshielded. In addition, IR spectroscopy revealed dis-
−
Cyclic voltammograms of 11–14 in CH3CN, with 0.1 M [nBu4N]+PF6
.
tinct CO bands for complexes 11–14, which allows evaluation
of the donor/acceptor strength of compounds SSS, PPP, PSS
and PPS. In the solid state, a sharp CO band at 1925 cm−1 is
uniformly observed for all complexes, and two additional CO order of SR < PR < NR and the thioether groups were reported
bands at lower wavenumbers (∼1830 cm−1 and ∼1790 cm−1
to be much weaker donors than their corresponding
depend on the donor sets of ligands SSS, PPP, PSS and PPS phosphines.
)
(Table 3). For complexes 11 and 12, which exhibit more sulfur
Electrochemical investigations on the Mo compounds fur-
than phosphorus atoms, the band at lower wavenumbers is thermore support this finding. Cyclic voltammograms of com-
most intense.
However, if the number of phosphorus atoms increases in couple,39 with gradually increasing redox couples from
plexes 11–14 in CH3CN reveal a fully reversible Mo0/MoI redox
−0.130 V (11), −0.072 V (12), 0.068 V (13), to 0.134 V (14) vs.
the ferrocene/ferrocenium couple (Fig. 3B and Table 3).
In contrast to thioethers, which act as pure σ-donors,53,54
the phosphines herein exhibit additional π-acceptor pro-
complexes 13 and 14, the CO band at higher wavenumbers
becomes dominant with the second band becoming
shoulder. This behavior indicates a weaker π-backbonding
character of the CO ligands trans to the P-donors and hence
a
supporting the previously mentioned π-acceptor character perties55 resulting in more anodic potentials due to the
of the phosphines (Fig. 3A). The observation of weaker decreased electron density at the metal center.
π-backbonding to the CO ligands trans to the stronger
κ2/κ3-Isomerization vs. CO release
for the related compounds (L)W(CO)3 and (L)Mo(CO)3 (L = RX– The κ3-coordination mode and altered electrochemical pro-
CH2CH2–Y–CH2CH2XR; Y, X = N, S, P; R = Me, Et).32 For both perties directed our interest towards a possible κ2/κ3-isomeriza-
complexes, the basicity of the metal center decreased in the tion vs. CO release in different solvents and upon oxidation of
π-acceptor phosphine ligands is consistent with observations
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
the Mo(CO)3-moieties. Consequently, the Mo-complexes 11–14 CH3CN is most stable with IR bands at 1894, 1798 and
were studied in polar (CH3CN) and unpolar solvents (THF) and 1773 cm−1. The IR bands for complex 13 in THF (1939, 1852,
changes in coordination were examined by IR spectroscopy 1834 cm−1) and CH3CN (1930, 1850, 1836 cm−1) show compar-
(Table 3). Such data provide valuable information on the sym- able frequencies assuming a similar conformation in both sol-
metry and electronic configuration of complexes with tripodal vents. Moreover, the IR spectrum of 13 in CH3CN shows
ligands, as was reported e.g. for the hemilabile ligand binding additional frequencies at 1817 and 1807 cm−1 also suggesting
at di-iron sites comprising thiol and thioether donors.56 For the presence of an intact κ3-isomer. This finding is supported
that, the CO bands are a valuable probe to judge the binding in terms of energy by the calculations suggesting the κ3-isomer
of the ligands or even loss of CO: while complexes with a κ3- to be the most stable one, followed by formation of a κ2-isomer
binding ligand should reveal a maximum of three CO stretch- rather than CO cleavage (Fig. S5, S6 and Table S1†).
ing frequencies, complexes with a κ2-bound ligand should
Compound 14 reveals significantly shifted bands at 1938
produce a more complex IR spectrum. In contrast, less CO and 1844 cm−1 in THF that are comparable with those of 13 in
bands or decreased signal intensity should be observable THF (Fig. S7†). In contrast, the spectrum in CH3CN suggests
when CO is released. In general, the IR spectra in solution that the ligand remains bound in a κ3-configuration with com-
comprised sharper bandwidths than those in the solid state. parable IR bands as observed in the solid IR (1929 and
Whilst complex 11 with a [SSS] donor set reveals CO stretching 1838 cm−1). Such findings are further supported by DFT calcu-
frequencies at 1919, 1898, 1828 and 1780 cm−1 in THF, only lations showing the intact κ3-isomers being energetically
two bands at 1919 and 1796 cm−1 are observed in CH3CN favored (Table S1†). While the observed spectroscopic changes
(Fig. S2 and S3†). In order to rationalize the number of bands, are small, it is evident that the solvent has a noticeable influ-
a DFT calculation was performed for 11. Likewise, calculations ence on the solution structure of the complexes, which is in
have been performed for 11 as κ2-isomer with and without co- agreement with previous reports on Mo0(CO)3 trithioether
ordinated solvent, and for 11 with dissociated CO. The calcu- complexes.57 Furthermore, the observation of the additional
lations in CH3CN indicate that the κ2-isomers with and CO bands in different solvents are in accordance with DFT cal-
without coordinating CH3CN are favored and within culations and suggests a κ2/κ3-isomerization of the complexes
1.5 kcal mol−1 of each other (Table S1†). Herein, the calculated 11, 12, 13 and 14 depending on the polar or non-polar solvent
IR bands are observed at 1895, 1774 and 1772 cm−1 and 1906, system rather than CO release.
