Synthesis, structure, NMR spectroscopy, and electrochemistry of the sterically congested triarylarsine Dipp3As: EPR characterization of its radical cation

Article abstract of DOI:10.1071/CH13244

The synthesis, NMR spectroscopy, single-crystal X-ray structure, and solution electrochemistry of the new compound [2,6-{CH(CH3)}2C6H3]3 As, abbreviated as Dipp3As, is reported. The molecule, prepared by reaction of AsCl3 with a pre-formed aryl copper reagent, Dipp4Cu4, crystallizes in the hexagonal space group R3 as a racemic twin. The sum of angles around As, ∠{CAsC}, is 329.13(3)° in the X-ray structure and 329.17° from an R-B3LYP/6-31G(d,p) hybrid density functional theory calculation. The aromatic rings are quite distorted with both the ipso carbon and especially the As atom significantly out of plane by 0.503(3)A. The ambient temperature NMR spectrum fits for C3 symmetry implying that inversion is slow on the NMR timescale. Cyclic voltammetry on a glassy carbon electrode in CH2Cl2 with 0.4M [nBu4N][PF6] over scan rates of 0.05-0.8Vs-1 and temperatures of 22±2°C produced one quasi-reversible process with Em1≤+0.43V at a scan rate of 0.20Vs-1 and a second irreversible process with a peak potential of +1.45V (v. Fc+/0). The diffusion coefficient has been measured as 3.3±0.1×10-6cm2s-1 in CH2Cl2 solution containing 0.4M [nBu4N][PF6]. Chemical oxidation with AgPF6 in CH2Cl2 in degassed solutions in sealed vessels allowed for recording of characteristic EPR spectra; at 293K, a (75As)≤26.1mT and g≤2.021. In frozen solution, an almost isotropic spectrum is obtained (g ≤2.000mT and g ≤2.003mT) and the hyperfine splitting constants are a ≤47.9 and a ≤19.0mT, leading to an estimate for the structure being slightly pyramidal with ∠{CAsC}≈351°.

Full text of DOI:10.1071/CH13244

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem.
Full Paper
http://dx.doi.org/10.1071/CH13244
Synthesis, Structure, NMR Spectroscopy, and
Electrochemistry of the Sterically Congested Triarylarsine
Dipp3As: EPR Characterization of its Radical Cation
A
A
A
Mona Taghavikish, Brock L. Price, Tracey L. Roemmele,
and Rene´ T. Boere´
A B
,
A
Department of Chemistry and Biochemistry, University of Lethbridge,
Lethbridge, Alberta, T1K3M4, Canada.
B
Corresponding author. Email: boere@uleth.ca
The synthesis, NMR spectroscopy, single-crystal X-ray structure, and solution electrochemistry of the new compound
[2,6-{CH(CH3)}2C6H3]3 As, abbreviated as Dipp3As, is reported. The molecule, prepared by reaction of AsCl3 with a pre-
formed aryl copper reagent, Dipp4Cu4, crystallizes in the hexagonal space group R3 as a racemic twin. The sum of angles
P
around As, +{CAsC}, is 329.13(3)8 in the X-ray structure and 329.178 from an R-B3LYP/6-31G(d,p) hybrid density
functional theory calculation. The aromatic rings are quite distorted with both the ipso carbon and especially the As atom
˚
significantly out of plane by 0.503(3) A. The ambient temperature NMR spectrum fits for C3 symmetry implying
that inversion is slow on the NMR timescale. Cyclic voltammetry on a glassy carbon electrode in CH2Cl2 with 0.4 M
[nBu4N][PF6] over scan rates of 0.05–0.8 V sꢀ1 and temperatures of 22 ꢁ 28C produced one quasi-reversible process with
Em1 ¼ þ0.43 V at a scan rate of 0.20 V sꢀ1 and a second irreversible process with a peak potential of þ1.45 V (v. Fcþ/0).
The diffusion coefficient has been measured as 3.3 ꢁ 0.1 ꢂ 10ꢀ6 cm2 sꢀ1 in CH2Cl2 solution containing 0.4 M
[nBu4N][PF6]. Chemical oxidation with AgPF6 in CH2Cl2 in degassed solutions in sealed vessels allowed for recording
of characteristic EPR spectra; at 293 K, a (75As) ¼ 26.1 mT and g ¼ 2.021. In frozen solution, an almost isotropic spectrum
is obtained (g! ¼ 2.000 mT and g> ¼ 2.003 mT) and the hyperfine splitting constants are a! ¼ 47.9 and a> ¼ 19.0 mT,
P
leading to an estimate for the structure being slightly pyramidal with +{CAsC} E 3518.
Manuscript received: 7 May 2013.
Manuscript accepted: 11 June 2013.
Published online: 2 August 2013.
Introduction
inversion compared with its congener Dipp3P. Consequently,
the solution 1H NMR spectrum at room temperature (RT)
resembles that of the phosphine at ꢀ808C. Another marker
of having more pyramidal character is greater stabilisation of
the highest occupied molecular orbital (HOMO); this results in
a first oxidation potential that is 0.32 V more positive for the
arsine compared with Dipp3P. The steric shielding provided by
the bulky Dipp group acts to protect the free radical Dipp3Asþꢃ
despite this facile oxidation potential. EPR characterisation in
combination with the I ¼ 3/2 spin of the 100 % abundant 75As
nucleus provides unambiguous confirmation for the formation
of the radical cation and provides a ready handle for monitoring
its presence, thereby establishingitasa new persistentradical.[15]
The most crowded triarylphosphines reported to date, using
the criterion of the sums of angles around the central atom
P
{+CPC}, are those with three 2,6-diisopropylphenyl sub-
stituents, namely Dipp3P and Tripp3P, where Dipp ¼ 2,6-
diisopropylphenyl and Tripp ¼ 2,4,6-triisopropylphenyl.[1,2]
Interest in the redox properties of this class of compounds is
on-going[3–8] and the first crystal structures of a phosphoniumyl
radical cation were recently reported for several salts of
Tripp3Pþꢃ [9]
There is also very strong current interest in the
.
