Synthesis and characterization of four cubical molybdenum(V) tetramers and their catalytic properties for the epoxidation of cis-cyclooctene using H 2O2

Article abstract of DOI:10.1016/j.ica.2011.03.060

Molybdenum tetramers: Mo₄(μ³-O) ₄[μ-O₂P(CH₂Cl)₂] ₄O₄ (1), Mo₄(μ³-O) ₄(μ-O₂P(CH₂OH)₂) ₄O

Full text of DOI:10.1016/j.ica.2011.03.060

Inorganica Chimica Acta 373 (2011) 85–92
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Synthesis and characterization of four cubical molybdenum(V) tetramers and
their catalytic properties for the epoxidation of cis-cyclooctene using H2O2
⇑
Linsheng Feng, John S. Maass, Rudy L. Luck
Department of Chemistry, Michigan Technological University, 1400 Townsend Drive, Houghton, MI 49931, USA
a r t i c l e i n f o
a b s t r a c t
3-O)4[
l
-O2P(CH2Cl)2]4O4 (1), Mo4(
3-O)4[
-O2P(o-C6H4(CH2)2)]4O4 (4) have been synthesized
l l-O2P(CH2OH)2)4O4 (2),
3-O)4(
Article history:
Molybdenum tetramers: Mo4(
l
Mo4(l
3-O)4[
l
-O2P(PhOMe)2]4O4 (3), and Mo4(
l
l
Received 28 January 2011
Received in revised form 20 March 2011
Accepted 26 March 2011
Available online 1 April 2011
and characterized by IR, UV–Vis, and 31P NMR spectroscopy. Molybdenum tetramers 1 and 4 along with
the ligands L2A and L4 were structurally characterized by single crystal X-ray crystallography. An infinite
2D polymeric sheet was formed via inter and intra hydrogen bonds in the crystals of L2A. The crystals of
L4 consist of infinite polymeric chains formed through hydrogen bonding. All molybdenum tetramers
were tested as catalysts for the epoxidation of cis-cyclooctene in the presence of H2O2. Compounds 1
and 2 resulted in more than 80% epoxide after 24 hours at 70 °C, and displayed superior catalytic activ-
ities over compounds 3 and 4 under identical conditions. The superior catalytic activities of compounds 1
and 2 may be attributed to their better solubility in the ethanol/H2O2 system.
Keywords:
Molybdenum tetramer
Distorted cube
X-ray crystal structures
Organophosphinic acid
Epoxidation
Ó 2011 Elsevier B.V. All rights reserved.
cis-Cyclooctene
1. Introduction
[23–25]. These compounds all contain distorted cubane core
arrangements which result from the formation of Mo–Mo single
Epoxides are an important source of starting materials in indus-
try and drug discovery [1]. A common method for producing epox-
ides from olefins is to react them with oxidants such as hydrogen
peroxide, oxygen, or alkyl hydroperoxides in the presence of a
high-valent transition-metal catalyst [2]. In particular, molybde-
num and tungsten complexes have been widely investigated be-
cause of their high activity for the epoxidation of olefins using
tert-butyl hydroperoxide (TBHP) or hydrogen peroxide [2–17].
The use of hydrogen peroxide is ideal because of its high atom-effi-
ciency as compared to TBHP and that the sole byproduct, water, is
environmentally benign [18,19]. The use of chlorinated solvents for
the epoxidation of olefins is becoming increasing unacceptable be-
cause of their negative environmental effect [20]. Catalytic systems
using environmentally friendly solvents such as ethanol or acetoni-
trile are being developed to replace the chlorinated solvent sys-
tems [11]. In general, the epoxidation of olefins using transition
metal catalysts with hydrogen peroxide in non-chlorinated sol-
vents is a very active area in chemistry.
bonds with Mo(V) atoms and the ability of the S2ꢀ and O2ꢀ anions
to act as triply bridging ligands [23,26]. These cubic arrangements
are stabilized by various ligands such as the monodentate dimeth-
ylamine and trimethylsilyloxy [26] and the bidentate bridging
3,4-dihydroxycyclobut-3-ene-1,2-dione [27], phosphate [25,28],
dimethylthiophosphinato [29], diphenylphosphinato [24], and
dimethylphosphinato [30]. In this paper, we extend this class
of compounds using di(chloromethyl)phosphinic acid (L1),
di(hydroxymethyl)phosphinic acid (L2), di(p-methoxyphenyl)-
phosphinic acid (L3) and 2-hydroxyisophosphindoline-2-oxide
(L4) as ligands to synthesize new molybdenum(V) tetramers 1–4,
Scheme 1. Compounds 1–4 were characterized by IR, UV–Vis and
NMR spectroscopic methods. The structure of the ligands L2A and
L4 and compounds 1 and 4 were determined by single X-ray diffrac-
tion. The catalytic capabilities of compounds 1–4 for the epoxida-
tion of cis-cyclooctene were assessed using hydrogen peroxide as
the source of oxygen.
Cubical compounds M4(l
3-X)4 , where M = Co, Fe, Mo, Mn and
2. Experimental
X = O or S have drawn interest not only for their unique structures
and electronic properties, but also for their potential applications in
catalysis [21,22]. Molybdenum cubical-like structures such as
2.1. General method
[Mo4(l , X = S or O have been synthesized before
3-X)4O4]4+
Infrared spectra were obtained on a PerkinElmer Spectrum one
FT-IR spectrometer. 31P NMR data were recorded on a Varian XL-
400 spectrometer. A Fisher–Johns melting point apparatus (Fisher
Scientific Company) was used for the melting point determina-
⇑
Corresponding author.
E-mail address: rluck@mtu.edu (R.L. Luck).
0020-1693/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2011.03.060

86
L. Feng et al. / Inorganica Chimica Acta 373 (2011) 85–92
O
O
HO
P
OH
Cl
P
Cl
OH
OH
di(hydroxymethyl)phosphinic acid, L2
di(chloromethyl)phosphinic acid, L1
O
O
P
O
P
OH
OH
O
di(p-methoxyphenyl)phosphinic acid,L3
2-hydroxyisophosphindoline-2-oxide, L4
O
O
O
O
Mo
Mo
O
O
O
O
PR2
R2P
O
O
PR2
Mo
O
O
O
O
O
R2P
Mo
O
Scheme 1. Illustration of the four ligands and the resulting structure of the molybdenum tetramer produced using four conjugate bases of each ligand L1–L4 to bind to the
[Mo4O8]4+ core producing complexes 1–4, respectively.
tions. A Shimadzu QP5050 GC–MS was used for the mass spectra
determinations and quantitative analysis for the epoxidation reac-
tions. The elemental analysis data were obtained from Galbraith
Laboratories, Inc, Knoxville, TN. Methylene chloride and N,N-diiso-
propylethylamine were dried from CaH2 and NaH, respectively.
Other solvents were used as received from commercial suppliers.
Most chemicals were purchased from Aldrich and used as received.