1786 and 1776 cm−1, respectively, and are in good agreement
with the experimental findings. Unfortunately, calculations
Spectroelectrochemistry
with a specific THF interaction turned out to be not feasible. In order to further investigate effects that trigger the κ2/κ3-iso-
The more complex band structure found in THF suggests that merization and CO release of tripodal ligands, we focused on
besides a κ3-isomer a second complex species is present in the alteration of the metal oxidation state as a possible
solution. Likewise, the band structure of 12 in THF differs pathway to change the coordination mode of the tripodal
clearly from the one in CH3CN (Fig. 4 and S4†). While com- ligand. We therefore performed spectroelectrochemical IR
pound 12 reveals bands at 1931, 1830 and 1801 cm−1 in measurements on complexes 11–14. Oxidation of the com-
CH3CN, comparable to the solid crystalline samples, bands at plexes 11–14 in both THF as well as in CH3CN result in signifi-
1939, 1919, 1837 and 1780 cm−1 were observed in THF, cant changes in the CO band structure. While complex 11
suggesting at least a partial formation of another unidentified shows a sharp CO band at 1919 cm−1 in CH3CN, the intensity
isomer of 12. Calculations for 12 reveal that the κ3-isomer in of this band fades in the oxidized state and an additional
band at 2037 cm−1 appears (Fig. S3†). This alteration is indica-
tive of a degradation (CO release) that is concomitant with the
oxidation of 11. While calculations indicate that formation of
the κ2-isomer is preferred, it is not unreasonable to assume
that a formed κ2-intermediate as suggested by DFT calcu-
lations can subsequently undergo CO release (Table S2†). In
THF, only a slight decrease of the CO band intensities at 1828
and 1785 cm−1 can be observed over time (Fig. S2†) likewise
suggesting degradation via an altered intermediate. While this
behavior suggests that upon oxidation CO is readily released
and is in line with previous observations on CORMs,58 it is
obvious that the solvent has a significant influence on the CO
release and the intermediates formed.
In contrast, oxidation of compound 12, dissolved in THF,
affords two new CO bands at 1969 and 1894 cm−1 (Fig. 4).
Calculations indicate that in this case, the κ3-isomer is almost
isoenergetic with the κ2-isomer with either the S or P arm
Fig. 4 IR spectrum of compound 12 in THF (black) and spectrocelec-
trochemical IR of 12 upon oxidation at −0.1 V vs. Fc/Fc+ (red).
being uncoordinated, as well as with the κ3-complex with
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Table 4 Catalytic conversion of N,N-dimethylbenzamide to N,N-di-
methylbenzylamine with 2 mol% Fe3(CO)12, Ph2SiH2 and in the pres-
ence/absence of compounds SSS, PSS, PPS or PPP. Reactions were per-
formed in toluene under reflux conditions and yields were determined
by GC-MS using an external authentic standard. Values were determined
in triplicates
uncoordinated CO (Table S2†). In contrast, oxidation of 12 in
CH3CN does not lead to an altered CO band structure. The
decrease in intensity also suggests a degradation of 12 upon
oxidation in CH3CN, but significantly slower compared to 11
(Fig. S4†). When complex 13 is oxidized in THF, the intensity
of the bands at 1939, 1852 and 1833 cm−1 decreases drastically
and a new band at 2019 cm−1 is formed (Fig. S5†). Likewise,
similar behavior is observed for the oxidation of complex 13 in
CH3CN (Fig. S6†). Again calculations suggest that the energy
differences between κ3, κ2 and CO uncoordinated species are
small and likewise indicate the formation of a complex
Yield of N,N-dimethylbenzylamine [%]
Complex
Ligand
After 1 h
After 3 h
After 16 h
Fe3(CO)12
Fe3(CO)12
Fe3(CO)12
—
14.1 2.1
26.3 2.2
4.7 1.2
4.8 0.9
10.9 0.9
24.4 0.5
35.8 0.4
14.5 1.3
12.2 0.8
17.0 0.1
100
100
30.7 1.8
28.0 1.4
90.2 0.8
SSS
PSS
PPS
PPP
mixture with all three species being present (Table S2†). Fe3(CO)12
Fe3(CO)12
During the oxidation of complex 14 in THF, the intensities of
the CO bands at 1938 and 1844 cm−1 decrease and an
additional band at 2021 cm−1 is observed (Fig. S7†).
Unfortunately, compound 14 in its oxidized form could not be
facilitated by Fe3(CO)12. Unfortunately we were not able to
isolate the corresponding Fe-complexes and thus performed
the reactions in mixtures of 2 mol% Fe3(CO)12 and the ligands
SSS, PPP, PSS or PPS. However, the alteration of the ligand set
is expected to result in a different catalytic performance.
investigated in CH3CN due to its very low solubility; calcu-
lations indicate that the κ3-complex is most stable, that is in
line with the stabilizing effect of the P atoms (Table S2†). The
altered number of CO stretching vibrations for the 11–14
suggests a change of the ligands’ coordination mode and all
complexes 11, 12, 13 and 14 show a ligand isomerization or
CO release/decomposition upon oxidation in polar solvents
resulting in different CO bands. However, the rate as well as
the pathway of decomposition/CO release is depending on the
number of P atoms in the ligand.
Although showing a similar trend for isomerization/CO
release, it is obvious from the altered CO shifts and electro-
chemical redox potentials for 11–14 that the binding strength
of the different ligands SSS, PPP, PSS and PPS towards the
central metal must be different. Such observations thus show
that the complexes’ ability to perform ligand isomerization
(most likely by adopting either κ2/κ3-bound ligands) as well as
CO release is controlled by the Triphos-based ligand systems.
Such properties can be easily modified by altering the donor
sets of the tripodal ligands and render the complexes prone to
κ2/κ3-isomerization or CO release/decomposition.
The respective conversion and yields were determined by
GC-MS after 1, 3 and 16 hours (Fig. S8–S11†). While in the
absence of any supporting ligand, 14.1 and 24.4% N,N-di-
methylbenzylamine were obtained after 1 and 3 hours, respect-
ively, full conversion was obtained after 16 hours (Table 4).
Notably, in the presence of SSS, the yield of N,N-dimethyl-
benzylamine is significantly increased after 1 hour reaction
time as compared to experiments in the absence of any ligand.