preparation, properties and applications of heavy p-block ele-
ment radicals,[10–16] but according to Konu and Chivers in 2010:
‘the chemistry of stable arsenic-, antimony-, and bismuth-
centered radicals has remained virtually unexplored’.[12]
Here we report on the synthesis of a new sterically congested
arsine, Dipp3As, with full characterisation and a thorough
examination of its redox properties by chemical and electro-
chemical methods. The structure was determined by X-ray
crystallography at low temperature (LT) which establishes its
Results and Discussion
Synthesis and Reactions
Dipp3As is prepared by the reaction of Dipp4Cu4 with AsCl3 in
a one-pot reaction starting from DippBr. The use of aryl copper
reagents is essential for successful introduction of the third
bulky aryl group. If Grignard or lithium reagents are employed
in this last step, there is a strong tendency towards reductive
coupling rather than substitution (by formation of products
such as, for example, Tripp2P–PTripp2 from the reaction of PCl3
claim to ‘most sterically congested’ by edging out Tripp3As[2]
P
based on the sum of angles metric, +{CAsC}, for flattening
at the central atom.[17] Despite the high steric pressure, the
intrinsically greater pyramidal character of arsenic(III) over
phosphorus(III) results in a substantially higher barrier to planar
Journal compilation Ó CSIRO 2013
www.publish.csiro.au/journals/ajc

B
M. Taghavikish et al.
with TrippMgBr as shown many years ago by its accidental
synthesis).[18]
to the strong repulsion among the endo C10–11–12 isopropyl
groups, and the latter to protection by the exo C7–8–9 isopropyl
groups of the As ‘lone pair’ electrons, which contribute the
most to the highest occupied molecular orbital (HOMO) of
Crystal Structure Determinations
The crystal structure of Dipp3As was determined at low tem-
perature (173 K). A summary of the crystal, structure solution,
and refinement data is presented in the Experimental section
and selected geometric parameters are compiled in Table 1.
The structure can be compared most fruitfully with Tripp3As
reported by Sasaki et al. in 2002 in a short communication.[2]
Similar to Dipp3P,[1] but unlike the related Tripp compounds
which have low symmetry space groups, it crystallises as a
racemic twin in the chiral space group R3 with a 62 : 38 ratio of
the two enantiomers defined by C3 rotation about the principal
Dipp3As.[1,17] Thedegree of steric pressureamongst R3As species
P
can be ascertained from the aforementioned
+{CAsC},
whereas the degree of steric shielding can be perused visually
(Fig. 1) byconsidering the flanking exo isopropyl groups. A figure
that includes the H atoms as well as a space-filling diagram is
available as Fig. S1 in the Supplementary Material; these show
that the As atom in Dipp3As is effectively hidden from external
influence. A survey of the 169 tri-carbon R3As structures reported
in the Cambridge Structure Database (CSD version 5.34 with
updates to February, 2013)[20] indicates that only Tripp3As
approximates the degree of steric pressure encountered in
˚
axis (Fig. 1). The equivalent As–C bonds are 1.9903(9) A long,
˚
compared with an average of 1.986(3) A in the two independent
molecules of the reported structure for Tripp3As; these values
are not significantly different at the 99 % confidence level.
Dipp3As. Indeed the next largest values encountered amongst
P
these structures for +{CAsC} is in di-tertbutyl-pentamethyl-
cyclopentadien-1-yl-arsenic at 322.48;[21] the structure of tris
(tertbutyl)arsine has apparently not been reported. Tertbutyl
groups induce very large amounts of steric pressure but are
P
The sum of angles around As, +{CAsC} ¼ 329.13(3)8,[17] is
P
significantly larger than +{CAsC} ¼ 327.70(4)8 in Tripp3As
at the 99 % confidence level. Similarly, the sum of angles
relatively poor at shielding.[17] Another relatively flattened arse-
P
P
around P in Dipp3P, +{CPC} ¼ 335.64(2)8, is larger (even at
the 99 % confidence level) than that reported around Tripp3P,
334.4(1)8; whether these effects reflect differences in inductive
electron donating ability caused by the ‘extra’ iPr group in the
Tripp ligands, or are merely due to fortuitous crystal packing
effects, remains unclear. There is considerable distortion to the
aryl rings due to the severe steric crowding at the central As; for
example, whereas the five ring atoms C2 . C6 define a rea-
nic centre, +{CAsC} ¼ 313.48, is encountered in the phos-
methylide.[22] This may be compared with the ‘base’ value in
phorus
ylide
triphenylphosphonium-bis(diphenylarsino)-
C9
C8
C7
C2
˚
As
sonable plane (root mean square (r.m.s.) deviation of 0.005 A),
the ipso atom C1 and the As deviate from this plane (in the same
C3
C4
C1
˚
direction) by 0.0383(16) and 0.503(3) A. We note that the
sums of the angles at As are 6.58–6.78 smaller than at P in
Dipp3P, reflecting the higher inherent pyramidality of
arsenic.[19] Molecules of Dipp3As pack densely (Fig. 2) with a
single kind of intermolecular contact less than the sum of van der
Waals’ radii between an iPr methyl hydrogen and a ring carbon
C6
C5
C12
C10
˚
atom at 2.870(1) A. In the chiral space group R3, all the C3As
C11
pyramids are oriented in the same direction (resulting in a high
overall polarity); the overall crystal dipole moment is likely
reduced as a result of the racemic twinning.
Earlier we introduced the distinction between steric pressure
and steric shielding in R3E compounds, where the former refers
Fig. 1. Displacement ellipsoids plot (30 %) of the molecular structure of
one enantiomer of Dipp3As as found in the crystal lattice at 173 K. All three
rings are numbered identically due to three-fold rotational crystal symmetry;
H atoms omitted for a less obstructed view.
Table 1. Comparative structural data for Dipp3As and Dipp3As1ꢃ from X-ray diffraction and computationA
Parameter
X-ray diffraction
In R/U-B3LYP/6-31G(d,p)
Dipp3As C3
Dipp3Asþꢃ C3
Dipp3As
Dipp3PB
˚
E–C1 [A]
˚
C1–C2 [A]
1.9904(10)
1.4179(15)
1.4050(17)
0.656
1.8507(16)
1.425(2)
1.418(2)
0.539
2.0136
1.4252
1.4152
0.663
1.9334
1.4181
1.4159
0.174
˚
C1–C6 [A]
E oop of C1C [A]
˚
E oop C2 . C6D [A]
0.5030(3)
0.430
0.406
0.047
˚
P
+{CEC} [deg.]8
C#1–E–C1 [deg.]
C2–C6–C1 [deg.]
C2–C1–As [deg.]
C6–C1–As [deg.]
329.13
335.64
329.17
357.59
109.71(3)
119.46(11)
112.16(7)
127.21(9)
111.88(5)
118.90(16)
113.10(12)
127.17(13)
109.72
119.31
112.64
127.37
119.20
122.24
116.56
121.18
ASymmetry transformation used to generate equivalent atoms #1 (–y, x – y, z).
BRef. [1].
COut of the plane of the three ipso C atoms.
DOut of the plane of the five ring C atoms excluding C1.

Dipp3As and its Radical Cation
C
P
tiphenylarsine of
+{CAsC}¼ 300.28.[23] Amongst these
is reported to react extremely slowly with oxygen.[24] The other
serious contender for highest steric shielding is {1-(8-Me2N)
C8H6}3As which has been shown to display internal coordina-
tion between the three pendant dimethylamino groups and the
central arsenic atom.[28] Although this particular compound
offers the possibility of electronic as well as steric protection
for the HOMO and may thus be especially stable, the relative
reactivity of this arsine does not seem to have been further
investigated.
structures there seem to be an unusually large number of R3As
compounds with strongly pyramidal geometries but bearing
substituents that provide a large degree of steric shielding of their
HOMOs. Comparative space filling models drawn to scale of 10
contenders for the most sterically shielded R3As are shown in Fig.
