The preparation of multimetallic complexes using sterically bulky N-centred tripodal dialkyl phosphino ligands

Article abstract of DOI:10.1016/j.jorganchem.2010.01.017

Sterically bulky monodentate and bidentate phosphines have been widely used as ligands for metal complexation and catalyst formation. Bulky tridentate phosphine ligands are however much rarer and have not been widely investigated even though they may be considered attractive ligands for coordination chemistry studies and catalysis. Here we report the synthesis of two new N-centred tripodal phosphine ligands bearing bulky cyclohexyl and tert-butyl groups. The coordination chemistry of the cyclohexyl triphosphine ligand N(CH₂PCy₂)₃ (4) was investigated and found to react with Mo and W hexacarbonyls preferentially forming bidentate metal tetracarbonyl complexes [Mo(CO)₄{N(CH₂PCy₂)₃-κ²P}] (6) and [W(CO)₄{N(CH₂PCy₂)₃-κ²P}] (7) over the expected facial capping tridentate complexes. The steric bulk of the cyclohexyl groups on the phosphorus atoms is sufficient to prevent the third arm of the ligand from coordinating and adopting the required geometry for facial coordination. This 'steric control' at the metal centre results in the third arm remaining freely available for further metal coordination. The coordination chemistry of this free phosphine arm on complexes 6 and 7 was investigated further and used to prepare a series of gold, platinum and silver multimetallic complexes. The X-ray crystal structures of the resulting mixed bi and trimetallic complexes [W(CO)₄{N(CH₂PCy₂)₃-κ²P}AuCl] (8), [[Mo(CO)₄{N(CH₂PCy₂)₃-κ²P}]₂(μ-PtCl₂)] (9) and [[W(CO)₄{N(CH₂PCy₂)₃-κ²P}]₂(μ-Ag)]ClO₄ (11) are reported.

Full text of DOI:10.1016/j.jorganchem.2010.01.017

Journal of Organometallic Chemistry 695 (2010) 1138–1145
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
The preparation of multimetallic complexes using sterically bulky N-centred
tripodal dialkyl phosphino ligands
*
Philip W. Miller , Andrew J.P. White
Department of Chemistry, Imperial College London, South Kensington, London SW7 2AZ, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Sterically bulky monodentate and bidentate phosphines have been widely used as ligands for metal com-
plexation and catalyst formation. Bulky tridentate phosphine ligands are however much rarer and have
not been widely investigated even though they may be considered attractive ligands for coordination
chemistry studies and catalysis. Here we report the synthesis of two new N-centred tripodal phosphine
ligands bearing bulky cyclohexyl and tert-butyl groups. The coordination chemistry of the cyclohexyl tri-
phosphine ligand N(CH2PCy2)3 (4) was investigated and found to react with Mo and W hexacarbonyls
Received 26 November 2009
Received in revised form 7 January 2010
Accepted 14 January 2010
Available online 21 January 2010
Keywords:
Phosphine
Triphosphine
Multimetallic
Tetracarbonyl
preferentially forming bidentate metal tetracarbonyl complexes [Mo(CO)4{N(CH2PCy2)3-
j
2P}] (6) and
[W(CO)4{N(CH2PCy2)3-
j
2P}] (7) over the expected facial capping tridentate complexes. The steric bulk
of the cyclohexyl groups on the phosphorus atoms is sufficient to prevent the third arm of the ligand from
coordinating and adopting the required geometry for facial coordination. This ‘steric control’ at the metal
centre results in the third arm remaining freely available for further metal coordination. The coordination
chemistry of this free phosphine arm on complexes 6 and 7 was investigated further and used to prepare
a series of gold, platinum and silver multimetallic complexes. The X-ray crystal structures of the resulting
mixed bi and trimetallic complexes [W(CO)4{N(CH2PCy2)3-
2P}]2( 2P}]2(
-PtCl2)] (9) and [[W(CO)4{N(CH2PCy2)3- -Ag)]ClO4 (11) are reported.
Ó 2010 Elsevier B.V. All rights reserved.
j
2P}AuCl] (8), [[Mo(CO)4{N(CH2PCy2)3-
j
l
j
l
1. Introduction
in the literature may be associated with difficulties in their synthe-
sis i.e. inefficient generation of alkylphosphides from the air sensi-
tive secondary phosphines and incomplete substitution of halide
leaving groups on the alkyl backbone which may result in poor
yields. A general, high yielding and straightforward synthetic route
to a broad range of triphos-type ligands, especially those contain-
ing dialkylphosphino groups, would therefore be of interest.
The previously reported amino bridged triphosphine ligand
tris(diphenylphosphinomethyl)amine (N(CH2PPh2)3, N-triphos, 2)
is structurally analogous to the parent triphos ligand except for a
nitrogen atom at the bridgehead instead of a C–CH3 group. Consid-
ering the similarity of these two ligands and widespread use of tri-
phos for metal complexation and catalytic studies, it is surprising
that there have been very few reports on the coordination chemis-
try or catalysis using this N-triphos ligand [25–28]. In comparison,
the more flexible tris(diphenylphosphinoethyl)amine ligand (3),
with an ethyl spacer between the bridgehead nitrogen and P
atoms, has received much more attention [3,29]. The greater length
and flexibility of the ethyl spacers of this ligand can result in the
central bridgehead N-atom competing with the phosphorus atoms
for metal coordination sites which can result in a mixture of iso-
mers. In contrast the bridgehead nitrogen atom of N-triphos may
be believed to behave in an ‘innocent’ fashion, with regards to
competition with the surrounding P atoms for metal coordination,
since the formation of a four-membered chelate ring would be
Triphosphine ligands have generated considerable interest for
the preparation of transition metal complexes and catalysts over
the past three decades [1–3]. The tripodal phosphine 1,1,1-
tris(diphenylphosphinomethyl)ethane (CH3(CH2PPh2)3, triphos, 1)
is one of the most widely studied triphosphine ligands and has
been shown to form an extensive range of metal complexes [4–
12] and a number of active catalytic species [13–18]. New tri-
phos-type ligands continue to attract interest for the preparation
of coordination complexes [19–24], however, their synthesis is
not always trivial. Triphos ligands are commonly prepared via
the nucleophilic substitution of a suitable alkyltrihalide backbone
with a dialkyl/diaryl phosphide generated from the corresponding
secondary phosphine. The necessary phosphide reagents can be
challenging to prepare and the complete substitution of all three
halide leaving groups can be difficult to achieve in high yields.