After 1 hour already a yield of 26.3% of N,N-dimethyl-
benzylamine is obtained. After 3 hours the yield is further
increased to 35.8% and after 16 hours it reaches 100%. While
we are not able to give mechanistic details, this example
clearly shows that in the presence of compound SSS, the cata-
lytic conversion is significantly faster than in the absence of
any ligand, which is also notable by the increased turnover fre-
quency (TOF) (1.1 × 10−3 s−1 for Fe3(CO)12 alone and 1.7 × 10−3 s−1
for Fe3(CO)12 + SSS determined after 3 hours reaction time
(Table 5)). While applying the ligand SSS has a beneficial
effect on the catalytic hydrosilylation, application of the more
phosphorus containing ligands PPP, PSS and PPS considerably
slow down the reaction. In the presence of compounds PSS
and PPS under the same experimental conditions as used for
compound SSS, even after 16 hours the desired reduction
Catalysis
As a result of the different metal-binding strength and the
observed binding isomerization of ligands SSS, PPP, PSS and
PPS in FeII-,24 CoII-23 as well as the herein reported Mo0-com-
plexes, we expected significant alterations also in catalytic
applications. As a proof-of-principle study, we also attempted
to show the influence of the different donor sets on catalytic
applications. While the Mo-complexes 11–14 were never
reported for any catalytic application and we did not manage
to find any catalytic application for such complexes, Ru-, Rh-
Table 5 Turnover numbers (TON) and frequencies (TOF) for the con-
version of N,N-dimethylbenzamide to N,N-dimethylbenzylamine with
and Fe-complexes based on the ligand PPP and its Si-derivative 2 mol% Fe3(CO)12, Ph2SiH2 and in the presence/absence of compounds
are frequently reported catalysts in literature.1,2 A prominent
SSS, PSS, PPS or PPP after 3 hours reaction time
example herein is the reduction of N,N-dimethylbenzamide to
Complex
Ligand
TON
TOF [s−1
]
N,N-dimethylbenzylamine by iron carbonyls.59,60 Since the
alteration of the substitution pattern clearly showed a differ-
ence on the solution stability as well as ligand isomerization
and CO release, we thus investigated the reduction of N,N-di-
methylbenzamide to N,N-dimethylbenzylamine with Ph2SiH2
Fe3(CO)12
Fe3(CO)12
Fe3(CO)12
Fe3(CO)12
Fe3(CO)12
—
12.2
17.9
7.2
6.1
8.5
1.11 × 10−3
1.66 × 10−3
0.67 × 10−3
0.57 × 10−3
0.79 × 10−3
SSS
PSS
PPS
PPP
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
product was solely obtained in 30.7 and 28.0% yield, respect-
ively (Table 4).
Experimental
General
Notably, the higher the phosphorus content, the lower was
the maximum yield for N,N-dimethylbenzylamine obtained All reactions were performed under a dry Ar atmosphere
after 16 hours. One might assume that the potentially more by using standard Schlenk techniques or by working in a
rigid binding of the ligand to the metal center as well as an Glovebox. Starting materials were obtained from commercial
increased steric demand (PPP > PPS > PSS > SSS) will suppress suppliers and used without further purification. Prior to use,
the coordination of the substrate. However, although not as all solvents were dried and degassed according to standard
effective as in the presence of SSS, reactions performed in the methods. Mo(CO)3(toluene), compounds SSS,44,61 746 and 1020
62
presence of compound PPP reveal an enhanced catalytic reac- as well as [Ni(CH3CN)6](BF4)2 were synthesized according to
tion and afford
a comparable catalytic performance as literature-known procedures.
Fe3(CO)12 itself (Table 4). It is thus currently unclear if the
different binding affinity of the ligands will either change the
metal complexes involved within catalysis or the substrate
binding towards the complexes. Turnover numbers (TON) and
(2-Methyl-2-((phenylthio)methyl)propane-1,3-diyl)-
bis-(phenylsulfane) SSS
A
suspension of 1,1,1-tris(chloro-methyl)ethane (1) (3 g,
frequencies (TOF) decrease according to SSS > PPP > PSS > PPS 17.09 mmol) and KI (8.51 g, 51.27 mmol) in 40 mL DMF was
(Table 5). Whilst we are not able to examine the reaction mech- cooled to 0 °C. Subsequently, thiophenol (5.23 mL, 5.65 g,
anism for the hydrosilylation, we demonstrate, for the first 51.27 mmol) was slowly added to a suspension of NaH (60% in
time, that the different coordination of the tripodal SSS, PPP, mineral oil, 2.05 g, 51.27 mmol) in 20 mL DMF. The thiolate
PSS and PPS ligands influences the catalytic hydrosilylation of suspension was then added dropwise to the reaction mixture
N,N-dimethylbenzylamine.
containing 1 and KI. The reaction was heated at 50 °C for 2 d,
followed by the addition of 160 mL ethyl acetate. The organic
phase was extracted with 2 × 200 mL 1 M NaOH/brine (1 : 1)
and dried over anhydrous Na2SO4. The solvent was removed in
vacuum and the oily residue purified by column chromato-
Conclusion
While the tripodal ligand Triphos is commonly used in graphy (hexanes/CHCl3, 10 : 3) to yield SSS as a colorless oil
1
catalysis,1 derivatives with altered donor sets were sparsely (3.86 g, 57%). H NMR (200 MHz, CDCl3): δ 1.14 (s, 3H, CH3),
reported.26,29,35–38 We herein describe a simple and straight- 3.13 (s, 6H, CH2), 7.07–7.32 (m, 15H, Ar). 13C{1H} NMR
forward two-step synthesis of Triphos homologs comprising (50 MHz, CDCl3): δ 23.98 (s, CH3), 41.60 (s, Cq), 43.69 (s, CH2),
mixed [PSS], [PPS] donor sets. Furthermore, we herein show, 126.22 (s, Ar), 128.99 (s, Ar), 129.79 (s, Ar), 137.09 (s, Ar).
based on their respective NiII-, FeII- and Mo0-complexes that IR (cm−1): 3074, 2961, 2912, 1584, 1481, 1438, 1405, 1156,
they comprise differences in their metal binding capability 1089, 854, 748, 702.