S2 in the Supplementary Material alongside similar pictures of
Dipp3As and the ‘base’ compound Ph3As.[3,21–29]
Perhaps the most exotic of these species is tris((1,2-dicarba-
closo-dodecaboran-1-yl)methyl)arsine, in which the central
As atom is flanked by three B10C2 carboranes; this compound
Solution Structure by NMR Spectroscopy
The NMR spectra taken in conventional solvents at ambient
conditions are consistent with C3 symmetry (Table 2, with
copies of spectra in the Supplementary Material), with all three
Dipp groups identical but in the slow-exchange limit for pyra-
midal inversion. This behaviour is quite different from that of
1
Dipp3P, which at ambient temperature displays H and 13C
NMR spectra consistent with D3 symmetry which is attributed
to fast pyramidal inversion.[1] The latter compound, however,
displays spectra at ꢀ808C that fit C3 symmetry and there is
a remarkable similarity between these LT spectra and those of
Dipp3As recorded at RT. This allows assignment of the signals
based on those of the phosphine where 31P coupling patterns are
a helpful aid (Table 2). With reference to the structure shown in
Fig. 1 from crystallography, the two isopropyl groups on any of
the three rings are shift inequivalent resulting in two methine 1H
resonances at 3.31 and 3.29 ppm. Each isopropyl group also has
inequivalent methyl environments resulting in four methyl 1H
doublets of equal intensity (9H each) at 1.28, 1.09, 1.05, and
0.51 ppm. The remarkably high-field resonance of the last listed
(assigned to the protons of C12) is caused by ‘anisotropic ring
shielding’ from the closer proximity of this particular methyl
group to the ring current of an adjacent aromatic ring. The
equivalent shift in the LT Dipp3P spectrum is þ0.36 ppm and a
reasonable explanation for the differences in chemical shift is
a reduced ring shielding influence in the arsine from longer
C–As compared with C–P bond lengths. Certainly in the solid
a
o
c
b
Fig. 2. Packing diagram showing the three-fold symmetry, with dashed
˚
lines indicating the only short intermolecular contacts of 2.870(1) A between
isopropyl methyl hydrogen H8C and the p-electron density of ring carbon
C3. View is up the c-axis.
Table 2. 1H and 13C NMR data for Dipp3As with comparison to Dipp3P at ambient and low temperatureA
Parameter
C1
C2
C6
C4
C3
C5
C10
C7
C9
C11
C8
C12
Dipp3PB
d(1H) þ308CC
7.29
7.68
0
7.11
7.68
3.28
3.50
6.72
5.20
1.17
6.72
0
0.71
6.72
0
JHH [Hz]
JHP [Hz]
d(1H) ꢀ808CC
7.27
7.76
0
7.07
7.58
3.3
7.07
3.30
3.14
1.13
6.1
0
0.99
6.3
0
0.90
6.3
0
0.36
6.2
0
JHH [Hz]
7.58 ,6
3.3
,6
,6.4
JHP [Hz]
0
Dipp3As
d(1H) þ228CD
7.26
7.5
7.04
7.5
7.04
7.5
3.42
6.6
3.28
1.28
6.6
1.09
6.7
1.05
6.7
0.51
6.6
JHH [Hz]
Dipp3PB
6.6
d(13C) þ208CD 135.37
153.64
18.6
129.38
0
124.37
3.9
32.37
17.6
32.19
0
32.37
17.6
23.30
24.82
JCP, [Hz]
25.9
0
0
d(13C) ꢀ808CC 134.26 153.00 151.94 128.64 125
122.47
0
31.40 21.77 22.65 23.59 24.41
32
JCP [Hz]
25
40
0
0
6
0
0
0
0
Dipp3As
d(13C) þ228CD 139.48 153.67 153.25 129.03 125.31 123.02
33.90
33.02 22.98 24.21 24.45 25.57
AThe atom numbering scheme is that used in the X-ray structure (Fig. 1) in which all three Dipp rings are identical by symmetry. For
H atoms, the attached C atom is designated.
BFrom Ref. [1].
CCD2Cl2.
DCDCl3. Gradient COSY confirms the isopropyl group assignments.

D
M. Taghavikish et al.
structures, the average methyl-hydrogen to ring-centroid dis-
tance is larger in the arsine (C12 to closest ring-centroid
˚
distances of 3.81 and 3.77 A for As and P, respectively).
‘scale’ to the internal Fcþ/0 reference scale we established pre-
viously for key phosphines (as presented on the right-hand side
of Table 4).[1] The values in Table 4 are therefore directly
comparable for all listed species in CH2Cl2 solutions; as indi-
cated in the footnotes, the CH3CN data of Romakhin et al. is also
in good agreement with these data since the latter solvent shifts
[1]
The presence of two distinct methine resonances (33.9 and
33.0 ppm) and four different methyl signals at 25.6, 24.5, 24.2
and 23.0 ppm is also observed in the 13C NMR spectrum.
