More user-friendly phosphines? Molecular structure of methylphosphine and its adduct with borane, studied by gas-phase electron diffraction and quantum chemical calculations

Article abstract of DOI:10.1039/b804780d

The molecular structures of methylphosphine (CH₃PH₂) and methylphosphine-borane (CH₃PH₂?BH₃) have been determined from gas-phase electron diffraction data and rotational constants, employing the SARACEN method. The experimental geometric parameters generally showed a good agreement with those obtained using ab initio calculations and previous microwave spectroscopy studies. In order to assess the accuracy of the calculated structures a range of ab initio methods were used, including the CCSD(T) method, with correlation-consistent basis sets. The structural environment around the phosphorus atom was found to change significantly upon complexation with borane, with the P-C bond length shortening and the bond angles widening. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b804780d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
More user-friendly phosphines? Molecular structure of methylphosphine and
its adduct with borane, studied by gas-phase electron diffraction and quantum
chemical calculations†
a
a
a
a
Robert Noble-Eddy, Sarah L. Masters (n e´ e Hinchley),* David W. H. Rankin, Derek A. Wann,
b
b
Brahim Khater and Jean-Claude Guillemin
Received 19th March 2008, Accepted 13th June 2008
First published as an Advance Article on the web 1st August 2008
DOI: 10.1039/b804780d
The molecular structures of methylphosphine (CH
3
PH
2
) and methylphosphine–borane (CH
3
PH
2
·BH
3
)
have been determined from gas-phase electron diffraction data and rotational constants, employing the
SARACEN method. The experimental geometric parameters generally showed a good agreement with
those obtained using ab initio calculations and previous microwave spectroscopy studies. In order to
assess the accuracy of the calculated structures a range of ab initio methods were used, including the
CCSD(T) method, with correlation-consistent basis sets. The structural environment around the
phosphorus atom was found to change significantly upon complexation with borane, with the P–C
bond length shortening and the bond angles widening.
Introduction
be present in larger molecules. 1 and 2 also present the opportunity
to conduct high-level ab initio calculations, which would be
prohibitively expensive for much larger phosphines. The GED
study of larger phosphines will rely more heavily on the use of ab
initio data via the SARACEN method, and so it is vital to gauge
the accuracy of various theoretical methods in order to identify
suitable computational techniques for future work.
Lewis acid–base adducts are also interesting in a more general
sense. Previous research has found large deviations between the gas
phase and crystalline structures of some Lewis acid–base adducts;
Relatively few studies have been conducted on the structures of
the primary aliphatic phosphines, R–PH and they are less widely
used throughout chemistry than the aryl phosphines. This is due
to their unfavourable properties; they are often pyrophoric, toxic,
unstable in air and possess unpleasant odours. Primary phosphines
2
5–7
1
have potential as starting materials for various applications
and therefore the development of more air-stable, user-friendly
phosphines is of great interest. Increased air and moisture
stability can be achieved by the addition of a bulky protecting
group, for example the mesityl group in mesitylphosphine, or by
complexation with a suitable Lewis acid. Complexes with borane
for example, in HCN–BF
3
the B–N bond length is found to be
2
8
8
3.5(31) pm shorter in the crystalline solid than in the gas phase.
It is therefore apparent that the bond between such a Lewis acid–
base complex can be significantly shortened by the effects of crystal
3
have been reported and the chemistry of the adducts explored
but only a few structural studies have been reported. Tertiary
4
9
packing and so it is vital that the structure is studied in the gas
phosphine–borane complexes are considered to be protected
free phosphines, with the free phosphine easily recovered by
reaction with excess amine. It has been shown that the same is
possible for primary phosphines, albeit with a reduced yield. The
gas phase electron diffraction (GED) study of methylphosphine
phase, free from the potential distortions present in the crystalline
solid.
The structures of both systems have previously been studied
3
to some degree, with a structure of 2, determined by microwave
10
(MW) spectroscopy, yielding a P–B bond length of 190.6(6) pm,
(
1) and methylphosphine–borane (2), as one of the simplest
close to the sum of the covalent radii for the two atoms. 1 has
examples of a primary phosphine and its borane adduct, provides
a starting point for the investigation of larger functionalised
primary phosphines. The degree of increased stability imparted
by formation of the adduct is of particular interest, especially with
regard to the structural changes accompanying the complexation.
Studying a simple system such as methylphosphine allows analysis
of these phenomena without the increased complexity that would
11
12
previously been studied by both GED and MW spectroscopy,
but both studies failed to yield complete structures. We have
revisited both compounds utilising modern GED techniques and
analysis methods to allow complete structural determination
for both molecules using a combination of GED data and
published rotational constants. Such an analysis allows us to
draw conclusions on the effects of complexation using comparable
results.
a
School of Chemistry, University of Edinburgh, West Mains Road, Edin-
burgh, UK EH9 3JJ. E-mail: s.masters@ed.ac.uk
b
Sciences Chimiques de Rennes, E´ cole Nationale Sup e´ rieure de Chimie de
Experimental
Rennes-CNRS, 35700 Rennes, France
†
Electronic supplementary information (ESI) available: Syntheses, GED
Synthesis
parameters, molecular geometries, interatomic distances, amplitudes of
vibration, least-squares correlation matrices, experimental gas-phase
electron diffraction coordinates and experimental–theoretical molecular
scattering intensities for (1) and (2). For ESI see DOI: 10.1039/b804780d
3,13
14
The syntheses of methylphosphine,
phosphine–borane are given in the ESI.†
diborane and methyl-
3,15
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 5041–5047 | 5041

View Article Online
Theoretical methods
6–311++G** level for 1 and 2 served both to confirm the nature of
the minima found by the optimisation and to provide vibrational
information for use in the SARACEN refinement.