and compare them with the literature known compounds
SSS and PPP. In addition, spectroelectrochemical IR
measurements were performed and revealed an altered
(2-Methyl-3-(phenylthio)-2-((phenylthio)methyl)propyl)-
diphenyl-phosphane PSS
tendency of ligands with mixed S/P donor sets to perform a KOtBu (0.38 g, 3.42 mmol) was suspended in 20 mL DMSO
κ2/κ3-isomerization as well as CO release, triggered by oxi- followed by the addition of HPPh2 (0.60 mL, 3.42 mmol) to
dation of the respective Mo0-complex. The experimental find- afford a red colored solution. The mixture was stirred for
ings are supported by DFT calculations and show a pro- 20 min at room temperature and added to a solution of
nounced tendency of the Mo0-complexes 11–14 to perform (2-(chloromethyl)-2-methylpropane-1,3-diyl)bis(phenylsulfane) (4)
κ2/κ3-isomerization upon oxidation prior to CO release. (1.00 g, 3.11 mmol). The reaction was stirred for 1 d at 100 °C
Based on the different solvent and oxidation behavior of while the solution turned colorless. Subsequently, the solvent
complexes 11–14, we expected an altered influence when was removed and the residue purified by column chromato-
using the ligands in catalytic applications. We subsequently graphy (hexane/Et2O, 9 : 1) to yield compound PSS (1.43 g,
tested the possibility to apply compounds SSS, PPP, PSS and 97%) as a colorless oil. 1H NMR (200 MHz, CDCl3): δ 1.09
PPS as ligands for Fe3(CO)12 in the hydrosilylation of N,N-di- (s, 3H, CH3), 2.44 (s, 2H, CH2PPh2), 3.18 (s, 4H, CH2SPh),
methylbenzamide to N,N-dimethylbenzylamine with Ph2SiH2. 7.10–7.47 (m, 20H, Ar). 31P{1H} NMR (101 MHz, CDCl3):
Clearly, the subtle exchange of the substituents has an effect δ −25.58 (s, –PPh2). 13C{1H} NMR (50 MHz, CDCl3): δ 25.69
on this reduction. While we solely presented one catalysis (d, J = 9.3 Hz), 39.42 (d, J = 15.6 Hz), 40.68 (d, J = 14.4 Hz),
example showing different reaction behavior, it is obvious 45.42 (d, J = 9.7 Hz), 110.67–139.04 (m, Ar). IR (cm−1): 3054,
that literature reported catalytic systems based on Triphos 3013, 2960, 2913, 1953, 1877, 1806, 1580, 1477, 1431, 736, 690.
(PPP) can be further altered in their efficiency and reactivity
(2-Methyl-2-((phenylthio)methyl)propane-1,3-diyl)-
bis(diphenylphosphane) PPS
for numerous processes when altering the donor sets. We
expect these findings to further stimulate the application of
compounds SSS, PPP, PSS and PPS in future catalytic HPPh2 (3.10 mL, 3.31 g, 17.78 mmol) was added dropwise to a
applications.
suspension of KOtBu (2.00 g, 17.78 mmol) in 20 mL DMSO to
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
produce an intense red colored solution. The mixture was NaH (60% in mineral oil, 1.38 g, 34.50 mmol) suspended in
stirred at room temperature for 20 min and added to (3-chloro- 15 mL DMF. The obtained thiolate suspension was then added
2-(chloromethyl)-2-methylpropyl)-(phenyl)sulfane (5) (2.10 g, dropwise to the mixture containing KI and 1. The reaction
8.47 mmol) in 30 mL DMSO. The reaction was heated at mixture was kept at 50 °C for 2 d. Subsequently, 160 mL ethyl
100 °C for 1 d and subsequently the solvent was removed in acetate were added to the reaction mixture, the organic phase
vacuum. The oily crude product was purified by column was extracted with 2 × 200 mL 1 M NaOH/brine (1 : 1) and
chromatography (hexane/diethyl ether, 9 : 1) to yield PPS as a dried over anhydrous Na2SO4. The solvent was removed in
colorless oily product (3.75 g, 80%). 1H NMR (200 MHz, vacuum and the residue purified by column chromatography
CDCl3): δ 1.04 (s, 3H, CH3), 2.47 (s, 4H, CH2PPh2), 3.18 (s, 2H, (hexane/CHCl3, 10 : 3) to yield compound 4 as colorless oily
CH2SPh), 7.22–7.51 (m, 25H, Ar). 13C{1H} NMR (50 MHz, product (2.30 g, 42%). 1H NMR (250 MHz, CDCl3): δ 1.07
CDCl3): δ 27.73 (t, J = 8.3 Hz, Cq.), 39.54 (t, J = 13.8 Hz), 41.29 (s, 3H, CH3), 3.04 (s, 4H, CH2), 3.55 (s, 2H, CH2Cl), 7.05–7.31
(q, J = 8.8 Hz), 47.00 (t, J = 10.6 Hz), 125.59–139.85 (m, Ar). (m, 10H, Ar). 13C{1H} NMR (63 MHz, CDCl3): δ 22.32 (s, CH3),
31P{1H} NMR (101 MHz, CDCl3): δ −25.54 (s, PPh2).
41.90 (s, Cq.), 42.33 (s, CH2SPh), 51.47 (s, CH2Cl), 126.51
(s, Ar), 129.10 (s, Ar), 130.10 (s, Ar), 136.77 (s, Ar). IR (cm−1): 3058,
2965, 1583, 1480, 1438, 1376, 1301, 1089, 1067, 1025, 735, 689.
