A closer look at the phosphorus-phosphorus double bond lengths in meta-terphenyl substituted diphosphenes

Article abstract of DOI:10.1016/j.ica.2010.07.018

The single crystal X-ray structure of DmpPPDmp (1, Dmp = 2,6-Mes ₂C₆H₃), which was previously reported to have a relatively short PP bond distance of 1.985(2) at room temperature, has been reexamined at variable temperatur

Full text of DOI:10.1016/j.ica.2010.07.018

Inorganica Chimica Acta 364 (2010) 39–45
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
A closer look at the phosphorus–phosphorus double bond lengths in
meta-terphenyl substituted diphosphenes
a
a
a
b
John D. Protasiewicz a, , Marlena P. Washington , Vittal B. Gudimetla , John L. Payton , M. Cather Simpson
*
a Department of Chemistry, Case Western Reserve University, Cleveland, OH 44106, USA
b Department of Chemistry and Physics and the Photon Factory, University of Auckland, Auckland, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 15 July 2010
The single crystal X-ray structure of DmpP@PDmp (1, Dmp = 2,6-Mes2C6H3), which was previously
reported to have a relatively short P@P bond distance of 1.985(2) Å at room temperature, has been reex-
amined at variable temperatures. Crystallographic analyses of 1 at 100 K allow for resolution of disorder
of the two phosphorus atoms (which is unresolvable from room temperature diffraction data), and for
determination of a more conventional P@P bond length of 2.029(1) Å. Single crystals of the closely related
diphosphene DxpP@PDxp (2, Dxp = 2,6-(2,6-Me2C6H3)2C6H3) show similar disorder in one of two crystal-
lographically independent molecules in the unit cell. A value of 2.0276(4) Å is found for the non-disor-
dered P@P bonds at 100 K for 2. A new diphosphene Ar0P@PAr0 (3, Ar0 = 2,6-Mes2-4-OMe–C6H3) has
been prepared and its structure has also been examined. The P@P bond length for 3 was determined
to be 2.0326(9) Å and relatively free of the effects of disorder.
Dedicated to Arnold L. Rheingold
Keywords:
Diphosphene
Main group
X-ray structure
Double bonds
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
to have a phosphorus–phosphorus bond length of 2.034(2) Å at
room temperature, representing a 8.4% reduction of bond length
from the average value of 2.22 Å for a PP single bond. A later deter-
mination of this structure by Cowley at 100 K, and with a larger
data set, yielded a value of 2.047(1) Å (0.17 Å, 7.8% reduction from
PP single bond average) [3]. While these contractions may not be
as great as for compounds having analogous N@N linkages, they
‘‘This all comes back to my long-held observation that I have
never met a competent crystallographer that wasn’t also a
first-rate chemist.”
-Professor Arnold Rheingold, 1-21-2010, commenting on the
Bruker User List for the need for reviewers who have compe-
tency in both of these areas.
still argue for the presence of significant
phosphorus atoms.
p-bonding between the
Our group reported the structure of the diphosphene
DmpP@PDmp bearing a sterically encumbered meta-terphenyl li-
gand Dmp (Dmp = 2,6-Mes2C6H3) [4]. A key metric determined
from this study was the P@P bond length of 1.985(2) Å (0.23 Å
shorter, or a 10.6% reduction, from PP single bond value). This bond
length is the shortest reported to date for two coordinate phospho-
rus atoms in acyclic RP@PR molecules. Even allowing for three sig-
ma units in standard deviation leads to a maximum bond length of
1.991 Å, a value that still lies shorter than those for most other
diphosphenes. Our continuing interest in these types of molecules
and recently upgraded diffraction facilities allowed us to reexam-
ine this material. Here we report on the results of new diffraction
experiments on 1 and two closely related derivatives. In particular,
the temperature dependence of the crystallography findings offer
important insights as to the interpretation of the short P@P bond
length originally reported for 1 and similar systems. Comparisons
are also drawn to structurally characterized meta-terphenyl shel-
tered diphosphenes 4–7 (Chart 1) showing P@P distances ranging
from 2.008(2) to 2.039(2) Å [5–8].
Professor Rheingold has been an active advocate for more
chemists to become competent crystallographers, and for crystal-
lographers to make sure the structures they produce are chemi-
cally reasonable. Chemists working on compounds having
multiple bonds between main group elements are often fixated
upon bond distances obtained from crystallographic studies. In
particular, a shorter bond can be an important indicator of multiple
bonding. For the heavier elements, however, the decrease in bond
lengths with increased bond order is much less dramatic than wit-
nessed for carbon–carbon or nitrogen–nitrogen bonds [1]. There is
thus need for extra care in examining such distances.
