An insight into the reversible dissociation and chemical reactivity of a sterically encumbered diphosphane

Article abstract of DOI:10.1002/ejic.201200633

The known 1,1',3,3'-tetrahydro-2,2'-bi-1,3,2-diazaphosphole 4a was for the first time synthesized on a preparative scale and characterized by spectroscopy. Equilibrium constants for the dissociation 4a [rlarr2] 2 5a were determined from quantitative variable-temperature electron paramagnetic resonance(VT-EPR) studies and were used to obtain thermochemical data for the P-P bond homolysis. The calculated bond dissociation energy of 79(1) kJ mol^-1 is substantially lower than the gas-phase dissociation energies of small diphosphanes. Modeling of the molecular structure of 4a by DFT calculations was feasible if dispersion effects were included, but the calculations still failed to give a reasonable account of the energetics of the dissociation process. Chemical reactions of 4a with alkynes proceeded, depending on substrate structure, either as a radical-induced diphosphanation of the triple bond or by a nonradical reaction to give an unprecedented ring metathesis. The tendency of sterically encumbered diphosphanes to dissociate at room temperature into phosphanyl radicals is explained by a low P-P bond enthalpy of 79 kJ mol ^-1. Still, reactions with alkynes do not necessarily proceed as radical additions to the triple bond but may proceed by direct attack of the diphosphane at the alkyne, yielding unexpected ring metathesis products with intact P-P bonds. Copyright