Calculations for methylphosphine were performed using a Linux
cluster, whilst those for methylphosphine–borane were performed
using a Silicon Graphics Altix 4700, both using the Gaussian 03
16
Gas electron diffraction measurements
program. All MP2 methods were frozen core (fc).
Geometry optimisations. An extensive search of the torsional
potential of each compound was undertaken at the RHF/3–21G*
Data were collected for methylphosphine and methylphosphine–
38
borane using the Edinburgh gas-diffraction apparatus. For each
molecule, an accelerating voltage of ca. 40 kV (electron wavelength
ca. 6.0 pm) was used. Sample and nozzle temperatures were
maintained at 220 and 293 K, respectively, for 1 and 300 and 320 K,
respectively, for 2. Scattering intensities were recorded on Kodak
electron image films at nozzle-to-plate distances of 127.8 and
284.6 mm for 1 and 92.0 and 249.2 mm for 2. The weighting points
for the off-diagonal weight matrices, correlation parameters and
scale factors for the two camera distances for each molecule are
given in Table S1† in the ESI, together with electron wavelengths,
which were determined from the scattering patterns of benzene
vapour, recorded immediately after the compound patterns and
analysed in exactly the same way to minimise systematic errors in
wavelengths and camera distances. The scattering intensities were
measured using an Epson Expression 1680 Pro flatbed scanner
and converted to optical densities as a function of the scattering
1
7–19
level
with C
to locate all minima. For each molecule one minimum
symmetry was located. Further geometry optimisations
s
20
21
were conducted for both molecules at the HF, MP2 and
CCSD(T)
22–25
levels of theory. At the MP2 and CCSD(T) levels
optimisations were conducted using both the Pople-type basis sets
26–28
29,30
(
6–31G*
sets of Dunning and co-workers.
and 2 with the atomic numbering schemes are shown in Fig. 1.
and 6–311G* ) and the correlation-consistent basis
31–35
The optimised structures of
1
39
variable, s, using an established program. Data reduction and
least-squares refinements were carried out using the ed@ed v2.4
40
41
program, employing the scattering factors of Ross et al.
Results
Ab initio calculations
Methylphosphine. The lowest-energy structure of methylphos-
phine on the potential-energy surface had a staggered conforma-
tion and possessed C symmetry. The effects of improving the
s
basis set and description of electron correlation on the structural
parameters were gauged by a series of calculations at the MP2 and
CCSD(T) levels of theory using both Pople-type and correlation-
consistent basis sets. The results of selected calculations are shown
in Table 1 with the full results from geometry optimisations given
in the ESI, in Table S2 and Table S3.†
Fig. 1 Molecular structures of methylphosphine (1) and methylphos-
phine–borane (2).
36,37
Frequency calculations. Analytic second derivatives
of the
The structure of 1 was found to be largely independent of
the level of theory used, with two main exceptions. In the
energies with respect to nuclear coordinates calculated at the MP2/
a
Table 1 Molecular geometries of the lowest-energy structures of methylphosphine (1) at the MP2 and CCSD(T) levels of theory
MP2
CCSD(T)
b
6
–311++G**
aug-cc-pVQZ
6–311++G**
aug-cc-pVQZ
rC(1)–H(3)/pm
rC(1)–H(4)/pm
rC(1)–P(2)/pm
rP(2)–H(6)/pm
109.2
109.1
185.6
141.0
109.0
113.8
108.7
107.5
97.6
108.7
108.5
185.1
141.1
108.6
113.4
109.2
107.7
97.8
109.6
109.4
186.7
141.7
109.0
113.6
108.7
107.6
97.2
109.1
108.9
185.9
141.7
108.9
113.2
109.2
107.8
97.6
◦
◦
∠
P(2)–C(1)–H(3)/
∠
P(2)–C(1)–H(4)/
◦
∠
H(3)–C(1)–H(4)/
◦
◦
◦
∠H(3)–C(1)–H(5)/
∠C(1)–P(2)–H(6)/
∠H(6)–P(2)–H(7)/
94.6
–381.8117
93.5
−381.9166
94.4
−381.8599
93.3
−381.9685
Energy/Hartrees
a
b
See Fig. 1 for atom numbering. Extrapolated using CCSD(T)/aug-cc-pVQZ = CCSD(T)/aug-cc-pVTZ + [MP2/aug-cc-pVQZ–MP2/aug-cc-pVTZ].