One of the earlier Group 15 elements to illustrate bond length
shortening upon double bond formation is the first structurally
*
*
characterized diphosphene Mes P@PMes by Yoshifuji in 1981
t
*
(Mes = 2,4,6- Bu3C6H2) [2]. This landmark compound was found
* Corresponding author. Tel.: +1 216 368 5060.
E-mail address: protasiewicz@case.edu (J.D. Protasiewicz).
0020-1693/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2010.07.018

40
J.D. Protasiewicz et al. / Inorganica Chimica Acta 364 (2010) 39–45
P
P
OMe
P
P
MeO
P
P
1
2
3
Et
Et
Cl
Et
Et
P
Cl
Cl
Et
Cl
P
P
P
P
P
P
P
P
Et
Cl
P
Et
Cl
Cl
Et
Cl
4
5
6
7
Chart 1.
2. Results and discussion
2.1. Structure analyses of DmpP@PDmp
Table 1
Important variable temperature diffraction information for compound 1.
Solution-data
set
1B
1C
1D
Five new crystals of compound 1, grown from pentane at ꢀ35 °C
in a drybox, were examined at a variety of temperatures. Two sam-
ples were examined independently at 100 K, a third at 190 K and at
290 K, a fourth at 300 K, and a fifth sample was measured at 300 K,
200 K, and at 100 K. As the overall results for all 5 samples are con-
sistent with one another (P21/c, similar cells, R1 values spanning
from 3–6%), only data for the fifth sample will be presented and
discussed. The orange crystal of 1 was mounted on a glass fiber
with epoxy and subjected to crystallographic analyses at 300 K,
200 K, and 100 K. Relevant data collection and solution details
are provided in Tables 1 and 2.
From 300 K to 100 K the cell volume shows a 4.1% decrease,
dominated by decreases in the a and c axis lengths (2.3% and
1.7%, respectively) and the change in the b angle (0.71%). The b axis
shows little change (0.12%). The 300 K data set yields a structural
solution (solution 1B0) comparable to the reported RT structure of
1, in that commonly acceptable R-factors and Goodness-of-fit are
produced, and a ‘‘short” P@P bond length of 1.9893(9) Å is realized.
Three views of thermal ellipsoid diagrams for the inner core atoms
of 1 are presented in Fig. 1. While the thermal ellipsoids for the
phosphorus atoms appear normal to the casual observer, submis-
sion of the Cif file for 1B0 to Check Cif [9] generates a Level B warn-
ing for the Hirshfeld rigid bond test [10] difference for P1–P2
(11.25 su). This test notes significant differences in anisotropic
thermal parameters along chemical bonds and can be an indicator
of disorder, among other things. The largest difference peaks in the
residual electron density map occur near P1 and P2 (0.79 and 0.56),
and one can introduce a disorder model of the type depicted in
Chart 2.
Empirical
formula
Formula
weight
Temperature
(K)
Wavelength
(Å)
Crystal
C48H50P2
688.82
300
C48H50P2
688.82
200
C48H50P2
688.82
100
0.71073
monoclinic
P21/c
0.71073
monoclinic
P21/c
0.71073
monoclinic
P21/c
system
Space group
Unit cell dimensions
a (Å)
b (Å)
c (Å)
10.9346 (1)
10.7974 (1)
22.2807 (2)
16.8506 (2)
90.00
90.756 (1)
90.00
4053.46 (7)
4
1.129
10.6802 (1)
22.2695 (3)
16.7013 (2)
90.00
90.404 (1)
90.00
3972.18 (8)
4
1.152
22.2979 (3)
16.9984 (2)
90.00
91.048 (1)
90.00
4143.83 (8)
4
1.104
a
(°)
b (°)
c
(°)
V (Å3)
Z
Dcalc (g cmꢀ3
Absorption
)
0.14
0.14
0.14
coefficient
(mmꢀ1
F(0 0 0)
)
1472
1472
1472
Crystal size
(mm)
0.35 ꢁ 0.30 ꢁ 0.20 0.35 ꢁ 0.30 ꢁ 0.20 0.35 ꢁ 0.30 ꢁ 0.20
Crystal color
and shape
h Range data
collection
Limiting
pale orange block
1.51–28.3
pale orange block
1.52–28.3
pale orange block
1.52–28.3
ꢀ14 < h < < 14
ꢀ29 < k < 29
ꢀ22 < l < 22
ꢀ14 < h < 14
ꢀ29 < k < 29
ꢀ22 < l < 22
99.9
ꢀ14 < h < 14
ꢀ29 < k < 29
ꢀ22 < l < 22
99.9
indices
Completeness 99.9
(%)
This approach reduces the final R-factors and residual peaks in
the difference map to give solution 1B00. The resulting pair of P@P
bond distances of 1.999(5) and 1.975(8) Å for the 53/47 model of
the disorder, however, remain below that of typical disphosphene
bond lengths. Furthermore, the thermal parameters for the second
set of phosphorus atoms are larger and less isotropic than for the
Reflections
collected
Independent
reflections
R(int)
52 637
51 361
50 220
10317 (0.034)
10074 (0.030)
9876 (0.027)

J.D. Protasiewicz et al. / Inorganica Chimica Acta 364 (2010) 39–45
41
Table 2
Key refinement and structural results for compound 1.