Di(chloromethyl)phosphinic acid [31,32] (L1), di(hydroxy-
methyl)phosphinic acid [31,32] (L2), and 2-hydroxyisophosphind-
oline-2-oxide [33] (L4) were prepared as described in the cited
literature. Anilinium di(hydroxymethyl)phosphinate (L2A) was
synthesized by refluxing an ethanol solution of ligand L2 and
aniline in the ratio of 1:1 for 2 h and then allowing the solution
to cool to room temperature at which point colorless crystals
formed.
2.2.2. Synthesis and characterization of Mo4(l l-
3-O)4(
O2P(CH2OH)2)4O4, 2
Di-(hydroxymethyl)phosphinic acid (0.12 g, 0.92 mmol) was
added to a solution of MoO2(acac)2 (0.30 g, 0.92 mmol) in 7 ml of
ethanol contained in a thick-walled tube. The tube was then sealed
and heated to 120 °C for 48 h to obtain a red-brown precipitate.
This mixture was cooled to ambient temperature and filtered.
The precipitate was vacuum dried overnight affording 0.16 g
(84%) of Mo4(l l-O2P(CH2OH)2)4O4. Anal. Calc. for C8H24Mo4-
3-O)4(
O
24P4ꢁ0.5(C2H5OH): C, 10.44; H, 2.63. Found: C, 10.66; H, 2.94%. 1H
NMR(d6-DMSO): d = 3.26 ppm (d, –CH2P, J = 5.2 Hz), d = 3.55(broad,
s, HO–CH2). 31P NMR (DMSO): d = 43.3, 52.7 and 59.8 ppm. IR (neat,
cmꢀ1): 3269 (broad, w), 2898 (w), 1420 (w), 1068 (w), 1008 (s),
966 (s), 851 (m), 723 (s) cmꢀ1
.
2.2.3. Synthesis and characterization of Mo4(l l-
3-O)4[
O2P(PhOMe)2]4O4, 3
2.2. Synthesis
Di(p-methoxyphenyl)phosphinic acid (0.26 g, 0.92 mmol) was
mixed with MoO2(acac)2 (0.30 g, 0.92 mmol) in 7 ml ethanol using
a similar procedure to that for compound 1 described above. The
resulting red-orange precipitate was filtered and vacuum dried at
room temperature for overnight affording 0.36 g of molybdenum
2.2.1. Synthesis and characterization of tetramer Mo4(l l-
3-O)4[
O2P(CH2Cl)2]4O4, 1
Di(chloromethyl)phosphinic acid 0.15 g (0.92 mmol) was added
to a solution of MoO2(acac)2 0.30 g (0.92 mmol) in 8 mL of ethanol
in a thick-walled tube. The tube was sealed with a Teflon cap and
heated to 120 °C for 48 h. The mixture went through a green stage
finally resulting in red crystals. This mixture was then cooled down
to ambient temperature and filtered by vacuum filtration. Red
crystals were collected and vacuum dried overnight affording
tetramer
3 (Yield 97%). Anal. Calc. for C56H56Mo4O24P4.0.5(-
C2H5OH): C, 40.91; H, 3.43. Found: C, 40.67; H, 3.46%. 31P NMR
(CH2Cl2): d = 47.6 ppm (s, 1P) relative to H3PO4. IR (neat, cmꢀ1):
2993 (w), 2840 (w), 1596 (m), 1504 (m), 1299 (m), 1256 (m),
1128 (s) 1017 (s), 980(s), 836 (m), 805 (m), 711 (m), 671(m) cmꢀ1
.
0.10 g of Mo4(l l-O2P(CH2Cl)2]4O4. Yield: 38% (based on
3-O)4[
Mo). Anal. Calc. for C8H16Cl8Mo4O16P4: C, 8.28; H, 1.39. Found: C,
8.44; H, 1.39%. 31P NMR (THF): d = 57.7 ppm (s, 1P) relative to
H3PO4. 1H NMR (d6-DMSO): d = 3.39 ppm (d, ClCH2P, J = 9.0 Hz) IR
(neat, cmꢀ1): 3014 (w), 2995 (w), 2940 (w), 1387 (m), 1225
(m), 1099 (m), 1015 (s), 977 (s), 847 (m), 779 (m), 715 (s), 663
2.2.4. Synthesis and characterization of Mo4(l l-O2P(o-
3-O)4[
C6H4(CH2)2)]4O4, 4
0.034 g (0.1 mmol) of MoO2(acac)2 was added to 5 mL of etha-
nol in a thick-walled tube. The tube was sealed with a Teflon cap
and heated to 120 °C for 4 h. It was then allowed to cool to room
temperature and 0.018 g (1.1 mmol) of 2-hydroxyisophosphindo-
(m) cmꢀ1
.

L. Feng et al. / Inorganica Chimica Acta 373 (2011) 85–92
87
line-2-oxide was added to the solution. It was then heated to
120 °C for 36 h which resulted in the formation of a brown precip-
itate. The precipitate was filtered and then dissolved in methylene
chloride. This solution was filtered to remove an insoluble by-
product to give a red filtrate. Hexanes were added to the filtrate
and the solution was kept at 4 °C overnight. A red precipitate
condenser and
a
magnetic stirrer. In
a
typical experiment,
1.82 mmol of cis-cyclooctene, 1 mol % catalyst (1.82 ꢂ 10ꢀ2 mmol),
2.0 mL ethanol, 0.28 ml of 30% H2O2 (2.72 mmol, 1.5 equiv.), and
0.100 g n-decane (0.703 mmol) as an internal standard were mixed
together and the mixture heated to 70 °C. Samples were taken peri-
odically and analyzed by a Shimadzu QP5050 GC–MS. The conver-
sion and the selectivity of the reaction were calculated using
calibration curves.
was collected the next day to afford 0.017 g of Mo4(
l l-
3-O)4[
O2P(o-C6H4(CH2)2)]4O4. Yield: 50% (based on Mo). Anal. Calc. for
C
32H32Mo4O16P4: C, 32.57; H, 2.73. Found: C, 32.98; H, 2.94%. 31P
NMR (CH2Cl2): d = 88.3 ppm (s, 1P) relative to H3PO4. 1H NMR
(CDCl3): d = 3.48-3.3.53 ppm (q, CH2) d = 7.27–7.33 ppm (m, C6H4).
3. Results and discussion
3.1. Synthesis and spectral characterization
2.3. X-ray crystallography
We have previously reported that Mo4(l l-O2P(Ph)2]4O4
3-O)4[
Suitable crystals of compound 1 were obtained by allowing the
slow cooling of the reaction mixture. Red crystals of compound 4
suitable for single X-ray diffraction were obtained by dissolving
the compound in dichloromethane, layering with hexanes, and
allowing this mixture to stand for several days at room tempera-
ture. Suitable crystals of L2A, L4, 1 and 4 were coated with epoxy
resin and mounted on a glass fiber. The instrument used was an En-
raf-Nonius Turbo CAD4 X-ray diffractometer. Final cell constants
and orientation matrix were obtained by collecting appropriate
preliminary data, selecting 25 reflections, centering and refinement
can be produced by heating an ethanol solution of MoO2(acac)2
with diphenylphosphinic acid at 120 °C in a Teflon capped glass
tube for 48 hours [30]. Compounds 1, 3 and 4 were prepared by fol-
lowing a similar procedure substituting the appropriate ligand in
place of diphenylphosphinic acid. Compounds 1, 3, and 4 were
characterized by 1H NMR, 31P NMR, IR and UV–Vis spectroscopy.
