Synthesis, characterization, and electrochemical studies of PPh 3-n(dipp)n (dipp = 2,6-diisopropylphenyl): Steric and electronic effects on the chemical and electrochemical oxidation of a homologous series of triarylphosphines and the reactivities of the corresponding phosphoniumyl radical cations

Article abstract of DOI:10.1021/ja403555d

Activation barriers to the electrochemical oxidation for the series PPh_3-n(dipp)_n (dipp = 2,6-diisopropylphenyl) in CH ₂Cl₂/Bu₄NPF₆ were measured using large amplitude FT ac voltammetry. Increasing substitution across this series, which offers the widest range of steric requirements across any analogous series of triarylphosphines reported to date, increases the energetic barrier to electron transfer; values of 18, 24, and 25 kJ mol^-1 were found for compounds with n = 1, 2, and 3, respectively. These values are significantly greater than those calculated for outer sphere activation barriers, with deviations between observed and calculated values increasing with the number of dipp ligands. This suggests that the steric congestion afforded by these bulky substituents imposes significant reorganizational energy on the electron transfer processes. This is the first investigation of the effect of sterics on the kinetics of heterogeneous electron transfer across a structurally homologous series. Increased alkyl substitution across the series also increases the chemical reversibility of the oxidations and decreases the oxidation peak potentials. As the compounds for which n = 1 and 2 are novel, the synthetic strategies employed in their preparation are described, along with their full spectroscopic, physical, and crystallographic characterization. Optimal synthesis when n = 1 is via a Grignard reagent, whereas when n = 2 an aryl copper reagent must be employed, as use of a Grignard results in reductive coupling. Chemical oxidation studies were performed to augment the electrochemical work; the O, S, and Se oxidation products for the parent triarylphosphines for which n = 1 and 2 were isolated and characterized.

Full text of DOI:10.1021/ja403555d

Article
pubs.acs.org/JACS
Synthesis, Characterization, and Electrochemical Studies of
PPh3−n(dipp)n (dipp = 2,6-Diisopropylphenyl): Steric and Electronic
Effects on the Chemical and Electrochemical Oxidation of a
Homologous Series of Triarylphosphines and the Reactivities of the
Corresponding Phosphoniumyl Radical Cations
John P. Bullock,*,† Alan M. Bond,*,‡ Rene T. Boere,*,§ Twyla M. Gietz,§ Tracey L. Roemmele,§
́
́
Sonja D. Seagrave,§ Jason D. Masuda,∥ and Masood Parvez⊥
†Division of Natural Sciences & Mathematics, Bennington College, Bennington, Vermont 05201, United States
‡School of Chemistry, Monash University, Clayton, Victoria 3800, Australia
§Department of Chemistry and Biochemistry, University of Lethbridge, Lethbridge, AB, Canada T1K3M4
∥The Maritimes Centre for Green Chemistry and the Department of Chemistry, Saint Mary’s University, Halifax, Nova Scotia,
Canada B3H 3C3
⊥Department of Chemistry, University of Calgary, Calgary, AB, Canada T2N1N4
S
* Supporting Information
ABSTRACT: Activation barriers to the electrochemical
oxidation for the series PPh3−n(dipp)n (dipp = 2,6-
diisopropylphenyl) in CH2Cl2/Bu4NPF6 were measured
using large amplitude FT ac voltammetry. Increasing
substitution across this series, which offers the widest range
of steric requirements across any analogous series of
triarylphosphines reported to date, increases the energetic
barrier to electron transfer; values of 18, 24, and 25 kJ mol−1
were found for compounds with n = 1, 2, and 3, respectively.
These values are significantly greater than those calculated for
outer sphere activation barriers, with deviations between observed and calculated values increasing with the number of dipp
ligands. This suggests that the steric congestion afforded by these bulky substituents imposes significant reorganizational energy
on the electron transfer processes. This is the first investigation of the effect of sterics on the kinetics of heterogeneous electron
transfer across a structurally homologous series. Increased alkyl substitution across the series also increases the chemical
reversibility of the oxidations and decreases the oxidation peak potentials. As the compounds for which n = 1 and 2 are novel, the
synthetic strategies employed in their preparation are described, along with their full spectroscopic, physical, and crystallographic
characterization. Optimal synthesis when n = 1 is via a Grignard reagent, whereas when n = 2 an aryl copper reagent must be
employed, as use of a Grignard results in reductive coupling. Chemical oxidation studies were performed to augment the
electrochemical work; the O, S, and Se oxidation products for the parent triarylphosphines for which n = 1 and 2 were isolated
and characterized.
interactions between donor and acceptor.2 There is, however, a
dearth of systematic studies concerning the influence of sterics
on heterogeneous electron transfer processes. As part of our
continuing study of sterically hindered phosphines, we
undertook a study on the electrochemistry of PPh3−n(dipp)n
(Chart 1, dipp = 2,6-diisopropylphenyl); given the relative sizes
of the Ph and dipp ligands, the title homologous series spans
INTRODUCTION
■
The influence of sterics on chemical reactions has long been
recognized. The kinetics and thermodynamics of organic1a and
inorganic1b substitution, Lewis acid/base,1c and radical reaction
pathways1d can all be strongly affected by the size of structural
features of either the substrate or an attacking species. Redox
reactions can likewise be influenced by sterics. Since any
the entire known steric range of triarylphosphines. Using a
homogeneous electron transfer reaction requires appropriate
powerful new implementation of large amplitude ac voltam-
orbital overlap between the donor and acceptor species,
metry,3 we measured the activation energy of the electrode
features that diminish such overlap will reduce the correspond-
ing rate constants. Moreover, the operative mechanism for any
particular redox reaction, i.e., the tendency to proceed via an
inner or an outer sphere process, can depend on the steric
Received: April 20, 2013
Published: June 28, 2013
© 2013 American Chemical Society
11205
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
routes: application of a standard Grignard method
(dippMgBr/THF) and use of the aryl copper reagent,
dipp4Cu4.9 Use of the Grignard was the optimal approach for
making 1 (Scheme 1A), as employment of dipp4Cu4 led
Chart 1
Scheme 1. Synthetic Routes to 1 (A) and 2 (B)
processes for complexes 1, 2, and 3 and found that the presence
of the ortho isopropyl substituents systematically increases the
activation barrier to oxidation. This represents, we believe, the
first such systematic study that demonstrates the effect of steric
crowding on the electrode processes across a series of
structurally analogous compounds.
The unexpected effect of sterics on the heterogeneous
electron transfer kinetics is not only interesting in its own right,
it is also of obvious relevance to the burgeoning field of the
steric modulation of reactions. Specifically, the use of bulky
substituents is now recognized as essential for the stabilization
and isolation of new modalities at the frontiers of chemistry,
especially main-group element chemistry. Numerous examples
can be cited: heavy element multiple bonds,4a−d bulky N-
heterocyclic carbenes,4e−j isolation of low-coordinate molecules
and radicals,4k−m heavy group 14 R3E free radicals,4n and the
first stable, isolable phosphanyl radical.4o Most recently, Pan
and co-workers successfully used a combination of the bulky
2,4,6-triisopropylphenyl (tripp) group and superbulky weakly
coordinating anions to crystallographically characterize P-
(tripp)3+• and [tripp2PPtripp2]+•, the first time phosphoniumyl
radical cations have been so examined.5 Protasiewicz and co-
workers have drawn attention to unexpected consequences of
sterically demanding substituents in phosphorus chemistry,6
showing, inter alia, that the lengths of PP bonds may be
affected by crystal disorder induced by the packing demands of
the bulky substituents.6a Finally, the work we describe herein is
also relevant to recent efforts in the area of frustrated Lewis
pairs, as the promising catalytic chemistry such systems exhibit
often involves oxidation of an exceptionally hindered moiety.7
The effects we describe may also be operative in such systems,
thereby influencing the kinetics of the corresponding catalytic
redox processes. We attribute the steric enhancement of the
oxidative activation barrier to an increase in the reorganiza-
tional energy of the compounds associated with the electron
transfer; the more enmeshed the alkyl groups are in the parent
compound, the more energetically demanding is the reorgan-
ization.
exclusively to the copper(I) chloride adduct 4 (δ(31P) = −20.9
ppm, br s; see Supporting Information for additional
characterization details and crystal structure description). This
compound, in turn, yielded 1 via ammoniacal workup, albeit in
low yield.
