Rapid phosphorus(III) ligand evaluation utilising potassium selenocyanate

Article abstract of DOI:10.1039/b712782k

Oxidative addition of SeCN^- to tertiary phosphine ligands has been investigated in methanol at 298 K by use of UV-Vis stopped-flow and conventional spectrophotometry. In most cases k_obsvs. [SeCN ^-] plots were linear with zero intercepts corresponding to a rate expression of k_obs = k₁[SeCN^-]. Reactions rates are dependent on the electron density of the phosphorus centre with k ₁ varying by five orders of magnitude from 1.34 ± 0.02 × 10^-3 to 51 ± 3 mol^-1 dm³ s^-1 for P(2-OMe-C₆H₄)₃ to PCy₃ respectively. Activation parameters range from 27 ± 1 to 49.0 ± 1.3 kJ mol^-1 for ΔH^? and -112 ± 9 to -140 ± 3 J K^-1 mol^-1 for ΔS ^? supporting a S_N2 mechanism in which the initial nucleophilic attack of P on Se is rate determining. Reaction rates are promoted by more polar solvents supporting the mechanistic assignment. Reasonable linear correlations were observed between log k₁vs. pK_a, ¹J_P-Se and χ_d values of the phosphines. The reaction rates are remarkably sensitive to the steric bulk of the substituents, and substitution of phenyl rings in the 2 position resulted in a decrease in the reaction rate. The crystal structures of SePPh₂Cy and SePPhCy₂ have been determined displaying Se-P bond distances of 2.111(2) and 2.1260(8) respectively. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b712782k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Rapid phosphorus(III) ligand evaluation utilising potassium selenocyanate†‡
a
b
Alfred Muller,§ Stefanus Otto*¶ and Andreas Roodt*b
Received 20th August 2007, Accepted 31st October 2007
First published as an Advance Article on the web 16th November 2007
DOI: 10.1039/b712782k
Oxidative addition of SeCN− to tertiary phosphine ligands has been investigated in methanol at 298 K
by use of UV-Vis stopped-flow and conventional spectrophotometry. In most cases kobs vs. [SeCN−] plots
were linear with zero intercepts corresponding to a rate expression of kobs = k1[SeCN−]. Reactions rates are
dependent on the electron density of the phosphorus centre with k1 varying by five orders of magnitude
from 1.34 0.02 × 10−3 to 51 3 mol−1 dm3 s−1 for P(2-OMe-C6H4)3 to PCy3 respectively. Activation
parameters range from 27 1 to 49.0 1.3 kJ mol−1 for DH‡ and −112 9 to −140 3 J K−1 mol−1 for
DS‡ supporting a SN2 mechanism in which the initial nucleophilic attack of P on Se is rate determining.
Reaction rates are promoted by more polar solvents supporting the mechanistic assignment. Reasonable
linear correlations were observed between log k1 vs. pKa, 1JP–Se and vd values of the phosphines. The
reaction rates are remarkably sensitive to the steric bulk of the substituents, and substitution of phenyl
rings in the 2 position resulted in a decrease in the reaction rate. The crystal structures of SePPh2Cy and
˚
SePPhCy2 have been determined displaying Se–P bond distances of 2.111(2) and 2.1260(8) A respectively.
In addition to computational studies, several experimental
Introduction
procedures have been developed that can be used for rapid ligand
evaluation, most notably measuring of IR stretching frequencies in
complexes such as [NiP(CO)3]31 and trans-[RhCl(CO)(P)2]32,33 or
by measuring of coupling constants between 31P and other NMR
active nuclei34–36 such as 11B, 195Pt or 77Se.
The evaluation of the electronic parameter of tertiary phosphines
is a study in organometallic chemistry that spans over several
decades, and even today is still attracting attention in chemical
societies1–21 due to its importance in e.g. catalysis.4,8,22,23 The
classical model of Dewar24 and Chatt25 describes the bonding of
phosphines to a metal centre as rdonation of the phosphorus lone
pair electrons to the metal and p back donation from the metal to
the phosphorus through the dp–dp bond. It was later shown by
empirical11,12 and theoretical studies19,26–29 that the r* orbitals on
the phosphorus are more likely to be the acceptor orbitals. The
partitioning of the r donor and p acceptor abilities of phosphorus
is however not that clear and remains a hot topic even today,
with several research groups attempting to separate r donor and
p acceptor effects empirically. The best developed method of these
is the QALE model30 (Quantitative Analysis of Ligand Effects)
to interpret phosphorus ligand effects. The model consists of four
stereo electronic parameters (three electronic and one steric). The
electronic properties are the r donor capacity (described by the
parameter vd), the Ear effect, which is not restricted to aryl groups,
and the p acidity (pp) of the phosphorus(III) ligand.
Both steric and electronic parameters of phosphines are sig-
nificant in determining the characteristics induced by the specific
ligand; these two properties are however quite difficult to separate
as they are closely related. As the substituents attached to
phosphorus increases in size it causes an increase in the C–P–
C intervalence angles thus effectively reducing the s-character of
the lone pair, making the ligand a better Lewis base. Considering
this interrelationship Allen and Taylor35 demonstrated that the
1JP–Se coupling constant was a surprisingly good measure of
the phosphine basicity irrespective of the size of the phosphine.