Although a range of aryl triphos derivatives have been prepared
in this way [20], relatively few alkyl triphos derivatives have ap-
peared in the literature to date even though they may be consid-
ered attractive ligands for coordination chemistry studies and
catalytic applications. The lack of alkyl triphos derivatives reported
* Corresponding author. Fax: +44 207 594 5804.
E-mail address: philip.miller@imperial.ac.uk (P.W. Miller).
0022-328X/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.01.017

P.W. Miller, A.J.P. White / Journal of Organometallic Chemistry 695 (2010) 1138–1145
1139
unfavourable. Thus the coordination chemistry of N-triphos li-
gands may be considered to behave in an analogous manner to that
of triphos. Additionally, compared to the synthetic routes used to
prepare standard triphos-type ligands, N-triphos-type ligands are
relatively straightforward to prepare and therefore present greater
opportunities for developing new triphosphine ligands with varied
diaryl or dialkyl phosphino groups. Here, we report the synthesis of
two new N-triphos ligands bearing bulky dicyclohexyl (4) and di-
tert-butyl (5) phosphino groups, prepared by a straightforward
and general synthetic route, and investigate the coordination
chemistry of the cyclohexyl ligand N(CH2PCy2)3 (4).
C39H72NP3: C, 72.30; H, 11.20; N, 2.16. Found: C, 72.39; H, 10.97;
N, 2.08%. IR (v/cmꢁ1): 1H NMR (CDCl3, 400 MHz): d 2.89 (s, br,
6H, CH2P), 1.83–1.25 (m, 66H, cyclohexyl). 31P{1H} NMR (CDCl3,
162 MHz): d ꢁ17.0 (s). 13C{1H} NMR (CDCl3, 100 MHz): d 54.8 (s,
1
2
br, CH2P), 32.7 (d, JPC = 12.4 Hz P-CHcyclohexyl), 30.0 (d, JPC
=
2
11.8 Hz CH2cyclohexyl), 29.8 (d, JPC = 11.5 Hz CH2cyclohexyl), 27.5
3
3
(d, JPC = 7.8 Hz CH2cyclohexyl), 27.3 (d, JPC = 8.8 Hz CH2cyclohexyl),
26.6 (s, CH2cyclohexyl). TOF-MS ES: m/z (%):734 (5), 718 (48), 696
(50), 680 (95), 482 (100).
2.4. Synthesis of N,N,N-tris(ditert-butylphosphinomethyl)amine (5)
2. Experimental
To a Schlenk flask was added ditert-butyl(hydroxymethyl)phos-
phonium chloride (243 mg, 1.0 mmol), ammonium chloride
(18 mg, 0.33 mmol), degassed methanol (5 ml) and triethylamine
(0.24 ml, 5 mmol). The mixture was heated under reflux overnight.
The volume was reduced to approximately one third by removal of
methanol in vacuo, which resulted in the precipitation of a white
solid. Recrystallisation of this mixture by heating and slow cooling
to room temperature resulted in the formation of colourless crys-
tals suitable for X-ray diffraction studies. Methanol was removed
via cannula filtration and the crystals dried in vacuo and stored un-
der a nitrogen atmosphere. (84 mg, 52%). Anal. Calc. for C27H60NP3:
C, 65.95; H, 12.30; N, 2.85. Found: C, 65.81 H, 12.53; N, 2.93%. IR (v/
cmꢁ1): 3054, 2951, 1472, 1421, 1366, 1062, 896, 811. 1H NMR
2.1. General considerations
All preparations were carried out using standard Schlenk line
techniques under an inert atmosphere of N2 unless otherwise sta-
ted. Solvents were dried over standard drying agents and freshly
distilled under nitrogen before use. All starting materials were of
reagent grade, purchased from either Aldrich Chemical Company
or Strem Chemicals. Aryl and alkyl (hydroxymethyl)phosphonium
chloride salts were prepared according to literature methods
[30]. Chromatographic separations were carried out on Kieselgel
60 SiO2. 1H, 13C and 31P{1H} NMR spectra were recorded on Bruker
Av-400, DRX-400, Av-500 spectrometers. Chemical shifts are re-
ported in ppm using the residual proton impurities in the solvents.
Pseudo-triplets which occur as a result of identical J-value coupling
to two chemically inequivalent nuclei are assigned as dd and are
recognised by the inclusion of only one J-value. Positive-ion FAB
and electron ionisation mass spectra were recorded on a Micro-
mass Autospec Q spectrometer using a 3-nitrobenzyl alcohol ma-
trix. Electron ionisation was carried out at 70 eV. Infrared spectra
(CDCl3, 400 MHz):
d 2.90 (s, br, 6H, CH2P), 1.18 (d, 54H,
2JPH = 12.4 Hz, tBu). 31P{1H} NMR (CDCl3, 162 MHz): d ꢁ12.5 (s).
13C{1H} NMR (CDCl3, 100 MHz): d 56.5 (d, JPC = 10.2 Hz, CH2P),
1
31.7 (d, 1JPC = 21.2 Hz, P-C(CH3)3), 30.0 (d, 2JPC = 14.0 Hz, PC(CH3)3).
TOF-MS ES: m/z (%): 581 (6), 565 (25), 549 (60), 533 (100).
2.5. Synthesis of [Mo(CO)4{N(CH2PCy2)3-j
2P}] (6)
were recorded on
a Perkin–Elmer 983G spectrophotometer
To Schlenk flask was added N(CH2PCy2)3 (360 mg, 0.56 mmol),
Mo(CO)6 (149 mg, 0.56 mmol) and DMF (5 ml). The reaction mix-
ture heated to 100 °C for 2 h during which the evolution of CO
gas was observed and colour change from colourless to light
brown. After this time the reaction was allowed to cool to ambient
temperature. White crystals suitable for X-ray diffraction formed
on standing overnight. DMF was removed via cannula filtration
and the crystals washed with diethyl ether (3 ꢀ 10 ml) and dried
in vacuo. (320 mg, 67%). Anal. Calc. for C43H72MoNO4P3: C, 60.34;
H, 8.48; N, 1.64. Found: C, 60.39; H, 8.50; N, 1.65%. IR (v/cmꢁ1):
2930, 2008 (s), 1888 (br). 1H NMR (CDCl3, 400 MHz): d 2.89 (s,
br, 4H, CH2P), 2.68 (s, br, 2H, CH2P), 2.04–1.27 (m, 66H, cyclohexyl).