In contrast to the above, reaction of PPhCl2 with 2 equiv of
Grignard reagent did not yield 2 but, instead, the corresponding
reductive coupling product, Ph(dipp)P-PPh(dipp), 5, as the
major product (Scheme 1B); similar coupling from the
reduction of bulky Ar2PCl species by organometallic reagents
has previously been reported.10 While 31P NMR of 5 exhibits a
singlet at −30.55 ppm, an attempted recrystallization from
ethanol in air resulted in isolation of a new product exhibiting
two doublets (δ = +39.43, −36.13 ppm, 1JPP = 279 Hz), which
we ascribe to the monoxide Ph(dipp)P(O)PPh(dipp), 6. We
found that using the versatile dipp4Cu4 prevented the undesired
reductive coupling and successfully yielded 2, although certain
aspects of the mechanism remain unclear. Specifically, we have
previously shown that treatment of PCl3 with dipp4Cu4
provides excellent yields of P(dipp)Cl2,9 but when dipp4Cu4
reacts with PhPCl2, the major product, as determined by 31P
NMR studies of reaction aliquots, is [ClCuPPh(dipp)Cl]2 (δ =
75.4 ppm, br s), 8. Evidently the formation of 8 does not
prevent the further substitution of the last chlorine on the
phosphorus, but it is not clear whether that reaction happens
while the chlorophosphine is coordinated to Cu(I) or whether
reaction requires decomplexation, such that 8 acts only as a
reservoir for the intermediate ClPPh(dipp). Furthermore, the
final product of the aryl transfer reaction is another copper(I)
As two of the compounds in the series, 1 and 2, are novel, we
also describe in this article the specific synthetic strategies
employed in their preparation, as well as details of their physical
and spectroscopic characterization, especially as they inform the
interpretation of the kinetic studies. In addition, we performed
a series of DFT calculations of the parent species and their
corresponding radical cations to further shed light onto the
structural changes that take place upon oxidation.
EXPERIMENTAL SECTION
Full descriptions of all synthetic and electrochemical procedures, as
well as details concerning instrumentation and digital simulation
techniques, are provided in the Supporting Information.
■
RESULTS AND DISCUSSION
■
Synthesis and Characterization of PPh2(dipp), 1, and
PPh(dipp)2, 2. Based on our previous experience with the
synthesis of 3,8 we attempted the syntheses 1 and 2 from the
corresponding chlorophenylphosphines, PPh3−nCln, by two
11206
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
Table 1. Physical Properties and Characterization Data for PPh3, 1, 2, 3, 10a−c, and 11a−c and DFT Structural Results for
a
PPh3+•, 1+•, 2+•, and 3+•
base phosphine
PPh3
1
2
3
b
g
q
d(P−C)av,
Å
1.833(3)
1.837(13)
0.739
1.854(14)
0.651
1.8507(16)
b
g
q
P oop of 3 Cipso’s,
Å
0.790
0.539
b
g
q
P oop of C≠ispo
,
Å
0.071
0.04−0.025
314.7
0.15−0.33
324.6
0.430
b
g
q
∑∠(CPC), deg
307.2
335.6
b
h
q
Stokes shift, kJ·mol−1
201
183
146
129
b
ij
−4.7 ,
q
δ(31P)
−20.9
+32.7
−28.4
+31.2
−49.7
c
k
δ(31P), E = O
+29.6, +30.4
d
k
δ(31P), E = S
+41.7, +44.5
+35.9, +36.5 ,
+34.8
+31.2
e
ik
δ(31P), E = Se
+25.2
+19.2
e
i
J(31P−77Se), Hz
732
734
690
c
l
ν(P→O), cm−1
1190/1189
1174/1175
610/610
560/559
322.4
1162/1162
603/603
547/545
328.0
d
k
ν(P→S), cm−1
631
e
k
ν(P→Se), cm−1
558
c
m
∑∠(CPC)E=O, deg
∑∠(CPC)E=S, deg
∑∠(CPC)E=Se, deg
( P), mT
∑∠(CPC)R3P+•, deg DFT
∑∠(CPC)R3P, deg DFT
Δ{∑∠(CPC)},
319.3
d
n
317.6
315.9
325.3
e
o
318.2
316.3
326.4
f
p
q
31
a
29.9
26.8
23.9
iso
f
r
342.0
307.6
34.4
346.6
317.2
29.4
354.7
358.9
b
326.0
336.6
22.3
b,c,d,e,f
deg DFT
28.7
a
The source data for this table is located in the Supporting Information except where indicated. DFT performed at (r/u)b3lyp/6-31G(d,p) levels.
b
c
d
e
f
g
PR3. OPR3. SPR3. SePR3. PR3+•. Bruckmann, J.; Kruger, C.; Lutz, F. Z. Naturforsch., B: Chem.Sci. 1995, 50, 351. Kooijman, H.; Spek, A. L.;
h
Bommel, K. J. C. van; Verboom, W.; Reinhoudt, D. N. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1998, 54, 1695. Changenet, P.; Plaza, P.;
Martin, M. M.; Meyer, J. Y.; Rettig, W. Chem. Phys. 1997, 221, 311. Allen, D. W.; Taylor, B. F. J. Chem. Soc., Dalton Trans., 1982, 51. Schraml, J.;
Capka, M.; Blechta, V. Magn. Reson. Chem. 1992, 30, 544. Chutia, P.; Kumari, N.; Sharma, M.; Woollins, J. D.; Slawin, A. M. Z.; Dutta, D. K.
Polyhedron 2003, 23, 1657. NIST Webbook: http://webbook.nist.gov/cgi/cbook.cgi?ID=C791286&Type=IR-SPEC&Index=1#IR-SPEC. Brock,
i
j
k
̌
l
m
n
C. P.; Schweizer, W. B.; Dunitz, J. D. J. Am. Chem. Soc. 1985, 107, 6964. C. R.de Arellano, G. Asensio, M. Medio-Simon, A. Cuenca, G. Mancha,
private communication, 2007 (Cambridge Structural Database refcode: TPPOSS07). Schulz, T.; Meindl, K.; Leusser, D.; Stern, D.; Graf, J.;
o
Michaelsen, C.; Ruf, M.; Sheldrick, G. M.; Stalke, D. J. Appl. Crystallogr. 2009, 42, 885. Jones, P. G.; Kienitz, C.; Thone, C. Z. Kristallogr. 1994, 209,
p
80. A. L. Rheingold, private communication, 2011 (Cambridge Structural Database refcode: TPPHSE02). Il’yasov, A. V.; Kargin, Yu. M.; Nikitin, E.
V.; Vafina, A. A.; Romanov, G. V.; Parakin, O. V.; Kazakova, A. A.; Pudovik, A. N. Phosphorus Sulfur 1980, 8, 259. Il’yasov, A. V.; Kargin Yu. M.;
q
Vafina, A. A. Russ. J. Gen. Chem., 1993, 63,1833. Boere,
́
R. T.; Bond, A. M.; Cronin, S.; Duffy, N. W.; Hazendonk, P.; Masuda, J. D.; Pollard, K.;
r
Roemmele, T. L.; Tran, P.; Zhang, Y. New J. Chem. 2008, 32, 214. 360.0° reported by Pan, X.; Chen, X.; Li, T.; Li, Y.; Wang, X. J. Am. Chem. Soc.,
2013, 135, 3414.
complex, 9, which we identified spectroscopically. Fortunately
9, unlike 4, affords free 2 in moderate yields upon treatment
with deoxygenated aqueous ammonia followed by extraction
into xylenes.
ground state is highly pyramidal.8,11,12 Stokes shifts of 1 and 2
(Table 1) are intermediate between those of PPh3 and 3.