Full text of DOI:10.1002/ejic.201200633

Job/Unit: I20633
/KAP1
Date: 05-07-12 16:04:49
Pages: 7
FULL PAPER
DOI: 10.1002/ejic.201200633
An Insight into the Reversible Dissociation and Chemical Reactivity of a
Sterically Encumbered Diphosphane
Daniela Förster,[a] Herbert Dilger,[b] Fabian Ehret,[a] Martin Nieger,[c] and
Dietrich Gudat*[a]
Keywords: Phosphanes / Radicals / Dimerization / Bond energy / Reaction mechanisms
The known 1,1Ј,3,3Ј-tetrahydro-2,2Ј-bi-1,3,2-diazaphosphole
4a was for the first time synthesized on a preparative scale
and characterized by spectroscopy. Equilibrium constants for
the dissociation 4a i 2 5a were determined from quantitative
variable-temperature electron paramagnetic resonance
(VT-EPR) studies and were used to obtain thermochemical
data for the P–P bond homolysis. The calculated bond disso-
ciation energy of 79(1) kJmol–1 is substantially lower than the
gas-phase dissociation energies of small diphosphanes. Mod-
eling of the molecular structure of 4a by DFT calculations
was feasible if dispersion effects were included, but the cal-
culations still failed to give a reasonable account of the ener-
getics of the dissociation process. Chemical reactions of 4a
with alkynes proceeded, depending on substrate structure,
either as a radical-induced diphosphanation of the triple
bond or by a nonradical reaction to give an unprecedented
ring metathesis.
Introduction
The continuous interest in sterically distorted or elec-
tronically polarized molecules is motivated by the apprecia-
tion that unusual geometric or electronic structures often
coincide with unprecedented reactivity that is not found for
undistorted molecules.[1] Paramount examples are sterically
encumbered disilanes (1a,[2] 1b;[3] Scheme 1) or diphospha-
nes (2a,[4] 2b[5]) where very bulky substituents induce sizea-
ble lengthening of E–E bonds [E = Si, P; Si–Si 2.43 Å in
1a, 2.697 Å in 1b, P–P 2.310 Å in 2a, 2.291(4) Å in 2b vs.
standard distances Si–Si 2.359(12) Å, P–P 2.214(22) Å[6]],
which facilitates the formation of long-lived or even persist-
ent radicals.[4,7–9]
Scheme 1. (Dipp = 2,6-iPr2C6H3).
As these radicals are considered key intermediates, which
control the reactivity of the E–E-bonded precursors,[8] an
understanding of the factors governing radical formation
has some impact for the general explanation of the reactiv-
ity and thermal stability of E–E-bonded frameworks, and
the tuning of these qualities by steric strain. Current expla-
nations are based on computational results and attribute
the low E–E bond dissociation energies and easy radical
formation to the reduction of the sizeable intrinsic E–E
bond energies by substituent repulsion or conformational
strain.[10,11] Although the structural aspects of E–E bond
homolysis in species like 1 and 2 were intensely studied and
experimental molecular structures of radicals and dimers
are known,[4,5,7] experimental data on the energetics of such
reactions are, to the best of our knowledge, unavailable.
During studies of the P–P bond reactivity of unsymmet-
rical N-heterocyclic diphosphanes like 3,[12] our attention
was attracted by the recent discovery that the symmetrical
derivative 4a dissociates at room temp. to give 7π-radical
5a.[13] Even if neither the equilibrium between 4a and 5a
nor their chemical properties were further studied, the very
low calculated dissociation energy for model 4b (ca.
3 kJmol–1) led to the description of 4a as a weakly σ-
bonded dimer formed by the π*–π* interaction of two radi-
cals.[13] We present here the results of reactivity studies on
4a, which provide the first experimental data on the ener-
[a] Institut für Anorganische Chemie, University of Stuttgart,
Pfaffenwaldring 55, 70550 Stuttgart, Germany
Fax: +49-711-685-64186
E-mail: gudat@iac.uni-stuttgart.de
Homepage: http://www.iac.uni-stuttgart.de/akgudat
[b] Institut für Physikalische Chemie, University of Stuttgart,
Pfaffenwaldring 55, 70550 Stuttgart, Germany
[c] Laboratory of Inorganic Chemistry, Department of Chemistry,
University of Helsinki,
P. O. Box 55 (A. I. Virtasen aukio 1), 00014 University of Hel-
sinki, Finland
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201200633.
Eur. J. Inorg. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: I20633
/KAP1
Date: 05-07-12 16:04:49
Pages: 7
D. Förster, H. Dilger, F. Ehret, M. Nieger, D. Gudat
FULL PAPER
getics of P–P bond homolysis in a sterically crowded di-
phosphane, and indicate that the reactions of 4a with al-
kynes do not necessarily follow a radical pathway, in spite
of its easy dissociation into radicals.
Results and Discussion
Although a crystal structure of 4a is known,[13] neither
its synthesis on a preparative scale nor its spectroscopic
data were reported. We found that multigram quantities of
4a are accessible in reproducible yield if the reported syn-
thesis[13] is combined with an improved workup procedure
(see Experimental Section). The isolated product was char-
acterized by single-crystal X-ray diffraction as a new poly-
morph with crystallographic C1 symmetry and a gauche
conformation rather than the previously reported transoid
conformation[13] of the R2P–PR2 moiety (see Experimental
Section and Supporting Information). Below –20 °C, the 1H
NMR spectrum shows a pattern of sharp resonances, which
reveals that the conformation established for solid 4a per-
sists in solution. The 31P chemical shift (δ = 150.4 ppm at
–20 °C) clearly exceeds reported values of other bi-diaza-
phospholes,[13,14] but its assignment is confirmed by com-