31P{1H} NMR (CDCl3, 162 MHz): d 22.8 (s, 2P, PCy2-Mo), ꢁ16.7 (s,
1P, PCy2). LSIMS: m/z (%): 856 (5) [M]+, 829 (5) [Mo(CO)3-
equipped with a Perkin–Elmer 3700 data station and recorded as
a solution in dichloromethane. Elemental analyses were carried
out by Mr. Stephen Boyer of the Department of Health and Human
Sciences, London Metropolitan University. X-ray diffraction analy-
sis was carried out by Dr. Andrew White of the Department of
Chemistry at Imperial College London.
2.2. Synthesis of N,N,N-tris(diphenylphosphinomethyl)amine (2)
To a Schlenk flask was added diphenyl(hydroxymethyl)phos-
phonium chloride (7.75 g, 27.4 mmol), ammonium chloride
(0.49 g, 9.14 mmol), degassed methanol (100 ml) and triethyl-
amine (7.4 ml, 55.0 mmol). The mixture heated under reflux for
2 h, during which a white precipitate formed. The precipitate was
filtered in air and washed with methanol (3 ꢀ 30 ml). (4.5 g,
81%). Characterisation data was consistent with that reported in
the literature [25,26]. 1H NMR (CDCl3, 400 MHz): d 7.40–7.22 (m,
30 H, PPh2), 3.82 (d, 6 H, 2JPH = 3.29 Hz, CH2P). 31P{1H} NMR (CDCl3,
162 MHz): d ꢁ28.9 (s).
{N(CH2PCy2)3-
j j
2P}]+, 799 (5) [Mo(CO)2{N(CH2PCy2)3- 2P}]+, 771
(10) [Mo(CO){N(CH2PCy2)3-j2P}]+, 211 (100).
2.6. Synthesis of [W(CO)4{N(CH2PCy2)3-j
2P}] (7)
This complex was prepared by an analogous method to that de-
scribed for the Mo complex using N(CH2PCy2)3 (340 mg,
0.53 mmol) and W(CO)6 (185 mg, 0.53 mmol). (400 mg, 80%). Anal.
Calc. for C43H72WNO4P3: C, 54.72; H, 7.69; N, 1.48. Found: C, 54.77;
H, 7.78; N, 1.52%. IR (v/cmꢁ1): 3056, 2931, 2854, 2004 (s), 1878
(br), 1448, 1421, 1274, 1266, 1258, 896. 1H NMR (CDCl3,
400 MHz): d 2.97 (s, br, 4H, CH2P), 2.67 (s, br, 2H, CH2P), 2.06–
1.26 (m, 66H, cyclohexyl). 31P{1H} NMR (CDCl3, 162 MHz): d 3.2
2.3. Synthesis of N,N,N-tris(dicyclohexylphosphinomethyl)amine (4)
To
a
Schlenk flask was added dicyclohexyl(hydroxy-
methyl)phosphonium chloride (588 mg, 2.0 mmol), ammonium
chloride (35 mg, 0.67 mmol), degassed methanol (10 ml) and tri-
ethylamine (0.45 ml, 10 mmol). The mixture heated under reflux
for 2 h. A white precipitate was noticed to form almost immedi-
ately on heating to reflux. Methanol was removed via cannula fil-
tration and the white solid rinsed with methanol (2 ꢀ 10 ml) and
dried in vacuo and isolated as a white powder that was stored un-
1
(s, 2P, PCy2-W) with W satellites (d, J183W,P = 217.3 Hz), ꢁ16.9 (s,
1P, PCy2). LSIMS: m/z (%): 943 (52) [M]+, 915 (15) [W(CO)3{N-
(CH2PCy2)3-j2P}]+, 887 (5) [W(CO)2{N(CH2PCy2)3-j2P}]+, 860 (7)
der
a
nitrogen atmosphere. (360 mg, 84%). Anal. Calc. for
[W(CO){N(CH2PCy2)3-j
2P}]+, 211 (100).

1140
P.W. Miller, A.J.P. White / Journal of Organometallic Chemistry 695 (2010) 1138–1145
2.7. Synthesis of [W(CO)4{N(CH2PCy2)3-
j
2P}AuCl] (8)
2.9. Synthesis of [[W(CO)4{N(CH2PCy2)3-
j l-PtCl2)] (10)
2P}]2(
To
a
solution of
7
[W(CO)4{N(CH2PCy2)3-j2P}] (94 mg,
To a solution of 7
[Mo(CO)4{N(CH2PCy2)3-j2P}] (94 mg,
0.1 mmol) in dichloromethane (4 ml) was added [AuCl(SMe)2]
(29.5 mg, 0.1 mmol) and the reaction mixture stirred overnight.
Colourless crystals were obtained by diffusion of diethyl ether into
this solution over a period of two days. The crystals were collected
by filtration, rinsed with diethyl ether and dried in vacuo. (100 mg,
85%). Anal. Calc. for C43H72AuClNO4P3W: C, 43.91; H, 6.17; N, 1.19;
Found: C, 43.90; H, 6.27; N, 1.13%. IR (v/cmꢁ1): 2932, 2361, 2006
(s), 1879 (br), 1262. 1H NMR (CDCl3, 400 MHz): d 2.90 (s, br, 4H,
CH2P), 2.65 (s, br, 2H, CH2P), 2.03–1.29 (m, 66H, cyclohexyl).
31P{1H} NMR (CDCl3, 162 MHz): d 13.8 (s, 1P), 6.8 (s, 2P, PCy2-W)
0.1 mmol) in dichloromethane (5 ml) was added [PtCl2(COD)]
(18.5 mg, 0.05 mmol) and the reaction mixture stirred for 1 h dur-
ing which time a white insoluble precipitate formed. This was fil-
tered, washed with diethyl ether and dried in vacuo (97 mg, 90%).
Anal. Calc. for C86H144Cl2N2O8P6PtW2: C, 47.96; H, 6.74; N, 1.30;
Found: C, 48.02; H, 6.65; N, 1.26%. IR (v/cmꢁ1) KBr Nujol mull:
2007, 1889, 1871, 1860, 1460, 1377, 1104, 722, 605. LSIMS: m/z
(%): 2153 (55) [M]+, 2125 [M]+ꢁCO (5), 746 (100).