Compound 2 on the other hand proved to be much harder to syn-
thesize. A red precipitate was obtained after heating MoO2(acac)2
with ligand L2 in ethanol for 48 h. The IR spectrum of this product
showed the signature stretching mode for a single Mo„O bond at
966 cmꢀ1 which is in agreement with other molybdenum tetra-
mers. The bands at 3269 and 1068 cmꢀ1 can be assigned to O–H
and P@O stretches, respectively. The proton NMR spectrum of 2
has a doublet at 3.26 ppm and a broad singlet at 3.55 ppm which
are due to the presence of hydrogen atoms on the methyl group
and hydroxyl group, respectively. The 31P NMR spectrum of 2 puz-
zled us as it consisted of several resonances ranging from 43.3 to
59.8 ppm. This indicates that precipitate 2 may be a mixture of
several molybdenum compounds. The mixture of 2 was imputed
by a least squares fit, Table 1. Mo Ka radiation (k = 0.71073 Å) was
used for data collection. The procedures used to collect the data,
solve and refine the structure were as detailed previously for other
complexes [30].
2.4. Epoxidation of cis-cyclooctene in the presence of H2O2
The reactions were carried out under an open atmosphere using
a two-necked 25 ml round bottom flask equipped with a reflux
Table 1
Crystal data and structure refinement details of ligands L2A, L4, 1 and 4.
Compound
L2A
L4
1
4
Molecular formula
Formula weight
Crystal size (mm)
Crystal color
T (K)
C8H14NO4P
219.17
0.50 ꢂ 0.40 ꢂ 0.20
colorless
291(2)
C8H9O2P
168.12
0.45x0.30x0.30
Colorless
291(2)
C8H16Cl8Mo4O16P4
1159.45
0.25x0.15x0.05
Red-Brown
291(2)
C32H32Mo4O16P4
1180.22
0.20x0.20x0.10
Red
291(2)
Crystal system
Space group
a (Å)
b (Å)
c (Å)
monoclinic
P2(1)/C
11.909(3)
8.555(5)
11.304(3)
90
110.72(2)
90
1077.2(7)
4
Monoclinic
P2(1)/C
8.843(3)
8.139(2)
11.788(3)
90
107.70(2)
90
808.3(4)
4
Triclinic
P1
Triclinic
P1
ꢀ
ꢀ
8.979(3)
9.379(5)
19.120(10)
101.34(4)
100.39(4)
98.81(4)
1522.9(12)
2
12.796(2)
13.751(4)
13.816(2)
111.25(2)
110.19(1)
102.63(2)
1955.5(9)
2
a
(°)
b (°)
c
(°)
V (Å)3
Z
Dcalc (g/cm3)
1.351
0.245
1.382
0.283
2.529
2.585
2.004
1.488
Absorption coefficient (mmꢀ1
)
F(0 0 0)
h (°)
464
1.83–22.47
1401
352
2.42–24.97
1412
1112
1.11–22.47
4268
1160
1.73–22.47
5387
Number of reflections measured
Number of observed data [I > 2
r
(I)]
1202
1232
3962
5109
Number of parameters
145
104
361
505
R1,a wR2 [I > 2
r
(I)]b
0.056, 0.143c
0.064, 0.153
1.056
0.032, 0.078d
0.039, 0.083
1.055
0.066, 0.173e
0.087, 0.196
1.106
0.023, 0.062f
0.029, 0.065
1.096
R1, wR2 (all data)
Goodness-of-fit (GOF) on F2
a
b
c
R1
=
R
||Fo| ꢀ |Fc||/
R|Fo|.
[w(Fo ꢀ Fc2)2]/
2
wR2 = [
R
R .
[w(Fo2)2]]1/2
w = 1/[2(Fo2)+(0.1082P)2 + 0.5002P] where P = (Fo2 + 2(Fc)2)/3.
w = 1/[2(Fo2)+(0.0313P)2 + 0.4430P].
d
e
f
w = 1/[2(Fo2)+(0.0853P)2 + 52.5061P].
w = 1/[2(Fo2)+(0.0375P)2 + 1.3582P].

88
L. Feng et al. / Inorganica Chimica Acta 373 (2011) 85–92
Table 2
Table 3
31P NMR chemical shift of ligands L1–L4, L2A and compounds 1–4 at 25 °C.
Selected bond distances (Å) and angles (°) for ligands L2A and L4.
Compound
31P NMR (ppm)a, (solvent)
L2A
L4
(ClCH2)2POOH, L1
(HOCH2)2POOH, L2
36.0 (s), (THF)
45.9 (s), (D2O)
Bond distances (Å)
P(1)–O(1)
1.507(2)
1.510(2)
1.807(4)
1.811(4)
1.396(5)
1.425(4)
1.452(4)
O(1)–P(1)
O(2)–P(1)
C(7)–P(1)
C(8)–P(1)
1.552(2)
1.485(2)
1.800(2)
1.796(2)
(MeOPh)2POOH, L3
(o-C6H4(CH2)2POOH, L4
PhNH3[OOP(HOCH2)2], L2A
33.4 (s), (CH2Cl2)
73.1 (s), (CDCl3)
42.7 (s), (EtOH)
P(1)–O(2)
P(1)–C(1)
P(1)–C(2)
C(1)–O(4)
C(2)–O(3)
N(1)–C(3)
Mo4(
Mo4(
Mo4(
Mo4(
l
l
l
l
3-O)4[
3-O)4(
3-O)4[
3-O)4[
l
l
l
l
-O2P(CH2Cl)2]4O4, 1
57.7 (s), (THF)
-O2P(CH2OH)2)4O4, 2
-O2P(PhOMe)2]4O4, 3
-O2P(o-C6H4(CH2)2)]4O4, 4
43.3, 52.7, 59.8 (m), (DMSO)
47.6 (s) , (CH2Cl2)
88.3(s), (CH2Cl2)
Bond angles (°)
O(1)–P(1)–O(2)
O(1)–P(1)–C(1)
O(2)–P(1)–C(1)
O(1)–P(1)–C(2)
O(2)–P(1)–C(2)
C(1)–P(1)–C(2)
O(4)–C(1)–P(1)
O(3)–C(2)–P(1)
a
116.22(14)
110.71(16)
106.66(15)
107.04(14)
109.12(15)
106.72(17)
111.6(2)
O(2)–P(1)–O(1)
O(2)–P(1)–C(8)
O(1)–P(1)–C(8)
O(2)–P(1)–C(7)
O(1)–P(1)–C(7)
C(8)–P(1)–C(7)
C(1)–C(7)–P(1)
113.22(9)
115.83(10)
107.99(9)
116.52(10)
104.46(10)
97.12(9)
s = singlet, m = multiple resonances.
initially to an impurity in ligand L2 (95% pure based on 31P NMR).