Refinement method
Solution
Full-matrix least-square on F2
1B0
1B00 w/disorder [occupancy] 1C0
1C00 w/disorder [occupancy] 1D0
1D00 w/disorder [occupancy]
Temperature (K)
300
300
200
200
100
100
Data/restraints/parameters
10 317/0/
463
10 317/0/482
10 074/0/
463
1.04
R1 = 0.0526
wR2 = 0.134
R1 = 0.0748
wR2 = 0.150
2.0098(7)
10 074/0/482
9876/0/463 9876/0/482
Goodness-of-fit (GOF) on F2 1.05
1.01
1.04
1.02
1.03
Final R indices [I > 2
R indices (all data)
dP@P (Å)
r
(I)]a,b
R1 = 0.0588
wR2 = 0.155
R1 = 0.1051
wR2 = 0.183
1.9893(9)
R1 = 0.0479
wR2 = 0.125
R1 = 0.0925
wR2 = 0.149
1.999(5) [53]
1.975(8) [47]
0.24
R1 = 0.0427
wR2 = 0.110
R1 = 0.0641
wR2 = 0.122
2.018(1) [90]
2.056(1) [10]
0.29
R1 = 0.0434
R1 = 0.0371
wR2 = 0.110 wR2 = 0.0913
R1 = 0.0546 R1 = 0.0482
wR2 = 0.117 wR2 = 0.0980
2.0263(5)
2.029(1) [78]
2.009(9) [22]
0.35
Highest residual peak
0.79
0.96
1.29
first set. While introduction of the restraints DELU and/or SIMU
produces slight changes in the final data for this structure, no sig-
nificant impact on the P@P bond lengths are realized and thus
these restraints were not used (similar results for use of SIMU/
DELU on solution 1B0 were found).
While the disorder model for the 300 K structure provides better
statistics, it does not address the key question of the anomalously
short P@P bond length in 1, and may be of questionable crystallo-
graphic rigour. When unusually short bond distances have been
determined unrelated to disorder, libration effects have been in-
voked. Libration, or large scale oscillation about bonds, can yield re-
duced bond length findings. These effects tend to be most important
for lighter atoms and for structures performed at room temperature.
Using the LIBR option in XP [11], an estimated impact of 0.003 Å (a
P@P length of 1.993 Å) was found for the structure of 1.
The crystal was cooled to 200 K and data collected using other-
wise identical conditions (1C). The solution of the structure with-
out (1C0) and with a disorder model (1C00) led to the data in
Tables 1 and 2. The results mirror that obtained for the 300 K struc-
ture, but with some subtle differences. First, the P@P distances
have increased slightly, and are closer to the range of other struc-
turally determined P@P distances. Second, the solution having the
disorder model (1C00) changed from 53/47 to 90/10 fit. This fact
suggests that the disorder within crystals 1 might be dynamic,
and temperature dependent.
The crystal of 1 was further cooled to 100 K, and a third data set
(1D) was collected under identical conditions. The corresponding
structural solutions without (1D0) and with (1D00) the disorder
model yielded results that are shown in Tables 1 and 2 as well.
The resulting P@P bond lengths have increased further, and are
only slightly shorter than those values of the aforementioned
*
*
Mes P@PMes molecule. Oddly, as opposed to solution 1B00 and
1C00, solution 1D00 refined to yield a 78/22 occupancies for the dis-
ordered phosphorus atoms. The fluctuations in occupancies among
the three temperatures might reflect a shallow potential surface or
an unknown artifact.
Fig. 1. Three thermal ellipsoid plots (30% probability) of 300 K solution (1B0) of 1 for
selected atoms. Hydrogen atoms, and all but ipso carbon atoms of mesityl groups,
omitted for clarity.
As compared to the previously published structure of 1, the
‘‘best” solution seems to be 1D00, and 2.029(1) Å should be taken
as the best value of the P@P bond length as it is based on the mol-
ecule having the highest occupancy (and more certainty). The value
for the minor component (P1A and P2A) not only contains greater
error (2.009(9) Å), but the Check Cif report indicates remaining is-
Ar
Ar
P(2a)
sues for the thermal parameters for P1A and P2A (large Ueq(max)
/
P(1)
P(2)
Ueq(min) range). Attempts to squeeze finer details from P1A and
P2A and remove this warning were unsuccessful. It is also impor-
tant to note that while the second set of phosphorus atoms are
not far displaced from the first (0.13 and 0.25 Å), notable changes
in PP distances occur upon these slight displacements. Regardless,
this structure provides another reminder of the importance of low
temperatures for optimal X-ray structural solutions [12].