2.10. Synthesis of [[W(CO)4{N(CH2PCy2)3- -Ag)]ClO4 (11)
j
2P}]2(
l
1
with W satellites (d, J183W,P = 219 Hz). LSIMS: m/z (%): 1175 (10)
[M]+, 1063 (100) [W(CO){N(CH2PCy2)3- 2P}AuCl].
j
To solution of 7 [W(CO)4{N(CH2PCy2)3-j
a
2P}] (94 mg,
0.1 mmol) in dichloromethane (5 ml) was added AgClO4
(10.4 mg, 0.05 mmol) and the reaction mixture stirred overnight.
Colourless crystals, suitable for X-ray diffraction, were obtained
by diffusion of diethyl ether into this solution over a period of four
days. The crystals were collected by filtration, rinsed with diethyl
ether and dried in vacuo. (68 mg, 64%) Anal. Calc. for C86H144AgCl-
N2O12P6W2: C, 49.31; H, 6.93; N, 1.34; Found: C, 49.22; H, 6.89; N,
1.29%. IR (v/cmꢁ1): 3055, 2930, 2305, 2006 (s), 1880 (br), 1421,
1297, 1272, 1269, 896. 31P{1H} NMR (CDCl3, 162 MHz): d 13.8
2.8. Synthesis of [[Mo(CO)4{N(CH2PCy2)3-
j l-PtCl2)] (9)
2P}]2(
To
a solution of 6
[Mo(CO)4{N(CH2PCy2)3-j2P}] (92 mg,
0.107 mmol) in dichloromethane (5 ml) was added [PtCl2(COD)]
(20 mg, 0.054 mmol) and the reaction mixture stirred for one hour
during which a white insoluble precipitate formed. The precipitate
was collected by filtration, washed with diethyl ether and dried in
vacuo (90 mg, 85%). The dichloromethane filtrate from this separa-
tion was collected and layered with diethyl ether. Crystals suitable
for X-ray diffraction formed after a period of 4 days (10 mg, 10%).
Anal. Calc. for C86H144Cl2N2O8P6PtMo2: C, 52.23; H, 7.34; N, 1.42;
Found: C, 52.26; H, 7.48; N, 1.32%. IR (v/cmꢁ1) KBr Nujol mull:
2010, 1895, 1882, 1863, 1460, 1377, 1170, 1104, 852, 7222, 618,
592. 31P{1H} NMR (CD2Cl2, 162 MHz): d 29.1 (s, 4P, PCy2-Mo),
16.3 (s, 2P, PCy2-Pt). LSIMS: m/z (%): 1977 (10) [M]+, 1750
1
1
(dd, J109Ag,P = 503 Hz, J107Ag,P = 460 Hz), 7.8 (s, 4P, PCy2-W) with
1
W satellites (d, J183W,P = 218 Hz). LSIMS: m/z (%): 1996 (25) [M]+,
1968 (5) [M]+ꢁCO, 1052 (60) [W(CO)4{N(CH2PCy2)3-
j
2P}Ag]+,
1022 (95) [W(CO)3{N(CH2PCy2)3-j2P}Ag]+.
2.11. X-ray crystallography
Table 1 provides a summary of the crystallographic data
for compounds 5, 6, 8, 9 and 11. Data were collected using Oxford
Diffraction Xcalibur PX Ultra (5, 8, 9 and 11) and Xcalibur 3 (6)
(40) [[Mo{N(CH2PCy2)3-j l
2P}]2( -PtCl2)]+, 1009 (100) [Mo{N-
(CH2PCy2)3-j2P}PtCl2]+.
Table 1
Crystal data, data collection and refinement parameters for compounds 5, 6, 8, 9 and 11.a
Data
5
6
8
9
11
Formula
Solvent
C27H60NP3
–
C43H72MoNO4P3
–
C43H72AuClNO4P3W
–
C86H144Cl2Mo2N2O8P6Pt
2CH2Cl2
[C86H144AgN2O8P6W2](ClO4)
–
Formula weight
Colour, habit
Crystal size (mm)
Temperature (K)
Crystal system
Space group
a (Å)
491.67
Colourless blocks
855.87
1176.19
2147.57
2094.87
Colourless tablets
0.42 ꢀ 0.26 ꢀ 0.10
173
Monoclinic
P21/c (no. 14)
15.11884(19)
12.87650(14)
23.4163(3)
–
Colourless shards
0.28 ꢀ 0.09 ꢀ 0.07
173
Monoclinic
P21 (no. 4)
11.01766(7)
17.61223(7)
12.96732(8)
–
Colourless blocks
0.28 ꢀ 0.16 ꢀ 0.07
173
Monoclinic
P21/n (no. 14)
14.15811(19)
14.1726(2)
24.9176(4)
–
Pale yellow plates
0.26 ꢀ 0.25 ꢀ 0.10
293
Monoclinic
P21/c (no. 14)
12.70098(11)
26.2168(2)
14.79270(13)
–
0.24 ꢀ 0.13 ꢀ 0.11
173
Orthorhombic
Pbca (no. 61)
11.51153(12)
15.27016(16)
37.7523(4)
–
b (Å)
c (Å)
a
(°)
b (°)
–
–
106.5290(14)
–
4370.25(10)
4
111.2489(7)
–
2345.18(3)
2
91.1687(12)
–
107.2617(9)
–
c
(°)
V (Å3)
6636.21(12)
8
0.984
4998.86(13)
4703.81(7)
Z
2d
2d
Dcalc (g cmꢁ3
)
1.301
1.666
1.427
1.479
Radiation used
(mmꢁ1
2h max (°)
Cu K
1.721
143
a
Mo K
0.450
65
a
Cu K
12.057
145
a
Cu K
7.366
143
a
Cu Ka
7.744
145
l
)
Number of unique reflections
Measured (Rint
)
6386 (0.037)
4477
281
0.038
0.104
14 259 (0.038)
11 350
469
0.039
0.104
7435 (0.028)
7294
488
0.030
0.077
9721 (0.030)
9004
575
0.027
0.072
8784 (0.024)
7477
548
0.036
0.100
obs, |Fo| > 4 (|Fo|)
r
Number of variables
R1(obs)b
wR2(all)c
a
b
c
Details in common: graphite monochromated radiation, refinement based on F2.