The anilinium salt of ligand L2 (L2A) was then made by refluxing
aniline with ligand L2 in ethanol for two hours and this compound
was then employed to make 2 by heating a mixture of MoO2(acac)2
and L2A in ethanol. In this case, the 31P NMR spectrum contained
resonances ranging from 43.3 to 62.6 ppm. This suggests that a
mixture of products was produced even using pure L2A. Corcoran
and Haushalter have reported 1D and 2D molybdenum phosphate
polymers were produced hydrothermally by mixing different
molybdenum sources with phosphoric acid under different reaction
conditions [25,28]. Also, like phosphoric acid, ligand L2 has four
active oxygen atoms which can bind to the molybdenum center.
Different coordination modes of ligand L2 may be the reason for
the resonances in the 31P NMR spectrum of 2 since a reasonable
elemental analysis was obtained. We have not been able to obtain
104.06(13)
112.0(2)
Table 4
Hydrogen bonds for L2A and L4 with HꢁꢁꢁA < (r(A) + 2.000 Å) and hDHAi 110°.
D–HꢁꢁꢁA
L2A
d(D–H)
d(HꢁꢁꢁA)
d(DꢁꢁꢁA)
h(DHA)
N1––H1CꢁꢁꢁO2a
N1––H1DꢁꢁꢁO3b
N1––H1EꢁꢁꢁO1c
O3–H3ꢁꢁꢁO2a
O4–H4AꢁꢁꢁO1d
L4
0.88(5)
1.00(4)
0.91(3)
0.82(6)
0.79(5)
1.84(5)
1.82(4)
1.78(3)
1.84(6)
2.02(6)
2.721(5)
2.810(4)
2.684(5)
2.649(4)
2.775(4)
174(4)
172(3)
174(4)
173(6)
161(5)
a
31P NMR spectroscopically pure sample of 2 despite employing
several different synthetic routes [20,23,25].
O1–H1ꢁꢁꢁO2e
0.84(3)
1.70(3)
2.521(2)
167(3)
The 31P chemical shifts of the ligands and compounds are listed
in Table 2. Free ligands L1–L4 and L2A resonate at 36.0, 45.9, 33.6,
73.1 and 42.7 ppm, respectively. A single resonance was observed
for 1, 3, and 4 at the position of 57.7, 47.6 and 88.4 ppm, respec-
tively. The downshifts of the resonance peaks of the coordinated li-
gands from the free ligands are comparable to what was observed
previously [30]. In addition, a doublet at 3.39 ppm and a quartet
at 3.51 ppm were observed in the 1H NMR spectrum of 1 and 4,
respectively which are due to the H-atoms on the methylene
groups.
Symmetry transformations used to generate equivalent atoms:
a
1 ꢀ x, ꢀ1/2 + y, 1/2 ꢀ z.
b
x, 1/2 ꢀ y, 1/2 + z.
c
x, ꢀ1 + y, z.
d
x, 3/2 ꢀ y, 1/2 + z.
e
ꢀx, 1/2 + y, 1/2 ꢀ z.
An ORTEP-3 [34] drawing of ligand L2A is shown in Fig. 1. Four
molecules of L2A crystallized in a monoclinic unit cell. The
phosphorous atom is in a distorted tetrahedral geometry with
the O–P–O and C–P–C bond angles equal to 116.22(14) and
106.72(17)°, respectively. The bond angles for O–P–C range from
106.66(15) to 110.71(16)° due to the different coordination envi-
ronments among oxygen atoms. The P–O bond distances at
1.507(2) and 1.510(2) Å are not significantly different, are interme-
diate between expected values for P@O and P–O bond lengths
[35,36] and suggestive of a multiple bond. A large degree of hydro-
gen bonding occurred in the crystal packing of L2A. In the refine-
ment of this molecule, the five H atoms involved in H-bonding
were freely refined (includes the three H atoms on the N atom) ex-
cept for the two attached to O atoms (on the OH moiety) which had
only their thermal parameters constrained. Two of the H atoms at-
tached to the N atom are each H-bonded to two phosphinate O
atoms and the remaining H atom is H-bonded to OH groups on
adjacent anions as listed in Table 4 and illustrated in Fig. S1. As
can be seen in the labeled anion in Fig. S1, the phosphinate O atoms
are also H-bonded to an OH moiety on an adjacent phosphinate
as well as to the H-atoms on the N atom on the anilinium cation.
This results in an interconnected planar 2-dimensional array of
molecules held together with these H-bonds which are stacked
up parallel to each other resulting in polar and nonpolar domains
in the crystal, see Fig. S2.
The IR spectra of compounds 1, 3 and 4 contain diagnostic bands
at 977, 980 and 980 cmꢀ1, respectively indicating the stretching
mode for
compounds 1, 3 and 4 can be ascribed to the
m
(Mo@O). The bands at 1125, 1128 and 1130 cmꢀ1 for
m(P@O) stretch which
are blue shifted compared to the free ligands. Absorptions for 1,
308; 2, 308; 3, 316; 4, 308 nm are observed for compounds 1–4 in
DMSO which can be attributed to ligand to metal charge transitions.
There are bigger differences observed in the weaker absorptions as-
cribed to d-d transitions (presumably
r
to r⁄ transitions) with 1,
482 nm; 2, 489 nm; 3, 467 nm, 4, 472 nm. This is perhaps related
to the ability of the ligand to donate electron density to the Mo
atoms allowing for good overlap in the Mo–Mo single bond as the
more electronegative ligands in complexes 1 and 2 would appear
to have a slightly weaker interaction in DMSO solution.
3.2. X-ray crystallographic data of C6H5NH3.O2P(CH2OH)2, L2A and
C8H8PO2H, L4
Colorless crystals of ligand L2A suitable for single X-ray diffrac-
tion were obtained by cooling an ethanol solution of the compound
overnight in a refrigerator. Colorless crystals of L4 were grown by
putting a saturated acetone solution in the freezer overnight. Se-
lected bond lengths and angles of compound L2A and L4 and a list-
ing of hydrogen bonds are given in Tables 3 and 4, respectively.
An ORTEP-3 [34] representation of ligand L4 is illustrated in
Fig. 2. Ligand L4 crystallizes in a monoclinic unit cell. The bond

L. Feng et al. / Inorganica Chimica Acta 373 (2011) 85–92
89
Fig. 1. ORTEP-3 [34] representation of L2A with selected atom numbering. Thermal
ellipsoids are drawn at the 50% probability level and hydrogen atoms are
represented by circles of arbritary radii. One H-bond is illustrated with a dashed
line.
Fig. 3. ORTEP-3 [34] representation of 1 with selected atom numbering. Thermal
ellipsoids are drawn at the 50% probability level and hydrogen atoms are omitted
for clarity.