Similarly, 31P NMR chemical shifts also change stepwise
between those of PPh3 and 3. However, each of these measures
shows a different degree of change with each added substituent,
indicating that subtle structure−property effects are operative
within the series.
Chemical Oxidation: Phosphine Chalcogenides 10a−c
and 11a−c. In the syntheses of 1 and 2, these compounds
exhibited sensitivity toward atmospheric oxygen that is higher
than either PPh3 or 3. In our previous work on 3, we
demonstrated its oxidation is induced strictly by electron
transfer processes because the steric shielding by the exo
isopropyl substituents blocks access of atom transfer reactions
to the phosphorus atom.8 PPh3, in contrast, is easily accessible
to oxidants but, as we discuss below, has a substantially higher
oxidation potential. Thus 1 and 2 have lower steric protection
than 3 and a more accessible oxidation potential than PPh3. To
further investigate this behavior, we systematically examined the
chemical oxidation of 1 and 2 with oxygen, sulfur, and
selenium. Both phosphines form thermally stable chalcogenides
with oxygen (using H2O2 in acetone), yielding PPh2(dipp)O,
10a, and PPh(dipp)2O, 11a, as well as sulfur and selenium (by
refluxing with the elements in xylenes), yielding PPh2(dipp)S,
We characterized 1 and 2, as well as their chalcogenide
adducts (see below), by a variety of spectroscopic and physical
methods (Table 1 and Supporting Information) and
determined their structures in the solid state using single-
crystal X-ray diffraction (Figure 1). Of particular importance is
the degree of steric pressure and steric protection experienced
in 1, 2, 3, and PPh3. Their respective ∑∠(C−P−C) values
show a systematic increase with number of dipp groups on the
central phosphorus. As we have shown previously, the sums of
angles around the P atom is a key parameter; substituents with
sufficient steric bulk cause flattening at the central atom which
raises the HOMO, thereby facilitating oxidation (vide infra).8
Various spectroscopic data indicate that the structural trends
described above for the solid PPh3 and 1−3 are also present in
solution. For example, it has been shown that photoexcitation
of phosphines involving promotion of a HOMO electron to an
excited state induces a change to planar or close-to-planar
geometries. Hence, the Stokes shift can be used to assess the
effective pyramidalization of triarylphosphines in solution; large
shifts, such as those reported for PPh3, are observed when the
11207
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
Figure 1. Thermal displacement plots (30% probability) of (a) 1 and
(b) one of the independent molecules of 2 as determined in their
crystal lattices. A symmetry code (−x, 1 − y, 1 − z) was applied to the
atom coordinates of 2 to emphasize the similarity in conformations of
the two phosphines. For full atom numbers and the second molecule
of 2 see the Supporting Information.
Figure 2. Thermal displacement plots (30% probability) of the
structures of (a) 10a, (b) 10b, (c) 10c (isostructural with 10b), (d)
11a, (e) 11b, and (f) one of the independent molecules of 11c, as
determined in their crystal lattices. Symmetry codes were applied to
the atom coordinates of 11b (1 − x, 1 − y, 1 − z) and 11c (1/2 + x,
−y, 1/2 + z) to emphasize the similarity in conformations among
related phosphine chalcogenides. For full atom numbers and the
second molecule of 11c see the Supporting Information.
10b, PPh2(dipp)Se, 10c, PPh(dipp)2S, 11b, and PPh(dipp)2Se,
11c. We fully characterized all six derivatives and determined
their structures by single-crystal X-ray crystallography (Figure
2). The solid-state structures show an increase in the steric
flattening parameter ∑∠(CPC) upon oxidation, especially for
the oxygen adducts. This is consistent with the formulation of
R3P+→E− for the bonding in the chalcogenides, which is
strongly indicated in such oxides by solid state 17O NMR
spectroscopy.13
Room Temperature Electrochemical Studies. Cyclic
voltammograms (19
1 °C) of the oxidation of PPh3 and
compounds 1−3 in CH2Cl2 (0.5 M Bu4NPF6) at scan rates of
250 mV s−1 are presented in Figure 3. The results for PPh3 are
consistent with previously published accounts which indicated
that the compound undergoes a diffusion-controlled but
chemically irreversible one electron oxidation in organic
solvents (eq 1).14,15
PPh3 → PPh3+· + e−
(1)
There is ample evidence that the electrogenerated phospho-
niumyl radical cation, PPh3+•, reacts rapidly with adventitious
water in the solvent/electrolyte medium to ultimately yield the
corresponding oxide, OPPh3, and the hydrophosphonium
Figure 3. Cyclic voltammograms of compounds PPh3−n(dipp)n in
CH2Cl2 (0.5 M Bu4NPF6) at a 3 mm diameter GC electrode. The scan
rate employed in all traces was 250 mV s−1; analyte concentrations are
between 0.8 and 1.0 mM. Inset: Oxidation peak potential as a function
of the number of dipp ligands; the best linear fit (red line) has a slope
of −190 mV per dipp substituent; peak potential measurements were
reproducible to within 5 mV.
+
cation, HPPh3 . If the concentration of water is high enough,
the overall stoichiochiometry of the oxidation is described by eq
2, although published reports indicate that the product
proportions will depend on the amount of water present in
the reaction medium.
addition, they clearly illustrate that regular substitution of
phenyl ligands with the sterically demanding dipp ligands
results in an inductive effect, which shifts the oxidation peaks to
lower potentials, and also bestows greater kinetic stability to the
corresponding radical cations, as evidenced by enhanced
3PPh3 + H2O → ΟPPh3 + 2HPPh3+ + 2e−
(2)
The voltammograms in Figure 3 are consistent with one-
electron oxidations of each of the triarylphosphines. In
11208
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
reversibility of the oxidation processes. These effects are
summarized in Table 2 and are discussed in greater detail
below.
for chemically reversible couples. Moreover, isopropyl groups
in the ortho positions have both electronic and structural
effects. Specifically, as the phosphorus becomes more
congested, it is forced to assume a more planar geometry
(∑∠(CPC) = 314.7°, 324.6°, and 335.6° for 1, 2, and 3,
respectively), resulting in a HOMO that becomes more 3p in
character and higher in energy.16,21 This decrease of
pyramidalization is manifested in a correspondingly lower
oxidation peak potential.
Table 2. Summary of Room Temperature Electrochemical
Data for the Oxidation of PPh3, P(mes)3, and Compounds
a
1−3 in CH2Cl2
b
b
c
compound
Ep
E1/2
ir/if
k°
PPh3
2.078
1.579
1.926
1.748
1.487
0
0.10
Electrochemical Mechanistic Studies. While there have
been numerous reports on the electrochemical oxidation of
PPh3,14,15,22 there is little rate information available on the rapid
reactions coupled to the electrode process. Given the utility of
such kinetic data to serve as a point of reference for the
corresponding reaction rates of radical cations, 1+•, 2+•, and 3+•,
we undertook a scan rate study of PPh3 in CH2Cl2 (Figure 4).
P(mes)3
1.547
0.88
0
0.063
0.081
0.031
0.067
1
2
3
1.710
1.449
0.77
0.97
a
0.5 M Bu4NPF6 electrolyte; GC working electrode; scan rate = 250
b
mV s−1.. Oxidation peak potentials, Ep, and half-wave potentials, E1/2
the average of the oxidation and reduction peak potentials for the
,
c
reversible couple, are reported in V vs the E1/2 of Cc+/0
. Room
temperature value of k°, in cm s−1; estimated by optimizing digital
simulations of ac voltammetric data.