The general procedure still widely used for the preparation of
phosphine selenides involves refluxing the phosphine with Se
metal in a high boiling solvent, typically toluene. However, the
rapid reaction between SeCN− and phosphines to produce the
corresponding phosphine selenides has been well documented.37
Previous attempts to do a detailed kinetic investigation proved
troublesome38 and some indirect methods were rather followed.39
In this paper we report our results of a comprehensive kinetic study
between SeCN− and selected phosphine ligands as a kinetic probe
of isolating the electronic parameters of phosphorus ligands from
their steric properties. This method of analysis could in principle
be incorporated into the QALE model, or be utilised on its own
as a rapid way of determining ligand properties.
aDepartment of Chemistry, University of Johannesburg, P. O. Box 524,
Johannesburg, 2006, South Africa
bDepartment of Chemistry, University of the Free State, P. O. Box 339,
Bloemfontein, 9300, South Africa. E-mail: roodta.sci@ufs.ac.za; Fax: +27
51 444 6384; Tel: +27 51 401 2547
† CCDC reference numbers 657196 & 657197. For crystallographic data
in CIF or other electronic format see DOI: 10.1039/b712782k
‡ Electronic supplementary information (ESI) available: Observed pseudo-
first order rate constants for all reactions investigated. See DOI:
10.1039/b712782k
Experimental
§ Currently employed at: University of the Free State, Bloemfontein 9300,
Chemicals
South Africa
¶ Currently employed at: Sasol Technology Research & Development,
P.O. Box 1, Sasolburg, 1947, South Africa. Tel.: +27 16 960 4456; Fax: +27
11 522 3218. E-mail: fanie.otto@sasol.com
Methanol, ethanol, 1-propanol and acetone solvents were of
analytical grade (SaarChem) and dried according to literature
650 | Dalton Trans., 2008, 650–657
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
procedures.40 The ligands PPh3, PPh2Cy, PPhCy2, PCy3, P(4-Cl-
C6H4)3, P(4-F-C6H4)3, P(2-Me-C6H4)3, P(3-Me-C6H4)3, P(4-Me-
C6H4)3, PPh(2-OMe-C6H4)2, P(4-OMe-C6H4)3, PPh2(CH2C6H4),
P(2-furyl)3 and KSeCN (99%+) were obtained from Aldrich and
used as received.
Cyclohexyldiphenylphosphine selenide
Methanol solutions (10–20 mL) containing KSeCN (5.3 × 10−3 g,
0.037 mmol) and PPh2Cy (1.0 × 10−2 g, 0.037 mmol) were prepared
and the SeCN− solution was added drop-wise over a period of
5 min to the PPh2Cy solution with stirring at room temperature.
The solution was left to evaporate until dry; the solid residues
were extracted with chloroform and filtered to remove KCN. The
product was allowed to slowly evaporate, forming crystals suitable
for single crystal data collection in quantitative yield.
NMR measurements
1H, 13C{1H} and 31P{1H} NMR spectra were recorded at 295 K in
CDCl3 on a Varian Gemini 300 MHz instrument and referenced
1
to residual solvent signals for H and 13C{1H}. 31P{1H} NMR
Spectroscopic data. 1H NMR (CDCl3): dH (300 MHz; CDCl3;
Me4Si): 7.92(m, 4H), 7.45 (m, 6H), 1.80, 1.71, 1.60 1.29 (m, 11H);
13C{H} NMR (CDCl3): dC (75 MHz; CDCl3; CHCl3): 132.1 (d,
spectra were referenced to external H3PO4, chemical shifts are
reported in ppm, J values are given in Hz.
2
1
4C, J
= 9.7), 131.3 (d, 6C, 3/4J
= 2.9), 128.5 (d, 2C, J
C–P
C–P
1
= 11.5), 37.6 (d, 1C, JC–P = 48.7), 26.2, 26.0, 25.9, 25.6 (m,
C–P
Preparation of phosphines
5C); 31P NMR (CDCl3): dP (121 MHz; CDCl3; H3PO4): 46.25 (d,
1P, 1JSe–P = 724.9); 46.25 (s, 1P)
Synthesis of all phosphines or phosphine selenides involving air-
or moisture-sensitive chemicals or intermediates were performed
using dried solvents and standard anaerobic techniques.
Dicyclohexylphenylphosphine selenide
PPh2(2-OMe-C6H4), P(2-OMe-C6H4)3 and PPh(2,4-OMe-
C6H3)2 were prepared either by direct ortho metallation of anisole
with BuLi followed by the addition of the appropriate chlorophos-
phine or by metal/halogen exchange between BuLi and 1-bromo-
2,4-dimethoxybenzene followed by the addition of PPhCl2 ac-
cording to established methods.41 Ferrocenyldiphenylphosphine,42
PPh2Fc, and 1,3,5-triaza-7-phosphatricyclo[3.3.1.13,7]decane,43
PTA, were synthesized according to the published procedures.
Synthesised phosphines were checked for purity with their known
characterisations as reported.
The title compound was prepared in a similar manner as SePPh2Cy
also yielding crystals suitable for single crystal data collection.