Fig. 2. ORTEP-3 [34] representation of L4 with selected atom numbering. Thermal
ellipsoids are drawn at the 50% probability level and hydrogen atoms are not
shown.
angles at the phosphorous atom range from 97.12(9)° to
116.52(10)°, with the lowest angle being for the C–P–P angle. There
are two significantly different P–O bond distances; the short P@O2
and longer P–(O1)–H which are 1.485(2) and 1.552(2) Å, respec-
tively. These values are in good agreement with the average P@O
and P–OH distances found in other organic phosphates [35–40].
An infinite one dimensional chain was formed by inter molecular
hydrogen bonding between the (O1)–H and O2 atoms as shown
in Fig. S3. In this case the H atom bonded to O1 was refined freely
and the rest constrained. The O1–O2 bond length and angle for
O2ꢁꢁꢁH–O1 are 2.521(2) Å and 167(3)°, respectively, see Table 4.
Fig. 4. ORTEP-3 [34] representation of 4 with selected atom numbering. Thermal
ellipsoids are drawn at the 50% probability level and hydrogen atoms are omitted
for clarity.
Table 5
Selected bond distances (Å) for compounds 1 and 4.
3.3. X-ray crystallographic data of Mo4(
and Mo4( -O2P(o-C6H4(CH2)2)]4O4, 4
3-O)4[
l l-O2P(CH2Cl)2]4O4, 1
3-O)4[
Compound 1
Compound 4
l
l
Mo(1)–O(1)
Mo(1)–O(5)
Mo(1)–O(6)
Mo(1)–O(15)
Mo(1)–O(13)
Mo(1)ꢁꢁꢁO(8)
Mo(1)–Mo(2)
Mo(2)–O(2)
Mo(2)–O(6)
Mo(2)–O(5)
Mo(2)–O(11)
Mo(2)–O(9)
Mo(2)ꢁꢁꢁO(7)
Mo(3)–Mo(4)
O(9)–P(1)
O(10)–P(1)
O(11)–P(2)
O(12)–P(2)
O(13)–P(3)
O(14)–P(3)
O(15)–P(4)
O(16)–P(4)
1.668(11)
1.933(10)
1.954(10)
2.058(11)
2.065(10)
2.443(11)
2.629(2)
1.661(10)
1.940(10)
1.971(10)
2.074(11)
2.082(10)
2.378(11)
2.624(2)
1.490(11)
1.536(11)
1.508(11)
1.499(12)
1.522(11)
1.487(12)
1.507(11)
1.523(12)
Mo(1)–O(111)
Mo(1)–O(2)
Mo(1)–O(1)
Mo(1)–O(5)
Mo(1)–O(11)
Mo(1)ꢁꢁꢁO(3)
Mo(1)–Mo(2)
Mo(2)–O(211)
Mo(2)–O(2)
Mo(2)–O(1)
Mo(2)–O(7)
Mo(2)–O(9)
Mo(2)ꢁꢁꢁO(4)
Mo(3)–Mo(4)
O(5)–P(1)
O(6)–P(1)
O(7)–P(2)
O(8)–P(2)
O(9)–P(3)
O(10)–P(3)
O(11)–P(4)
O(12)–P(4)
1.657(2)
1.965(2)
1.968(2)
2.057(2)
2.072(2)
2.430(3)
2.6433(6)
1.666(2)
1.958(2)
1.961(2)
2.062(2)
2.085(2)
2.412(3)
2.6264(7)
1.523(3)
1.542(3)
1.529(3)
1.526(3)
1.525(3)
1.533(3)
1.529(3)
1.526(3)
Crystal data and details of structural refinement for compounds
1 and 4 are listed in Table 1. ORTEP [34] representations of com-
pounds 1 and 4 are shown in Figs. 3 and 4. Both 1 and 4 crystallized
ꢀ
in the triclinic space group P1 with one entire molecule of each
comprising the asymmetric unit. The structures of 1 and 4 consist
of distorted [Mo4O4(l
3-O)4]4+ cubic cores which are assembled by
two [Mo2O4]2+ units and four bridging phosphinate ligands
attached to four of the six cubic faces. Distances and angles corre-
sponding to one of these [Mo2O4]2+ units are listed in Tables 5 and
6, respectively for 1 and 4. The other halves, while not identical, are
not different enough to warrant inclusion and can be obtained
from the CIF files. Each molybdenum atom is coordinated in a dis-
torted octahedral fashion by six oxygen atoms: one terminal triply-
bonded oxygen atom, three
l
3-bridging oxygen atoms, and two
oxygen atoms from two different ligands. The bond lengths of
the molybdenyl group (Mo„O) range from 1.661(10)–1.669(11)
and 1.657(2)–1.668(3) Å for 1 and 4, respectively. Much longer
interactions were found for the MoꢁꢁꢁO bonds in the position trans
to the molybdenyl group ranging from 2.366(11) to 2.443(11) Å in
1 and 2.412(3)–2.430(3) Å for 4. These large MoꢁꢁꢁO(oxo) interac-
tions are caused by the trans effect of the triply-bonded terminal
oxo ligands [25,27,30]. The Mo–O(oxo) bond distances on the

90
L. Feng et al. / Inorganica Chimica Acta 373 (2011) 85–92
3-O)4]4+ cubes in 1 and 4 also contain two Mo–Mo
at 2.629(2) and
Table 6
The [Mo4O4(
single bonds; similar bond distances for
l
Selected bond angles (°) for compounds 1 and 4.