The effect of steric crowding about the phosphorus center on
the stabilities of triaryl phosphoniumyl cations has been the
subject of previous work.8,10,16−19 For example, the series
PPh3−n(mes)n16 exhibits similar trends to those observed in this
work, with the oxidation of P(mes)3 showing the greatest
chemical reversibility. While oxidation of P(mes)3 in our hands
showed moderate chemical reversibility in dry, deoxygenated
CH2Cl2 (ir/if = 0.88 at 250 mV s−1, where ir/if is the ratio of the
return to forward peak currents), the inertness of 3+• afforded
by the bulky isopropyl substituents makes the oxidation of the
parent compound, 3, even more chemically reversible (ir/if =
0.97 at 250 mV s−1). Similarly, the oxidation of 2 is coupled to a
much larger cathodic return (ir/if = 0.77 at 250 mV s−1) than
P(mes)2Ph (ir/if = 0.17 at 5 V s−1).16 The relative inertness of
2+• made it possible to characterize this species using EPR
spectroelectrochemistry (Table 1 and Figure S11 in the
Supporting Information). Interestingly, the previously reported
Figure 4. Dc voltammetric scan rate study of the oxidation of PPh3
(2.6 mM) at a 3 mm glassy carbon electrode in CH2Cl2 (0.5 M
Bu4NPF6); scan rates employed were 0.1 (orange), 0.5 (green), 2
(blue), 4 (magenta), 8 (red), and 12 V s−1 (black). Potentials are
relative to the Cc+/0 couple.
The voltammograms exhibited very little cathodic current
coupled to the bulk oxidation at lower scan rates (ν < 2 V s−1).
Some return current is observed at higher scan rates, but the
peak shapes are inconsistent with a simple EC mechanism.
Rather, the elongation suggests that the first chemical step
involves an equilibrium process that allows for the chemical
regeneration of the reducible electrogenerated species during
the reverse sweep.
To estimate the kinetic parameters of the coupled
homogeneous reactions, we fit these voltammograms to the
mechanism outlined in Scheme 2. This scheme, which is similar
to that proposed for the attack of water on P(mes)3+• proposed
by Gronchi and co-workers,19 involves the radical abstraction of
hydrogen from water by the phosphoniumyl cation, followed by
attack of the resulting hydroxyl radical on a second electro-
generated phosphoniumyl cation. The final triarylphosphonium
oxide is generated via rapid deprotonation by water; we did not
consider a scheme involving deprotonation by the bulk
phosphine, as previous work demonstrated that this process
is slow on the cyclic voltammetry time scale.22 Water
concentrations from 0.5 to 2 mM were considered in the
simulations, reasonable values given the routine steps taken in
drying the solvent/electrolyte.
+•
reactivity of P(mes)3 toward oxygen, which results in a
virtually complete loss of peak reversibility and current
attenuation by a factor of 0.5,19 is shared by 2+•, but not 3+•.
This observation, along with the peak current ratios mentioned
above, implies that the accessibility of the phosphorus to
+•
chemical attack in 2+• and P(mes)3 is similar, i.e., the degree
of steric protection afforded by three mesityl groups is
comparable to that by the combination of one phenyl and
two dipp ligands. This is consistent with the chemical oxidation
studies reported above resulting in both 10a and 11a; similarly
P(mes)3 can form a stable oxide.20 Finally, like PPh3,
electrooxidation of 1 is irreversible; apparently the steric
protection afforded to the central phosphorus by a single dipp
ligand is insufficient to protect it from rapid attack by
adventitious water.
While the reversibility of the oxidation processes of
PPh3−n(dipp)n species makes qualitative comparisons of the
relative stabilities of the corresponding radical cations
straightforward, interpretation of the peak potentials is less
so. The oxidation peak potentials, Ep, show an average shift of
−190 mV per dipp ligand (Figure 3, inset), but exhibit a small
deviation from linearity that we attribute to a number of effects.
Specifically, the rapid homogeneous chemical processes
coupled to the electron transfer for PPh3 and 1 decrease the
corresponding peak potentials from what would be observed
The observed peak shapes could be satisfactorily simulated
by employing a relatively wide range of rate constants, with k1,f
values ranging from about 1 × 106 M−1 s−1 up to the diffusion
11209
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
Scheme 2. Generation and Reactivity of
Triarylphosphoniumyl Cations (Ar = dipp or Ph)
Figure 5. Experimental (colored) and digital simulations (gray) of the
second (black), third (red), fourth (violet), fifth (blue), and sixth
(green) harmonics of room temperature ac voltammograms of the
oxidation of 1.0 mM PPh3 in CH2Cl2. Experimental parameters: f =
14.16 Hz, ΔE = 100 mV, ν = 59.60 mV s−1, T = 19 °C, A = 0.047 cm2;
the switch time for the potential sweep direction is at 13.4 s.
Simulation parameters: Ru = 200 Ω, k° = 0.1 cm s−1, α = 0.5, diffusion
coefficients for all species except water, D = 1.5 × 10−5 cm2 s−1, DH O
=
2
2.2 × 10−5 cm2 s−1, CDL constants: Ec = 1.6 V, c0 = 1.8 × 10−4 F cm−2,
c1 = 1.8 × 10−5 F cm−2 V−1, c2 = −1.4 × 10−5 F cm−2 V−2.
We discuss these results briefly below. As previously
mentioned, substitution of one phenyl ligand with a dipp
ligand does not appreciably affect the reversibility of the bulk
oxidation. The optimized rate constants for the reaction of 1+•
with residual water are slightly lower than those for PPh3, but
appear to be within an order of magnitude. The k° values are
also similar. In contrast, the corresponding rate constant for 2+•
decreases by more than 3 orders of magnitude compared to 1+•.
While we used the same operative mechanism to fit the
voltammogram of 2, it is worth noting that equally good fits
could be obtained by a simple EC mechanism owing to the
slow rates of the reactions following the initial attack by water.
Furthermore, the heterogeneous rate constant for the oxidation
of 2 is somewhat lower than those of the other compounds in
the series, a characteristic that has a significant effect on its low
temperature ac voltammetric response, as discussed more fully
below. Finally, 3+• is unreactive toward adventitious water on
the time scale of these experiments.8
Effect of Temperature on Dc and Ac Voltammetry.
Previous studies of 3 in our laboratories indicated that this
compound, despite its chemically reversible oxidation, under-
goes electron transfer slowly enough for the anodic process to
be considered only quasi-reversible.8 It therefore falls in a
kinetic regime that makes fundamental studies concerning the
mechanism of its electron-transfer attractive using FT ac
voltammetry. The determination of electron-transfer activation
parameters using dc cyclic voltammetry is usually difficult
because the temperature changes that affect the electrode
kinetics also affect solution resistance and diffusion coefficients.
We have shown, however, that large amplitude ac techniques
are ideal for distinguishing and quantifying these effects.23,24b,c
In our previous work using this technique to measure the
activation parameters of electrode processes we focused on
highly chemically reversible couples of metallocenes;26 the
limit. The fitting is complicated by the large non-Faradaic
currents that result from use of high scan rates with the
macroelectrode. To get additional insight into the kinetics of
+•
the attack of water on PPh3 we performed large amplitude
Fourier-transformed ac voltammetry. This technique, which has
been extensively employed by one of our laboratories in the
investigation of a variety of electrode processes,3,23−26 can
provide reliable kinetic information on fast homogeneous
coupled reactions without the use of microelectrodes.25 By
simultaneously fitting the current signals of all of the accessible
harmonics of the forward and reverse sweeps of the
voltammograms (Figure 5) we estimated k1,f (3 × 106 M−1
s−1) more precisely than was possible from the dc voltammetric
results. Other optimized simulation parameters are provided in
Scheme 2 and the caption of Figure 5. Because several of the
kinetic and electron transfer parameters can compensate for
one another in the ac voltammetry simulations, we found that
simulations of similar quality could be obtained using k1,f values
between 1 × 106 and 1 × 107 M−1 s−1 and standard rate
constant, k°, values between 0.08 and 0.15 cm s−1. For
comparison, this places k° in a range comparable to that which
we found for the oxidation of ferrocene using similar
techniques.26
Large amplitude ac voltammetric experiments were also
performed on compounds 1−3 to measure their k° values and
the reaction rates of their respective radical cations at ambient
temperature. Fit optimization of the resulting voltammograms
(vide infra) yielded the kinetic parameters shown in Scheme 2.