R1
=
R
||Fo| ꢁ |Fc||/
R|Fo|.
2
wR2 = {
R
[w(Fo ꢁ Fc2)2]/
R = r
[w(Fo2)2]}1/2; wꢁ1 2(Fo2) + (aP)2 + bP.
d
The complex has crystallographic Ci symmetry.

P.W. Miller, A.J.P. White / Journal of Organometallic Chemistry 695 (2010) 1138–1145
1141
diffractometers, and the structures were refined based on F2 using
3. Results and discussion
the SHELXTL and SHELX-97 program systems [31]. The structure of
8 was shown to be a partial racemic twin by a combination of
R-factor tests [R1+ = 0.0342, R1ꢁ = 0.0370] and by use of the Flack
parameter [x+ = +0.366(9), xꢁ = +0.634(9)]; the lower residuals
given in the table are from refinements where the Flack parameter
was included as a variable.
3.1. Ligand synthesis
N-triphos ligands bearing diphenylphosphino, dicyclohexyl-
phosphino and ditert-butylphosphino groups were prepared by a
modified procedure to that reported by Markl et al. [26]. The air
stable diphenyl or dialkyl hydroxylmethylphosphonium chloride
salts were found to be particularly convenient starting materials
and easily prepared via the reaction of the required secondary
phosphine with 2 equiv of formaldehyde in the presence of hydro-
chloric acid [30]. The phosphonium chloride salts could be conve-
niently deprotonated in situ, using excess triethylamine, to the
corresponding hydroxymethylphosphine and reacted in a phos-
phorus based Mannich condensation reaction [32–34] with ammo-
nium chloride (Scheme 1). The N-triphos ligands 2 and 4 were
conveniently isolated as white precipitates in high yield (>80%)
after several hours at reflux by simply filtering off the methanol
solvent. The tert-butyl ligand 5 was found to be much more soluble
in methanol and was consequently isolated in lower yield (52%) as
colourless crystals on recrystallisation from methanol. 31P{1H} and
1H NMR spectra for 2 were consistent with literature data, while
the 31P{1H} spectra for 4 and 5 gave single resonances at -17.0
and ꢁ12.5 ppm, respectively. The X-ray crystallographic structure
of 5 (Fig. 1) revealed molecular threefold symmetry about the cen-
tral bridgehead nitrogen atom with the bulky tert-butyl groups ori-
entated away from each other.
Scheme 1. The formation of N-triphos ligands bearing diphenylphosphino (2),
dicyclohexylphosphino (4) and ditert-butylphosphino (5) via the phosphorus based
Mannich-type reaction.
3.2. Coordination chemistry
The cyclohexyl ligand 4 was selected for coordination chemistry
investigations because of its high yielding synthesis and the great-
er availability of the dicyclohexylphosphine starting material. Tri-
phos-type ligands are well known to chelate in a tridentate
fashion and found to commonly occupy facial coordination sites
Fig. 1. The molecular structure of 5.
Scheme 2. Reaction of the dicyclohexylphosphine N-triphos ligand (4) with Mo(CO)6 and W(CO)6, followed by coordination of the free phosphine arm to form a series of
heteromultimetallic gold, platinum and silver complexes.

1142
P.W. Miller, A.J.P. White / Journal of Organometallic Chemistry 695 (2010) 1138–1145
Table 2
of octahedral, square pyramidal and trigonal bipyramidal metal
centres [1]. Group 6 hexacarbonyls are widely used to react with
phosphine ligands, via carbonyl displacement, and may be antici-
pated to form facially capping tricarbonyl complexes with triphos-
phines [20]. Reaction of equimolar amounts of 4 with Mo(CO)6 in
DMF at 100 °C (Scheme 2) gave the new complex [Mo-
(CO)4{N(CH2PCy2)3-j2P}] (6). The 31P{1H} spectrum of 6 showed
resonances at 22.8 ppm and ꢁ16.7 ppm in an approximate 2:1
integration. This indicated that a bidentate complex had formed in-
stead of the expected facially capping tricarbonyl complex. Two
dicyclohexylphosphino arms of ligand 4 chelate to the Mo centre
accounting for the higher frequency resonance at 22.8 ppm while
the third arm remains uncoordinated which accounts for the lower
frequency signal at ꢁ16.7 ppm. This was further supported by the
IR spectrum complex 6 which showed a strong absorption at
2008 cmꢁ1 and a broad peak with shoulders at 1888 cmꢁ1 charac-
teristic of a cis-tetracarbonyl complex [35]. Reaction of 4 with
W(CO)6 under the same reaction conditions gave [W(CO)4{N-
Selected bond lengths (Å) and angles (°) for 6 and 8.
6
8
Mo–P(1)
Mo–P(2)
Mo–C(41)
Mo–C(42)
Mo–C(43)
Mo–C(44)
2.5145(5)
2.5303(5)
1.987(2)
1.988(2)
2.040(2)
2.045(2)
Au–Cl
2.2921(18)
2.2370(16)
2.5275(15)
2.5218(16)
2.022(7)
1.988(7)
2.052(6)
2.044(6)
Au–P(3)
W–P(1)
W–P(2)
W–C(41)
W–C(42)
W–C(44)
W–C(43)
P(1)–Mo–P(2)
P(1)–Mo–C(41)
P(1)–Mo–C(42)
P(1)–Mo–C(43)
P(1)–Mo–C(44)
P(2)–Mo–C(41)
P(2)–Mo–C(42)
P(2)–Mo–C(43)
P(2)–Mo–C(44)
87.476(15)
172.07(7)
86.80(6)
87.90(6)
97.07(6)
93.97(7)
172.62(6)
88.82(6)
91.17(6)
Cl–Au–P(3)
178.12(6)
88.42(5)
178.2(2)
94.7(2)
95.5(2)
88.1(2)
P(1)–W–P(2)
P(1)–W–C(41)
P(1)–W–C(42)
P(1)–W–C(43)
P(1)–W–C(44)
P(2)–W–C(41)
P(2)–W–C(42)
P(2)–W–C(43)
P(2)–W–C(44)
90.4(2)
176.8(2)
97.64(19)
89.9(2)
(CH2PCy2)3-j
2P}] (7) which displayed a similar pattern in its
31P{1H} spectrum, showing a higher frequency chemical shift to
3.2 ppm, with 183W satellites (1J = 217 Hz), and a lower frequency
resonance at ꢁ16.9 ppm accounting for the free phosphorus arm.