1
2.624(2) Å and significantly different ones in 4 at 2.6433(6) and
2.6264(7) Å. The non-bonded MoꢁꢁꢁMo distances range from
Compound 1
Compound 4
O(1)–Mo(1)–O(5)
O(1)–Mo(1)–O(6)
O(5)–Mo(1)–O(6)
O(1)–Mo(1)–O(15)
O(5)–Mo(1)–O(15)
O(6)–Mo(1)–O(15)
O(1)–Mo(1)–O(13)
O(5)–Mo(1)–O(13)
O(6)–Mo(1)–O(13)
O(15)–Mo(1)–O(13)
O(1)–Mo(1)–O(8)
O(5)–Mo(1)–O(8)
O(6)–Mo(1)–O(8)
O(15)–Mo(1)–O(8)
O(13)–Mo(1)–O(8)
O(1)–Mo(1)–Mo(2)
O(8)–Mo(1)–Mo(2)
O(2)–Mo(2)–O(6)
O(2)–Mo(2)–O(5)
O(6)–Mo(2)–O(5)
O(2)–Mo(2)–O(11)
O(6)–Mo(2)–O(11)
O(5)–Mo(2)–O(11)
O(2)–Mo(2)–O(9)
O(6)–Mo(2)–O(9)
O(5)–Mo(2)–O(9)
O(11)–Mo(2)–O(9)
O(2)–Mo(2)–O(7)
O(6)–Mo(2)–O(7)
O(5)–Mo(2)–O(7)
O(11)–Mo(2)–O(7)
O(9)–Mo(2)–O(7)
O(2)–Mo(2)–Mo(1)
O(7)–Mo(2)–Mo(1)
Mo(1)–O(5)–Mo(2)
Mo(1)–O(5)–Mo(3)
Mo(2)–O(5)–Mo(3)
Mo(2)–O(6)–Mo(1)
Mo(2)–O(6)–Mo(4)
Mo(1)–O(6)–Mo(4)
P(1)–O(9)–Mo(2)
P(2)–O(11)–Mo(2)
P(3)–O(13)–Mo(1)
P(4)–O(15)–Mo(1)
O(9)–P(1)–O(10)
110.5(5)
109.6(5)
88.9(4)
98.4(5)
150.7(4)
85.3(4)
98.5(5)
83.7(4)
151.7(4)
88.0(4)
169.9(5)
77.0(4)
76.5(4)
73.7(4)
75.3(4)
98.9(4)
91.1(2)
110.2(5)
109.2(5)
88.2(4)
97.1(5)
152.4(4)
86.2(4)
98.0(5)
82.8(4)
152.8(4)
90.0(4)
169.6(5)
77.1(4)
77.7(4)
75.3(4)
75.3(4)
98.4(4)
91.9(3)
84.6(4)
104.2(4)
101.9(4)
84.9(4)
103.0(4)
103.4(4)
127.8(7)
132.3(6)
128.7(7)
132.1(7)
115.6(6)
O(111)–Mo(1)–O(2)
O(111)–Mo(1)–O(1)
O(2)–Mo(1)–O(1)
O(111)–Mo(1)–O(5)
O(2)–Mo(1)–O(5)
O(1)–Mo(1)–O(5)
O(111)–Mo(1)–O(11)
O(2)–Mo(1)–O(11)
O(1)–Mo(1)–O(11)
O(5)–Mo(1)–O(11)
O(111)–Mo(1)–O(3)
O(2)–Mo(1)–O(3)
O(1)–Mo(1)–O(3)
O(5)–Mo(1)–O(3)
O(11)–Mo(1)–O(3)
O(111)–Mo(1)–Mo(2)
O(3)–Mo(1)–Mo(2)
O(211)–Mo(2)–O(2)
O(211)–Mo(2)–O(1)
O(2)–Mo(2)–O(1)
O(211)–Mo(2)–O(7)
O(2)–Mo(2)–O(7)
O(1)–Mo(2)–O(7)
O(211)–Mo(2)–O(9)
O(2)–Mo(2)–O(9)
O(1)–Mo(2)–O(9)
O(7)–Mo(2)–O(9)
O(211)–Mo(2)–O(4)
O(2)–Mo(2)–O(4)
O(1)–Mo(2)–O(4)
O(7)–Mo(2)–O(4)
O(9)–Mo(2)–O(4)
O(211)–Mo(2)–Mo(1)
O(4)–Mo(2)–Mo(1)
Mo(2)–O(1)–Mo(1)
Mo(2)–O(1)–Mo(3)
Mo(1)–O(1)–Mo(3)
Mo(2)–O(2)–Mo(1)
Mo(2)–O(2)–Mo(4)
Mo(1)–O(2)–Mo(4)
P(1)–O(5)–Mo(1)
P(2)–O(7)–Mo(2)
P(3)–O(9)–Mo(2)
P(4)–O(11)–Mo(1)
O(5)–P(1)–O(6)
111.28(11)
109.34(11)
88.71(10)
98.62(11)
84.77(10)
151.75(10)
98.03(11)
150.61(10)
83.45(9)
88.87(10)
169.43(10)
76.95(9)
76.72(9)
75.03(9)
Mo(2)ꢁꢁꢁMo(3) = 3.378(2) Å to Mo(1)ꢁꢁꢁMo(4) = 3.421(2) Å for
1
and Mo(2)ꢁꢁꢁMo(4) = 3.4232(5) Å to Mo(1)ꢁꢁꢁMo(3) = 3.4379(4) Å
for 4. Therefore some equivalent Mo–Mo bond distances and
non-bonded interactions in 1 are significantly shorter than those
in 4 indicating that 1 is a smaller cube. This conclusion is also sup-
ported by the fact that three of the long MoꢁꢁꢁO(oxo) interactions
which connect the two [Mo2O4]2+ units in 1 are significantly short-
er than those in 4; specifically Mo(2)ꢁꢁꢁO(7) = 2.378(11) Å,
Mo(3)ꢁꢁꢁO(5) = 2.366(11) Å, Mo(4)ꢁꢁꢁO(6) = 2.389(10) Å in 1 are all
significantly shorter than the equivalent in 4 of Mo(2)ꢁꢁꢁO(4) =
2.412(3) Å, Mo(3)ꢁꢁꢁO(1) = 2.413(2) Å, Mo(4)ꢁꢁꢁO(2) = 2.418(3) Å
73.69(9)
99.69(9)
90.78(6)
with comparable Mo(1)ꢁꢁꢁO(8) = 2.443(11) Å in
1 longer than
Mo(1)ꢁꢁꢁO(3) = 2.430(3) Å in 4. The difference may arise out of crys-
tal packing forces as the phenyl substituents inꢀ4 enforce a more
rigid environment around the donating R2PO2 entity and may
be forced to twist in order to minimize steric forces resulting in
an elongation of the cube. The Mo–Mo bond distances are similar
with those in molybdenum tetramers previously reported [30],
but are significantly longer than those in [Mo4O8(C4O4)]4ꢀ, i.e.,
110.65(12)
109.30(11)
89.12(10)
98.92(11)
83.83(10)
151.59(10)
96.41(12)
152.74(10)
84.90(9)
88.88(10)
169.16(11)
77.48(9)
77.27(9)
74.34(9)
Mo1–Mo2 = 2.5904(4) Å
and
Mo3–Mo4 = 2.5932(4) Å
[27],
[Mo4O4( 3-O)4(OSi(Me)3)4(HN(Me)2]4, i.e., Mo1–Mo2 = 2.607(4)
l
and Mo3–Mo4 = 2.606(4) Å [26].
There are two different bond angles within each [Mo2O4]2+ unit
of the [Mo4O4(l l
3-O)4]4+ cube. The Mo– 3-O–Mo angles range from
84.3(4) to 104.2(4)° and 84.33(9) to 102.83(10)° for 1 and 4,
75.26(9)
99.16(9)
91.63(6)
84.54(9)
respectively which are not significantly different. The other one
involving the Mo atoms at the center (i.e.,
l l
3-O–Mo– 3-O) range
from 76.5(4) to 110.5(5)° for 1 and 76.72(9)–111.28(11)° for 4
are also not significantly different suggesting that the [Mo2O4]2+
units in 1 and 4 are similar.