(3-Chloro-2-(chloromethyl)-2-methylpropyl)-
diphenylphosphane 2
(3-Chloro-2-(chloromethyl)-2-methylpropyl)(phenyl)sulfane 5
HPPh2 (1.00 mL, 1.06 g, 5.70 mmol) was added dropwise to
a suspension of KOtBu (0.64 g, 5.70 mmol) in 20 mL DMSO Compound 5 was synthesized according to the procedure
to produce an intense red colored solution. The mixture described for compound 4. When compound 1 (2.5 g,
was stirred at room temperature for 30 min and added 14.25 mmol), KI (2.37 g, 14.25 mmol), thiophenol (1.45 mL,
to 1,3-dichloro-2-(chloromethyl)-2-methylpropane (1) (1.00 g, 1.57 g, 14.24 mmol) and NaH (60% in mineral oil, 0.57 g,
5.70 mmol) in 60 mL DMSO. The reaction was heated at 14.24 mmol) reacted under otherwise identical conditions,
100 °C for 1 d and subsequently the solvent was removed in compound 4 was obtained as a colorless oil (1.94 g, 55%)
vacuum. The oily crude product was purified by column 1H NMR (200 MHz, CDCl3): δ 1.15 (s, 3H, CH3), 3.08 (s, 2H,
chromatography (hexane/toluene, 2 : 1) to yield 2 as colorless CH2SPh), 3.59 (d, 4H, J = 2.5 Hz, CH2Cl), 7.14–7.43 (m, 5H).
1
oily product (1.33 g, 72%). H NMR (200 MHz, CDCl3): δ 1.03 13C{1H} NMR (50 MHz, CDCl3): δ 20.76 (s, CH3), 41.01 (s, Cq.),
(s, 3H; CH3), 2.27 (d, J = 3.98 Hz, 2H, CH2PPh2), 3.58 (q, J = 42.06 (s, CH2S), 49.88 (s, CH2Cl), 126.80 (s, Ar), 129.20 (s, Ar),
11.03 Hz, 4H, CH2Cl), 7.12–7.50 (m, Ar). 13C{1H} NMR 130.38 (s, Ar), 136.44 (s, Ar). IR (cm−1): 3075, 3059, 2970, 2951,
(50 MHz, CDCl3): δ 22.15 (d, J = 9.92 Hz, CH3), 35.87 (d, J = 1583, 1480, 1459, 1438, 1409, 1300, 1089, 1025, 772, 702.
17.61 Hz, Cq.), 40.68 (d, J = 14.28 Hz, CH2PPh2), 51.47 (d, 10.76
[Ni(PPP)(CH3CN)3](BF4)2 7
Hz, CH2Cl), 128.63–138.33 (m, Ar). 31P{1H} NMR (100 MHz,
CDCl3): δ −26.24 (s, PPh2).
Compound 7 was synthesized following a modified literature
reported procedure.46 Compound PPP (100 mg, 0.16 mmol)
was dissolved in 4 mL CH3CN followed by the addition of
[Ni(CH3CN)6](BF4)2 (70 mg, 0.16 mmol) dissolved in 2 mL
(2-(Chloromethyl)-2-methylpropane-1,3-diyl)
bis(diphenyl-phosphane) 3
KOtBu (1.28 g, 11.40 mmol) was suspended in 20 mL DMSO CH3CN and stirred for 24 h at room temperature. The resulting
followed by the addition of HPPh2 (2.00 mL, 2.12 g, dark red solution was evaporated to dryness and the residue
11.40 mmol) to afford a red colored solution. After stirring at then dissolved in CH3CN. Crystals of 7 were obtained by slow
room temperature for 15 min the solution was added slowly to diffusion of diethyl ether into the CH3CN solution of 7. The
a mixture of 1,3-dichloro-2-(chloromethyl)-2-methylpropane (1) dark red crystals obtained were filtered off, washed with
(1.00 g, 5.70 mmol) in 60 mL DMSO. The mixture was stirred hexane and diethyl ether and dried in a vacuum to yield
at 100 °C for 1 d and the solvent was removed in vacuum. The 137 mg (86%) of compound 7. ESI-MS. Calcd for [C41H39NiP3 +
obtained oily residue was purified by column chromatography H2O]+: 701.4. Found: 701.1. IR (KBr, cm−1): 3053, 1483, 1435,
(hexane/toluene, 1 : 1) to yield colorless oily 3 (0.65 g, 24%). 1284, 1121, 1060, 993, 797, 744, 695. UV/vis: 422 nm (ε = 1490
1H NMR (200 MHz, CDCl3): δ 0.98 (s, 3H; CH3), 2.35 (s, 4H, L mol−1 cm−1), 728 nm (ε = 460 L mol−1 cm−1).
CH2PPh2), 3.68 (s, 2H, CH2Cl), 7.15–7.46 (m, Ar). 13C{1H} NMR
[Ni(PPS)(CH3CN)2](BF4)2 8
(50 MHz, CDCl3): δ 21.82 (s, CH3), 26.05 (t, J = 9.31 Hz, Cq.),
39.99 (m, J = 8.91 Hz, CH2PPh2), 55.29 (m, 11.65 Hz, CH2Cl), Compound PPS (23 mg, 0.042 mmol) was dissolved in CH3CN
125.68–139.34 (m, Ar). 31P{1H} NMR (100 MHz, CDCl3): followed by the addition of Ni(CH3CN)6(BF4)2 (20 mg,
δ −25.60 (s, PPh2).
0.042 mmol). The resulting dark red solution was stirred at
room temperature overnight. Subsequently, the reaction
mixture was filtered and the solvent removed. The solid was
recrystallized from CH3CN and diethyl ether to afford complex
(2-(Chloromethyl)-2-methylpropane-1,3-diyl)bis-
(phenyl-sulfane) 4
A suspension of compound 1 (3 g, 17.24 mmol) and KI (5.73 g, 6 as red crystalline solid (20 mg, 55%). 1H NMR (200 MHz,
34.50 mmol) in 40 mL DMF was cooled to 0 °C. Subsequently, CDCl3): δ 1.13 (s, 3H, CH3), 2.49 (s, 4H, CH2PPh2), 2.82 (s, 4H,
thiophenol (3.60 mL, 3.80 g, 34.50 mmol) was slowly added to CH2SPh), 7.48 (m, Ar). 13C{1H} NMR (50 MHz, CDCl3): δ 15.19
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
(s, Cq.), 25.18 (s, CH3), 38.42 (s, CH2PPh2), 47.88 (s, CH2SPh), was stirred for 24 h at room temperature. Subsequently, the
57.35 (s, CH3), 180.35–189.82 (m, Ar). 31P{1H} NMR (100 MHz, solvent was removed and the obtained residue dissolved in
CDCl3): δ 15.20 (s, PPh2). IR(cm−1): 3067, 2989, 2930, 2339, THF. Addition of hexane led to precipitation of crude 13,
1427, 1048, 734, 699. UV/vis: 428 nm (ε = 532 L mol−1 cm−1). which was then recrystallized by diffusion of THF into a
Calcd for [C39H40N2NiP2S]: C 67.94%; H 5.85%; N 4.06%; hexane solution (1.26 g, 95%). 1H NMR (200 MHz, CDCl3):
S 4.65%. Found: C 67.2%; H 5.1%; N 4.5%; S 3.9%.