Spectroscopic data. 1H NMR (CDCl3): dH (300 MHz; CDCl3):
7.88 (m, 2H), 7.49 (m, 3H), 2.27, 2.08, 1.69, 1.29 (m, 22H); 13C{H}
NMR (CDCl3): dC (75 MHz; CDCl3; CHCl3): 133.1 (d, 2C, 3JC–P
8.6), 131.4 (s, 1C), 128.3 (d, 3C, 1/2JC–P = 10.5), 36.6 (d, 2C, 1JC–P
=
=
4
3
43.0), 26.4 (d, 2C, JC–P = 2.9), 26.3 (d, 4C, JC–P = 11.4), 25.88
2
(d, 4C, JC–P = 26.2); 31P NMR (CDCl3): dP (121 MHz; CDCl3,
H3PO4): 55.87 (d, 1P, 1JSe–P = 701.2), 55.87 (s, 1P)
Preparation of phosphine selenides
General procedure
Crystallography
The X-ray intensity data for the compounds SePPh2Cy and
SePPhCy2 were measured on a Bruker SMART 1 K CCD area
detector diffractometer equipped with a graphite monochromator
and Mo Ka fine-focus sealed tube (k = 0.71073 A) operated
The phosphine selenide derivatives were prepared according
to a generalised procedure as described below for cyclo-
hexyldiphenylphosphine selenide (SePPh2Cy). 31P{H} NMR anal-
ysis indicated complete conversions to the corresponding phos-
phine selenide in all cases. Table 1 contains a summary of the
NMR data for the ligands investigated.44
˚
at 2.0 kW power (50 kV, 40 mA). The detector was placed at
a distance of 4.0 cm from the crystal. Data collections were
performed at room temperature (293 K). The initial unit cell
determinations and data collections were carried out by the
SMART45 software package. A total of 1315 frames with a frame
width of 0.3◦ were collected in x using an exposure time of 10 s
frame−1 in both cases. The frames were integrated using a narrow-
frame integration algorithm and reduced with the Bruker SAINT-
Plus and XPREP software packages46 respectively. Analysis of
the data showed no significant decay during the data collection.
Data were corrected for absorption effects using the multi-scan
technique SADABS.47
Table 1 31P NMR data for P(III) and P(V) compounds in CDCl3
No. Compound
dP(III) (ppm) dP(V) (ppm) 1JSe–P/Hz
1
PPh3
PPh2Cy
PPhCy2
PCy3
P(2-Me–C6H4)3
P(3-Me–C6H4)3
P(4-Me–C6H4)3
P(4-Cl–C6H4)3
P(4-F–C6H4)3
PPh2Fc
PPh2(CH2C6H5)
PPh2(2-OMe–C6H4)
PPh(2-OMe–C6H4)2
P(2-OMe–C6H4)3
−5.31
−1.33
4.98
36.77
46.25
55.87
59.18
28.98
36.29
34.60
34.09
33.23
32.66
35.24
33.10
28.42
20.45
26.57
32.46
−21.7
−35.62
728.9
724.9
701.2
672.9
704.6
721.9
717.6
746.9
740.2
731.1
735.1
720.7
717.5
720.2
706.8
711.3
786.5
753.8
2
3
4
11.13
5
−28.71
−20.01
−7.14
−7.66
−8.28
−16.18
−15.21
−15.68
−26.41
−36.54
The structures were solved by the direct methods package
SIR9748 and refined using the WinGX49 software, incorporat-
ing SHELXL.50 The final anisotropic full-matrix least-squares
refinement was performed on F2. The methine, methylene and
aromatic protons were placed in geometrically idealized positions
6
7
8
9
10
11
12
13
14
15
16
17
18
˚
(C–H = 0.98, 0.97 and 0.93 A) and constrained to ride on
their parent atoms with Uiso(H) = 1.2 Ueq(C). Non-hydrogen
atoms were refined with anisotropic displacement parameters. The
graphics were generated using the DIAMOND51 program with
30% probability ellipsoids for all non-hydrogen atoms. Details
of the crystal data, intensity measurements and data processing
PPh(2,4-OMe-C6H3)2 −29.08
P(4-OMe–C6H4)3
P(2-furyl)3
PTA
−29.12
−76.93
−97.96
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 650–657 | 651
©

View Article Online
Table 2 Crystal structure and refinement data for SePPh2Cy and SePPhCy2
SePPh2Cy
SePPhCy2
Empirical formula
Formula weight/g mol−1
Temperature/K
Crystal system
C18H21PSe
347.28
293(2)
Monoclinic
P21/c
15.140(2)
11.6421(19)
10.1580(17)
C18H27PSe
353.33
293(2)
Monoclinic
P21/c
11.100(2)
13.668(3)
11.773(2)
90
98.26(3)
90
1767.6(6)
4
1.328
2.205
Space group
˚
a/A
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
90
107.741(3)
90
1705.3(5)
4
1.353
2.285
3
˚
V/A
Z
Dc/Mg m−3
l/mm−1
F(000)
712
736
Crystal size/mm
0.32 × 0.23 × 0.21
0.28 × 0.22 × 0.14
1.9–28.3
99.2
h range/◦
1.4–28.3
Completeness for collection (%)
99.2
Limiting indices
−20 ≤ h ≤ 13, −15 ≤ k ≤ 13, −13 ≤ l ≤ 13
11246/4207/2441
0.0794
0.7113/0.3717
4207/0/181
−11 ≤ h ≤ 14, −18 ≤ k ≤ 18, −13 ≤ l ≤ 15
12106/4364/2688
0.0428
0.7477/0.5772
4364/0/181
Reflections collected/unique/observed (I > 2r(I))
R(int)
T
max/Tmin
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices (I > 2r(I))
R indices (all data)
0.913
1.004
R1 = 0.0368, wR2 = 0.0809
R1 = 0.0845, wR2 = 0.0938
0.598 and −0.568
R1 = 0.0375, wR2 = 0.0706
R1 = 0.0851, wR2 = 0.0820
0.258 and −0.280
−3
˚
Dqmax; Dqmin/e A
◦
˚
decomposition keeping in mind the actual time of the studied
reaction. The decomposition of the phosphines as well as for the
SeCN− solutions were negligible over the reaction period studied.