5
042 | Dalton Trans., 2008, 5041–5047
This journal is © The Royal Society of Chemistry 2008

View Article Online
a
Table 2 Molecular geometries of the lowest-energy structure of methylphosphine–borane (2) at the MP2 and CCSD(T) levels of theory
MP2
CCSD(T)
b
Parameter
6–311++G**
aug-cc-pVQZ
6–311++G**
aug-cc-pVQZ
rC(1)–H(3)/pm
rC(1)–H(4)/pm
rC(1)–P(2)/pm
rP(2)–H(6)/pm
rP(2)–B(8)/pm
rB(8)–H(9)/pm
rB(8)–H(11)/pm
109.2
109.1
182.1
140.1
192.8
120.9
120.6
108.6
112.3
109.5
108.2
103.2
100.3
115.0
103.3
106.1
113.7
114.3
−408.3568
108.7
108.5
181.3
140.0
191.4
120.6
120.2
108.1
112.4
109.9
108.3
103.5
99.8
114.5
102.8
106.3
113.8
114.7
−408.4922
109.5
109.4
182.9
140.5
193.9
121.3
121.0
108.6
112.2
109.5
108.4
103.1
100.3
115.2
103.4
106.0
113.8
114.3
−408.4268
109.0
108.9
182.2
140.5
192.5
121.1
120.6
108.3
112.0
109.9
108.4
103.4
99.7
114.7
102.9
106.4
113.7
114.7
−408.5644
◦
◦
∠
P(2)–C(1)–H(3)/
∠
P(2)–C(1)–H(4)/
◦
∠
H(3)–C(1)–H(4)/
◦
◦
◦
◦
◦
∠H(3)–C(1)–H(5)/
∠
C(1)–P(2)–H(6)/
∠H(6)–P(2)–H(7)/
∠
C(1)–P(2)–B(8)/
∠P(2)–B(8)–H(9)/
◦
∠P(2)–B(6)–H(11)/
◦
◦
∠H(9)–B(8)–H(10)/
∠H(9)–B(8)–H(11)/
Energy/Hartrees
a
b
See Fig. 1 for atom numbering. Extrapolated using CCSD(T)/aug-cc-pVQZ = CCSD(T)/aug-cc-pVTZ + [MP2/aug-cc-pVQZ–MP2/aug-cc-pVTZ].
MP2 optimisations using Pople-type basis sets the P–H distance
shortens by 0.7 pm when polarisation functions are added to the
H atoms, indicating the need for such functions to describe this
bond accurately. The C–P bond length is also sensitive to the level
of theory, varying by 1.6 pm in the calculations shown in Table 1.
The structure of 1 was defined in terms of eight independent
geometric parameters, comprising three bond lengths and five
bond angles (Table 3, atom numbering shown in Fig. 1). The C, P
and H(4) atoms lie on the mirror plane of the molecule. The C–H
bond lengths were modelled by a single parameter (p
) and P–H (p ) bond lengths were also included. The two P–C–
H angles were defined by the average (p ) and difference (p ), given
1
) and the P–C
(
p
2
3
Methylphosphine–borane. The lowest-energy structure of
methylphosphine–borane on the potential energy surface was an
4
5
s
all-staggered conformation with C symmetry, and the same series
of calculations was performed as for 1.
Table 3 Refined and calculated geometric parameters for methylphos-
phine (1) from the GED study and rotational constants, in MHz, used in
A selection of the structural parameters and energies from these
calculations is given in Table 2 with the full results from the series
of optimisations given in Table S4 and Table S5 in the ESI.† The
absolute values of most of the structural parameters were found to
be independent of the level of theory and basis set used. However,
the C–P and P–B bond lengths were found to vary significantly
across the range of calculations performed. In calculations with
the largest basis sets the C–P bond length varied from 181.3 pm
to 182.9 pm and the P–B bond varied from 191.4 pm to 193.9 pm.
In both cases the CCSD(T) calculations produced longer bond
lengths than MP2, and the Pople-type basis sets produced longer
bond lengths than the correlation-consistent basis sets.
a,b
the GED refinements
MP2/
6–311++G**
SARACEN
Parameter
(rh1
)
Restraint
Independent parameters
p
p
p
p
1
2
3
4
rC(1)–H(3)/pm
rC(1)–P(2)/pm
rP(2)–H(6)/pm
∠P(2)–C(1)–Hav.
109.2
185.6
141.0
111.4
2.4
108.7
97.6
108.0(1)
185.72(6)
142.1(1)
111.7(4)
2.4(4)
109.0(9)
97.4(8)
109.1(10)
—
—
—
2.4(4)
—
◦
/
◦
p5
6
7
8
∠P(2)–C(1)–Hdiff./
◦
p
p
p
∠H(3)–C(1)–H(4)/
◦
∠C(1)–P(2)–H(6)/
97.6(10)
94.6(10)
∠H(6)–P(2)–H(7)/◦
94.6
94.2(8)
Dependent parameters
Gas phase electron diffraction refinements
dp
dp
1
2
∠P(2)–C(1)–H(3)/◦
∠P(2)–C(1)–H(4)/◦
109.0
113.8
109.4(4)
114.1(6)
—
—
Methylphosphine. On the basis of the ab initio calculations
described above, electron-diffraction refinements were carried out
using a model with appropriate C
s
symmetry. The calculations
symmetry,
Rotational constants
Constant Experimental
show that the CH group does not possess local C
3
3
with the C–H(3) and C–H(4) bond lengths differing by only 0.2
GED
Exp.–GED
Uncertainty
◦
pm and the H(3)–C–H(4) and H(3)–C–H(5) angles by ∼1 . The
B
B
z
z
11795.10
–61129.20
113.00
11795.08
−61082.05
113.01
0.02
−47.15
−0.01
0.25
114.00
0.20
difference in the C–H bond lengths of 0.2 pm is too small to
be differentiated by the diffraction experiment, so the C–H bond
–A
z
z
B –C
z
lengths in the CH group were modelled by a single parameter.