δ 1.13 (s, 3H, CH3), 2.36 (s, 4H, CH2PPh2), 2.75 (s, 2H,
CH2SPh), 7.19–7.53 (m, Ar). 13C{1H} NMR (50 MHz, CDCl3):
δ 12.68 (s, Cq.), 7.00 (s, CH3), 49.33 (s, CH2SPh), 53.49
[Fe(PPS)Cl2] 9
Compound PPS (50 mg, 0.0912 mmol) was suspended in 2 mL (s, CH2PPh2), 172.88–191.23 (m, Ar). 31P{1H} NMR (100 MHz,
THF and FeCl2 (12 mg, 0.0912 mmol) was added in one CDCl3): δ 18.08 (s, PPh2). IR (cm−1): 3054, 2972, 2924, 2859,
portion. The formed dark red solution was then stirred at 1925, 1829, 1818, 1474, 1432, 1090, 746, 693. Calcd for
room temperature overnight. The reaction mixture was filtered [C38H34MoO3P2S]: C 62.64%; H 4.70%; S 4.40%. Found:
and the solvent removed in vacuum. The resulting dark red C 62.6%; H 5.0%; S 4.1%.
solid was recrystallized from THF by diethyl ether diffusion.
[Mo(PPP)(CO)3] 14
The crystals were separated and dried to afford 34 mg (55%) of
complex 9. 13C{1H} NMR (50 MHz, CDCl3): δ −18.36 (s, CH3),
−14.57, −12.68, 6.25, 100.89–104.65 (Ar). IR(cm−1): 3053, 2922,
1580, 1478, 1432, 1237, 1184, 1093, 1024, 991, 822, 738, 691.
Calcd for [C35H34FeP2SCl2]: C 62.24%; H 5.07%; S 4.75. Found:
C 62.1%; H 5.0%; S 4.2%.
Compound PPP (1.00 g, 1.60 mmol) was dissolved in 10 mL
THF followed by the addition of Mo(CO)3(toluene) (435 mg,
1.60 mmol) dissolved in 5 mL THF. The obtained yellow solu-
tion was stirred at room temperature for additional 24 h. The
solvent was removed in vacuum and the obtained residue dis-
solved in THF. Crude complex 14 was precipitated by addition
of hexane. The product was then recrystallized by diffusion of
THF into a hexane solution of 14 and afforded 515 mg (40%)
of the desired product as yellow crystalline solid. 1H NMR
(200 MHz, CDCl3): δ 1.44 (s, 3H; CH3), 2.41 (s, 6H, CH2PPh),
7.07–7.35 (m, 30H, Ar). 31P{1H} NMR (100 MHz, CDCl3):
[Mo(SSS)(CO)3] 11
Compound SSS (150 mg, 0.38 mmol) was dissolved in 4 mL
THF followed by the addition of Mo(CO)3(toluene) (102 mg,
0.38 mmol) in 2 mL dry THF. The obtained light yellow solu-
tion was stirred at room temperature. After 24 h, the solvent
was reduced to ∼1/5 of its original volume and maintained at
0 °C to obtain yellow crystals. The obtained crystals were then
washed with diethyl ether and dried in vacuum to afford com-
pound 11 as bright yellow crystalline compound (173 mg,
δ
18.32 (s, PPh2). IR (cm−1): 3078, 3042, 2948, 2894,
2859, 1925, 1831, 1803, 1480, 1432, 1084, 740, 687. Calcd
for [C26H24MoO3S3]: C 65.68%; H 4.89%. Found: C 65.6%;
H 4.4%.
1
79%). H NMR (CDCl3, 200 MHz): δ 1.11 (s, 3H, CH3), 3.10 (s,
6H, CH2SPh), 7.12–7.54 (m, Ar). 13C{1H} NMR (50 MHz, Catalytic reduction of N,N-dimethylbenzamide to
CDCl3): δ 8.41 (s, Cq.), 25.63 (s, CH3), 43.65 (s, CH2SPh), N,N-dimethylbenzylamine
126.14–137.00 (m, Ar). IR (cm−1): 3060, 2966, 2924, 2865,
Fe3(CO)12 (2 mol%) and the respective ligand (2 mol%) were
dissolved in 8 mL of toluene. After 5 min of stirring at room
temperature, N,N-dimethylbenzamide (100 mg, 0.67 mmol)
and 2 equiv. diphenylsilane were added. The reaction mixture
1925, 1821, 1791, 1439, 1065, 1013, 740, 687. Calcd for
[C26H24MoO3S3]: C 54.16%; H 4.20%; S 16.68%. Found:
C 53.7%; H 4.1%; S 16.2%.
was stirred for 16 h at 100 °C. The conversion to N,N-dimethyl-
[Mo(PSS)(CO)3] 12
benzylamine was detected via GC-MS and yields determined
Compound PSS (50 mg, 0.106 mmol) was dissolved in 6 mL
by external calibration with N,N-dimethylbenzylamine.