102.54(10)
102.92(10)
84.70(9)
102.43(10)
102.79(10)
130.44(15)
127.85(14)
129.71(14)
125.27(14)
114.03(14)
3.4. Application in epoxidation catalysis
Molybdenum complexes are used as catalysts for the epoxida-
tion of olefins with tert-butyl hydrogen peroxide in non-aqueous
solvents. However, efficient Mo catalysts with H2O2 as the source
of oxygen are limited [11,12,41–45] either because the presence
of water poisons the catalyst [46] or the use of H2O2 reduces the
catalytic activities or selectivies [47]. We did previously report
on the catalytic potential of molybdenum or tungsten monomeric
compounds such as MO2Cl2(OPPh2Me)2, MO2Cl2dppmO2, and MO2-
Cl2(OPPh2CH2OH)2 (M = Mo or W) in the presence of H2O2 which
displayed high activity and selectivity [11,12]. We have now ex-
tended our investigation to molybdenum tetramers 1 through 4
equatorial plane range from 1.933(10) to 1.971(10) Å and 1.951(2)
to 1.968(2) Å for 1 and 4, respectively and are shorter than the Mo–
O(phosphinate) bond distances which range from 2.058(11) to
2.088(11) Å in 1 and 2.057(2)–2.085(2) Å for 4. There is a greater
range in P–O bond distances in 1 at 1.49(1)–1.54(1) Å compared
to the range in 4 at 1.523((3)–1.542(3) Å though the average P–O
bond distances on 1 at 1.51(2) Å is not significantly different than
that on 4 at 1.529(6) Å.
and the previously prepared Mo4(
l l-O2P(CH2C6H5)2]4O4, 5
3-O)4[
[48] and Mo4( -O2P(C6H5)2]4O4, 6 [30]. In order to compare
l
3-O)4[
l
Table 7
Epoxidation of cis-cyclooctene with catalysts in the presence of H2O2 in ethanol solvent.a.
Entry
Catalyst
Cyclooctene epoxide yield (%)
TOFb (mol mol catꢀ1 hꢀ1
)
6h
24h
1
2
3
4
5
6
Compound 1
Compound 2
Compound 3
Compound 4
33.8
54.8
1.7
83.4
90.8
32.5
44.7
16.2
<1
2.9
26.8
0
1.8
0
28.6
2.5
Mo4(l
l
3-O)4[
3-O)4[
l
l
-O2P(CH2C6H5)2]4O4, 5c
-O2P(C6H5)2]4O4, 6d
Mo4(
NDe
0
a
Experiment condition: catalyst/cis-cyclooctene/H2O2 mole ratio equals 1:100:150. T = 70 °C. GC–MS quantitative analysis using decane as
internal standard.
b
The TOF were calculated after 30 min reaction time and described as mol epoxide/mol catalyst time (h).
Ref. [48].
Ref. [30].
None detected.
c
d
e

L. Feng et al. / Inorganica Chimica Acta 373 (2011) 85–92
91
the catalytic attributes of these molybdenum compounds with that
of related compounds made before, the reactions were carried out
using conditions identical to those utilized previously [11,12].
The results for the epoxidation of cis-cyclooctene with the eth-
anol/H2O2 system using complexes 1–4, 5 and 6 as catalysts are
summarized in Table 7. It should be noted that all reactions are
remarkably clean as only the epoxide was the final produced. Con-
trol experiments confirmed that none of epoxide was detected in
the absence of catalyst. It is clear from the results listed in Table
7 that compound 2 has the highest initial catalytic activity
(TOF = 26.8) compared to other compounds. In contrast, none of
the epoxide was detected after 30 min when compounds 3, 5 and
6 were used as catalysts. Compound 2 kept its high efficiency
and resulted in 54.8% conversion to the epoxide after 6 h and
reached 90.8% after 24 h. Interestingly the TOF and conversion to
the epoxide for compounds 1 and 4 are not significantly different
after 6 h. However, 83.4% of the epoxide was produced after 24 h
for compound 1 which is almost twice as much as that of com-
pound 4. Compound 3 has the lowest catalytic property among
the compounds 1–4, and resulted in 32.5% epoxide after 24 h.
The catalytic potential of the previously reported 5 and 6 was also
evaluated under identical conditions. Almost no catalytic activity
was found for compound 6. Complex 5 showed slightly better cat-
alytic activity than 6 as 16.2% epoxide conversion was noted after
24 h. A clear orange/red solution was obtained when compound 1
or 2 was used in the catalytic reaction. However, compounds 3–6
all resulted in an orange/red suspension in the ethanol/H2O2 sys-
tem. The big difference in catalytic activity may be attributed to
the better solubility of 1 and 2 in the ethanol/H2O2 system. Other
reported monomeric molybdenum complexes such as Mo(O)2-
(Cl)2(OPMePh2)2, Mo(O)2(Cl)2dppmO2 [11] and MoO2Cl2(OPPh2-
CH2OH)2 [12] resulted in 43%, 26% and 36.5%, respectively after
6 h under similar conditions and the yield of epoxide remained
constant for Mo(O)2(Cl)2(OPMePh2)2 and Mo(O)2(Cl)2dppmO2 even
with longer reaction times. The conversion to the epoxide only
increased to 67.7% for MoO2Cl2(OPPh2CH2OH)2 after 24 h. Com-
pounds 1 and 2 appear to be the best epoxidation catalysts for
cis-cyclooctene under these conditions.
system has been examined. Compounds 1 and 2 showed superior
catalytic capability over 3 and 4. This was attributed to the better
solubility of these compounds under the ‘‘green’’ catalytic
conditions.
Acknowledgements
We thank Michigan Technological University for supporting
this research.
Appendix A. Supplementary material
CCDC 809490–809493 contain the supplementary crystallo-
graphic data for complexes L2a, L4, 1 and 4, respectively. Figs. S1
and S2 depicting hydrogen bonding and unit cell packing for L2A
and Fig. S3 showing the 1-D polymeric chain formed by L4. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ica.2011.03.060.
References
[1] A.K. Yudin, Aziridines and Epoxides in Organic Synthesis, Weinheim, Wiley-
VCH, 2006.
[2] K.A. Jorgensen, Chem. Rev. 89 (1989) 431.
[3] A.U. Barlan, A. Basak, H. Yamamoto, Angew. Chem., Int. Ed. Enl. 45 (2006) 5849.
[4] C. Freund, M. Abrantes, F.E. Kühn, J. Organomet. Chem. 691 (2006) 3718.
[5] F.R. Fronczek, R.L. Luck, G. Wang, Inorg. Chim. Acta 342 (2003) 247–254.
[6] F.E. Kühn, A.M. Santos, M. Abrantes, Chem. Rev. 106 (2006) 2455.
[7] G. Wang, L. Feng, R.L. Luck, D.G. Evans, Z. Wang, X. Duan, J. Mol. Catal. A: Chem.
241 (2005) 8.
[8] M. Herbert, A. Galindo, F. Montilla, Catal. Commun. 8 (2007) 987.
[9] S.K. Maiti, S. Dinda, S. Banerjee, A.K. Mukherjee, R. Bhattacharyya, Eur. J. Inorg.
Chem. (2008) 2038.
[10] S.K. Maiti, K.M.A. Malik, S. Gupta, S. Chakraborty, A.K. Ganguli, A.K. Mukherjee,
R. Bhattacharyya, Inorg. Chem. 45 (2006) 9843.
[11] A. Jimtaisong, R.L. Luck, Inorg. Chem. 45 (2006) 10391.