The deviation in the angle is more pronounced and was therefore
included in the model.
3
a
Figures in parentheses are the estimated standard deviation of the last
b
digits. See text for parameter definitions.
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 5041–5047 | 5043

View Article Online
by [P(2)–C(1)–H(4) + P(2)–C(1)–H(3)]/2 and [P(2)–C(1)–H(4)–
P(2)–C(1)–H(3)]/2, respectively. The remaining bond angles were
correction for B and 10% of the difference between the corrections
for the two differences. The 10% figure is standard for vibrational
corrections to rotational constants, based on our experience of
how these quantities vary with the computational method used.
The weights applied to all the data depended on the uncertainties
∠
H(3)–C(1)–H(5) (p
6
), ∠C(1)–P(2)–H(6) (p
7
) and H(6)–P(2)–H(7)
s
symmetry the H(3)–
(
p
8
). To allow for deviations from perfect C
C(1)–P(2)–X dihedral angle was included as a parameter, where X
is defined as the H(6)–P(2)–H(7) bisector. However, no significant
5–7
of the observations, in accordance with the SARACEN method.
Methylphosphine–borane. The refinement of the structure of
methylphosphine–borane was conducted using a model of C
deviations were found from the C
s
symmetry and so this parameter
was not included in the final refinement.
s
42
The starting parameters for the rh1 refinement were taken
from the theoretical geometry optimised at the MP2/6–311++G**
level. A theoretical Cartesian force field was obtained at this level
and converted into a force field described by a set of symmetry co-
symmetry. The model and method used were similar to those used
for 1, with full details, including the rotational constants used,
given in the ESI.† The success of the final refinement, for which
R
G
= 0.100 (R = 0.080), can be assessed on the basis of the radial
D
42
ordinates using the SHRINK program, which generated both the
amplitudes of vibration (uh1) and the curvilinear corrections (kh1).
All eight geometric parameters and four groups of vibrational
amplitudes were then refined. Flexible restraints were employed
distribution curve (Fig. 3) and the molecular scattering intensity
curves (Fig. S2).† The final refined parameters are listed in Table 4.
5–7
during the refinement using the SARACEN method. Altogether,
four geometric restraints (Table 3) and one amplitude restraint
(
Table S6)† were employed. The success of the final refinement,
for which R
G
= 0.081 (R
D
= 0.055), can be assessed on the
basis of the radial distribution curve (Fig. 2) and the molecular
scattering intensity curves (Fig. S1).† The final refined parameters
are listed in Table 3. The ESI contains the interatomic distances
and the corresponding amplitudes of vibration (Table S6), with the
least-squares correlation matrix (Table S7) and the experimental
coordinates from the GED analysis (Table S8).†
Fig. 3 Experimental and difference (experimental–theoretical) radial
distribution curve, P(r)/r, for methylphosphine–borane (2). Before Fourier
2
inversion the data were multiplied by s × exp(−0.00002s )/(Z
C
− f
C
)(Z
P
−
P
f ).
Discussion
The molecular structures of methylphosphine and methyl-
phosphine–borane have been investigated in the gas phase by GED
supplemented by ab initio calculations and rotational constants via
the SARACEN method. An independent theoretical investigation
of the structures was also undertaken.
Fig. 2 Experimental and difference (experimental–theoretical) radial
The experimental structure of 1 is found to be generally in
good agreement with both the theoretical structures and previous
experimental results. For comparison, a selection of parameters
from previous GED and MW studies are given in Table 5. It
should be noted that neither the previous MW nor the earlier GED
structure is complete, with each offering only a partial structure
or requiring the use of fixed parameters.
distribution curves, P(r)/r, for methylphosphine (1). Before Fourier
2
inversion the data were multiplied by s × exp(−0.00002s )/(Z
C
− f
C
)(Z
P
−
f
P
).
Three rotational constants for 1 were combined with the GED
12
data. The rotational constants A0,
B
0
and C
0
were corrected to Az,
B
z
and C for the structural refinements using values calculated by
z
42
SHRINK, based on the MP2/6–311++G** force field. The three
corrected rotational constants were included in the refinements as
The calculated structure was found to be generally independent
of the level of theory and basis set used, with parameters generally
◦
the absolute value of B
These constants are given in Table 3. The vibrational corrections
to the rotational constants, which transform, for example, A into
, are summations of the corrections for each of the normal
modes of 1. For the rotational constant A this correction was
141.3 MHz and for B and C it was 2.5 and 4.4 MHz, respectively.
z
, and the differences B
z
–A
z
and B
z
–C
z
.
varying by less than 1 pm in the case of the bond lengths or 1
for the angles, despite the wide range of calculations conducted.