THF followed by the addition of Mo(CO)3(toluene) (28 mg,
0.106 mmol) in 6 mL THF. The mixture turned dark yellow and
was stirred overnight. The reaction mixture was filtered and
the solvent removed in vacuum. The yellow solid was recrystal-
Acknowledgements
lized from THF by diethyl ether diffusion. The obtained
crystals were washed with diethyl ether and dried in vacuum
to afford complex 12 as crystalline solid (20 mg, 28%). 31P{1H}
NMR (100 MHz, CDCl3): δ 18.50 (s, PPh2). IR (cm−1): 3054,
2960, 2924, 2859, 1926, 1837, 1788, 1432, 1090, 729, 693. Calcd
for [C32H29MoO3PS2]: C 58.89%; H 4.48%; S 9.82%. Found:
C 58.6%; H 4.22%, S 9.54%.
This work was supported by the Fonds der Chemischen
Industrie (Liebig grant to U.-P. A.) and through the Deutsche
Forschungsgemeinschaft (Emmy Noether grant to U.-P. A.,
AP242/2-1) as well as through the Alexander von Humboldt
Foundation (Humboldt Research Fellowship to R. Miller). We
thank M. Heller and Dr B. Mallick for valuable support.
[Mo(PPS)(CO)3] 13
Notes and references
Compound PPS (1.00 g, 1.82 mmol) was dissolved in 10 mL
THF followed by the addition of Mo(CO)3(toluene) (496 mg,
1.82 mmol) in 5 mL dry THF. The formed light yellow solution
1 A. Phanopoulos, P. W. Miller and N. J. Long, Coord. Chem.
Rev., 2015, 299, 39–60.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
2 A. Petuker, M. L. Reback and U.-P. Apfel, Eur. J. Inorg. 25 J. Krahmer, H. Broda, C. Naether, G. Peters, W. Thimm and
Chem., 2017, DOI: 10.1002/ejic.201700388. F. Tuczek, Eur. J. Inorg. Chem., 2011, 4377–4386.
3 B. J. Sarmah and D. K. Dutta, J. Organomet. Chem., 2010, 26 S.-T. Liu, H.-E. Wang, M.-C. Cheng and S.-M. Peng,
695, 781–785. J. Organomet. Chem., 1989, 376, 333–342.
4 F. M. A. Geilen, B. Engendahl, M. Hölscher, 27 S. T. Liu and K. J. Liu, Inorg. Chem., 1990, 29, 4576–
J. Klankermayer and W. Leitner, J. Am. Chem. Soc., 2011,
133, 14349–14358.
5 S. Imm, S. Bähn, M. Zhang, L. Neubert, H. Neumann,
4579.
28 H. Heidel, G. Huttner and G. Helchem, Z. Naturforsch.,
B: J. Chem. Sci., 1993, 48, 1681–1692.
F. Klasovsky, J. Pfeffer, T. Haas and M. Beller, Angew. 29 W. Hsin-Ell, C. Ming-Chu, L. Gene-Hsiang, P. Shie-Ming
Chem., Int. Ed., 2011, 50, 7599–7603.
and L. Shiuh-Tzung, J. Organomet. Chem., 1993, 445, 171–
179.
30 H. Heidel, J. Scherer, A. Asam, G. Huttner, O. Walter and
L. Zsolnai, Chem. Ber., 1995, 128, 293–301.
6 K. Beydoun, G. Ghattas, K. Thenert, J. Klankermayer and
W. Leitner, Angew. Chem., Int. Ed., 2014, 53, 11010–11014.
7 T. vom Stein, M. Meuresch, D. Limper, M. Schmitz,
M. Hölscher, J. Coetzee, D. J. Cole-Hamilton, 31 H. Heidel, G. Huttner and L. Zsolnai, Z. Naturforsch.,
J. Klankermayer and W. Leitner, J. Am. Chem. Soc., 2014,
136, 13217–13225.
B: J. Chem. Sci., 1995, 50, 729–734.
32 O. P. Siclovan and R. J. Angelici, Inorg. Chem., 1998, 432–
8 S. Wesselbaum, V. Moha, M. Meuresch, S. Brosinski,
444.
K. M. Thenert, J. Kothe, T. vom Stein, U. Englert, 33 R. Soltek, G. Huttner, L. Zsolnai and A. Driess, Inorg. Chim.
M. Holscher, J. Klankermayer and W. Leitner, Chem. Sci.,
2015, 6, 693–704.
9 E. J. Derrah, M. Hanauer, P. N. Plessow, M. Schelwies,
M. K. da Silva and T. Schaub, Organometallics, 2015, 34,
1872–1881.
10 M. Meuresch, S. Westhues, W. Leitner and J. Klankermayer,
Angew. Chem., Int. Ed., 2016, 55, 1392–1395.
11 B. Erb, E. Risto, T. Wendling and L. J. Gooßen,
ChemSusChem, 2016, 9, 1442–1448.
Acta, 1998, 269, 143–156.
34 H. W. Yim, L. M. Tran, E. E. Pullen, D. Rabinovich,
L. M. Liable-Sands, T. E. Concolino and A. L. Rheingold,
Inorg. Chem., 1999, 38, 6234–6239.
35 S. Batke, T. Kothe, M. Haas, H. Wadepohl and J. Ballmann,
Dalton Trans., 2016, 45, 3528–3540.
36 A. Muth, O. Walter, G. Huttner, A. Asam, L. Zsolnai and
C. Emmerich, J. Organomet. Chem., 1994, 468, 149–163.
37 V. Körner, G. Huttner, L. Zsolnai, M. Büchner, A. Jacobi
and D. Günauer, Chem. Ber., 1996, 129, 1587–1601.
12 I. Mellone, M. Peruzzini, L. Rosi, D. Mellmann, H. Junge,
M. Beller and L. Gonsalvi, Dalton Trans., 2013, 42, 2495– 38 S. Beyreuther, A. Frick, J. Hunger, G. Huttner,
2501.
B. Antelmann, P. Schober and R. Soltek, Eur. J. Inorg.
Chem., 2000, 597–615.
13 C. Bianchini, A. Meli, M. Peruzzini, F. Vizza, Y. Fujiwara,
T. Jintoku and H. Taniguchi, J. Chem. Soc., Chem. 39 M. A. Fox, K. A. Campbell and E. P. Kyba, Inorg. Chem.,
Commun., 1988, 299–301. 1981, 20, 4163–4165.