[12] L. Feng, E. Urnezius, R.L. Luck, J. Organomet. Chem. 693 (2008) 1564.
[13] J.C. Alonso, P. Neves, d.S.M.J. Pires, S. Quintal, P.D. Vaz, C. Silva, A.A. Valente, P.
Ferreira, M.J. Calhorda, V. Felix, M.G.B. Drew, Organometallics 26 (2007) 5548.
[14] C.I. Fernandes, N.U. Silva, P.D. Vaz, T.G. Nunes, C.D. Nunes, Appl. Catal., A 384
(2010) 84.
[15] Y.-D. Li, X.-K. Fu, B.-W. Gong, X.-C. Zou, X.-B. Tu, J.-X. Chen, J. Mol. Catal. A:
Chem. 322 (2010) 55.
[16] A. Rezaeifard, I. Sheikhshoaie, N. Monadi, M. Alipour, Polyhedron 29 (2010)
2703.
[17] J. Zhao, K.R. Jain, E. Herdtweck, F.E. Kuehn, Dalton Trans. (2007) 5567.
[18] K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S. Hikichi, N. Mizuno,
Science 300 (2003) 964.
[19] R. Noyori, M. Aok, K. Sato, Chem. Commun. (2003) 1977.
[20] J.J.C. Constable, P.J. Dunn, J.D. Hayler, G.R. Humphrey, J.L.J. Leazer, R.J.
Linderman, K. Lorenz, J. Manley, B.A. Pearlman, A. Wells, A. Zaksh, T.Y.
Zhang, Green Chem. 9 (2007) 411.
The stability of the catalysts 1–4 was examined using 31P NMR
spectroscopy. The color of the reaction mixtures had changed to
blue from orange/red and 31P NMR spectra revealed that ligand dis-
sociation occurred for compounds 1, 2, and 4 after 24 h in the eth-
anol/H2O2 system. The color of the reaction mixture for compounds
3, 5 and 6 was unchanged after 24 h. This can also be attributed to
the different solubilities for these compounds in the ethanol/H2O2
system since presumably if the complexes do not dissolve, they
are not decomposing under the conditions for catalysis.
[21] A.S. Foust, L.F. Dahl, J. Am. Chem. Soc 92 (1970) 7337.
[22] T.R. Amarante, P.c. Neves, A.C. Coelho, S. Gago, A.A. Valente, F.A. Almeida Paz,
M. Pillinger, I.S. Gonçalves, Organometallics 29 (2010) 883.
[23] P.D. Williams, M.D. Curtis, Inorg. Chem. 25 (1986) 4562.
[24] W. Schirmer, U. Floerke, H.J. Haupt, Z. Anorg. Allg. Chem. 574 (1989) 239.
[25] E.W. Corcoran, Inorg. Chem. 29 (1990) 157.
[26] G.S. Kim, D.A. Keszler, C.W. DeKock, Inorg. Chem. 30 (1991) 574.
[27] B. Modec, J.V. Brencic, E.M. Burkholder, J. Zubieta, Dalton Trans. (2003) 4618.
[28] L. Mundi, R.C. Haushalter, J. Am. Chem. Soc. 113 (1991) 6340.
[29] R. Mattes, K. Muehlsiepen, Z. Naturforsch, B: Anorg. Chem., Organ. Chem. 35B
(1980) 265.
[30] A. Jimtaisong, L.S. Feng, S. Sreehari, C.A. Bayse, R.L. Luck, J. Cluster Sci. 19
(2008) 181.
[31] L. Maier, J. Organomet. Chem. 178 (1979) 157.
[32] P.W. Morgan, B.C. Herr, J. Am. Chem. Soc. 74 (1952) 4526.
[33] E.A. Boyd, M.E.K. Boyd, F. Kerrigan, Tetrahedron Lett. 37 (1996) 5425.
[34] L.J. Farrugia, J. Appl. Crystallogr. 1997 (1997) 565.
[35] M.E. Druyan, A.H. Reis Jr., E. Gebert, S.W. Peterson, G.W. Mason, D.F. Peppard, J.
Am. Chem. Soc. 98 (1976) 4801.
4. Conclusions
Several new molybdenum tetramers bearing functionalized or-
ganic phosphinates have been synthesized. Pure compounds 1, 3
and 4 were prepared in good yields. However, despite reasonable
elemental analyses, the precise nature of compound 2 has not
yet been determined. We believe that the ligand
l-O2P(CH2OH)2
is capable of exhibiting coordination isomerism in a bidentate
mode binding either through the two phosphinate O atoms or via
one of these atoms and an O atom from the hydroxyl group. Li-
gands L2A, L4 and compounds 1 and 4 were also structurally char-
acterized by single crystal X-ray diffraction. The crystal structure of
L2A were built up from infinite 2D polymeric sheets formed via in-
ter and intra hydrogen bonding among oxygen and nitrogen atoms.
An infinite one dimensional polymeric chain was formed by inter
molecular hydrogen bonding in L4. The ability of complexes 1–4
to catalyze the epoxidation of cis-cyclooctene in an ethanol/H2O2
[36] F. Giordano, A. Ripamonti, Acta Crystallogr., Sect. 22 (1967) 678.
[37] R.A. Sachleben, J.H. Burns, G.M. Brown, Inorg. Chem. 27 (1988) 1787.
[38] I.L. Karle, K. Britts, Acta Crystallogr., Sect. 20 (1966) 118.
[39] M. Calleri, J.C. Speakman, Acta Crystallogr., Sect. 17 (1964) 1097.
[40] R.P. Singh, B. Twamley, J.n.M. Shreeve, J. Chem. Soc., Dalton Trans. (2000) 4089.
[41] J. Li, G. Wang, Z. Shi, M. Yang, R.L. Luck, Struct. Chem. 20 (2009) 869.

92
L. Feng et al. / Inorganica Chimica Acta 373 (2011) 85–92
[42] V.V.K.M. Kandepi, J.M.S. Cardoso, B. Royo, Catal. Lett. 136 (2010) 222.
[43] N. Gharah, S. Chakraborty, A.K. Mukherjee, R. Bhattacharyya, Chem. Commun.
(2004) 2630.
[44] C. Dinoi, M. Ciclosi, E. Manoury, L. Maron, L. Perrin, R. Poli, Chem. Eur. J. 16
(2010) 9572. S9572/9571-S9572/9579.
[46] M. Abrantes, A.M. Santos, J. Mink, F.E. Kuehn, C.C. Romao, Organometallics 22
(2003) 2112.
[47] A.M. Martins, C.C. Romao, M. Abrantes, M.C. Azevedo, J. Cui, A.R. Dias, M.T.
Duarte, M.A. Lemos, T. Lourenco, R. Poli, Organometallics 24 (2005) 2582.
[48] J.S. Maass, M. Zeller, D. Holmes, C.A. Bayse, R.L. Luck, J. Cluster Sci. 22,
doi:10.1007/s10876-011-0373-7.
[45] J.S. Costa, C.M. Markus, I. Mutikainen, P. Gamez, J. Reedijk, Inorg. Chim. Acta
363 (2010) 2046.