P(1a)
Ar
Ar
Chart 2.

42
J.D. Protasiewicz et al. / Inorganica Chimica Acta 364 (2010) 39–45
2.2. Structure analysis of DxpP@PDxp (2)
two independent molecules in the asymmetric unit, only one of
which possesses the disorder displayed by compound 1.
Preliminary diffraction studies of 2 were mentioned in a previous
report [4]. Relatively poor crystal quality and disorder issues, how-
ever, hampered successful refinement at that time. Armed with bet-
ter knowledge of the disorder in crystals of 1 and better equipment,
this structure was again investigated. Single crystals of 2 were thus
examinedat 100 K and200 Kand theresultsofX-ray determinations
presented in Tables 3 and 4. Interestingly, the structure of 2 contains
The data obtained at 100 K can be refined successfully without
disorder modeling (solution 2B0) to provide reasonable R factors
and what might be acceptable PP bond lengths of 2.0276(5) and
2.0109(6) Å for each of the two independent molecules in the unit
cell. Closer examination, however, shows that one of these two
molecules showed behavior akin to that of 1 above. Introduction
of the same type of disorder model gives solution 2B00 and revealed
two sets of P@P atoms in a 87:13 ratio. The resulting P@P bond
lengths of 2.009(1) and 2.006(1) Å, however, have greater error
and differ somewhat from the value of 2.0276(4) for the P@P bond
in the second non-disordered molecule. Fig. 2 shows the resulting
diagrams for this refinement.
Table 3
Important diffraction information for compounds 2 and 3.
Solution-data
set
2A
2B
3
At 200 K the results are less satisfactory. Without disorder mod-
Empirical
formula
Formula
weight
Temperature
(K)
C44H42P2
632.72
200
C44H42P2
632.72
100
C50H54O2P2
748.87
eling (solution 2A0), the P@P distances were found to be 2.0176(8)
100
Wavelength
(Å)
Crystal
system
Space group
0.71073
triclinic
0.71073
triclinic
0.71073
monoclinic
P21/n
ꢀ
ꢀ
P1
P1
Unit cell dimensions
a (Å)
b (Å)
c (Å)
11.2740 (2)
11.2056 (2)
17.6872 (2)
19.4574 (3)
111.767 (1)
92.666 (1)
91.053 (1)
3574.90 (9)
4
2.2796 (6)
19.458 (1)
18.382 (1)
90.00
103.062 (4)
90.00
4278.5 (4)
4
1.163
17.7924 (3)
19.5055 (4)
111.624 (1)
92.352 (1)
91.101 (1)
3631.7 (3)
4
a
(°)
b (°)
c
(°)
V (Å3)
Z
Dcalc (g cmꢀ3
Absorption
)
1.157
0.15
1.176
0.15
0.14
coefficient
(mmꢀ1
F(0 0 0)
)
1344
1344
1600
Crystal size
(mm)
0.45 ꢁ 0.20 ꢁ 0.20 0.45 ꢁ 0.20 ꢁ 0.20 0.14 ꢁ 0.12 ꢁ 0.07
Crystal color
and shape
h Range data
collection
Limiting
orange block
1.12–30.68
orange block
1.13–30.48
orange plate
2.00–28.3
ꢀ16 < h < 16
ꢀ25 < k <25
ꢀ28 < l <27
ꢀ15 < h <15
ꢀ25 < k <25
ꢀ27 < l <27
99.4
ꢀ16 < h < 14
ꢀ25 < k < 18
ꢀ24 < l < 24
99.1
indices
Completeness 98.6
(%)
Reflections
collected
Independent
reflections
104 715
102 423
38 306
Fig. 2. Two thermal ellipsoid plots (50% probability) of 100 K solution (2B00) of non-
disordered molecule of 2. The second independent molecule in cell not shown. (a)
All non-hydrogen atoms (b) hydrogen atoms and all but ipso carbon atoms of xylyl
groups, omitted for clarity.
22 201 (0.0446)
21 626 (0.0358)
10 567 (0.0979)
(Rint
)
Table 4
Key Refinement and Structural Solution Results for 2 and 3.