14 M. Rosales, G. Chacon, A. Gonzalez, I. Pacheco, 40 T. Maina, A. Pecorale, A. Dolmella, G. Bandoli and
P. J. Baricelli and L. G. Melean, J. Mol. Catal., 2008, 287,
110–114.
15 M. Rosales, R. Vallejo, J. Jose Soto, L. Jhonatan Bastidas,
K. Molina and P. J. Baricelli, Catal. Lett., 2010, 134, 56–62.
U. Mazzi, J. Chem. Soc., Dalton Trans., 1994, 2437–2443.
41 N. Salvarese, N. Morellato, A. Venzo, F. Refosco,
A. Dolmella and C. Bolzati, Inorg. Chem., 2013, 52, 6365–
6377.
16 A. B. Chaplin and P. J. Dyson, Eur. J. Inorg. Chem., 2007, 42 C. Strohmann, S. Lüdtke and O. Ulbrich, Organometallics,
4973–4979. 2000, 19, 4223–4227.
17 A. B. Chaplin and P. J. Dyson, Inorg. Chem., 2008, 47, 381– 43 P. K. Dornan, P. L. Leung and V. M. Dong, Tetrahedron,
390. 2011, 67, 4378–4384.
18 H. Broda, J. Krahmer and F. Tuczek, Eur. J. Inorg. Chem., 44 S. Herold, A. Mezzetti, L. M. Venanzi, A. Albinati, F. Lianza,
2014, 3564–3571.
T. Gerfin and V. Gramlich, Inorg. Chim. Acta, 1995, 235,
215–231.
45 E. N. Tsvetkov, N. A. Bondarenko, I. G. Malakhova and
M. I. Kabachnik, Synthesis, 1986, 198–208.
46 A. Petuker, S. Mebs, N. Schuth, P. Gerschel, M. L. Reback,
B. Mallick, M. van Gastel, M. Haumann and U.-P. Apfel,
Dalton Trans., 2017, 46, 907–917.
19 H. Broda, S. Hinrichsen, J. Krahmer, C. Nather and
F. Tuczek, Dalton Trans., 2014, 43, 2007–2012.
20 T. J. Korstanje, J. Ivar van der Vlugt, C. J. Elsevier and B. de
Bruin, Science, 2015, 350, 298–302.
21 J. Schneidewind, R. Adam, W. Baumann, R. Jackstell and
M. Beller, Angew. Chem., Int. Ed., 2017, 56, 1890–1893.
22 L. Söncksen, C. Gradert, J. Krahmer, C. Näther and 47 E. J. Hawrelak, W. H. Bernskoetter, E. Lobkovsky,
F. Tuczek, Inorg. Chem., 2013, 52, 6576–6589.
23 K. Heinze, G. Huttner, L. Zsolnai and P. Schober, Inorg.
Chem., 1997, 36, 5457–5469.
24 A. Petuker, K. Merz, C. Merten and U.-P. Apfel, Inorg.
Chem., 2016, 1183–1191.
G. T. Yee, E. Bill and P. J. Chirik, Inorg. Chem., 2005, 44,
3103–3111.
48 R. Langer, F. Bönisch, L. Maser, C. Pietzonka, L. Vondung
and T. P. Zimmermann, Eur. J. Inorg. Chem., 2015, 2015,
141–148.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
49 A. Barton, J. Connolly, W. Levason, A. Mendia-Jalon, 57 M. C. Durrant, B. Goerdt, C. Hauser, T. Krawinkel and
S. Orchard and G. Reid, Polyhedron, 2000, 19, 1373–
1379.
50 D. Sellmann and J. Schwarz, J. Organomet. Chem., 1983,
241, 343–361.
R. L. Richards, Transition Met. Chem., 1995, 20, 583–589.
58 J. D. Seixas, A. Mukhopadhyay, T. Santos-Silva,
L. E. Otterbein, D. J. Gallo, S. S. Rodrigues, B. H. Guerreiro,
A. M. L. Goncalves, N. Penacho, A. R. Marques,
A. C. Coelho, P. M. Reis, M. J. Romao and C. C. Romao,
Dalton Trans., 2013, 42, 5985–5998.
59 S. Zhou, K. Junge, D. Addis, S. Das and M. Beller, Angew.
Chem., Int. Ed., 2009, 48, 9507–9510.
60 B.-H. Zhu, G.-H. Xu and Z.-W. Xia, in Adv. Chem.
Engineering, PTS 1-3, ed. Y. X. Wen, and F. H. Lei, Guangxi
51 S. Dilsky, J. Organomet. Chem., 2007, 692, 2887–2896.
52 A. C. Kautz, P. C. Kunz and C. Janiak, Dalton Trans., 2016,
45, 18045–18063.
53 N. A. F. Al-Rawashdeh, S. Chatterjee, J. A. Krause and
W. B. Connick, Inorg. Chem., 2014, 53, 294–307.
54 S. V. Timofeev, O. B. Zhidkova, E. M. Mosolova, I. B. Sivaev,
I. A. Godovikov, K. Y. Suponitsky, Z. A. Starikova and
V. I. Bregadze, Dalton Trans., 2015, 44, 6449–6456.
55 P. Dyer, J. Fawcett, M. Hanton, R. Kemmitt, R. Padda and
N. Singh, Dalton Trans., 2003, 104–113.
Univ, China; Wuhan Univ Sci
&
Technol, China;
Queensland Univ Technol, Australia, 2012, vol. 396–398,
pp. 2485–2488.
61 A. Petuker, C. Merten and U.-P. Apfel, Eur. J. Inorg. Chem.,
2015, 2139–2144.
56 F. Xu, C. Tard, X. Wang, S. K. Ibrahim, D. L. Hughes,
W. Zhong, X. Zeng, Q. Luo, X. Liu and C. J. Pickett, Chem. 62 B. J. Hathaway, D. G. Holah and A. E. Underhill, J. Chem.
Commun., 2008, 606–608. Soc., 1962, 2444–2448.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017