The oxidative addition reactions were studied extensively in
methanol by observing the absorbance changes at a suitable
fixed wavelength in the UV region (200–355 nm) for the various
phosphorus containing compounds as selected from pre-recorded
absorbance vs. wavelength time resolved spectra. Reactions with
selected phosphines were in addition also investigated in ethanol,
1-propanol and acetone. The SeCN− solutions were of ten to fifty-
fold excess compared to the phosphines ensuring pseudo first-order
conditions in all cases. Observed rate constants determined by
stopped-flow techniques are given as the average of at least three
individual runs. Fig. 1 shows a typical reaction between PPh3 and
SeCN− to illustrate the methods as described above.
Table 3 Selected bond lengths (A) and angles ( ) for SePPh2Cy and
SePPhCy2
SePPh2Cy
SePPhCy2
Se–P
2.1172(7)
1.834(2)
1.814(2)
1.825(3)
2.1271(8)
1.835(2)
1.835(2)
1.828(2)
P–C11
P–C21
P–C31
C11–P–Se
C21–P–Se
C31–P–Se
113.38(8)
112.60(8)
113.12(9)
111.07(8)
112.97(8)
113.28(8)
for the two structures are summarized in Table 2 and important
geometrical parameters in Table 3.
CCDC reference numbers 657196 and 657197 for SePPh2Cy
and SePPhCy2 respectively.
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b712782k
The kinetic traces of all the reactions showed well defined first-
order behaviour corresponding to the reaction given in eqn (1):
k
1
SeCN− + PR3 SePR + CN−
(1)
ꢀ
3
Kinetic measurements
k
−1
UV-Vis measurements were performed using a Varian Cary 100
spectrophotometer for slow reactions, while faster reactions (t1/2
<
The observed rate constants were obtained from least-squares
fits of the absorbance vs. time traces to a first-order model
using Scientist.52 The observed pseudo first-order rate constants
were fitted vs. the SeCN− concentration using the linear model.
Temperature variations for selected phosphines were performed
between 288.2 and 318.2 K and the corresponding activation
parameters were calculated using the exponential form of the
Eyring equation. Complete raw data for the individual kinetics
are given as ESI.‡
1 min) were performed on a Hi-Tech Scientific SF-61 DX2
double mixing stopped-flow instrument equipped with a methanol
cooled cell for sub-zero kinetic runs as well as a diode-array
system to evaluate kinetic spectra of fast reactions. All apparatus
were equipped with constant temperature water baths (Julabo)
regulating the reaction temperature to within 0.1 ◦C.
Initial kinetic experiments were performed using freshly pre-
pared phosphine and SeCN− solutions and were checked for
652 | Dalton Trans., 2008, 650–657
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
studied in all cases. In principle this correlation could also be
extended to the corresponding phosphine oxides and sulfides
but falls beyond the scope of the current investigation. The
graph shows three distinct regions corresponding to the type
of substituents on the phosphorus centre, i.e. 2–4 = alkyl, 17–
18 = heteroatom and 1, 5–16 = aryl substituents. The gradient
of the least-squares fit is 1.15 0.06, indicating that both steric
and electronic properties of the different type of substituents are
influenced in a similar way.
Crystallography
In addition to NMR analysis of the reaction products, SePPh2Cy
and SePPhCy2 were also characterised using single crystal X-ray
diffraction. The numbering schemes and molecular structures are
illustrated in Fig. 3.
Fig. 1 Typical time resolved absorbance vs. wavelength spectra used to
determine first order reaction rates. Data in absorbance vs. time graph from
k = 310 nm is fitted to exponential first-order kinetic equation. [SeCN−] =
9.715 × 10−3 mol dm−3; [PPh3] = 9.5 × 10−4 mol dm−3; temperature =
298.2 K; solvent = MeOH; kobs = 9.26 0.04 × 10−3 s−1
.
Results and discussion
Synthesis
The reactions between tertiary phosphines and SeCN− proceed
at moderate to fast rates to produce the corresponding phosphine
selenides in virtually quantitative yields. All reaction products were
unambiguously characterised by means of their characteristic 31
P
NMR spectra, clearly exhibiting the coupling associated with the
P–Se interaction (77Se 7.58% spin 1/2), see Table 1.
NMR
1
The JP–Se coupling constants increase as the substituents on
phosphorus become more electron withdrawing, indicating an
increased s-character for the phosphorus orbital involved in
bonding to selenium, in accordance with Bent’s rule.53 The 31P
shifts between the d{P(III)} and d{P(V)} centres show a linear
relationship (see Fig. 2) confirming that the same reaction was
Fig. 3 Molecular diagrams showing the numbering scheme and displace-
ment ellipsoids (30% probability) for (a) SePPh2Cy and (b) SePPhCy2.
Numbering scheme of the C-rings: first digit indicates ring number; second
digit indicates number of the atom in the ring.