The largest deviations were observed for the P–C distance, which
varied by 1.5 pm. Therefore, it can be stated that the molecule is
adequately described by the MP2/6–311++G** level of theory.
Comparing our experimental structure to the theoretical results,
the P–C distance of 185.72(6) pm is in the middle of the range of the
calculated values, closest to the MP2/6–311++G** calculations.
0
A
z
1
The uncertainties of the vibrational corrections to the rotational
constants were taken as 10% of the value of the vibrational
5
044 | Dalton Trans., 2008, 5041–5047
This journal is © The Royal Society of Chemistry 2008

View Article Online
1
1
Table 4 Refined and calculated geometric parameters for methyl-
phosphine–borane (2) from the GED study and microwave rotation
Table 5 A selection of structural parameters from previous GED
12 10
and MW studies of methylphosphine (1) and a MW study of
a,b
a,b
constants, in MHz, used in the GED refinements
methylphosphine–borane (2)
MP2/
6–311++G**
SARACEN
Parameter
Previous GED
MW
This study
Parameter
(rh1
)
Restraint
Methylphosphine
rP–C/pm
Independent parameters
e
e
e
e
185.8(3)
109.4(8)
142.3(7)
—
186.3
109.3
141.4
93.2
—
185.72(6)
108.0(1)
142.1(1)
94.2(8)
c
p
p
p
p
p
p
p
p
p
p
p
p
p
p
1
rP–C/rP–Bav. /pm
rP–C/rP–Bdiff./pm
rC–H/pm
187.5
5.4
109.2
140.1
120.8
110.4
1.9
104.7
1.4
109.5
103.2
115.0
114.3
100.3
186.1(1)
5.0(2)
108.7(5)
139.3(5)
119.9(5)
111.1(6)
1.8(4)
106.2(5)
1.3(5)
110.2(8)
102.6(8)
115.0(1)
114.4(6)
99.8(9)
—
—
rC–H/pm
2
rP–H/pm
◦
3
109.2(8)
140.1(8)
120.8(10)
110.5(10)
1.9(4)
104.7(7)
1.4(5)
109.5(10)
103.2(10)
—
114.3(7)
100.3(10)
∠H–P–H/
∠C–P–H/◦
96.5
d
97.4(8)
4
rP–H/pm
5
rB–H/pm
◦
6
∠P–C–Hav.
/
◦
Methylphosphine–borane
7
∠P–C–Hdiff.
/
◦
8
∠P–B–Hav.
/
◦
rP–C/pm
rP–B/pm
rP–H/pm
∠H–P–H/
∠C–P–B/◦
—
—
—
—
—
—
180.9(6)
190.6(6)
140.4(6)
99.4(9)
103.2(6)
115.7(4)
181.1(2)
191.1(2)
139.3(5)
99.8(9)
102.6(8)
115.0(1)
9
∠P–B–Hdiff.
/
◦
10
11
12
13
14
∠H(3)–C(1)–H(4)/
◦
◦
∠C–P–H/
◦
◦
∠C–P–B/
◦
∠C–P–H/
∠H(9)–B(8)–H(10)/
∠H(7)–P(2)–H(6)/◦
a
Figures in parentheses are the estimated standard deviation of the last
Dependent parameters
b
c
digits. See text for parameter definitions. In both studies the methyl
d
group was assumed to possess local C symmetry. This parameter was
fixed at this value and not allowed to refine. No estimated standard
deviations were reported for this study.
3
dp
dp
dp
dp
dp
dp
1
2
3
4
5
6
rC(1)–P(2)/pm
182.1
192.8
108.6
112.3
106.1
103.3
181.1(2)
191.1(2)
109.3(6)
112.9(7)
107.5(7)
104.9(6)
—
—
—
—
—
—
e
rP(2)–B(8)/pm
◦
◦
∠P(2)–C(1)–H(3)/
∠P(2)–C(1)–H(4)/
◦
∠P(2)–B(8)–H(11)/
∠P(2)–B(8)–H(10)/◦
◦
◦
angle is found to be 94.2(8) , significantly smaller than the 96.5
angle found in PH
in tertiary phosphines. The reduction in cone angle corresponds
to an increase in s character in the phosphorus lone pair and a
decrease in basicity, and is expected to lead to different properties
43
◦
3
and the >100 angle that is generally found
Rotational constants
Constant Experimental GED
Exp.–GED Uncertainty
C
C
C
C
C
C
C
C
C
z
z
z
z
z
z
z
z
z
(I)–A
z
(I)
(I)
−12739.5
−1011.7
4985.8
−12758.3
−1056.8
−151.9
−12731.1 −8.4
−1011.3 −0.4
12.8
1.6
0.8
13.0
1.5
0.1
12.5
1.5
0.1
1
when such molecules are used as ligands.
(I)–B
z
The previous GED experiment assumed a fixed C–P–H angle of
(I)
4985.4
−12767.9
−1057.6
−152.0
0.4
9.6
0.9
0.1
◦
◦
(II)–A
(II)–B
(I)–C
(III)–A
(III)–B
(I)–C (III)
z
(II)
96.5 , whilst the current study found the angle to be 97.4(8) . The
(II)
◦
z
calculated angle was consistently around 97.5 . It is therefore likely
z
(II)
that the previous study underestimated the C–P–H angle, which
may have influenced the experimentally determined parameters.