Moreover, all six phenyl ring carbon resonances are distinct,
indicative of frozen or at least slowed rotation about the Cipso–As
bond. The signals in Dipp3As lack the diagnostic presence of
phosphorus nuclear couplings and hence the specificassignments
in Table 2 are tentative and are based largely on analogy to those
obtained in the LT Dipp3P spectrum. The enforced geometry of
these highly sterically encumbered molecules and the excellent
fit between the solution NMR and solid-state structural data
suggests that the solution structures retain most of the features
found in the solid state. That the arsine should still be in the slow
exchange limit at RT when the phosphine requires cooling by as
much as one hundred degrees to achieve the same state is
consistent with estimates of the barriers to pyramidal inversion
that are 54–58 kJ molꢀ1 higher at As compared with P.[30]
Table 3. Summary of cyclic voltammetry data obtained for Dipp3As
at a glassy carbon electrode and at T 5 228C
Obtained in CH2Cl2 solution containing 0.4 M [nBu4N][PF6] measured
against (Ccþ/0) cobaltocenium redox couple and expressed v. E8 . See text
for definition of the parameters described
Fcþ/0
n [V sꢀ1
]
Epa1 [V] Epc1 [V] Em1
ffi
DEp1 [mV] Eap2 [V] Ipred/Ipox
E8 [V]
f1
0.05
0.10
0.20
0.40
0.80
0.470
0.470
0.502
0.534
0.565
0.351
0.351
0.354
0.330
0.312
0.410
0.410
0.428
0.432
0.438
119
119
148
204
253
1.402
1.402
1.447
1.514
1.565
0.96
0.91
0.93
1.00
0.98
Cyclic Voltammetry (CV) Studies
Epa2
Dipp3As was studied by CV on a glassy carbon (GC) electrode in
CH2Cl2 with 0.4 M [nBu4N][PF6] over scan rates of 0.05–
20 V sꢀ1 and temperatures of 22 ꢁ 28C (Table 3). Two oxidation
processes appear upon sweeping the potential in the
anodic direction from the rest potential, designated as Eap1 and
Eap2. Voltammetric data were calibrated using internal cobalto-
cenium hexafluorophosphate (i.e. the Ccþ/0 redox couple) and
expressed on the ferrocene (Fcþ/0) scale as zero by subtracting
1.35 V,[31] and are reported in Table 3. Representative cyclic
voltammograms obtained at a scan rate of 0.2 V sꢀ1 are shown in
Fig. 3. The first oxidation process displayed close to electro-
chemically reversible behaviour, with anodic (Eap1) to cathodic
(Ecp1) peak-to-peak separations, DEp1, in the range of 119 to
253 mV, and is chemically reversible on the timescale of CV and
thus indicative of substantial stability of the cation radical. That
this process involves one-electron transfer is evidenced by cur-
rents comparable to those from equal molar amounts of internal
reference (not shown in the figure) and by the EPR spectroscopic
results (see below). The second, irreversible oxidation process
labelled Eap2 in Fig. 3 is observed at a considerably more anodic
potential, which is similar to what has been reported for
Dipp3P.[1] For Dipp3As, fast scanning (up to 20 V sꢀ1) has no
effect on the relative size of the return wave for this second
process which remains irreversible under all conditions that we
studied. Moreover, cycling of the potential through both redox
couples serves to decrease the size of the return wave (Ecp1) for
the first oxidation process. The Irped/Ipox ratio is independent of
the scan rate within experimental error for the Eap1/Epc1 process.
120
Epa1
80
40
0
᎐40
Epc1
0.2
0.6
1.0
1.4
1.8
Potential [V]
Fig. 3. Cyclic voltammetry of Dipp3As (5 mM) in CH2Cl2 solution
containing 0.4 M [nBu4N][PF6] under an atmosphere of dry argon at a scan
rate of 0.2 V sꢀ1 on a glassy carbon electrode at 228C: (a) through the first
oxidation process (red line), and (b) through the first and second oxidation
processes (black line).
Table 4. Comparative redox potential data in CH2Cl2 for R3E
(E 5 As, P)A
See text for definition of the parameters listed
P
7.9F
6.3F
6.5F
6.7F
Compound
Em1
ffi
Compound
Em1
ffi
DEm1
s
D[ +
{CEC}]P-As
E8 [V]
f1
E8 [V]
f1
Determination of the Diffusion Coefficients of Dipp3As
Ph3AsB,C
Mes3AsB,D
Dipp3As
(0.96)
0.51
0.42
0.28
Ph3PE
(0.88)
0.19
0.08
0.32
0.33
0.34
Rotated disk electrode (RDE) measurements on Dipp3As were
undertaken in order to determine the diffusion coefficient D by
making use of the Levich equation.[32] Solving for D yields an
estimate of (3.3 ꢁ 0.1) ꢂ 10ꢀ6 cm2 sꢀ1 for Dipp3As. The value
for kinematic viscosity (nk) was taken to be that of the solvent
CH2Cl2 at 208C rather than a measured value for the solution.[33]
Mes3PE
Dipp3PE
Tripp3PE
0.09
Tripp3AsB
ꢀ0.06
AMidpoint potentials from cyclic voltammetry or rotated disk electrode
(RDE) measurements, scan rate ¼ 0.2 V sꢀ1; values in parenthesis are Eox
peak potentials.
BRef. [2], corrected by ꢀ0.22 V based on Mes3P potentials, see Ref. [1].
Interpretation of Voltammetric Data
C
þ1.11 V in CH3CN, Ref. [34].
D
We are aware of only one report that provides comparative
redox potential data for numerous triorganoarsines.[34] Since
these authors also report analogous phosphine data, we can
þ0.62 V in CH3CN, Ref. [34].
ERef. [1].
FFrom the published crystal structures and this work; see also Ref. [17].

Dipp3As and its Radical Cation
E
phosphine oxidation potentials þ0.07–0.10 V more positive.[1]
If we disregard the Ph3E species, since for such irreversible
redox processes the potentials may be strongly affected by the
rates of follow-up reactions, for the other three species with
reversible or close-to-reversible behaviour we find that corre-
sponding arsines are þ0.32 to þ0.34 V more difficult to oxidise
than phospines. This is fully consistent with lower-lying
HOMOs as a result of the more pyramidal geometry at the
central atom. The structural studies already discussed inform us
that the change for Ar3E from P to As results in a decrease in the
sum of angles around the pnictogen of 6.38–6.78 (Table 4) for the
three with reversible Em1 values.
intensity ratio. However, what is observed are four lines with
different intensities and different spacing.
The unequal intensity pattern is a consequence of ‘slowed
tumbling’ from incomplete averaging of the g and a tensors.[36]
This partial averaging of the signals can be accounted for by a
linear fit to the correlation time for molecular motion which
accurately fits the peak-to-peak line widths DBp–p to three
constants A, B, and C (Table 5). The unequal spacing of
the lines is a consequence of the large size of the 75As hyperfine
splitting (on the order of 26 mT, ,8 % of the resonant field
strength); when the energy of the hyperfine interaction
approaches that of the Zeeman interaction the two are not fully
separable. This is commonly referred to as a ‘higher order’
interaction, since an accurate solution to the spin Hamiltonian
by perturbation methods requires second or third order terms.[37]
Importantly, the magnitude of both the hyperfine coupling and
the Zeeman energy are affected, and hence a full simulation is
required to recover both the a and g terms accurately. The effects
of the two corrections are illustrated by the full lineshape
simulations shown in Fig. 4. Fig. 4b incorporates both correc-
tions while Fig. 4c shows the corrected line positions but has the
linewidth correction turned off so as to better illustrate the
separate effects.
Fluid Solution EPR Characterisation of Dipp3As1ꢃ
The chemically reversible Dipp3Asþ/0 redox couple occurs at an
average potential of þ0.43 V v. Fc0/þ, and no further oxidation
occurs under CV conditions until þ1.45 V v. Fc0/þ (with onset
of current flow for this second process .1.2 V). The list of
potential chemical oxidising agents provided by Connelly and
Geiger advises that the AgI ion in CH2Cl2 solution should be a
suitable oxidant for one-electron oxidation, with little chance of
initiating double oxidation to the presumably unstable dication
(the Agþ/0 redox couple is reported as þ0.65 V v. Fc0/þ in this
solvent).[35] This is an attractive reagent because the by-product
is expected to precipitate as solid silver. Therefore AgPF6 was
used to generate solutions of [Dipp3As]1ꢃ PF6ꢀ salts in CH2Cl2.