Refinement method
Solution
Full-matrix least-square on F2
2A0
2A00 w/disorder (occupancy)
2B0
2B00 w/disorder (occupancy)
3
Temperature (K)
200
200
100
100
100
Data/restraints/parameters
22 201/0/846
22 201/0/864
21 626/0/845
21 626/0/864
10 567/0/
501
1.01
R1 = 0.0596
wR2 = 0.133
R1 = 0.1279
wR2 = 0.170
2.0326(9)
Goodness-of-fit (GOF) on F2 1.04
1.02
R1 = 0.0505
wR2 = 0.126
R1 = 0.0875
1.03
1.03
R1 = 0.0438
wR2 = 0.111
R1 = 0.0595
Final R indices [I > 2
R indices (all data)
dP@P (Å)
r
(I)]a,b
R1 = 0.0685
wR2 = 0.173
R1 = 0.1070
wR2 = 0.200
2.0176(8),
1.975(1)
R1 = 0.0490
wR2 = 0.124
R1 = 0.0648
wR2 = 0.135
2.0276(5),
2.0109(6)
1.8
wR2 = 0.147
wR2 = 0.122
2.0173(6)/1.991(3)*/1.979(6)* [67/
2.0276(4), 2.009(1)*, 2.006(1)* [87/
33]
0.31
13]
0.50
Highest residual peak
2.19
0.75
*
Disordered PP bonds of other molecule in cell.

J.D. Protasiewicz et al. / Inorganica Chimica Acta 364 (2010) 39–45
43
and 1.975(1) Å. Introduction of the disorder model (2A00) for one of
the two independent molecules in the cell resulted in P@P dis-
2.4. Comparison to other meta-terphenyl sheltered diphosphenes, and
predicted structures
tances of 1.991(3) and 1.979(6) Å, and
a P@P distance of
2.0173(6) Å for the non-disordered second molecule. Introduction
of the restraints SIMU and/or DELU made little difference, and still
yielded P@P distances below 2 Å.
Selected data for meta-terphenyl diphosphenes 1–7 (Chart 1)
that have been structurally characterized are summarized in
Table 5. The P@P bond lengths fall between 2.008(2)–2.039(2) Å.
The most significant differences among 1–7 lie in the range of C–
P–P bond angles. Values range from 95° to 109°. Additionally, the
PPCC torsion angles vary greatly, with molecules like 1 and 7 hav-
ing one aromatic ring nearly coplanar with P@P unit and the other
ring nearly orthogonal, while other compounds have either a mix
of angles or both rings nearly orthogonal. While packing forces
likely play a role in the solid state structures of these materials,
examination of the packing of these materials did not reveal any
The presence of two independent molecules in the unit cell of 2
provides a unique internal calibration for evaluating the success of
these refinements. As the P@P bond distances from the disordered
molecule in 2 seem ‘‘short” and contain greater error, it seems
most appropriate to accept the value of 2.0276(4) Å as the best va-
lue for the P@P bond length in compound 2, taken from only the
non-disordered molecule from the refinement 2B00. This value
agrees well with that determined for compound 1.
trends or evidence for significant p-stacking.
While computationally predicted structures for rather simple
diphosphenes such HP@PH or PhP@PPh have been reported [13–
15], few calculations have examined the undeniable role that the
sterically encumbered ligands play upon structure. As the power
of computational chemistry has grown over the recent years, the
opportunity to tackle larger systems has become increasingly possi-
ble. For example, Tokitoh has reported on DFT (B3LYP) studies
that predict P@P bond lengths of 2.088 and 2.068 Å for TbtP@PTbt
and BbtP@PBbt, respectively (Tbt = 2,4,6-tris[bis(trimethylsilyl)-
methyl]phenyl; Bbt = 2,6-bis[bis(trimethylsilyl)methl]phenyl) [16].
These values compare favorably with the corresponding experi-
mentally determined values of 2.051(2) and 2.043(1) Å.
Similar calculations (B3LYP/3-21G(d)) have been performed on
the full compound 1 (Fig. 4). A predicted P@P bond distance of
2.039 Å indeed compares well with the best experimental value
of 2.029(1) Å for the major component of the disordered P@P unit
of 1. The C–P–P bond angles of 96.9 and 107.8° also agree well with
those found experimentally (95.38(6) and 109.43(7)°).
2.3. Structure analysis of Ar0P@PAr0 (3)
Single crystals of 3 were subjected to X-ray diffraction analysis
at 100 K and the results are presented in Tables 3 and 4. The mol-
ecule crystallizes in the P2(1)/n with one molecule in the asym-
metric unit. Unlike the crystallography of 1 and 2, attempts to
find and refine a structural solution with the same type of disor-
der model were unsuccessful. The best unsuccessful attempt sug-
gested at most a 6% contamination by a second set of phosphorus
atoms could be estimated. Further, the solution without disorder
modeling provides a P@P bond length of 2.0326(9) Å that is rea-
sonable and falls into the expected range for such values. As such,
it appears that this structure is reliable and the results are not
clouded by disorder issues. Fig. 3 provides two views of the
solution.