Selected geometrical parameters from these compounds, calcu-
lated according to the Tolman model,31 are compared in Table 4
to relevant structures from the literature and the Cambridge
Structural Database.54 There are two important aspects to note
about the Tolman cone angles (hT) reported here: (i) Actual
geometrical crystal data is used for the calculation and not
optimised CPK models as those used by Tolman,31 and (ii) in
accordance with Tolman’s definition, a dummy atom was placed
˚
at a distance 2.28 A from phosphorus in line with the P–Se bond
thus ignoring any effect the P–Se bond distance may have on this
value.
31
Fig. 2 Graph showing the linear relationship between d{ P(III)} and
d{ P(V)} centers obtained from 31P NMR spectra in Table 1.
31
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 650–657 | 653
©

View Article Online
Table 4 Selected geometrical parameters from structural studies of
phosphine selenides
effectively separating the electronic and steric parameters in the
process.
hT/◦
Reference
55
a
˚
Phosphine selenide
SePPh3
Se–P/A
Kinetic investigation
2.105/2.107
2.111/2.113
2.111(2)
2.1260(8)
2.108(1)
2.052
2.116(1)
2.109(5)
2.1119(5)
176/175
176/179
168
Graphs of the observed pseudo first-order rate constants for PPh3
vs. [SeCN−] concentration at various temperatures are and in
various solvents at 298.2 K are reported in Fig. 4(a) and 4(b)
respectively.
56
b
SePPh2Cy
SePPhCy2
SePCy3
SeP(OPh)3
SeP(2-Me–C6H4)3
SeP(3-Me–C6H4)3
SeP(4-Me–C6H4)3
SeP(4-F–C6H4)3
SePPh2Fc
b
169
168
183
190
186
169
57
58
59
60
61,c
2.1149(5)/2.1113(6) 166, 168
62
d
2.1101(5)
2.1117(8)
SePPh(2,4-OMe–C6H4)2 2.1219(5)
170
167
186
SePPh2(CH2C6H5)
63
d
SeP(2,6-OMe–C6H3)3
SeP(2,4,6-OMe–C6H2)3
SeP(2-furyl)3
2.136(2)
2.120
2.094
2.0991(19)
2.100(2)
208
196
141, 125
112
64
64
36, 65
66
66
SePTA
[SeMePTA]Ie
115
a Tolman cone angles from actual crystallographic data of Se derivatives
with Se–P distances adjusted to 2.28 A. b This work. c Data recollected to
˚
obtain geometrical parameters as no 3D coordinates were available from
CSD. d Future work, to be published elsewhere. e 1-Methyl-1-azonia-3,5-
diaza-7-phosphatricyclo(3.3.1.13,7)decane-7 selenide iodide.
In combination with increased s-character of the phosphorus
lone pair, associated with electron withdrawing substituents, the P–
Se bond is expected to shorten due to more effective orbital overlap
as reflected by the increasing 1JP–Se coupling constant. Even though
a direct correlation between the P–Se bond distance and 1JP–Se was
claimed previously67 for a narrow selection of structurally related
phosphines, this trend becomes significantly less pronounced when
considering a wider selection of phosphine selenides as presented
here. Notably, no definite trend is observed in the series PPh3,
PPh2Cy, PPhCy2 and PCy3 or that of P(2-Me-C6H4)3, P(3-Me-
C6H4)3 and P(4-Me-C6H4)3.
It is important to note than even though a cyclohexyl group is
regarded to have more steric bulk that a phenyl, a discrepancy is
noted for the cone angles in tertiary phosphine ligands containing
cyclohexyl moieties, all showing a tendency to average at ca.
168◦ regardless of the number of cyclohexyl substituents on the
phosphorus atom. This can be explained by the substituents ability
to rotate around the P–C axis in an attempt to occupy all free
space in the proximity of the phosphorus atom. This “free space”
appears to have a threshold value of ca. 168◦ and corresponds to
that observed from previous studies68 involving PPh3 where the
cone angles for PPh3 were reported to have a mean value of 148◦
with a standard deviation of 5◦ and a spread from 129 to 168◦.
The total spread of P–Se bond distances found in the literature
Fig. 4 Plots of kobs vs. [SeCN−] for the reaction with PPh3 at (a) various
temperatures in MeOH and (b) in various solvents at 298.2 K. [PPh3] =
9.535 × 10−4 mol dm−3, k = 310 nm except for plot in acetone (k = 330 nm).
All reactions displayed linear relationships between kobs and
[SeCN−] and zero intercepts (within experimental error) were
obtained for all ligands, except 5 and 17. A non-zero intercept
was also noted when acetone is used as solvent. This observation
could be due to several influences such as (i) the variable stability
(ii) solubility of KSeCN in various solvents as well as (iii) ion
pairing effects in solution. With this in mind, all data were
restrained to intercept at zero where experimental data approached
a zero intercept. All reactions (see Fig. 1) could be described
by a two-term rate law, as shown in eqn (2), containing SeCN−
concentration dependent and independent terms, respectively:
˚
covers only 0.12 A and even though sensitive to some extent to
the electronic properties of the ligand, vide infra, it is remarkably
unaffected by the size of the ligands, indicating steric effects to
1
be significantly less important. In contrast, the JP–Se coupling
constants cover a range larger than 100 Hz, indicating a high
sensitivity towards the electronic properties of the phosphine
ligands. In this regard it is well known that the rate of oxidative
addition reactions are favoured by increased electron density of
the reductant and in the absence of steric factors it should be a
reliable kinetic parameter of electron density of the phosphines,
ꢀ
ꢁ
kobs = k1 SeCN− + (k−1 + ks)
(2)
654 | Dalton Trans., 2008, 650–657
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 5 Data of the reactions between SeCN− and phosphines in MeOH at 298 K
a
No.