The IR spectrum of this complex also indicated a tetracarbonyl
complex while the LSIMS spectrum showed a characteristic pattern
of carbonyl loss. Crystals of the Mo complex 6 suitable for X-ray
diffraction were obtained by cooling a hot solution of 6 to ambient
temperature. The molecular structure of 6 (Fig. 2) confirmed the
tetracarbonyl and the expected six-membered chelate ring formed
by the bidentate phosphine arms with a P(1)–Mo–(P2) bite angle of
87.476(15)° (Table 2). The space filling diagram (Fig. 3) shows the
extent of the steric crowding around the metal centre as a result of
the bulky dicyclohexylphosphine groups which clearly prevent the
coordination of the third phosphine arm. Attempts to enforce coor-
dination of the free arm by heating to reflux in DMF only resulted
in recovery of the bidentate complex. Experiments under identical
reaction conditions with the less sterically hindered triphos or N-
triphos ligands exclusively produced the facial capping tridentate
LMo(CO)3 or LW(CO)3 complexes and therefore confirm that steric
Fig. 3. Space-filling representation of 6 showing the steric bulk of the cyclohexyl
groups around the Mo(CO)4 centre.
interactions prevent the third arm of ligand 4 from coordinating.
The formation of bidentate triphos transition metal complexes,
resulting in two bound and one free phosphorus donor, are rarer
and generally dictated by the choice of metal cation and not by
the steric constraints of the ligands [36–40]. Comparable bidentate
complexes, using the triphos ligand, have been formed under mild
reaction conditions using the precursor metal carbonyl complexes
([M(CO)4(piperidine)2] M = Mo or W) which contain two labile
piperidine groups [41].
This steric control enforced by ligand 4 at the Mo and W metal
centres, which results in a freely available phosphorus arm, pre-
sents a simple and straight forward route to the formation multi-
metallic complexes using one simple ligand scaffold. To
investigate the scope of coordination chemistry available to this
free phosphine arm, complexes 6 and 7 were reacted with a range
of transition metal salts (Scheme 2) with aim of preparing a series
of heteromultimetallic complexes. Reaction of complex 7 with one
equivalent [AuCl(SMe)2] at room temperature for 2 h resulted in
the formation of the mixed Au:W complex [W(CO)4{N-
(CH2PCy2)3-j
2P}AuCl] (8). The 31P{1H} NMR spectrum of 8 showed
a characteristic high frequency shift of the uncoordinated phos-
Fig. 2. The molecular structure of 6. The cyclohexyl groups have been omitted and
replaced with ‘‘Cy” for clarity.
phine arm from ꢁ16.9 ppm to 13.8 ppm on reaction with gold

P.W. Miller, A.J.P. White / Journal of Organometallic Chemistry 695 (2010) 1138–1145
1143
chloride. Crystals suitable for single crystal analysis were obtained
by slow diffusion of diethyl ether into a dichloromethane solution
of 8 overnight. The molecular structure of 8 (Fig. 4) revealed the
expected coordination of the lone dicyclohexylphosphine arm to
gold chloride in a linear fashion, the Cl–Au–P(3) bond angle being
178.12(6)° with an Au–P(3) bond distance of 2.2370(16) Å
(Table 2).
In an attempt to form a mixed trimetallic complex 2 equiv of
the complex 6 or 7 were reacted with 1 equiv of [PtCl2(COD)] in
CH2Cl2 (Scheme 2). This resulted in the immediate precipitation
of insoluble white solids that were found to be the bisligated spe-
cies [[Mo(CO)4{N(CH2PCy2)3-
j l-PtCl2)] (9) and [[W(CO)4{N-
2P}]2(
(CH2PCy2)3-j2P}]2(
l
-PtCl2)] (10) based on their elemental analysis
and LSIMS spectra which showed molecular ion peaks at 1977 m/z
and 2153 m/z, respectively. The 31P{1H} NMR spectrum of 9 (Fig. 5)
showed resonances at 29.1 and 16.3 ppm, however the Pt–P cou-
pling constant could not be resolved owing to its poor solubility
and resulting low concentration in all solvents. Unfortunately com-
plex 10 was found to be insoluble and 31P NMR could not be ob-
tained. A small amount of crystalline solid of complex 9, suitable
for X-ray diffraction studies, was obtained when the filtrate of this
reaction was allowed to slowly evaporate at room temperature
over a period of 4 days. The molecular structure of 9 revealed an
almost perfectly square planar platinum centre with the two bulky
P(3) dicyclohexylphosphino groups in a trans configuration with
Pt–P(3) bond distances of 2.3237(6) Å. The chelating bis-phosphino
molybdenum moieties of the complex are oriented away from each
other above and below the plane of the PtCl2 centre, presumably
for steric reasons.
Fig. 5. The molecular structure of 9. The cyclohexyl groups have been omitted and
replaced with ‘‘Cy” for clarity.
Reaction of 2 equiv of the tungsten complex 7 with 1 equiv of
AgClO4 in DCM resulted in the formation the bisligated silver com-
plex [[W(CO)4{N(CH2PCy2)3-j2P}]2(
l-Ag)]ClO4 (11) (Scheme 2).
This complex was found to be non-labile on the NMR timescale
at room temperature. The 31P{1H} NMR spectrum of 11 (Fig. 6) dis-
played a distinctive 107Ag,P, l09Ag,P coupling pattern of two dou-
blets centred at 13.8 ppm in addition to the low frequency
resonance at 7.8 ppm for the chelating dicyclohexylphosphino
arms of the complex. The magnitudes of the 1J(107Ag,P) and
1J(109Ag,P) coupling constants are a useful indicator of the coordi-
nation number at the metal centre [42]. In the case of complex
11 the 1J(107/109Ag,P) coupling constants were 460 and 503 Hz,
respectively, which indicates that two phosphine groups are
bonded to the silver centre. The structurally related complex
[Ag(PCy3)2]O3SCF3, which is also non-labile on the NMR timescale
at room temperature, displays similar 1J(107/109Ag,P) coupling con-
stants of 463 and 534 Hz to that of 11 [43]. Single crystals of 11
suitable for X-ray diffraction were obtained by slow diffusion of
diethyl ether into a dichloromethane solution of 11 over a period
of 3 days. The molecular structure of 11 (Fig. 7) bears several
similar features to that of complex 9. The steric bulk of the two
Fig. 4. The molecular structure of 8. The cyclohexyl groups have been omitted and
replaced with ‘‘Cy” for clarity.