2.5. Comparisons to stilbenes
Phosphorus is often called a ‘‘carbon copy” for its ability to pro-
vide analogs of compounds where multiply bonded carbon atoms
are exchanged for the isolobal phosphorus containing equivalents
[17]. Does the ‘‘carbon copy” concept span to the X-ray structural
world? In fact, single crystal structures of stilbenes (and azobenz-
enes) can often show unusually short C@C bond lengths [18]. Most
relevant to the structures of 1 and 2 are molecules such as E-2,20-
dimethylstilbene. The structure of this stilbene shows ethylenic CC
distances that vary from 1.321(2) Å at 188 K to 1.283 (3) at 298 K
[19–22]. A longer distance of 1.353 Å would be expected from
MMP2 calculations. Unraveling the disorder that occurs in these
systems is challenging, and in some cases, can not be discerned
from X-ray structural data analyses alone. In addition, a dynamic
motion in the crystalline state via a ‘‘pedal motion” is also present.
Recent studies, performed at various temperatures and sometimes
by shock cooling, have revealed that disorder/motion is likely pres-
ent at all temperatures, even when low occupancy factors of minor
conformers essentially make the disorder ‘‘invisible”.
In the analysis of compound 1, there was no evidence for these
other types of disorder. The rather immense size of the meta-ter-
phenyls used in the present study make movements of the large
aryl groups in these diphosphene crystal less likely, but allow for
displacements of attached atoms. The location of the phosphorus
atoms in the crystalline ensemble give the (albeit not necessarily
correct) impression that they can ‘‘rattle around” inside the protec-
tive cage provided by the sterically demanding ligands. Such disor-
der has often been observed [23]. In particular, the structure of
DmpGe(Cl)Ge(Cl)Dmp exhibits nearly the same type of disorder
described here for the germanium atoms [23b]. In addition, other
types of crystallographic/chemical problems have been noted in
Fig. 3. Two thermal ellipsoid plots (50% probability) of 100 K solution of 3. (a) All
non-hydrogen atoms. (b) Hydrogen atoms and all but ipso carbon atoms of mesityl
groups, omitted for clarity

44
J.D. Protasiewicz et al. / Inorganica Chimica Acta 364 (2010) 39–45
Table 5
Structural comparisons for diphosphenes bearing meta-terphenyl ligands.
a
a
Structure
1D00
2B00
3
4b,c
5c
6
7
Temperature (K)
dP@P (Å)
C–P–P angles (°)
PPCC torsion angles
100
2.029(1)
95.38(6), 109.43(7)
8.2, 75.0
100
100
233
300
2.008(2)
97.5 (2), 106.3(2)
83.9, 30.1
173
2.024(1)
101.2(1), 98.0(1)
82.5, 84.1
173
2.0276(4)
98.33(4), 104.26(4)
63.0, 38.5
2.0326(9)
97.62(8),109.32(8)
73.2, 21.1
2.039(2), 2.021(2)
98.5(2), 98.1(2)
88.7, 88.2
2.017(2)
98.0, 110.7
7.1, 76.3
a
b
c
For best determined values (see text).
Two independent molecules in asymmetric unit.
Center of symmetry in molecule.
length of 2.0326(9) Å and no resolvable disorder of the phosphorus
atoms. For reasons that are unclear, these systems are prone to
yield erroneous P@P bond lengths and thus caution should be ta-
ken in structural analyses of meta-terphenyl substituted diphosph-
enes or other main group compounds.
4. Experimental
4.1. Materials
Compounds 1 and 2 were prepared as previously reported
[4,25]. Compound 3 was prepared by reduction of the correspond-
ing ArPCl2 (Ar = 4-OMe-2,6-Mes2–C6H2) [26] with magnesium me-
tal as follows. In a 10 mL flask, 0.100 g (0.23 mmol) of 4-OMe-2,6-
Mes2–C6H2PCl2 and 0.006 g (0.24 mmol) of magnesium metal was
taken into 5 mL of dry THF under dinitrogen. The solution was sub-
jected to ultrasound for 10 minutes to produce a dark orange col-
ored solution. The flask was taken into a glove box and the solvent
was removed under vacuum. The orange colored material was dis-
solved into 10 mL of pentanes and filtered through celite using a
glass fritted funnel. The solvent was removed under vacuum leav-
ing 0.053 g (66%) of compound 3 as a dark orange colored solid. 1H
NMR (CDCl3): d 6.71 (s, 8H), 6.49 (s, 4H), 3.71 (s, 6H), 2.30 (s, 12H),
1.67 (s, 24H). 31P{1H} NMR (CDCl3): d 489.4.
Fig. 4. Minimized structure of compound 1 (B3LYP/3-21G(d)), hydrogen atoms
omitted for clarity.