Compound
k1/mol dm−3 s−1
Relative rate to PPh3
pKa
DH‡/kJ mol−1
DS‡/J K−1 mol−1
−124.8 1.1
1
PPh3
PPh2Cy
PPhCy2
PCy3
1.05 0.02
4.9 0.3
14.1 0.4
1.00
4.67
13.43
48.57
0.0015
1.57
2.14
0.59
0.90
1.26
2.73
35.7 0.4
2
3
27
47
1
1
−132
−140
3
4
51
3
9.65
5
P(2-Me-C6H4)3
P(3-Me-C6H4)3
P(4-Me-C6H4)3
P(4-Cl-C6H4)3
P(4-F-C6H4)3
PPh2Fc
PPh2(CH2C6H5)
PPh2(2-OMe-C6H4)
PPh(2-OMe-C6H4)2
P(2-OMe-C6H4)3
PPh(2,4-OMe-C6H3)2
P(4-OMe-C6H4)3
P(2-furyl)3
0.00162 0.00009
1.65 0.04
2.25 0.05
3
6
3.3
7
3.84
1.03
1.97
32
39
2
1
−131
−118
7
4
8
0.621 0.007
0.947 0.005
1.32 0.02
0.700 0.009
0.184 0.005
0.0203 0.0009
0.00134 0.00002
0.0520 0.0019
3.90 0.09
9
10
11
12
13
14
15
16
17
18
0.67
0.18
39.0 0.3
46
−128
1
0.019
0.0013
0.050
3.71
0.012
0.35
3
0–124 10
40.5 0.3
30.7 0.8
49.0 1.3
−133.6 0.9
4.57
−131
−117
−112
2
4
9
0.0121 0.0004
0.371 0.013
PTA
42
3
a pKa values obtained from ref. 69–72.
The intercept obtained from the plot of kobs vs. [SeCN−] may in
addition comprise of a reversible contribution (k−1 as indicated in
eqn (1)) as well as/or a contribution from a concurrent solvent
assisted pathway (ks). Reversibility of eqn (1) is not expected
under normal reaction conditions though and a solvent assisted
contribution may in addition be likely in the cases of the exceptions
mentioned above. For P(2-furyl)3 interaction between O and
Se was previously presented as an explanation for its unusual
behaviour while Me and OMe substitution on the 2 position of
aryl rings showed through-space interactions (close proximity)
with Se, resulting in a deviation from general trends.36
Temperature variations on selected ligands were used to deter-
mine the activation parameters for the forward step, see Table 5.
Both the large negative values for DS‡ and the increase in reaction
rate with increasing polarity of the solvent (Fig. 4b) are classical
indications of an SN2 oxidative addition mechanism. The initial
nucleophilic attack of the phosphine on the Se atom should thus be
rate determining while breaking of the Se–C bond should follow in
a fast consecutive step. A variable that was not investigated during
this study was the use of alternative soluble sources of selenium,
which is also a topic of future investigation.
Data from the kinetic investigations for this study are sum-
marised in Table 5 and is organized according to related R groups
on the phosphorus atom. The rate for PPh3 was normalised to
unity to facilitate comparison with other phosphine ligands.
It is clear that the rate of reaction is highly dependent on the
electronic nature of the R groups attached to the phosphorus
atom, and spans a five order of magnitude range. This sensitivity
is visually illustrated by constructing a logarithmic graph of pseudo
first-order rate constants vs. [SeCN−] for a selected range of
phosphines (see Fig. 5). Rates of reaction for 5–7 and 12–16 show
the electronic effect induced by substituents on various positions of
an aromatic ring. It should be taken into account that the positions
of these substituents not only influence the electronic nature of
the phosphorus atom, but also the effective steric bulk, hence the
observed deviation from expected reaction rates for phosphines
containing substituents in the 2 position. Slower reactions are
Fig. 5 Selected plots of kinetic data on a logarithmic graph of kobs vs.
[SeCN−] at 298.2 K in MeOH illustrating the five orders of magnitude
difference in reactivity.
observed for electron withdrawing groups such as halides while
electron donating groups such as methyl and methoxy on the 3
and 4 positions increase the reaction rate.
Several correlations between the rate constants, k1, and a mea-
sure of the electronic properties of the ligands can be constructed.
In this regard excellent relationships are demonstrated in Fig. 6
1
between log k1 and the pKa of the phosphines and JP–Se of the
resulting phosphine selenides respectively. The graph in Fig. 6b
show outliers, i.e. 5 and 12–15, which are primarily all phosphines
containing varying degrees of substitution in the 2 position. This
observation supports the conclusion that the reaction rates are
not only governed by electronic effects but a steric component
also becomes significant under certain conditions. In principle
these graphs could be used to estimate the pKa values of new
phosphine ligands, provided no aryl groups, substituted in the 2
position, are present. In similar fashion a graph of log k1 vs. vd can
be constructed showing comparable results with Fig. 6a and b.73
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 650–657 | 655
©

View Article Online
Acknowledgements
Financial support from the research funds of the University of
Johannesburg, the University of the Free State, the South African
NRF [GUN Nr. 2067416], SASOL and THRIP are gratefully
acknowledged. The University of the Witwatersrand (Professor
D. Levendis and Dr D. Billing) is thanked for the use of its
diffractometer on data collections for SePPh2Cy and SePPhCy2.