Fig. 6. The 31P{1H} NMR spectrum of (11).

1144
P.W. Miller, A.J.P. White / Journal of Organometallic Chemistry 695 (2010) 1138–1145
Table 3
Selected bond lengths (Å) and angles (°) for 9 and 11.
9
11
Pt–Cl
Pt–P(3)
Mo–P(1)
Mo–P(2)
Mo–C(41)
Mo–C(42)
Mo–C(43)
Mo–C(44)
Cl–Pt–Cl0
Cl–Pt–P(3)
Cl–Pt–P(30)
P(3)–Pt–P(3’)
P(1)–Mo–P(2)
2.3057(7)
2.3237(6)
2.5549(7)
2.5455(7)
1.987(3)
1.989(3)
2.047(3)
2.021(3)
180
Ag–P(3)
W–P(1)
W–P(2)
W–C(41)
W–C(42)
W–C(43)
W–C(44)
2.3814(10)
2.5399(10)
2.5497(10)
1.979(5)
1.975(5)
2.030(5)
2.028(5)
P(3)–Ag–P(30)
P(1)–W–P(2)
180
92.18(3)
92.16(2)
87.84(2)
180
91.07(2)
Table 4
The P–C–N-lone pair torsion angles (°) for 5, 6, 8, 9 and 11.
5
6
8
9
11
P(1)–C(1)–N(4)–l.p.
P(2)–C(2)–N(4)–l.p.
P(3)–C(3)–N(4)–l.p.
ꢁ49.7
ꢁ53.0
ꢁ49.1
+38.9
ꢁ42.7
+45.1
ꢁ46.0
ꢁ34.7
+17.1
ꢁ11.1
+173.2
ꢁ42.4
+3.3
+52.3
ꢁ173.6
This method is limited only by the availability of the secondary
phosphine starting material, however many secondary phosphines
are now commercially available or reported in the literature by
well described syntheses [44]. Our initial coordination chemistry
studies using the cyclohexyl N-triphos ligand (4) have demon-
strated that the steric bulk of this ligand dictates the type of com-
plex that is formed. Bidentate Mo and W tetracarbonyl complexes
were formed exclusively in preference to the expected facially cap-
ping tricarbonyl complexes which are common for triphos-type li-
gands. The resulting uncoordinated phosphorus arm was used to
prepare a series of mixed bimetallic and trimetallic complexes that
would otherwise be difficult to synthesise using less sterically
bulky ligands. Bulky N-triphos ligands therefore present an inter-
esting scaffold for the straightforward preparation of mixed metal-
lic complexes by simple exploitation of the steric bulk. We are
currently investigating the scope of this ligand and the tert-butyl
triphos ligand for coordination to other transition metals, catalytic
applications and the preparation of coordination polymers.
Fig. 7. The molecular structure of 11. The cyclohexyl groups have been omitted and
replaced with ‘‘Cy” for clarity.
opposing lone dicyclohexylphosphino arms P(3) results in a per-
fectly linear P(3)–Ag–P(30) arrangement at the silver centre with
a bond angle of 180° (Table 3).
The three P–C–N–l.p. torsion angles can be used to describe the
conformational changes in the solid state structures of the free li-
gand (5) and metal complexes 6, 8, 9 and 11, where the lone pair
on the nitrogen has been assumed to occupy the ‘empty’ tetrahe-
dral site. An approximate gauche conformation can be seen for
the free tert-butyl ligand 5 and also for complexes 6 and 8 (Table 3),
whilst complexes 9 and 11 display an anti conformation for the
non-chelating coordinated P(3) arm. The chelating phosphine
arm P(2) of complexes 9 and 11 is syn, and the P(1) torsion angle
is intermediate between gauche and syn in 9, and approximately
syn in complex 11. This reflects the change in the orientation of
the N–l.p. vector with respect to the proximal P2M metal coordina-
tion plane (and consequently the Nꢂ ꢂ ꢂM distance). In complexes 6
and 8 this vector is almost perpendicular to the plane, being in-
clined by ca. 87° and 77° respectively, with Nꢂ ꢂ ꢂM separations of
ca. 4.02 and 4.07 Å, respectively. In contrast, for complexes 9 and
11 the N–l.p. vector is inclined by only ca. 23° and 26°, respectively,
to the P2M plane, and the Nꢂ ꢂ ꢂM distances are ca. 3.67 and 3.70 Å,
respectively (Table 4).
Acknowledgements
P.W.M. is grateful to the EPSRC for the award of a Life Sciences
Interface fellowship (EP/E039278/1) and to Prof. Nicholas J. Long
for helpful discussions during the preparation of this manuscript.
Appendix A. Supplementary material
CCDC 744593–744597 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.jorganchem.2010.01.017.
4. Conclusions
References
[1] J.-C. Hierso, R. Amardeil, E. Bentabet, R. Broussier, B. Gautheron, P. Meunier, P.
Kalck, Coord. Chem. Rev. 236 (2003) 143.
[2] H.A. Mayer, W.C. Kaska, Chem. Rev. 94 (1994) 1239.
[3] C. Bianchini, A. Meli, M. Peruzzini, F. Vizza, F. Zanobini, Coord. Chem. Rev. 120
(1992) 193.
We have reported the preparation of two new nitrogen centred
cyclohexyl and tert-butyl triphosphine ligands via a general syn-
thetic route that should allow the synthesis of a wide range of
other N-triphos-type ligands bearing different phosphino groups.

P.W. Miller, A.J.P. White / Journal of Organometallic Chemistry 695 (2010) 1138–1145
1145
[4] P. Aslanidis, P.J. Cox, K. Kapetangiannis, A.C. Tsipis, Eur. J. Inorg. Chem. (2008)
[25] O. Walter, G. Huttner, R. Kern, Z. Naturforsch. (B) 51 (1996) 922.
[26] G. Markl, G.Y. Jin, Tetrahedron Lett. 22 (1981) 1105.