For 2, a selection of structural parameters from a previous MW
study is shown in Table 5. As was the case with 1, the structure of 2
was found to be largely independent of the level of theory used. The
only major exception to this is the P–B bond length, which varies
by >2.5 pm across the range of calculations performed. This bond
length is likely to be sensitive to the nature of the charge transfer
from the phosphorus lone pair to the empty p orbitals of the boron
atom, so it is understandable that the parameter is dependent on
both the description of electron correlation employed, and the
basis set used.
z
(III) −12687.2
−12679.6 −7.6
−976.9 −0.7
z
(III)
−977.7
z
121.0
121.0
0.0
a
Figures in parentheses are the estimated standard deviation of the last
b
digits. See text for parameter definitions.
The relatively small uncertainty of 0.06 pm on the experimental
distance brings in to question the precision of the calculated
distances, and no clear convergence is found as the level of
theory and basis set is improved. The experimental distance is
also comparable to the previous GED structure.
The P–H distance, which for this study is 142.1(1) pm, is in
agreement with the largest calculated values, and is consistent
with the value obtained from the previous GED study. The
longest calculated bond lengths were obtained using correlation-
consistent basis sets, with the P–H distance being 141.7 pm at
CCSD(T)/aug-cc-pVQZ, and longer still when double- or triple-
zeta basis sets are used. In contrast, the experimental C–H bond
distances were found to agree with the shortest experimental bond
lengths, corresponding to the MP2/aug-cc-pVQZ level of theory.
The remaining parameters, all bond angles, are found to be in
good agreement with theoretical structures, although it must be
noted that they all refer to the angles involving hydrogen atoms and
theoretical restraints were used for some parameters. The H–P–H
The experimental and computed structures show a good level
of agreement. The P–B bond length is found by experiment to be
191.1(2) pm, a value which is consistent with the shortest of the
calculated bond lengths, lying just below the MP2/6–311++G**
value. As both bond lengths calculated at the CCSD(T) level of
theory deviated from the experimental value by a large amount,
this result seems to suggest that the MP2 level of theory is better
at predicting this bond length. The experimental P–C bond length
is 181.1(2) pm, again at the shorter end of the theoretical values,
closer to the values calculated at the MP2 level of theory.
For 1 the experimental P–H bond length was found to be closest
to the longer theoretical distances, whereas in the complex the
reverse is true. The distance from this study, 139.3(5) pm, is shorter
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 5041–5047 | 5045

View Article Online
Table 6 The changes in structural parameters defined as methylphosphine–borane (2) minus methylphosphine (1), from this study and ab initio
a,b
calculations at the MP2/6–311++G**, MP2/aug-cc-pVQZ, CCSD(T)/6–311++G** and CCSD(T)/aug-cc-pVTZ levels of theory
MP2
CCSD(T)
c
This study
6–311++G**
aug-cc-pVQZ
6–311++G**
aug-cc-pVTZ
Average from theory
rC(1)–H(3)/pm
rC(1)–H(4)/pm
rC(1)–P(2)/pm
rP(2)–H(6)/pm
0.7(5)
0.7(5)
0.0
0.0
−3.5
−0.9
−0.4
−1.5
5.6
0.0
0.0
−3.8
−1.1
−0.5
−1.1
5.7
−0.1
0.0
−3.8
−1.2
−0.4
−1.4
5.9
−0.1
0.0
−3.8
−1.3
−0.4
−1.0
5.8
−0.05(5)
0.0(0)
−3.7(2)
−1.1(2)
−0.42(3)
−1.3(3)
5.7(1)
−4.6(2)
−2.8(5)
−0.1(7)
−1.2(9)
5.2(11)
5.6(12)
◦
◦
◦
∠
P(2)–C(1)–H(3)/
P(2)–C(1)–H(4)/
C(1)–P(2)–H(6)/
∠
∠
◦
∠H(6)–P(2)–H(7)/
5.8
6.3
6.0
6.5
6.1(3)
a
Figures in parentheses are the estimated standard deviation of the last digits. For the theoretical average the standard deviation was calculated from the
b
c
four sets of calculations shown. See text for parameter definitions. For the SARACEN refinement no distinction was made between rC(1)–H(3) and
rC(1)–H(4).
than all the theoretical values and the value from the MW study,
although this is only 1–2r smaller than the calculated values
at the MP2/6–311++G** and MP2/aug-cc-pVQZ levels. The
experimental B–H distance is again consistent with the shortest of
the calculated values, being around 1r from the computed value
at the MP2/aug-cc-pVQZ level of theory.
The remaining experimental bond angles and torsion angles
are all reasonably close to the calculated parameters, with the
exception of the P–B–H bond angles, which are slightly larger
than the calculations suggest.
To investigate the structural changes in 1 upon complexation
to form 2, the differences between comparable parameters were
calculated. The changes in parameters from the SARACEN
refinements and a selection of theoretical calculations are shown
in Table 6.
C atom has a negative charge. When the P atom donates electron
density to boron in 2 it must become more positive, and therefore
more strongly bound to the negatively charged C, shortening the
P–C bond.