11210
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
current work demonstrates the approach can be applied more
generally, that is to systems with various degrees of chemical
irreversibility.
To illustrate how the ac and dc techniques differ with respect
to their temperature responses, we present in Figure 6 a
over a roughly 40 °C range; for the sake of clarity, only the
fourth harmonic of each ac voltammogram is presented, but
others show similar effects as can be seen in Figures 7 and 8.
P(mes)3 is included in these analyses because its electrode
kinetics and activation energy are similar in some ways to that
of ferrocene and therefore serves as a useful comparison for the
other systems. What is particularly noteworthy of the
voltammograms in Figure 6 is the fact that, uniformly,
temperature has relatively small effects on the dc response
but shows a larger effect on the ac data, significantly so in some
cases. This sensitivity to temperature in the ac data enables the
estimation of the relevant standard rate constants as illustrated
below.
As temperature is lowered, the dc voltammograms of
P(mes)3 show only a slight decrease in current, owing to a
decrease in the diffusion coefficient, and a small shift in E1/2
value relative to the Cc0/+ internal standard. Little discernible
effect attributable to slower electron transfer kinetics or
increased solution resistance is observed. This stands in
contrast to the more significant changes seen in the ac
voltammograms due to increases in uncompensated resistance,
Ru and decreases in k° which result from the lowered
temperatures. Note that the two envelopes of peaks in the ac
voltammograms correspond to the forward and reverse sweeps
of the cycle; in the cases of P(mes)3 and 3 the two sets of peaks
are very similar in shape and size as would be expected for a
chemically reversible couple. By simultaneously fitting all of the
available harmonics obtained at each temperature (Figures 7
and 8), relatively precise estimates of Ru and k° values are
obtained. Moreover, because fitting the data requires diffusion
coefficients, we performed double-potential step chronocoul-
ometry experiments at each temperature and used Anson plots,
Q vs t1/2, to obtain those values.27 With respect to the
Figure 6. Dc (left) and ac (right columns, fourth harmonic only)
voltammograms obtained for CH2Cl2 solutions of, from top, P(mes)3
(0.9 mM), 3 (1.7 and 0.6 mM in dc and ac experiments, respectively),
2 (0.8 mM), and 1 (0.9 mM) at 19 1 °C (black), 1 1 °C (red),
and −19
1 °C (blue). Scan rates were 250 mV s−1 in all dc
experiments; in ac experiments scan rates between 60 and 75 mV s−1
were used with ΔE = 100 mV and f = 14 0.3 Hz. Potentials for the
ac data are relative to the E° of the analyte as estimated from iterative
fitting (see text).
comparison of voltammograms of the oxidation of, from top to
bottom, P(mes)3, 3, 2, and 1, obtained by the two methods
Figure 7. Experimental (colored) and digital simulations (gray) of the fundamental (black), second (red), third (violet), fourth (blue), fifth (green),
and sixth (orange) harmonics of ac voltammograms of the oxidation of 0.86 mM P(mes)3 in CH2Cl2 obtained at 19°, 1°, and −18 °C. Experimental
parameters: f = 13.78 Hz, ΔE = 100 mV, ν = 59.60 mV s−1, A = 0.050 cm2; the switch time for the potential sweep direction is at 13.4 s. Optimized
simulation parameters employed at each temperature are listed: (19 °C) Ru = 175 Ω, k° = 0.063 cm s−1, α = 0.48, D = 1.13 × 10−5 cm2 s−1; CDL
constants, Ec = 0.9 V, c0 = 1.26 × 10−4 F cm−2, c1 = 5.8 × 10−5 F cm−2 V−1, c2 = 2 × 10−6 F cm−2 V−2; (1 °C) Ru = 225 Ω, k° = 0.048 cm s−1, α =
0.46, D = 0.97 × 10−5 cm2 s−1; CDL constants, Ec = 0.7 V, c0 = 9.4 × 10−5 F cm−2, c1 = 5.0 × 10−5 F cm−2 V−1, c2 = 6 × 10−6 F cm−2 V−2; (−18 °C) Ru
= 300 Ω, k° = 0.031 cm s−1, α = 0.44, D = 0.68 × 10−5 cm2 s−1; CDL constants, Ec = 0.7 V, c0 = 7.0 × 10−5 F cm−2, c1 = 4.4 × 10−5 F cm−2 V−1, c2 = 9
× 10−6 F cm−2 V−2.
11211
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
Figure 8. Experimental (colored) and digital simulations (gray) of, from top to bottom, the fundamental through sixth harmonics of ac
voltammograms of the oxidation of (a) 0.62 mM 3, (b) of 0.80 mM 2, and (c) 0.90 mM 1, all in CH2Cl2 (0.5 M Bu4NPF6), at the temperatures
indicated. Specific experimental and optimization parameters: (a) f = 14.31 Hz, ΔE = 100 mV, ν = 67.06 mV s−1, electrode area = 0.057 cm2; switch
time for the potential sweep direction is 13.4 s. Optimized simulation parameters employed at each temperature: (18 °C) Ru = 200 Ω, k° = 0.076 cm
s−1, α = 0.50, DO,R = 1.35 × 10−5 cm2 s−1, CDL constants, Ec = 0.8 V, c0 = 9.5 × 10−5 F cm−2, c1 = 4.6 × 10−5 F cm−2 V−1, c2 = 6 × 10−6 F cm−2 V−2; (2
°C) Ru = 275 Ω, k° = 0.036 cm s−1, α = 0.49, DO,R = 1.02 × 10−5 cm2 s−1; CDL constants, Ec = 0.8 V, c0 = 7.8 × 10−5 F cm−2, c1 = 4.2 × 10−5 F cm−2
V−1, c2 = 1.2 × 10−5 F cm−2 V−2; (−18 °C) Ru = 290 Ω, k° = 0.0148 cm s−1, α = 0.485, DO,R = 0.75 × 10−5 cm2 s−1; CDL constants, Ec = 0.2 V, c0 =
3.6 × 10−5 F cm−2, c1 = 1.2 × 10−5 F cm−2 V−1, c2 = 1.2 × 10−6 F cm−2 V−2. (b) f = 14.04 Hz, ΔE = 100 mV, ν = 74.51 mV s−1, electrode area = 0.047
cm2; switch time for the potential sweep direction is 13.4 s. Optimized simulation parameters employed at each temperature: (18 °C) Ru = 180 Ω, k°
= 0.031 cm s−1, α = 0.50, DO,R = 1.50 × 10−5 cm2 s−1, [H2O] = 0.7 mM, DH O = 2.2 × 10−5 cm2 s−1, kC1,f = 120 M−1 s−1, kC1,r = 1 M−1 s−1, kC2 = 1 ×
2
104 M−1 s−1; CDL constants, Ec = 0.8 V, c0 = 1.17 × 10−4 F cm−2, c1 = 5.4 × 10−5 F cm−2 V−1, c2 = 0 F cm−2 V−2; (1 °C) Ru = 225 Ω, k° = 0.0175 cm
s−1, α = 0.47, DO,R = 1.15 × 10−5 cm2 s−1, [H2O] = 0.7 mM, DH O = 1.4 × 10−5 cm2 s−1, kC1,f = 15 M−1 s−1, kC1,r = 1 M−1 s−1, kC2 = 1 × 104 M−1s−1;
2
C
DL constants, Ec = 0.8 V, c0 = 1.00 × 10−4 F cm−2, c1 = 4.7 × 10−5 F cm−2 V−1, c2 = 0 F cm−2 V−2; (−18 °C) Ru = 650 Ω, k° = 0.0076 cm s−1, α =
0.47, DO,R = 0.75 × 10−5 cm2 s−1, [H2O] = 0.7 mM, DH O = 1.0 × 10−5 cm2 s−1, kC1,f = 0 M−1 s−1; CDL constants, Ec = 0.8 V, c0 = 9.1 × 10−5 F cm−2,
2
c1 = 4.3 × 10−5 F cm−2 V−1, c2 = 5.0 × 10−5 F cm−2 V−2. (c) f = 14.16 Hz, ΔE = 100 mV, ν = 59.60 mV s−1, electrode area = 0.045 cm2; switch time
for the potential sweep direction is 13.4 s. Optimized simulation parameters employed at each temperature: (19 °C) Ru = 180 Ω, k° = 0.081 cm s−1,
α = 0.50, DO,R = 1.2 × 10−5 cm2 s−1, [H2O] = 0.7 mM, DH O = 2.2 × 10−5 cm2 s−1; CDL constants, Ec = 1.0 V, c0 = 1.48 × 10−4 F cm−2, c1 = 4.3 × 10−5
2
F cm−2 V−1, c2 = −6 × 10−6 F cm−2 V−2; (1 °C) Ru = 220 Ω, k° = 0.054 cm s−1, α = 0.50, DO,R = 9 × 10−6 cm2 s−1, [H2O] = 0.7 mM, DH O = 1.4 ×
2
10−5 cm2 s−1; CDL constants, Ec = 1.0 V, c0 = 1.36 × 10−4 F cm−2, c1 = 4.5 × 10−5 F cm−2 V−1, c2 = −2 × 10−6 F cm−2 V−2; (−18 °C) Ru = 300 Ω, k°
= 0.028 cm s−1, α = 0.50, DO,R = 6.5 × 10−6 cm2 s−1, [H2O] = 0.7 mM, DH O = 1.0 × 10−5 cm2 s−1; CDL constants, Ec = 1.0 V, c0 = 1.28 × 10−4
F
2
cm−2, c1 = 3.8 × 10−5 F cm−2 V−1, c2 = −3 × 10−6 F cm−2 V−2.