References
1 S. Song and E. C. Alyea, Comments Inorg. Chem., 1996, 18, 145.
2 E. C. Alyea and S. Song, Comments Inorg. Chem., 1996, 18, 189.
3 R. H. Crabtree, The Organometallic Chemistry of the Transition Metals,
2nd edn, Wiley, New York, 1994.
4 L. H. Pignolet, Homogeneous Catalysis with Phosphine Complexes,
Plenum, New York, 1983.
5 Comprehensive Organometallic Chemistry, ed. G. Wilkinson,
F. G. A. Stone and E. W. Abel, Pergamon Press, Oxford, UK, 1982,
vol. 3.
6 G. Wilkinson, R. D. Gillard, and J. A. McCleverty, Comprehensive
Coordination Chemistry, ed. G. Wilkinson, R. D. Gillard and J. A. Mc-
Cleverty, Pergamon Press, Oxford, UK, 1987, vol. 2.
7 O. Stelzer, Topics in Phosphorus Chemistry, ed. E. J. Griffith,
M. Grayson, Interscience, New York, 1977, vol. 9.
8 E. C. Alyea, Catalytic Aspects of Metal Phosphine Complexes,
in Advances in Chemistry Series 196, American Chemical Society,
Washington, DC, 1982.
9 J. A. Tossell, J. H. Moore and J. C. Giordan, Inorg. Chem., 1985, 24,
1100.
10 J. Rolke and C. E. Brion, Chem. Phys., 1996, 207, 173.
11 A. G. Orpen and N. G. Connelly, J. Chem. Soc., Chem. Commun., 1985,
1310.
12 A. G. Orpen and N. G. Connelly, Organometallics, 1990, 9, 1206.
13 B. J. Dunne, R. B. Morris and A. G. Orpen, J. Chem. Soc., Dalton
Trans., 1991, 653.
14 M. Caffery and T. L. Brown, Inorg. Chem., 1991, 30, 3907.
15 K. J. Lee and T. L. Brown, Inorg. Chem., 1992, 31, 289.
16 T. L. Brown, Inorg. Chem., 1992, 31, 1286.
17 M.-G. Choi and T. L. Brown, Inorg. Chem., 1993, 32, 1548.
18 M.-G. Choi and T. L. Brown, Inorg. Chem., 1993, 32, 5603.
19 G. Pacchioni and P. S. Bagus, Inorg. Chem., 1992, 31, 4391.
20 O. D. Haberlen and N. Rosch, J. Phys. Chem., 1993, 97, 4970.
21 R. Schmid, W. A. Herrmann and G. Frenking, Organometallics, 1997,
16, 701.
22 B. Cornils and W. A. Herrmann, Applied Homogeneous Catalysis with
Organometallic Chemistry, VCH Publishers, New York, 1996.
23 P. W. N. M. van Leeuwen, Homogeneous Catalysis—Understanding the
Art, Kluwer Academic Publishers, The Netherlands, 2004.
24 M. J. S. Dewar, Bull. Soc. Chim. Fr., 1951, 18, C71.
25 J. Chatt and L. A. Duncanson, J. Chem. Soc., 1953, 2939.
26 Y. Ruiz-Morales and T. Ziegler, J. Phys. Chem. A, 1998, 102, 3970.
27 S. Xiao, W. C. Trogler, D. E. Ellis and Z. Berkovitch-Yellin, J. Am.
Chem. Soc., 1983, 105, 7033.
28 D. S. Marynik, J. Am. Chem. Soc., 1984, 106, 4064.
29 P. Fantucci, Comments Inorg. Chem., 1992, 13, 241.
30 For detailed description on QALE see: http://www.bu.edu/qale and
references therein.
Fig. 6 Plots of (a) log k1 vs. pKa and (b) log k1 vs. 1JSe–P. Reaction rates are
at 298.2 K in MeOH; pKa values from Table 5, 1JSe–P values from Table 1,
1JSe–P constants in CDCl3.
It is anticipated that other correlations could be constructed
using other measures of electronic properties, such as the car-
bonyl IR stretching frequencies in [Ni(PR3)(CO)3]31 and trans-
[RhCl(CO)(PR3)2],32 and indeed, reasonable relationships exist.
Comparison of the rate constants with the cone angles given in
Table 5 clearly demonstrates the significant effect of 2-substitution
on aryl rings on the observed reactivity of the ligands. It seems that
a threshold value for steric influence is somewhere in the region
of 190◦ although this is a rough estimate which requires more
investigation.
Conclusion
31 C. A. Tolman, Chem. Rev., 1977, 77, 313.
32 S. Otto and A. Roodt, Inorg. Chim. Acta, 2004, 357, 1.
33 A. Roodt, S. Otto and G. Steyl, Coord. Chem. Rev., 2003, 245, 121.
34 A. H. Cowley and M. C. Damasco, J. Am. Chem. Soc., 1971, 93, 6815.
35 D. W. Allen and B. F. Taylor, J. Chem. Soc., Dalton Trans., 1982, 51.
36 D. W. Allen, I. W. Nowel and B. F. Taylor, J. Chem. Soc., Dalton Trans.,
1985, 2505.