The gas phase dissociation energy of methylphosphine–borane,
defined as MePH
2
·BH
3
→ BH
3
+ MePH
2
, has been assessed
at the MP2/6–311++G** level of theory and was found to
−
1
be 95 kJ mol at the MP2/6–311++G** level of theory. The
counterpoise method was used to account for BSSE and ZPE
46,47
corrections were included.
For comparison, the same quantity
·BH , was previously estimated
to be 146(3) kJ mol . The enthalpy of formation, defined as
in the nitrogen analogue, MeNH
2
3
−
1 48
1
2
B
2
H
6
+ MePH
2
→ MePH
2
·BH
3
, was calculated in the same way
−
1
and found to be −27.8 kJ mol , whilst for MeNH
2
·BH
3
the value
−
1 49
is −73(2) kJ mol . The weaker bonding in the phosphine is
5
0
Significant differences between the geometry of 1 and 2 are
found, particularly around the phosphorus atom. The P–C bond
is found to shorten by around 4 pm by both experiment and
theory, whilst both the H(6)–P(2)–H(7) and C(1)–P(2)–H(6) angles
expected due to its lower basicity.
Conclusion
◦
increase by approximately 6 . This represents a move towards a
Primary phosphines have received relatively little attention in the
literature and are not widely used by chemists. However, complex-
ation with borane results in a protected, less volatile compound
which can be used in further synthesis more easily than the free
phosphine. The complete gas-phase structures of methylphosphine
and methylphosphine–borane have been determined for the first
time and the structural changes that occur on complexation have
been assessed. Large changes in geometry around the phosphorus
atom are found, including a widening of the H–P–H angle and a
shortening of the P–C bond upon complexation. The P–B distance
is found to be 191.1(2) pm.
more regular tetrahedral geometry, with the coneangles now found
around phosphorus being close to those found in the more stable
secondary and tertiary phosphines. (For example, the C–P–C angle
◦
44
in trimethylphosphine is 98 .)
The borane group, previously having D3h symmetry, becomes
pyramidal upon complexation and does not possess local C3v
symmetry as the complex has overall C
bond length is about 1.5 pm shorter in the adduct (based on
the calculated B–H bond length in BH ), and the H–B–H angles
s
symmetry. The B–H
3
◦
◦
reduce from 120 to around 114 . The newly formed P–B–H angles
◦
are around 107 and 104 .
The shortening of the P–H bond upon complexation, estimated
by calculation to be 1.1(2) pm, is found to be 2.8(5) pm in this
study. The experimentally determined P–H bond length was found
to be at the longer end of the calculated bond lengths for 1 and at
the shorter end for 2; these results combine to produce the larger
experimental change in bond length.
Acknowledgements
SLM is grateful to the Royal Society of Edinburgh for the award
of a RSE/BP personal research fellowship and to the EPSRC
(EP/D057167) for a research grant. RNE acknowledges the EP-
SRC (EP/D057167) and the School of Chemistry at The Univer-
sity of Edinburgh for funding. We also thank the British Council
and Egide (Alliance number 12141YB) for funding the collabora-
tion between The University of Edinburgh and E´ cole Nationale
Sup e´ rieure de Chimie of Rennes, and the EPSRC National Service
for Computational Chemistry Software (http://www.nsccs.ac.uk)
The P–C bond is found to shorten dramatically [−4.6(2) pm
by experiment, −3.7(2) pm by theory] upon complexation. Such
a shortening can be rationalised by considering charge transfer.
The relative electronegativities of P and C (Pauling scale; C =
45
2
.55, P = 2.19) suggest the P–C bond is polarised such that the
5
046 | Dalton Trans., 2008, 5041–5047
This journal is © The Royal Society of Chemistry 2008

View Article Online
for computing time on Magellan. This work has made use of the
resources provided by the EaStCHEM Research Computing Facil-
ity (http://www.eastchem.ac.uk/rcf), which is partially supported
by the eDIKT initiative (http://www.edikt.org). J-CG thanks the
PID-OPV (INSU-CNRS) for financial support.
18 M. S. Gordon, J. S. Binkley, J. A. Pople, W. J. Pietro and W. J. Hehre,
J. Am. Chem. Soc., 1982, 104, 2797.
9 W. J. Pietro, M. M. Francl, W. J. Hehre, D. J. DeFrees, J. A. Pople and
1
J. S. Binkley, J. Am. Chem. Soc., 1982, 104, 5039.
20 C. C. J. Roothaan, Rev. Mod. Phys., 1951, 23, 69.
21 C. Møller and M. S. Plesset, Phys. Rev., 1934, 46, 618.
2
2
2 J. Cizek, Adv. Chem. Phys., 1969, 14, 35.
3 G. D. Purvis, III and R. J. Bartlett, J. Chem. Phys., 1982, 76, 1910.
References
24 G. E. Scuseria, C. L. Janssen and H. F. Schaefer, III, J. Chem. Phys.,
1
988, 89, 7382.
1
2
M. Brynda, Coord. Chem. Rev., 2005, 249, 2013.
G. Becker, O. Mundt, M. Roessler and E. Schneider, Z. Anorg. Allg.
Chem., 1978, 443, 42.