uncertainties for the k° values, they are considerably smaller
than that for PPh3 owing to the lack of coupled chemistry and
the correspondingly smaller number of input parameters
involved in fitting the data. We estimate optimized values of
k° for P(mes)3, 2, and 3 are correct to within about 10%; those
for 1 are somewhat larger, about 25%.
11212
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
Given the relatively small effects of temperature on the dc
voltammograms in Figure 6, the range of effects on the ac data
is remarkable. Oxidation of 3 is similar to that of P(mes)3 in
terms of its chemical reversibility, but the current attenuation at
lower temperatures is more pronounced, reflecting a greater
temperature dependence of the heterogeneous kinetics. An
even more dramatic decrease in oxidative current is seen for 2.
In addition, the lower reaction rate of 2+• toward water at lower
temperatures largely compensates for the slower electrode
kinetics so that the current maxima of the return sweeps
observed at 18 and 1 °C are nearly equivalent. Finally, the
chemical irreversibility of the oxidation of 1 has the effect of
significantly distorting the peak shapes25 and diminishing all of
the currents compared to those of the other systems shown.
Given the relatively small differences in the fourth harmonic
responses for the oxidation of 1, it would be difficult, on the
basis of this harmonic alone, to distinguish the effect of
temperature on diffusion characteristics, solution resistance, and
electrode and chemical kinetics. Fortunately, since the ac
experiment yields numerous harmonics simultaneously, each of
which differs in its response to these various effects,24a,25,26 it is
possible to estimate these parameters with greater precision
than the data in Figure 6 would suggest.
εop,r, and εs,r, are the permittivity of free space, the relative
permittivity of the solvent at optical frequencies, and the
corresponding relative static permittivity, respectively. The
temperature dependence of the latter two terms has been
shown to be small and is neglected in the following analysis.30
The terms rA and re are the radius of the reactive species and
the imaging distance to the electrode, usually assumed to be
infinity.31 We employed the following constants28 in eq 4: for
dichloromethane, εop,r = 2.03, and εs,r = 9.0, and for P(mes)3, rA
= 0.54 nm;32 a theoretical value, ΔG⧧ = 12.2 kJ mol−1, is
os
thereby obtained.
ΔG‡ = ΔG‡ + ΔG‡
(3)
ET
is
os
⎛
⎝
⎞
⎠
Ne2
32πε0
1
1
1
1
εs,r
⎛
⎜
⎝
⎞
⎟
⎠
ΔG‡
=
−
−
⎜
⎜
⎟
⎟
os
r
r
ε
A
e
op,r
(4)
The close agreement between the experimentally determined
E
act value and the calculated value of ΔG⧧os for P(mes)3 stands
in contrast to the results we obtained for the dipp-substituted
compounds. Figure 8a−c shows the fundamental through sixth
harmonics of the Fourier transformed ac voltammograms for
compounds 3, 2, and 1, respectively, at temperatures similar to
those employed in the examination of P(mes)3. As mentioned
above, P(mes)3 and 3 show similar voltammetric characteristics
except that the ac currents of the latter are considerably smaller
at lower temperatures. Simulations of the data reveal that this
difference is due to a greater temperature sensitivity of k° for 3.
The resulting Arrhenius plot of the best fit k° values (Figure
9b) yields an activation barrier of 25 kJ mol−1. This compares
Determination of Electron Transfer Activation Param-
eters. We used the standard rate constants obtained by the fit
optimizations presented in Figure 7 for the oxidation of
P(mes)3 to estimate the activation energy, Eact, of the electrode
process at the glassy carbon electrode by means of an Arrhenius
plot (Figure 9a); the data show good linearity, and the resulting
to a calculated value of ΔG⧧ of only 11.4 kJ mol−1 (obtained
os
using an average radius of 0.58 nm8). The inner sphere
component of the barrier to electron transfer must therefore be
much higher for 3 than P(mes)3. We attribute this to the
greater steric congestion afforded by the ortho substituents of
the dipp ligand relative to those of mesityl. We return to this
point after discussing Arrhenius analyses of the oxidation of 2
and 1, below.
The k° values obtained for 2 and 1 yielded activation
energies of 24 and 18 kJ mol−1, respectively (Figures 9c and
9d). The corresponding calculated values of ΔG⧧ are only
os
11.8 and 13.0 kJ mol−1. As was the case with PPh3, the
estimates of k° values for 1 are less precise owing to the rapid
coupled chemistry. While the uncertainty warrants more
caution in the interpretation of this plot (Figure 9d), it appears
the presence of even just one dipp ligand introduces a
significant barrier to electron transfer beyond that expected for
a simple outer sphere process. Furthermore, some comment on
the unusually large current diminution for 2 compared to 3
warrants mention. While such a contrast in temperature
response may seem incongruous with the similar activation
barriers we find for these compounds, the k° values of 2 are in a
kinetic regime in which the higher harmonic currents are
particularly sensitive to changes in the heterogeneous rate
constants.26 Thus, what appears to be a “collapse” of the
current at lower temperatures, especially in the higher
harmonics in this data series, is not due to a larger activation
barrier but, instead, is a result of the consistently lower values of
k° across this temperature series relative to those of 3. We also
note that the dc voltammogram for 2 at −18° (Figure 6, third
row, left) is unique in that series as it is the only one in which
the value of k° becomes small enough to make the peak
Figure 9. Arrhenius plots and linear fits of the standard rate constants
for (a) P(mes)3, (b) 3, P(dipp)3, (c) 2, PPh(dipp)2, and (d) 1,
PPh2(dipp); error bars indicate the estimated uncertainties for the rate
constants.
slope yields an activation energy of 12 kJ mol−1. As explained
below, this value is very close to the outer sphere component of
the electron transfer activation energy according to Marcus−
Hush theory, suggesting that the inner sphere component, i.e.,
the energetic barrier to bond length and other structural
changes, is quite small. For comparison, the inner sphere
component for the oxidation of ferrocene is estimated to be
about 0.6 kJ mol−1.28
The activation energy for any electron transfer process is the
sum of the inner and outer sphere components (eq 3). The
outer sphere component can be calculated using eq 4,29 where
N is the Avogadro constant, e is the charge on the electron, ε0,
11213
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
separation noticeably larger than those obtained at higher
temperatures.