Oxidised solutions prepared under vacuum in sealed tubes
subsequent to freeze–thaw degassing of the solvent displayed
persistent fluid solution EPR spectra (Fig. 4a). Based on
experience with corresponding phosphines and evidence from
density functional theory (DFT) calculations (see below), the
dominant hyperfine coupling is expected to be from the single
arsenic nucleus (75As, I ¼ 3/2, 100 % abundant), which ought to
result in a four-line spectrum with equal spacing and a 1 : 1 : 1 : 1
Excellent agreement with the experimental spectrum is
thereby achieved and the corrected a and g values are presented
in Table 5. At 293 K, a (75As) ¼ 26.1 mT and g ¼ 2.045. There is
significant temperature dependence of the EPR linewidth para-
meters consistent with slowed re-orientational motion on cool-
ing. Such a dependence has been observed previously for
Dipp3Pþꢃ (R. T. Boere´, Y. Zhang, unpubl. data). In their
preliminary report, Sasaki et al. did not comment on temperature
dependence for Tripp3Asþꢃ [2]
but the published ambient tem-
,
perature spectrum indicates that this radical is also in the slowed
tumbling regime; they report a g-value of 2.042 (which would
appear not to have been corrected for the higher-order coupling)
and an apparent isotropic a(75As) of 26.4 mT. These authors also
report EPR data for Mes3Asþꢃ of g ¼ 2.038 and a(75As) ¼ 27.
5 mT but here too it is not clear if the appropriate corrections
were made.
(a)
(b)
Frozen Solution EPR Characterisation of Dipp3As1ꢃ
In frozen CH2Cl2 solutions at 183 K, a characteristic axially
symmetric powder pattern was obtained for Dipp3Asþꢃ with
Table 5. Fluid solution EPR data for Dipp3As1ꢃ from full lineshape
simulationsA
See text for definition of the parameters listed
Temp. [K] g-value a(75As) [mT]
A
B [mT]
B
B [mT]
C
B [mT]
293
273
263
243
233
2.0210
2.0210
2.0210
2.0220
2.0220
26.11C
26.20
26.20
26.25
26.25
1.30
1.45
1.74
2.03
2.36
ꢀ0.16
ꢀ0.23
ꢀ0.34
ꢀ0.42
ꢀ0.53
0.24
0.36
0.55
0.75
0.97
(c)
ASimulations undertaken with SimFonia, v. 1.26 (Bruker Biospin, 1996).
BLinewidth parameters A, B, and C determined from the simulations by
fitting to DBp–p ¼ A þ BmI þ Cm2I . A, B, and C depend linearly on the
correlation time and on the anisotropic a and g tensors. A and C are always
positive, while B can be positive or negative and determines which line is
broadened most in the spectrum, Ref. [36]. For a plot of the A constant v.
temperature, see Fig. S5 in the Supplementary Material.
290
310
330
350
370
390 [mT]
Fig. 4. (a) Experimental fluid solution EPR spectrum of Dipp3Asþꢃ
obtained after chemical oxidation determined at 293 K. (b) Best simulated
spectrum (third-order perturbation theory and line-widths corrected for
tumbling). (c) Simulation as in (b) without the tumbling correction.
CFrom the simulation; measured spacings of the lines at 293 K: 24.126,
26.136, and 28.272 mT.

F
M. Taghavikish et al.
parallel and perpendicular hyperfine splitting (HFS) tensor
components of 47.9 and 19.0 mT (Fig. 5a). The parallel and
perpendicular g tensors are almost the same, so the EPR spec-
trum is close to being a g-isotropic system (g! ¼ 2.000 mT and
g> ¼ 2.003 mT). The simulation achieved is in reasonable
agreement with the observed spectrum and the fit is extremely
sensitive to alteration of the parameters (Fig. 5b). There is a
paucity of comparative data for triorgano-arsoniumyl radical
cations in the literature but what we have been able to locate is
presented in Table 6. There seems to be only a single report
about Ph3Asþꢃ, in which it was produced by radiolysis in
sulfuric acid.[38] By contrast, when Ph3As reacts with oxygen-
based free radicals, adducts of the type Ph3As–ORꢃ are the major
products.[39,40] Two studies on Mes3Asþꢃ report somewhat
divergent parameters.[2,41] The values of a! and a> allow for the
calculation of the relative s and p spin densities at As using the
calculated full-spin hyperfine parameters of Morton and Pre-
ston.[37,42] From this value the Coulson equation permits an
estimation of the pyramidal character of the radical cations
(further details on this procedure are found in the Supplementary
Material).[43] As expected, Dipp3Asþꢃ is greatly flattened upon
P
oxidation (a change in +{CAsC} of 228 from that in the
neutral molecule). This parameter is only marginally smaller in
Mes3Asþꢃ (using the Il’yasov data) but Ph3Asþꢃ is, as expected,
noticeably more pyramidal.[7,8]
DFT Calculations on Dipp3As1ꢃ/0
DFT calculations were undertaken in order to obtain informa-
tion about the structure of the radical cation as well as to gain
insights into the changes in electronic structure that accompany
the oxidation of Dipp3As (details of methodology and atom
coordinates are provided in the Supplementary Material.) Only
the neutral and þ1 oxidation state have been considered as we
yet have little experimental evidence about a putative dication;
if in fact Dipp3As2þ is formed at the potential of the second
observed CV peak, it would seem to be very electrophilic and is
destroyed by adventitious moisture or by other pathways.
However, the radical cation is persistent and solutions prepared
under vacuum last for days or more. For neutral Dipp3As the
agreement between the calculated and experimental X-ray
geometry (Table 1) is excellent. Key strained bond distances
(a)
(b)
agree within 1 % and angles within 0.5 %. The calculated and
P
experimental {+CAsC} is identical within experimental
error. The most significant deviation occurs in the amount by
which the As atom lies above the plane of the five non-ipso
phenyl ring C atoms; the gas-phase calculation predicts less
distortion from ideality than the solid state structure.
This agreement provides some confidence in assessing the
(unknown) radical structure. The Coulson equation provides an
260
280
300
320
340
360
380
400
420 [mT]
P
estimate of +{CAsC} ¼ 351.08 for Dipp3Asþꢃ from the EPR
Fig. 5. (a) Experimental frozen solution EPR spectrum of Dipp3Asþꢃ
obtained by chemical oxidation in CH2Cl2 at 183 K and (b) simulation
obtained using second order perturbation theory (line width (LW) for the
simulation x ¼ y ¼ 1.5; z ¼ 3.5 mT).
data (Table 6). The DFT calculated value is even more planar
at 357.68. Three crystal structures recently reported for
P
Tripp3PþꢃXꢀ salts have measured +{CPC} ranging from
359.648 to 359.998.[9] These authors also report that their DFT
Table 6. EPR and structural parameters in Dipp3As1ꢃ and related radicals
See text for definition of the parameters listed
P
P
Species
giso
g!
g>
aiso [mT]
a! [mT]
a> [mT]
r
4p/r4s
+{CAsC} [deg.]