25 G. E. Scuseria and H. F. Schaefer, III, J. Chem. Phys., 1989, 90,
3700.
26 W. J. Hehre, R. Ditchfield and J. A. Pople, J. Chem. Phys., 1972, 56,
2257.
3
4
5
K. Bourumeau, A.-C. Gaumont and J.-M. Denis, J. Organomet. Chem.,
1
997, 529, 205.
27 P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28, 213.
28 M. S. Gordon, Chem. Phys. Lett., 1980, 76, 163.
29 A. D. McLean and G. S. Chandler, J. Chem. Phys., 1980, 72, 5639.
30 R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys.,
1980, 72, 650.
31 D. E. Woon and T. H. Dunning, Jr., J. Chem. Phys., 1993, 98, 1358.
32 R. A. Kendall, T. H. Dunning, Jr. and R. J. Harrison, J. Chem. Phys.,
1992, 96, 6796.
33 T. H. Dunning, Jr., J. Chem. Phys., 1989, 90, 1007.
34 K. A. Peterson, D. E. Woon and T. H. Dunning, Jr., J. Chem. Phys.,
1994, 100, 7410.
35 A. Wilson, T. van Mourik and T. H. Dunning, Jr., THEOCHEM, 1996,
388, 339.
36 M. J. Frisch, M. Head-Gordon and J. A. Pople, Chem. Phys. Lett.,
1990, 166, 275.
37 M. J. Frisch, M. Head-Gordon and J. A. Pople, Chem. Phys. Lett.,
1990, 166, 281.
38 C. M. Huntley, G. S. Laurenson and D. W. H. Rankin, J. Chem. Soc.,
Dalton Trans., 1980, 954.
39 H. Fleischer, D. A. Wann, S. L. Hinchley, K. B. Borisenko, J. R. Lewis,
R. J. Mawhorter, H. E. Robertson and D. W. H. Rankin, Dalton Trans.,
2005, 3221.
40 S. L. Hinchley, H. E. Robertson, K. B. Borisenko, A. R. Turner, B. F.
Johnston, D. W. H. Rankin, M. Ahmadian, J. N. Jones and A. H.
Cowley, Dalton Trans., 2004, 2469.
41 A. W. Ross, M. Fink and R. Hilderbrandt, in International Tables for
Crystallography, ed. A. J. C. Wilson, Kluwer Academic Publishers,
Dordrecht, Netherlands, 1992, p. 245.
42 V. A. Sipachev, THEOCHEM, 1985, 121, 143.
43 O. N. Ulenikov, E. S. Bekhtereva, G. A. Onopenko and E. A. Sinitsin,
J. Mol. Spectrosc., 2002, 216, 252.
44 P. S. Bryan and R. L. Kuczkowski, J. Chem. Phys., 1971, 55, 3049.
45 J. Emsley, The Elements, Oxford University Press, Oxford, 1989.
46 S. Simon, M. Duran and J. J. Danneberg, J. Chem. Phys., 1996, 105,
11024.
H. Dorn, R. A. Singh, J. A. Massey, J. M. Nelson, C. A. Jaska, A. J.
Lough and I. Manners, J. Am. Chem. Soc., 2000, 122, 6669.
A. J. Blake, P. T. Brain, H. McNab, J. Miller, C. A. Morrison, S. Parsons,
D. W. H. Rankin, H. E. Robertson and B. A. Smart, J. Phys. Chem.,
1
996, 100, 12280.
6
P. T. Brain, C. A. Morrison, S. Parsons and D. W. H. Rankin, J. Chem.
Soc., Dalton Trans., 1996, 4589.
7
8
N. W. Mitzel and D. W. H. Rankin, Dalton Trans., 2003, 3650.
S. Reeve, W. A. Burns, F. J. Lovas, R. D. Suenram and K. R. Leopold,
J. Phys. Chem., 1993, 97, 10630.
9
K. R. Leopold, M. Canagaratna and J. A. Phillips, Acc. Chem. Res.,
1
997, 30, 57.
1
1
1
0 P. S. Bryan and R. L. Kuczkowski, Inorg. Chem., 1972, 11, 553.
1 L. S. Bartell, J. Chem. Phys., 1960, 32, 832.
2 T. Kojima, E. L. Breig and C. C. Lin, J. Chem. Phys., 1961, 35,
2
139.
1
1
1
1
3 W. L. Jolly, Inorg. Synth., 1968, 11, 124.
4 W. L. Jolly, Inorg. Synth., 1968, 11, 15.
5 A. B. Burg and R. I. Wagner, J. Am. Chem. Soc., 1953, 75, 3872.
6 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
47 S. F. Boys and F. Bernardi, Mol. Phys., 1970, 19, 553.
48 A. Haaland, Angew. Chem., Int. Ed. Engl., 1989, 28, 992.
49 R. E. McCoy and S. H. Bauer, J. Am. Chem. Soc., 1956, 78, 2061.
50 S. Ikuta and P. Kebarle, Can. J. Chem., 1983, 61, 97.
(
Revision C.02), Gaussian, Inc., Wallingford, CT, 2004.
7 J. S. Binkley, J. A. Pople and W. J. Hehre, J. Am. Chem. Soc., 1980, 102,
39.
1
9
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 5041–5047 | 5047