4.2. Crystallography
meta-terphenyl main group compounds. For example, Power and
colleagues has observed that while the diarsene DmpAs@AsDmp
is not disordered (and exhibits a very different conformation than
DmpP@PDmp), the distibene Ar00Sb@SbAr00 (Ar00 = 2,6-Trip2C6H3) is
plagued by cocrystallization with the partially reduced distibane
Ar00Sb(Cl)Sb(Cl)Ar00 [24].
Crystals of compounds 1 and 2 used for structural determina-
tions were grown from concentrated pentane solutions of the cor-
responding compounds held at ꢀ35 °C. X-ray quality crystals of 3
were grown from a concentrated solution in diethyl ether at
ꢀ35 °C. The X-ray intensity data were measured at varying tem-
peratures (Oxford Cryostream 700) on a Bruker SMART Apex II
CCD-based X-ray diffractometer system equipped with a Mo-target
X-ray tube (k = 0.71073 Å) operated at 1500 W power. For samples
that were analyzed only at 200 K and colder, crystals were frozen
onto a MiTeGen micromount. Samples that were analyzed at high-
er temperatures were epoxied to glass fibers. The detector was
placed at a distance of 6.00 cm from the crystal. Data were mea-
sured using omega scans of 0.5° per frame for 10 s. The frames
3. Conclusion
In summary, a reevaluation of the structure of compound 1 at
several temperatures using diffraction data collected on a modern
CCD-equipped diffractometer has been undertaken. The results
indicate the presence of disorder that involves the phosphorus
atoms. The resulting P@P bond lengths are influenced by the disor-
der, and are only of ‘‘conventional” values (i.e., fall into the range of
previously determined P@P bond lengths) at low temperature.
Thus, while the first structural report (room temperature) on 1
yielded a P@P distance of 1.985(2) Å, recent and better data at
300 K for 1 yield values of 1.999(5) and 1.975(8) Å with a disorder
model in place, and 1.9893(9) Å without the disorder model. At
100 K, values of 2.029(1) and 2.010(9) Å are obtained with a disor-
der model in place, and 2.0263(5) Å without the disorder model.
The values obtained at lower temperature for the major compo-
nent (2.029(1) Å) are the best available for 1 should thus be used
in describing the P@P bond length. Similar disorder exists in one
of the two independent molecules of 2 in the crystal state, and
again longer P@P bond lengths are determined at 100 K versus
200 K. The structure of 3, determined at 100 K, features a P@P bond
Ó
were integrated with the Bruker SAINT build in APEX II software
package using a narrow-frame integration algorithm, which also
corrects for the Lorentz and polarization effects. Absorption correc-
tions were applied using AXScale. The structures were solved and
refined using Bruker SHELXTL (Version 6.14) software. Hydrogen
atoms were introduced at idealized positions and refined as riding
to attached carbon atoms.
Ó
4.3. Computations
All calculations employed the GAUSSIAN 03 [27] software package
and were performed in parallel on Intel Pentium 4 Xeon EM64T
quad-core processors. Kohn–Sham formalized density functional
theory (DFT) was employed to calculate ground state structure

J.D. Protasiewicz et al. / Inorganica Chimica Acta 364 (2010) 39–45
45
[13] S. Nagase, S. Suzuki, T. Kurakake, J. Chem. Soc., Chem. Commun. 15 (1990)
1724.
[14] T.L. Allen, A.C. Scheiner, Y. Yamaguchi, H.F. Schaefer III., J. Am. Chem. Soc. 108
(1986) 7579.
[15] H.-L. Peng, J.L. Payton, J.D. Protasiewicz, M.C. Simpson, J. Phys. Chem. A 113
(2009) 7054.
[16] T. Sasamori, N. Takeda, N. Tokitoh, J. Phys. Org. Chem. 16 (2003) 450.
[17] K.B. Dillon, F. Mathey, J.F. Nixon, Phosphorus: The Carbon Copy, John Wiley
and Sons, New York, 1998.
with the B3LYP hybrid functional [28,29]. The global minima for 1
was fully structurally optimized at the B3LYP/3-21G(d) level of
theory. Frequency calculations were performed finding all positive
eigenvalues for the Hessian on the 3-21G(d) surface. The B3LYP/3-
21G(d) level of theory have proven to rather accurate on larger sys-
tems like fullerenes [30].
[18] J. Harada, K. Ogawa, J. Am. Chem. Soc. 123 (2001) 10884.
[19] K. Ogawa, H. Suzuki, T. Sakurai, K. Kobayashi, A. Kira, K. Toriumi, Acta
Crystallogr., Sect. C: Cryst. Struct. Commun. C44 (1988) 505.
[20] K. Ogawa, T. Sano, S. Yoshimura, Y. Takeuchi, K. Toriumi, J. Am. Chem. Soc. 114
(1992) 1041.
[21] K. Ogawa, J. Harada, S. Tomoda, Acta Crystallogr., Sect. B: Struct. Sci. B51
(1995) 240.