+{CAsC}DFT [deg.]
Dipp3AsþꢃA
Tripp3AsþꢃB
Mes3AsþꢃB
Mes3AsþꢃC
Ph3AsþꢃD
2.021
2.042
2.038
2.016
2.010
2.000
2.002
2.004
1.96
2.003
2.020
2.056
2.044
2.017
26.1
26.4
27.5
27.9
30.9
47.9
48.3
45.6
46.2
46.5
19.0
19.25
17.7
19.3
23.1
14. 8
14.7
15.1
13.9
11.1
351.0
350.9
351.2
350.5
348.4
357.6
–
–
–
–
1.995
AThis work; 183 K.
BRef. [2]; the magnitude of the g factors suggest that the parameters are uncorrected for second order effects.
CRef. [42].
DRef. [39]; frozen solution 77 K; italicized isotropic values are calculated from the anisotropic data.
Table 7. Summary of density functional theory calculations for Dipp3E1ꢃ/0 (E 5 P, As)A
See text for definition of the parameters listed
P
P
P
Species
{+CEC} (neutral) [deg.]
{+CEC} (cation) [deg.]
D
{+CEC} [deg.] Exp aiso [mT] Calc aiso [mT] rE
Dipp3As
Dipp3P[1]
329.17
336.6
357.6
359.7
28.4
23.1
26.1
23.9
18.7
13.4
0.787
0.858
AR/U B3LYP/631G(d,p) density functional theory calculations using Gaussian 03W (full details in the Supplementary Material).

Dipp3As and its Radical Cation
G
calculations on Dipp3Pþꢃ only converge when planarity is
imposed. We found no difficulty in optimising the non-planar
Dipp3Asþꢃ structure which is confirmed by a frequency calcu-
lation. As Pan et al. also conclude, the bending mode for
flattening this class of radical cation with highly congested
substituents is probably very small.[9] Nevertheless the substan-
tial size of the HFS values measured for such R3Eþꢃ species,
whether E is As or P, strongly supports solution-phase structures
that have minima with pyramidal geometries and indeed sub-
stantial lifetimes in these pyramidal states, however low the
barrier to pyramidal inversion may be.
filtration under vacuum. Thin layer chromatography (TLC) of
the tan coloured residue after solvent removal indicated two
components. Column chromatography on silica with hexanes
eluted the leading band, which was recovered and recrystallised
from hot n-heptane to afford crystalline Dipp3As (1.59 g,
2.76 mmol, 43 % yield), mp 339–3458C, determined to be pure
by NMR spectroscopy. A second crystallisation from n-heptane
afforded blocks suitable for X-ray diffraction. For NMR data see
Table 2. m/z (ESI, CH3CN) Calc. for C36H51As (Mþꢃ
)
558.32012. Found 558.31885. Anal. Calc. for C36H51As: C
77.39, H 9.20. Found: C 77.41, H 9.23 %.
The calculated spin density on P in Dipp3Pþꢃ is 0.858
(Table 7), compared with 0.787 on As in Dipp3Asþꢃ (the
remaining spin density is distributed over the C and H atoms,
which contributes significantly to line broadening) (R. T. Boere´,
Y. Zhang, unpubl. data). The ratio of the solution phase isotropic
HFS of 23.9 mT in Dipp3P and 26.1 mT in Dipp3As, is 0.916.
The ratio of the calculated full-spin hyperfine parameters Aiso
for 31P and 75As of 474.79 and 523.11 mT is 0.908.[37,42] The
similarity in these values also suggests that the amount of
s-orbital spin density on Dipp3Pþꢃ and Dipp3Asþꢃ is more
congruent than indicated by the (relatively low-level) DFT
calculations.
X-Ray Crystallographic Study
A colourless prism of Dipp3As (0.24 ꢂ 0.18 ꢂ 0.15 mm3) was
coated in Paratone oil, mounted on a glass fibre, and cooled to
ꢀ1008C for data collection on a Bruker Apex II CCD area
˚
detector diffractometer (MoKa1, l 0.71073 A) using v–f scans
(y range ¼ 2.458 to 28.818), followed by data correction and
reduction using SAINT-Plus.[45] A total of 9509 reflections were
measured of which 2635 were unique (Rint 0.05). The structure
was solved with dual-space methods (SHELX-M) and refined
with SHELX-L.[46] The compound, C36H51As, formula weight
558.69, was found to crystallise in the hexagonal space group
˚
˚
R3 (#146) [a 16.6193(16) A, c 10.1772(16) A], with three
molecules per unit cell. A semi-empirical absorption correction
Experimental
Materials and General Methods
(SADABS) from equivalents was applied (m ¼ 0.14 mmꢀ1
,
transmission factors 0.856 and 0.784).[45] The data is complete
Unless otherwise stated, all work was carried out under a dry
nitrogen atmosphere, using standard Schlenk techniques or an
MBraun Glove box. AsCl3 (Aldrich) was used as received.
Magnesium turnings (Aldrich) were stored in a glove box and
activated by grinding before use. Solvents for synthesis were
collected from a solvent purification system (MBraun) or dis-
tilled under an N2 atmosphere and degassed by freeze–pump–
thaw degassing. Drying agents for solvent purification include
sodium/benzophenone (tetrahydrofuran) and sodium (n-
heptane). 1H and 13C NMR spectra were recorded on a Bruker/
Tecmag AC250 spectrometer (1H, 250.13 MHz; 13C, 62.9 MHz)
and referenced relative to 0 ppm for tetramethylsilane (1H, 13C)
using either deuterated chloroform or deuterated benzene as
secondary references. Electrospray ionisation (ESI) mass spectra
were measured using a Thermo Instruments Exactive orbitrap
spectrometer. For the electrochemical experiments, dichlor-
omethane (BDH, reagent grade) was purified by distillation from
CaH2 and was purged with dry argon before use. Electrochemical
grade tetra-n-butylammonium hexafluorophosphate [nBu4N][PF6]
(Fluka) was used as the supporting electrolyte and was kept in
a desiccator before use. Bis(cyclopentadienyl)cobalt(III) hexa-
fluorophosphate (Aldrich) was used as the potential reference.