37 P. Nicpon and D. W. Meek, Inorg. Chem., 1966, 5, 1297.
38 J. Songstad and L. J. Stangeland, Acta Chem. Scand., 1970, 24, 804.
39 (a) L. J. Stangeland, T. Austad and J. Songstad, Acta Chem. Scand.,
1973, 27, 3919; (b) D. H. Brown, R. J. Cross and R. Keat, J. Chem. Soc.,
Dalton Trans., 1980, 871.
This study reports the first direct systematic kinetic study of the
oxidative addition reaction between phosphines and SeCN−. The
reaction rates are significantly dependent on the electronic proper-
ties of the phosphine ligands and in general, good correlations with
other measures of electron density, were obtained. Substituents
in the 2-position of aromatic rings have a retarding effect on
the reaction rates while bulky trialkyl phosphines are unaffected.
Kinetic studies of this nature may serve as an additional step
towards finding the ‘magic bullet’ to rapidly evaluate the electron
density on a phosphorus donor ligand.
40 D. D. Perrin and W. L. F. Armarego, Purification of Laboratory
Chemicals, 3rd edn, Pergamon Press, Oxford, UK, 1988.
656 | Dalton Trans., 2008, 650–657
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
41 H. Riihima¨ki, T. Kangas, P. Suomalainen, H. K. Reinius, S.
Ja¨a¨skela¨inen, M. Haukka, A. O. I. Krause, T. A. Pakkanen and J. T.
Pursiainen, J. Mol. Catal. A: Chem., 2003, 200, 81.
56 P. G. Jones, C. Kienitz and C. Thone, Z. Kristallogr., 1994, 209, 80.
57 J. A. Davies, S. Dutremez and A. A. Pinkerton, Inorg. Chem., 1991, 30,
2380.
42 G. P. Sollot and H. Edgar, (Jr.), J. Org. Chem., 1962, 4034.
43 D. J. Daigle, Inorg. Synth., 1998, 32, 40.
58 J. Clade and M. Jansen, Z. Kristallogr.–New Cryst. Struct., 2005, 220,
234.
44 A typical shift observed for the oxidative addition product and the 77Se
satellites can be found in the ESI‡.
59 T. S. Cameron and B. Dahlen, J. Chem. Soc., Perkin Trans. 2, 1975,
1737.
45 Bruker SMART-NT Version 5.050. Bruker AXS Inc., Area-Detector
60 T. S. Cameron, K. D. Howlett and K. Miller, Acta Crystallogr., Sect.
Software Package, Madison, WI, USA, 1998.
B, 1978, B34, 1639.
46 Bruker SAINT-Plus Version 6.02 (including XPREP), Bruker AXS
Inc., Area-Detector Integration Software, Madison, WI, USA, 1999.
47 Bruker SADABS Version 2004/1. Bruker AXS Inc., Area Detector
Absorption Correction Software, Madison, WI, USA, 1998.
48 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacov-
azzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and R. Spagna,
J. Appl. Crystallogr., 1999, 32, 115.
61 Zhdanov +, Dokl. Akad. Nauk SSSR (Russ.) (Proc. Natl. Acad. Sci.
USSR), 1953, 92, 983.
62 A. Muller and R. Meijboom, Acta Crystallogr., Sect. E, 2007, E63,
o4055.
63 A. Muller, Acta Crystallogr., Sect. E., 2007, E63, o317.
64 D. W. Allen, N. A. Bell, L. A. March and I. W. Nowell, Polyhedron,
1990, 9, 681.
65 R. E. Marsh, Acta Crystallogr., Sect. B, 1997, B53, 317.
66 S. Otto, A. Ionescu and A. Roodt, J. Organomet. Chem., 2005, 690,
4337.
49 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
50 G. M. Sheldrick, SHELXL97. Program for Crystal Structure Refine-
ment, University of Go¨ttingen, Germany, 1997.
51 K. Brandenburg and M. Berndt, DIAMOND Release 2.1e. Visual
Crystal Structure Information System, CRYSTAL IMPACT, Postfach
1251, D-53002, Bonn, 2001.
52 SCIENTIST Version 2.1, Program for Least-Squares Parameter Opti-
mization; MicroMath Scientific Software, Utah, 1995.
53 H. A. Bent, Chem. Rev., 1961, 61, 275.
54 Cambridge Structural Database (CSD), Version 5.26, February 2006
update; F. H. Allen, Acta Crystallogr., Sect. B, 2002, B58, 380.
55 P. W. Codding and K. A. Kerr, Acta Crystallogr., Sect. B, 1979, B35,
1261.
67 J. Malito and E. C. Alyea, Phosphorus, Sulfur Silicon Relat. Elem., 1990,
54, 95.
68 K. A. Bunten, L. Chen, A. L. Fernandez and A. J. Poe¨, Coord. Chem.
Rev., 2002, 233–234, 41.
69 W. A. Henderson and C. A. Streuli, J. Am. Chem. Soc., 1960, 82, 5791.
70 T. Allman and R. G. Goel, Can. J. Chem., 1982, 60, 716.
71 R. A. Jackson, R. Kanluen and A. Poe¨, Inorg. Chem., 1984, 23, 523.
72 C. A. Streuli, Anal. Chem., 1959, 31, 1652.
73 The plot of log k1 vs. vd for selected phosphorus(III) compounds can be
found in the ESI‡.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 650–657 | 657
©