[22] J. Harada, K. Ogawa, J. Am. Chem. Soc. 126 (2004) 3539.
[23] for some examples of structures where attached atoms to Dmp are disordered
see (a) M.M. Olmstead, L. Pu, R.S. Simons, P.P. Power, Chem. Commun. (1997)
1595;
Acknowledgments
We acknowledge the inspirational career of one of Case Wes-
tern Reserve University’s greatest chemist alums, Professor Arnold
L. Rheingold (B.S. 1962, M.S. 1963) from what was then proudly
known as Western Reserve University before its merger with the
Case Institute of Technology. We also thank the National Science
Foundation (CHE-0202040, CHE-0518510, and CHE-0748982) for
support of this work and for funds for purchase of the single crystal
X-ray diffractometer (CHE-0541766), and thank Professor Phil
Power (UC Davis) for thoughtful comments that inspired the cur-
rent investigation.
(b) R.S. Simons, L. Pu, M.M. Olmstead, P.P. Power, Organometallics 16 (1997)
1920;
(c) N.J. Hardman, B. Twamley, P.P. Power, Angew. Chem., Int. Ed. 39 (2000)
2771;
(d) A.F. Richards, M. Brynda, M.M. Olmstead, P.P. Power, Organometallics 23
(2004) 2841;
(e) H.J. Breunig, E. Lork, O. Moldovan, C.I. Rat, J. Organomet. Chem. 693 (2008)
2527;
Appendix A. Supplementary material
(f) T. Matsumoto, Y. Matsui, M. Ito, K. Tatsumi, Chem. Asian J. 3 (2008)
607;
(g) D.V. Partyka, M.P. Washington, J.B. Updegraff, X. Chen, C.D. Incarvito, A.L.
Rheingold, J.D. Protasiewicz, J. Organomet. Chem. 694 (2009) 1441;
(h) S. Spirk, F. Belaj, J. Baumgartner, R. Pietschnig, Z. Anorg. Allg. Chem. 635
(2009) 1048.
CCDC 769606, 769607 and 769608 contain the supplementary
crystallographic data for compounds 1, 2 and 3, respectively. 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.2010.07.018.
[24] B. Twamley, C.D. Sofield, M.M. Olmstead, P.P. Power, J. Am. Chem. Soc. 121
(1999) 3357.
[25] S. Shah, S.C. Burdette, S. Swavey, F.L. Urbach, J.D. Protasiewicz, Organometallics
16 (1997) 3395.
[26] V.B. Gudimetla, L. Ma, M.P. Washington, J.L. Payton, M.C. Simpson, J.D.
Protasiewicz, Eur. J. Inorg. Chem. (2010) 854.
References
[27] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R.
Cheeseman, J. Montgomery, J.A., 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.W. 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. Johnson,
W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, in: I. Gaussian (Ed), GAUSSIAN
03, Wallingford, CT, 2004.
[1] P.P. Power, J. Chem. Soc., Dalton. Trans. (1998) 2939.
[2] M. Yoshifuji, I. Shima, N. Inamoto, K. Hirotsu, T. Higuchi, J. Am. Chem. Soc. 103
(1981) 4587.
[3] A.H. Cowley, A. Decken, N.C. Norman, C. Krüger, F. Lutz, H. Jacobsen, T. Ziegler,
J. Am. Chem. Soc. 119 (1997) 3389.
[4] E. Urnezius, J.D. Protasiewicz, Main Group Chem. 1 (1996) 369.
[5] R.C. Smith, T. Ren, J.D. Protasiewicz, Eur. J. Inorg. Chem. (2002) 2779.
[6] S. Shah, T. Concolino, A.L. Rheingold, J.D. Protasiewicz, Inorg. Chem. 39 (2000)
3860.
[7] R.C. Smith, E. Urnezius, K.-C. Lam, A.L. Rheingold, J.D. Protasiewicz, Inorg.
Chem. 41 (2002) 5296.
[8] B. Buster, A.A. Diaz, T. Graham, R. Khan, M.A. Khan, D.R. Powell, R.J.
Wehmschulte, Inorg. Chim. Acta 362 (2009) 3465.
[9] International Union of Crystallography, <http://checkcif.iucr.org/>.
[10] F.L. Hirshfeld, Acta Crystallogr., Sect. A: Cryst. Phys. Diffr. Theor. Gen.
Crystallogr. Sect. A 32(Pt. 2) (1976) 239.
[28] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[29] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys. 37 (1988)
785.
[11] G.M. Sheldrick, SHELXTL, Version 6.14, Bruker-Nonius AXS, Madison, WI.
[12] H. Hope, Prog. Inorg. Chem. 41 (1994) 1.
[30] M.E. Zandler, F. D’Souza, C. R. Chim. 9 (2006) 960.