˚
(99.9 %) down to 0.77 A; R1 0.0494, wR2 0.1138 for 2635
observed data and a goodness of fit (GOF) of 1.064, with
largest peaks and hole in the final difference map of 0.29 and
ꢀ0.28 e Aꢀ3. In the course of refinement the distribution
˚
of intensity data was found not to match that expected for
this chiral space group. Subsequently, the structure was suc-
cessfully refined as a racemic twin (twinning by merohedry) in
a 61 : 39 ratio. Mercury 3.1 was employed for crystallographic
figures and post-refinement data analysis.[47] CCDC 937361
contains the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Cyclic Voltammetry
Cyclic voltammograms were obtained at room temperature in
CH2Cl2 containing 0.4 M [nBu4N][PF6] as the supporting elec-
trolyte. These solutions were purged with dry argon for 10 min
directly before use, and argon gas flowed over the solutions
during experiments. The cell design utilised a conventional
three electrode setup with a 3.0 mm diameter GC working
electrode, a Pt wire auxiliary and a silver wire quasi-reference
electrode. The working electrode was polished with a 0.3 micron
Al2O3 (Buehler, 0.05 mm) slurry on a clean polishing cloth,
rinsed with distilled water and acetone, and dried with tissue
paper before use. The reference electrode was separated from
the bulk solution by a fine porosity frit. Experiments com-
menced with initial background scans to characterise the size of
the electrochemical window and provide an estimate of
the likely background current. Cyclic voltammograms were
obtained over scan rates of 0.050–20 V sꢀ1. All potentials are
Preparation of Dipp3As
Activated (grinding and heating in vacuum) magnesium turn-
ings (0.56 g, 22.5 mmol), dibromoethane (0.1 mL), and tetra-
hydrofuran (20 mL) were heated at reflux for 30 m. To the
cooled mixture, DippBr (5.41 g, 22.5 mmol) was added, fol-
lowed by vigorous stirring and heating at reflux for 2.5 h. The
mixture was then stirred overnight at RT. The resultant solution
of DippMgBr[44] was cooled to ꢀ708C and CuCl (98 %, 2.40 g,
23.8 mmol) was added followed by stirring for 1 h at ꢀ708C, and
then warmed to RT and shielded from direct light. AsCl3 (98 %,
1.19 g, 6.44 mmol) was added and the mixture was stirred for
30 min at ꢀ708C, warmed to RT and heated to reflux overnight,
and then cooled to RT. Precipitated CuCl was removed by
reported versus the operative formal potential, E8 , for
Fcþ/0
the Fcþ/0 redox couple. The Ccþ/0 (Cc ¼ cobaltocene) redox
couple is known to appear at ꢀ1.35 V in CH2Cl2; data have been
converted into the Fcþ/0 scale by subtraction of 1.35 V.[31]

H
M. Taghavikish et al.
RDE Voltammetry
[2] S. Sasaki, K. Sutoh, F. Murakami, M. Yoshifuji, J. Am. Chem. Soc.
2002, 124, 14830. doi:10.1021/JA027493N
RDE voltammetry measurements on CH2Cl2 solutions 5.73 mM
in Dipp3As were performed on a Princeton Applied Research
PARSTAT 2273 potentiostat in conjunction with a PINE
Model AFMSRXE Modulated Speed Rotator. The cell used for
RDE measurements replaced the central size-10 joint with a
60 mm ꢂ 15 mm cylinder to accommodate the rotating elec-
trode. A 5.0 mm diameter GC macrodisc electrode (area ¼
1.86 ꢂ 10ꢀ1 cm2) was employed for RDE measurements. Dif-
fusion coefficients were determined from the limiting current
values obtained from RDE voltammetry and use of the Levich
equation.[32] The value for kinematic viscosity (nk) was taken to
[3] S. Sasaki, F. Murakami, M. Yoshifuji, Organometallics 2006, 25, 140.
doi:10.1021/OM050522T
[4] S. Sasaki, M. Yoshifuji, Curr. Org. Chem. 2007, 11, 17. doi:10.2174/
138527207779316507
[5] S. Sasaki, R. Chowdhury, M. Yoshifuji, Tetrahedron Lett. 2004, 45,
9193. doi:10.1016/J.TETLET.2004.10.084
[6] F. Chalier, Y. Berchadsky, J. Finet, G. Gronchi, S. Marque, P. Tordo,
J. Phys. Chem. 1996, 100, 4323. doi:10.1021/JP952854P
[7] C. Palau, Y. Berchadsky, F. Chalier, J. Finet, G. Gronchi, P. Tordo,
J. Phys. Chem. 1995, 99, 158. doi:10.1021/J100001A028
[8] M. Culcasi, Y. Berchadsky, G. Gronchi, P. Tordo, J. Org. Chem. 1991,
56, 3537. doi:10.1021/JO00011A018
be that of the solvent CH2Cl2 at 208C, 0.003318 cm2 sꢀ1 [33]
.
[9] X. Pan, X. Chen, T. Li, Y. Li, X. Wang, J. Am. Chem. Soc. 2013, 135,
Rotation rates from 1000 to 2250 rpm were applied over a
potential range of ꢀ0.5 to þ1 V. Plots of the limiting current
(IꢀL 1) v. the square root of the rotational velocity (vꢀ1/2) yielded
3414. doi:10.1021/JA4012113
[10] R. T. Boere´, Electron Paramag. Reson. 2013, 23, 22. doi:10.1039/
9781849734837-00022
[11] D. Martin, M. Soleilhavoup, G. Bertrand, Chem. Sci. 2011, 2, 389.
doi:10.1039/C0SC00388C
[12] J. Konu, T. Chivers, in Stable Radicals: Fundamentals and Applied
Aspects of Odd-Electron Compounds (Ed. R. G. Hicks) 2010,
pp. 381–406 (John Wiley: Chichester).
a straight line with slope ¼ (0.62nFACnꢀ1/6
D )
2/3 ꢀ1 (where n is
the number of electrons in the charge-transfer process, F is the
Faraday constant, A is the electrode area (cm2), C is the con-
centration (mol cmꢀ3), and D is the diffusion coefficient).
Solving for D yields an estimate of 3.3 ꢁ 0.1 ꢂ 10ꢀ6 cm2 sꢀ1 for
Dipp3As.
[13] A. Armstrong, T. Chivers, R. T. Boere´, ACS Symp. Ser. 2006, 917, 66.
doi:10.1021/BK-2005-0917.CH005
[14] S. Marque, P. Tordo, Top. Curr. Chem. 2005, 250, 43.
[15] P. P. Power, Chem. Rev. 2003, 103, 789. doi:10.1021/CR020406P
[16] M. Geoffroy, Recent Res. Dev. Phys. Chem. 1998, 2, 311.
[17] R. T. Boere´, Y. Zhang, J. Organomet. Chem. 2005, 690, 2651.
doi:10.1016/J.JORGANCHEM.2004.11.043
EPR Spectroscopic Studies of Chemically
Oxidized Dipp3As1ꢃ
An excess of AgPF6 was added to solid Dipp3As in a ‘T’ cell
consisting of a 4 mm Pyrex EPR tube fused at a 758 angle to an
8 mm Pyrex reaction tube. Dry degassed CH2Cl2 was vacuum
distilled onto the solid mixture in the reaction tube and the tube
was flame sealed. Upon melting, an immediate colour change
was observed. The concentration of radical in the EPR tube was
controlled by internal distillation of the solvent so that the
effects of concentration broadening on the EPR linewidths could
be monitored. All spectra were obtained on a Bruker EMX Plus/
10 instrument operating at X band frequencies (9.8 GHz) and
variable temperature (VT) experiments were performed using
the ER4141 VT accessory. Modulation amplitude was varied
between 1 to 5 G to optimize signal to noise for each experiment
and the sweep width was varied between 150 and 300 mT.