5029.
[5] G. Guilera, G.S. McGrady, J.W. Steed, R.P.L. Burchell, P. Sirsch, A.J. Deeming,
New J. Chem. 32 (2008) 1573.
[27] F. Cecconi, C.A. Ghilardi, S. Midollini, A. Orlandini, Inorg. Chem. Commun. 3
(2000) 350.
[6] P. Barbaro, M. Caporali, A. Ienco, C. Mealli, M. Peruzzini, F. Vizza, Eur. J. Inorg.
Chem. (2008) 1392.
[28] R. Fischer, J. Langer, A. Malassa, D. Walther, H. Gorls, G. Vaughan, Chem.
Commun. (2006) 2510.
[7] F. Karadas, E.J. Schelter, M. Shatruk, A.V. Prosvirin, J. Bacsa, D. Smirnov, A.
Ozarowski, J. Krzystek, J. Telser, K.R. Dunbar, Inorg. Chem. 47 (2008) 2074.
[8] E.J. Schelter, F. Karadas, C. Avendano, A.V. Prosvirin, W. Wernsdorfer, K.R.
Dunbar, J. Am. Chem. Soc. 129 (2007) 8139.
[29] C. Bianchini, D. Masi, A. Meli, M. Peruzzini, F. Zanobini, J. Am. Chem. Soc. 110
(1988) 6411.
[30] J. Fawcett, P.A.T. Hoye, R.D.W. Kemmitt, D.J. Law, D.R. Russell, J. Chem. Soc.,
Dalton Trans. (1993) 2563.
[9] E. Iengo, E. Zangrando, E. Baiutti, F. Munini, E. Alessio, Eur. J. Inorg. Chem.
(2005) 1019.
[10] A. Bencini, E. Bill, A. Scozzafava, F. Totti, Chem. Eur. J. 9 (2003) 3015.
[11] X.J. Wang, T. Langetepe, C. Persau, B.S. Kang, G.M. Sheldrick, D. Fenske, Angew.
Chem., Int. Ed. 41 (2002) 3818.
[12] S. Dilsky, J. Organomet. Chem. 692 (2007) 2887.
[13] A.B. Chaplin, P.J. Dyson, Inorg. Chem. 47 (2008) 381.
[14] A.A. Nunez, G.R. Eastham, D.J. Cole-Hamilton, Chem. Commun. (2007) 3154.
[15] C. Bianchini, A. Meli, M. Peruzzini, F. Vizza, Y. Fujiwara, T. Jintoku, H. Taniguchi,
J. Chem. Soc., Chem. Commun. (1988) 299.
[31] SHELXTL PC version 5.1, Bruker AXS, Madison, WI, 1997; G. Sheldrick, SHELX-97,
Institut Anorg. Chemie, Tammannstr. 4, D37077 Göttingen, Germany, 1998.
[32] G.M. Brown, M.R.J. Elsegood, A.J. Lake, N.M. Sanchez-Ballester, M.B. Smith, T.S.
Varley, K. Blann, Eur. J. Inorg. Chem. (2007) 1405.
[33] D.A. Clarke, P.W. Miller, N.J. Long, A.J.P. White, Dalton Trans. (2007) 4556.
[34] P.W. Miller, N.J. Long, A.J.P. White, Dalton Trans. (2009) 5284.
[35] D.J. Darensbourg, R.L. Kump, Inorg. Chem. 17 (1978) 2680.
[36] M. Kandiah, G.S. McGrady, A. Decken, P. Sirsch, Inorg. Chem. 44 (2005) 8650.
[37] P. Sevillano, A. Habtemariam, M.I.G. Seijo, A. Castineiras, S. Parsons, M.E.
Garcia, P.J. Sadler, Aust. J. Chem. 53 (2000) 635.
[16] C. Bianchini, A. Meli, M. Peruzzini, F. Vizza, P. Frediani, J.A. Ramirez,
Organometallics 9 (1990) 226.
[38] R.M. Kirchner, R.G. Little, K.D. Tau, D.W. Meek, J. Organomet. Chem. 149 (1978)
C15.
[17] A.B. Chaplin, P.J. Dyson, Eur. J. Inorg. Chem. (2007) 4973.
[18] H.T. Teunissen, C.J. Elsevier, Chem. Commun. (1998) 1367.
[19] P.W. Miller, M. Nieuwenhuyzen, X.L. Xu, S.L. James, Chem. Commun. (2002)
2008.
[20] A. Muth, O. Walter, G. Huttner, A. Asam, L. Zsolnai, C. Emmerich, J. Organomet.
Chem. 468 (1994) 149.
[21] M. Sulu, L.M. Venanzi, Inorg. Chim. Acta 293 (1999) 70.
[22] A. Pastor, A. Galindo, J. Chem. Soc., Dalton Trans. (1997) 3749.
[23] T. Suzuki, K. Isobe, K. Kashiwabara, J. Fujita, S. Kaizaki, J. Chem. Soc., Dalton
Trans. (1996) 3779.
[24] E.G. Thaler, K. Folting, J.C. Huffman, K.G. Caulton, J. Organomet. Chem. 376
(1989) 343.
[39] D.H. Gibson, H.Y. He, M.S. Mashuta, Organometallics 20 (2001) 1456.
[40] E.J. Fernandez, M.C. Gimeno, A. Laguna, M. Laguna, J.M. Lopez de Luzuriaga, E.
Olmos, J. Organomet. Chem. 514 (1996) 169.
[41] E.J. Fernandez, M.C. Gimeno, P.G. Jones, A. Laguna, M. Laguna, E. Olmos, J.
Chem. Soc., Dalton Trans. (1996) 3603.
[42] J.G. Verkade, L.D. Quin, Phosphorus-31 NMR Spectroscopy in Stereochemical
Analysis Organic Compounds and Metal Complexes, VCH Publishers, 1987.
[43] T.G. Driver, K.A. Woerpel, J. Am. Chem. Soc. 126 (2004) 9993.
[44] C.A. Busacca, J.C. Lorenz, N. Grinberg, N. Haddad, M. Hrapchak, B. Latli, H. Lee,
P. Sabila, A. Saha, M. Sarvestani, S. Shen, R. Varsolona, X.D. Wei, C.H.
Senanayake, Org. Lett. 7 (2005) 4277.