Digital simulations of the dc and ac voltammograms are
consistent with a radical mechanism involving abstraction of
hydrogen from water. In addition to the above effects, the
presence of dipp ligands also has a pronounced effect on the
activation barrier to electrooxidation, an effect we attribute to
the unique energetics associated with conformational changes
of the phosphines upon oxidation that result from the extreme
crowding from the ortho isopropyl substituents. The greater
reactivity of 1 and 2 compared to both PPh3 and 3 has been
confirmed through chemical synthesis of chalcogenides 10a−c
and 11a−c.
In contrast to P(mes)3, Eact values for the oxidation of 1, 2,
and 3 are all substantially larger than their respective outer
sphere components as calculated by eq 4. To aid in the
interpretation of these observations, we performed a series of
DFT calculations to determine the structural changes that
should take place upon oxidation of these compounds (Table
1). All of the compounds should “flatten” to some extent, a
process that could involve considerable rearrangement of the
intermeshing isopropyl substituents for the 30/+ couple, but
relatively little of the methyl groups in the less congested
0/+
ASSOCIATED CONTENT
* Supporting Information
P(mes)3 system. The flattening, expressed as the increase in
■
∑∠(CPC) upon oxidation, was calculated to be 38.5°, 28.7°,
and 22.2° for 1, 2, and 3, respectively (Table 1), and 23.9° for
P(mes)3+. The smaller degree of flattening calculated for
P(mes)3 and 3+• is a result of the steric congestion of the
parent compounds, i.e., these compounds start out flatter in
their reduced forms and so undergo less dramatic changes in
S
Crystallographic data in electronic (CIF) format. Synthetic
procedures and spectroscopic characterization (1H, 13C, and 31P
NMR, UV−vis, IR, Raman, fluorescence) of compounds 1, 2, 4,
9, 10a−c, and 11a−c. Details of the crystal structure
determinations with selected interatomic distances for 1, 2, 4,
10a−c, and 11a−c. Tables and assignments of vibrational
(FTIR and FT-Raman), electronic, and fluorescence peaks.
Spectroelectrochemical EPR characterization of 2+•. Minimized
geometries determined in the DFT calculations. This material is
available free of charge via the Internet at http://pubs.acs.org.
pyramidalization when oxidized. The very small ΔG⧧ for the
is
oxidation of P(mes)3 that our results imply indicates that ortho
methyl groups on adjacent aryl rings undergo a minimum of
steric intermeshing upon flattening of the molecule. The
situation is dramatically different for triarylphosphines with
bulky ortho substituents. Our activation data indicate that
ΔG⧧is is actually lowest for 1 and increases with increased dipp
substitution, despite the calculated decrease of flattening that
occurs upon oxidation across the series. A reasonable
explanation for this unexpected trend is that the greatest
contribution to inner sphere reorganization is not the
magnitude of flattening per se, but rather the intermeshing of
the ortho isopropyl groups with neighboring substituents; the
size of this effect for the redox couples is 30/+ > 20/+ > 10/+
(space filling views of the structures are provided in Figure S3
in the Supporting Information to illustrate these interactions).
Hence the apparent similarity in the total steric crowding of the
central phosphorus atoms of 2 and P(mes)3, reflected by the
aforementioned similar ir/if ratios of the dc voltammograms,
does not result in similar activation profiles because the total
steric congestion does not reflect regions of locally greater
steric interactions that can inhibit structural reorganization and,
thereby, affect the kinetics of the electron transfer.
AUTHOR INFORMATION
Corresponding Author
jbullock@bennington.edu; boere@uleth.ca, alan.bond@
monash.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
R.T.B. thanks the Natural Sciences and Engineering Research
Council of Canada, the University of Lethbridge, and the
Canada Foundation for Innovation for operating and equip-
ment funding. Dr. M. Gerken is thanked for the use of his FT-
Raman instrument. The financial support of the Australian
Research Council is gratefully acknowledged.
REFERENCES
■
These results represent the first systematic investigation of
sterics on heterogeneous electron transfer across any
homologous series. While there is no reason to anticipate
that this is not a more general phenomenon, its relevance to the
chemistry of congested phosphines is particularly acute.
Unexpected consequences of sterically demanding substituents
in phosphorus chemistry have been recognized previously from
structural and reactivity studies.6 To the best of our knowledge,
our work is the first to detect such effects from activation
energies of redox reactions.
(1) The following are seminal papers concerning the role of sterics in
a range of reaction types: (a) Dostrovsky, I.; Hughes, E. D.; Ingold, C.
K. J. Chem. Soc. 1946, 173−194. (b) Basolo, F.; Chatt, J.; Gray, H. B.;
Pearson, R. G.; Shaw, B. L. J. Chem. Soc. 1961, 2207−2215. (c) Brown,
H. C.; Sclesinger, H. I.; Cardon, S. Z. J. Am. Chem. Soc. 1942, 64, 325−
329. (d) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1976, 9, 13−19.
(2) Hubig, S. M.; Rathore, R.; Kochi, J. K. J. Am. Chem. Soc. 1999,
121, 617−626.
(3) Bond, A. M.; Duffy, N. W.; Guo, S.-X.; Zhang, J.; Elton, D. M.
Anal. Chem. 2005, 77, 186A−195A.
(4) (a) Fischer, R. C.; Power, P. P. Chem. Rev. 2010, 110, 3877−
3923. (b) Power, P. P. Chem. Rev. 1999, 99, 3463−3503. (c) Li, L.;
Fukawa, T.; Matsuo, T.; Hashizume, D.; Fueno, H.; Tanaka, K.;
Tamao, K. Nat. Chem. 2012, 4, 361−365. (d) Power, P. P. Nat. Chem.
2012, 4, 343−344. (e) Wang, Y.; Quillian, B.; Wei, P.; Wannere, C. S.;
Xie, Y.; King, R. B.; Schaefer, H. F.; Schleyer, P. V.; Robinson, G. H. J.
Am. Chem. Soc. 2007, 129, 12412−12413. (f) Wang, Y. Z.; Xie, Y. M.;
Wei, P. R.; King, R. B.; Schaefer, H. F.; Schleyer, P. V.; Robinson, G.
H. Science 2008, 321, 1069−1071. (g) Wang, Y. Z.; Xie, Y. M.; Wei, P.
R.; King, R. B.; Schaefer, H. F.; Schleyer, P. V.; Robinson, G. H. J. Am.
Chem. Soc. 2008, 130, 14970−14971. (h) Kinjo, R.; Donnadieu, B.;
Celik, M. A.; Frenking, G.; Bertrand, G. Science 2011, 333, 610−613.
(i) Valente, C.; Calimsiz, S.; Hoi, K. H.; Mallik, D.; Sayah, M.; Organ,
CONCLUSION
■
In this work we describe the synthesis and characterization of a
series of triarylphosphines that spans the limits of the steric
possibilities of this class of compounds. All members of the
series undergo one-electron oxidation processes, the potentials
of which exhibit a pronounced induction effect. The kinetic
stabilities of the resulting radical cations depend on the steric
congestion about the central phosphorus; more congested
compounds are resistant to attack by adventitious water, while
those with more accessible phosphorus centers react rapidly.
11214
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215

Journal of the American Chemical Society
Article
M. G. Angew. Chem., Int. Ed. 2012, 51, 3314−3332. (j) Balogh, J.;
Slawin, A. M. Z.; Nolan, S. P. Organometallics 2012, 31, 3259−3263.
(k) Rekken, B. D.; Brown, T. M.; Fettinger, J. C.; Tuononen, H. M.;
Power, P. P. J. Am. Chem. Soc. 2012, 134, 6504−6507. (l) Summer-
scales, O. T.; Olmstead, M. M.; Power, P. P. Organometallics 2011, 30,
Zhang, J.; Guo, S.-X.; Elton, D. M.; Bond, A. M. Anal. Chim. Acta
2009, 652, 205−214. (h) Fleming, B. D.; Zhang, J.; Elton, D. M.;
Bond, A. M. Anal. Chem. 2007, 79, 6516−6526. (i) Lertanantawong,
B.; O’Mullane, A. P.; Zhang, J.; Surareungchai, W.; Somasudrum, M.;
Bond, A. M. Anal. Chem. 2008, 80, 6515−6525.