[18] F. J. Brady, C. J. Cardin, D. J. Cardin, D. J. Wilcock, Inorg. Chim. Acta
2000, 298, 1. doi:10.1016/S0020-1693(99)00343-6
[19] B. Twamley, C.-S. Hwang, N. J. Hardman, P. P. Power, J. Organomet.
Chem. 2000, 609, 152. doi:10.1016/S0022-328X(00)00162-5
[20] F. H. Allen, Acta Crystallogr. 2002, B58, 380.
[21] P. Jutzi, S. Pilotek, B. Neumann, H.-G. Stammler, J. Organomet.
Chem. 1998, 552, 221. doi:10.1016/S0022-328X(97)00588-3
[22] H. Schmidbaur, P. Nusstein, G. Muller, Z. Naturforsch. B: Chem. Sci.
1984, 39, 1456.
[23] A. N. Sobolev, V. K. Belsky, N. Yu. Chernikova, F. Yu. Akhmadulina,
J. Organomet. Chem. 1983, 244, 129. doi:10.1016/S0022-328X(00)
98593-0
[24] A. S. King, G. Ferguson, J. F. Britten, J. F. Valliant, Inorg. Chem. 2004,
43, 3507. doi:10.1021/IC049947I
[25] O. bin Shawkataly, I. A. Khan, C. S. Yeap, H.-K. Fun, Acta Crystal-
logr. 2009, E65, o2772.
[26] S. C. Chmely, T. P. Hanusa, A. L. Rheingold, Organometallics 2010,
Supplementary Material
Full crystallographic data with comparison to the Dipp3P
structure; supplementary crystallographic figure; figure of space
filling models showing steric shielding in R3As structures; H
and 13C NMR spectra for Dipp3As; EPR linewidth temperature
dependence plot; use of the Coulson equation; and DFT
computational results and references are available on the Jour-
nal’s website.
29, 5551. doi:10.1021/OM100474M
1
[27] H. Lang, L. Zsolnai, Z. Naturforsch. B: Chem. Sci. 1990, 45, 1529.
[28] S. Kamepalli, C. J. Carmalt, R. D. Culp, A. H. Cowley, R. A. Jones,
N. C. Norman, Inorg. Chem. 1996, 35, 6179. doi:10.1021/IC960516C
[29] A. N. Sobolev, I. P. Romm, N. Yu. Chernikova, V. K. Belsky,
E. N. Guryanova, J. Organomet. Chem. 1981, 219, 35. doi:10.1016/
0022-328X(81)85007-3
[30] J. D. Andose, A. Rauk, K. Mislow, J. Am. Chem. Soc. 1974, 96, 6904.
doi:10.1021/JA00829A016
Acknowledgements
[31] R. S. Stojanovic, A. M. Bond, Anal. Chem. 1993, 65, 56. doi:10.1021/
AC00049A012
[32] A. J. Bard, L. R. Faulkner, Electrochemical Methods: Fundamentals
and Applications, 2nd edn 2001 (John Wiley & Sons: New York, NY).
[33] J. Nath, A. P. Dixit, J. Chem. Eng. Data 1984, 29, 317. doi:10.1021/
JE00037A027
The authors thank the University of Lethbridge and the Natural Sciences and
Engineering Research Council of Canada for financial support of this work.
The diffractometer was purchased with the help of both these agencies. The
EPR and mass spectrometers were purchased with support from the Canada
Foundation for Innovation, the Government of Alberta, Bruker Canada, Ltd
(EPR) and Thermo Fisher Canada Inc. (mass spectrometry).
[34] A. S. Romakhin, E. V. Nikitin, O. V. Parakin, Y. A. Ignat’ev,
B. S. Mironov, Y. M. Kargin, J. Gen. Chem. U.S.S.R. 1987, 2298.
[35] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877. doi:10.1021/
CR940053X
References
[1] R. T. Boere´, A. M. Bond, S. Cronin, N. W. Duffy, P. Hazendonk,
J. D. Masuda, K. Pollard, T. L. Roemmele, P. Tran, Y. Zhang, New
J. Chem. 2008, 32, 214. doi:10.1039/B709602J
[36] D. M. Murphy, in Metal Oxide Catalysis (Eds S. D. Jackson,
J. S. J. Hargreaves) 2009, Vol. 1, pp. 1–50 (Wiley-VCH: Weinheim).

Dipp3As and its Radical Cation
I
[37] J. A. Weil, J. R. Bolton, J. E. Wertz, Electron Paramagnetic Reso-
nance: Elementary Theory and Practical Applications, 2nd edn 1994
(John Wiley & Sons: New York, NY).
[43] C. A. Coulson, in Volume Comme´moratif Victor Henri, Contributions
a` l’Etude de la Structure Mole´culaire 1948, pp. 12–32 (Desoer: Lie`ge).
[44] R. R. Schrock, M. Wesolek, A. H. Liu, K. C. Wallace, J. C. Dewan,
Inorg. Chem. 1988, 27, 2050. doi:10.1021/IC00285A012
[45] APEX2, SAINT-Plus and SADABS 2008 (Bruker AXS Inc.: Madison,
WI).
[38] G. W. Eastland, M. C. R. Symons, J. Chem. Soc. Perkin 1977, II, 833.
[39] E. Furimsky, J. A. Howard, J. R. Morton, J. Am. Chem. Soc. 1972, 94,
5932. doi:10.1021/JA00771A088
[40] E. Furimsky, J. A. Howard, J. R. Morton, J. Am. Chem. Soc. 1973, 95,
[46] G. M. Sheldrick, Acta Crystallogr. 2008, A64, 112.
[47] C. F. Macrae, I. J. Bruno, J. A. Chisholm, P. R. Edgington, P. McCabe,
E. Pidcock, L. Rodriguez-Monge, R. Taylor, J. van de Streek, P. A.
Wood, J. Appl. Cryst. 2008, 41, 466. doi:10.1107/
S0021889807067908
6574. doi:10.1021/JA00801A009
[41] A. V. Il’yasov, Y. M. Kargin, E. V. Nikitin, A. A. Vafina, G. V.
Romanov, O. V. Parakin, A. A. Kazakova, A. N. Pudovik, Phosphorus
Sulfur 1980, 8, 259. doi:10.1080/03086648008078199
[42] J. R. Morton, K. F. Preston, J. Magn. Reson. 1978, 30, 577.