(25) Lee, C.-Y.; Bullock, J. P.; Kennedy, G. F.; Bond, A. M. J. Phys.
Chem. A 2010, 114, 10122−10134.
3468−3471. (m) Woodul, W. D.; Carter, E.; Muller, R.; Richards, A.
̈
F.; Stasch, A.; Kaupp, M.; Murphy, D. M.; Driess, M.; Jones, C. J. Am.
Chem. Soc. 2011, 133, 10074−10077;. (n) Sheberla, D.; Tumanskii, B.;
Bravo-Zhivotovskii, D.; Molev, G.; Molev, V.; Lee, V. Y.; Takanashi,
K.; Sekiguchi, A.; Apeloig, Y. Organometallics 2010, 29, 5596-−5606;.
(o) Ishida, S.; Hirakawa, F.; Iwamoto, T. J. Am. Chem. Soc. 2011, 133,
12968−12971..
(26) Bullock, J. P.; Mashkina, E.; Bond, A. M. J. Phys. Chem. A 2011,
115, 6493−6502.
(27) Bard, A. J.; Faulkner, L. R. Electrochemical Methods:
Fundamentals and Applications, 2nd ed.; Wiley: New York, 2001;
Chapter 5.
(28) Gennett, T.; Milner, D. F.; Weaver, M. J. J. Phys. Chem. 1985,
89, 2787−2794.
(29) Marcus, R. A. J. Chem. Phys. 1965, 43, 679−701.
(30) Richardson, J. N.; Harvey, J.; Murray, R. W. J. Phys. Chem. 1994,
98, 13396−13402.
(31) Fawcett, W. R.; Opallo, M. Angew. Chem., Int. Ed. Engl. 1994, 33,
2131−2143.
(32) The value for the average molecular radius is calculated from X-
ray crystallographic data by treating the molecules as oblate spheres
and using the van der Waals surface at points of greatest separation.
For P(mes)3, dimensions of 14.1 and 7.3 Å were used. See: Bott, R. C.;
Bowmaker, G. A.; Buckley, R. W.; Healy, P. C.; Perera, M. C. S. Aust. J.
Chem. 2000, 53, 175−181. Blount, J. F.; Camp, D.; Hart, R. D.; Healy,
P. C.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1994, 47, 1631−
1639.
(5) (a) Pan, X.; Chen, X.; Li, T.; Li, Y.; Wang, X. J. Am. Chem. Soc.
2013, 135, 3414−3417. (b) Pan, X.; Su, Y.; Chen, X.; Li, Y.; Zuo, J.;
Wang, X. J. Am. Chem. Soc. 2013, 135, 5561−5564.
(6) (a) Protasiewicz, J. D.; Washington, M. P.; Gudimetla, V. B.;
Payton, J. L.; Simpson, M. C. Inorg. Chim. Acta 2010, 364, 39−45.
(b) Dutan, C.; Shah, S.; Smith, R. C.; Choua, S.; Berclaz, T.; Geoffroy,
M.; Protasiewicz, J. D. Inorg. Chem. 2003, 42, 6241−6251. (c) Smith,
R. C.; Gleason, L. B.; Protasiewicz, J. D. J. Mater. Chem. 2006, 16,
2445−2452. (d) Smith, R. C.; Ren, T.; Protasiewicz, J. D. Eur. J. Inorg.
Chem. 2002, 2779−2783. (e) Smith, R. C.; Shah, S.; Urnezius, E.;
Protasiewicz, J. D. J. Am. Chem. Soc. 2003, 125, 40−41.
(7) (a) Stephan, D. W. Org. Biomol. Chem. 2008, 6, 1535−1539.
(b) Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed. 2010, 49, 46−76.
́
(c) Menard, G.; Hatnean, J. A.; Cowley, H. J.; Lough, A. J.; Rawson, J.
M.; Stephan, D. W. J. Am. Chem. Soc. 2013, 135, 6446−6449.
(8) Boere, R. T.; Bond, A. M.; Cronin, S.; Duffy, N. W.; Hazendonk,
́
P.; Masuda, J. D.; Pollard, K.; Roemmele, T. L.; Tran, P.; Zhang, Y.
New J. Chem. 2008, 32, 214−231.
́
(9) Boere, R. T.; Masuda, J. D.; Tran, P. J. Organomet. Chem. 2006,
691, 5585−5591.
(10) Sasaki, S.; Yoshifuji, M. Curr. Org. Chem 2007, 11, 17−31.
(11) Changenet, P.; Plaza, P.; Martin, M. M.; Meyer, J. Y.; Rettig, W.
Chem. Phys. 1997, 221, 311−322.
(12) Rozanel’skaya, N. A.; Bokanov, A. I.; Uzhinov, B. M.; Stepanov,
B. I. J. Gen. Chem. USSR 1975, 45, 277−280.
(13) Bryce, D. L.; Eichele, K.; Wasylishen, R. E. Inorg. Chem. 2003,
42, 5085−5096.
(14) Schiavon, G.; Zercchin, S.; Cogoni, G.; Bontempelli, G. J.
Electroanal. Chem. 1973, 48, 425−431.
(15) Ohmori, H.; Nakai, S.; Masui, M. J. Chem. Soc., Perkin Trans. 1
1978, 1333−1335.
(16) Culcasi, M.; Berchadsky, Y.; Gronchi, G.; Tordo, P. J. Org.
Chem. 1991, 56, 3537−3542.
(17) Tordo, P. In The Chemistry of Organophosphorus Compounds;
Hartley, F. R., Ed., Wiley: Chichester, 1990; Vol. 1, Chapter 6.
(18) (a) Sasaki, S.; Sutoh, K.; Murakami, M.; Yoshifuji, M. J. Am.
Chem. Soc. 2002, 124, 14830−14831. (b) Sasaki, S.; Chowdhury, R.;
Yoshifuji, M. Tetrahedron Lett. 2004, 45, 9193−9196.
(19) Merzougui, B.; Berchadsky, Y.; Tordo, P.; Gronchi, G.
Electrochim. Acta 1997, 42, 2445−2453.
(20) Zagumennov, V. A.; Nikitin, E. V. Russ. J. Electrochem. 2003, 39,
1236−1239.
(21) Palau, C.; Berchadsky, Y.; Chalier, F.; Finet, J.-P.; Gronchi, G.;
Tordo, P. J. Phys. Chem. 1995, 99, 158−163.
(22) Caram, J. A.; Vasini, E. J. Electrochim. Acta 1994, 39, 2395−
2400.
(23) Bond, A. M.; Duffy, N. W.; Elton, D. M.; Fleming, B. D. Anal.
Chem. 2009, 81, 8801−8808.
(24) (a) Zhang, J.; Guo, S.-X.; Bond, A. M. Anal. Chem. 2007, 79,
2276−2288. (b) Sher, A. A.; Bond, A. M.; Gavaghan, D. J.; Harriman,
K.; Feldberg, S. W.; Duffy, N. W.; Guo, S.-X.; Zhang. J. Anal. Chem.
2004, 76, 6214−6228. (c) Zhang, J.; Guo, S.-X.; Bond, A. M. Anal.
Chem. 2007, 79, 2276−2288. (d) O’Mullane, A. P.; Zhang, J.; Brajter-
Toth, A.; Bond, A. M. Anal. Chem. 2008, 80, 4614−4626. (e) Guo, S.;
Zhang, J.; Elton, D. M.; Bond, A. M. Anal. Chem. 2004, 76, 166−177.
(f) Fleming, B. D.; Barlow, N. L.; Zhang, J.; Bond, A. M.; Armstrong,
F. A. Anal. Chem. 2006, 78, 2948−2956. (g) Lee, C. Y.; Fleming, B. D.;
11215
dx.doi.org/10.1021/ja403555d | J. Am. Chem. Soc. 2013, 135, 11205−11215