parison with the data of solid 4a: a 31P cross-polarization
magic-angle-spinning (CP-MAS) NMR spectrum displays
two signals for the crystallographically independent phos-
phorus atoms with shifts of δ = 147.8 and 151.7 ppm and
Figure 1. (a) Superimposed EPR spectra of a 1.2 mm solution of
4a/5a recorded at 250, 280, 310 and 340 K drawn to identical verti-
cal scale; (b) van’t Hoff plot of Rln(KDiss) vs. 1/T.
The equilibrium constant for the equilibrium 4a i 2 5a
(KDiss = [5a]2/[4]) can be computed if one measures the con-
centration of 5a in a sample and then calculates the concen-
tration of 4a from the a priori known composition. The
radical concentration is readily obtained from the double
integral of the EPR signal in relation to that of a reference
sample with a known number of spins.[15] Following this
approach, we determined the amount of 5a in a solution of
4a (1.2 mm initial concentration) against a calibrated sam-
295
ple of ultramarine blue at 295 K to obtain KDiss
=
1.8(5)ϫ10–5 molL–1 (see Supporting Information). This
value implies a degree of dissociation of 0.12 and a free
295
enthalpy ΔGDiss of 26.8(8) kJmol–1. Measuring the rela-
tive variation of double integrals in a series of VT-EPR
spectra then allowed us to evaluate the temperature depen-
dence of KDiss, and revealed that equal concentrations of 4a
and 5a are reached at 338 K. A van’t Hoff plot (Figure 1b)
yielded ΔHDiss = 78.8(8) kJmol–1 for the enthalpy and
ΔSDiss = 176(15) JK–1 mol–1 for the entropy of dissociation.
These experimental data do not agree well with a com-
puted dissociation energy of only 3 kJmol–1 for model
4b.[13] In order to rule out substituent effects as a cause of
the deviation, we computed thermochemical data for the
gas-phase reactions 4a,b i 2 5a,b by DFT methods. Geom-
etry optimization of 4a using the popular B3LYP functional
yielded an intolerably long P–P bond, presumably due to
overestimation of substituent repulsion,[17] and this func-
tional was thus considered unsuitable to model reaction
energetics. In contrast, the ωB97X-D[18] and M06-2X[19]
functionals, which include long-range and dispersion cor-
rections,[20] produced molecular structures in close agree-
1
a scalar coupling of JPP = 365 Hz. The solution NMR
spectra vary strongly when the temperature is raised. The
1H NMR spectrum shows coalescence phenomena indicat-
ing a similar librational motion of the R2P fragments as in
tBu2P–PtBu2;[11] unexpectedly, however, the merged lines do
not sharpen but become broader when temperatures in-
crease further. The 31P NMR signal appears as a sharp sin-
glet (Δν1/2 = 4 Hz) at –20 °C but widens exponentially with
temperature (Δν1/2 = 1100 Hz at 30 °C) until the line broad-
ens beyond visibility above 40 °C. The extreme broadening
1
of the H and 31P NMR lines is a direct consequence of
dynamic exchange between 4a and radical 5a, which is too
slow at low temperatures to cause spectral changes but per-
ceptibly shortens transversal relaxation times T2 when its
rate increases with temperature. The implication of these
findings – that the true radical lifetime is much shorter than
that indicated by the decay time of the electron paramag-
netic resonance (EPR) signal – may also be of importance
for other persistent phosphanyl radicals.
A similar dissociation equilibrium as observed for 4a/5a
had also been inferred in the case of 2b and implies that, as
well as the exchange rate, the equilibrium composition also
varies with temperature.[5,8] In line with this assumption,
Table 1. P–P bond lengths for 4a,b and zero-point corrected elec-
tronic energy ΔEzpe, standard enthalpy ΔH298 and standard free
enthalpy ΔG298 for reactions 4a,b i 2 5a,b.
[c]
[c]
[c]
Method[a]
P–P[b]
ΔEzpe
ΔH298
ΔG298
4a/5a
ωB97X-D
M06-2X
exp.
2.323
2.346
2.320(1)
120.5
98.8
–
129.9
106.6
78.8(8)
5.9
–18.4
26.4(8)
variable-temperature
electron-paramagnetic-resonance
(VT-EPR) studies on samples prepared by the dissolution
of fixed amounts of 4a in hexane or xylene revealed a per-
ceptible and fully reversible increase in signal amplitude
(and thus in the concentration of radical 5a with tempera-
ture (Figure 1a).
4b/5b
ωB97X-D
M06-2X
2.258
2.260
94.4
101.4
95.8
103.3
33.3
39.4
[a] cc-pVDZ basis set used for all calculations. [b] Bond length for
4a,b in Å. [c] In kJmol–1 including correction for basis set superpo-
sition error.
2
www.eurjic.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 0000, 0–0

Job/Unit: I20633
/KAP1
Date: 05-07-12 16:04:49
Pages: 7
Properties of a Sterically Encumbered Diphosphane
ment with the experimental structures. The computed P–P sively (Z)-alkenes.[12,23] The long reaction time (several
distance in 4b is shorter than that in 4a, reflecting released hours at 50 °C, very slow at room temp.) indicates that radi-
steric strain (Table 1 and Supporting Information).
In contrast to the reported data,[13] the dissociation of
4a,b remains endothermic and (at least for 4b) also ender-
gonic. The M06-2X/cc-pVDZ reaction enthalpy is by some
20 kJmol–1 less positive than that obtained with the
ωB97X-D functional, but still exceeds the experimental
value by some 27 kJmol–1. The deviation between the calcu-
lated ΔG298 values is still larger, and as it has been claimed
that the entropy contribution to ΔG in solution at room
temp. is approximately half of that in the gas phase,[21] we
consider a comparison with experimental values as prob-
lematic.
cal formation is obviously not the rate-limiting step.
The M06-2X dissociation energies and the (free) enthal-
pies of 4a,b are similar, whereas ωB97X-D calculations pre-
dict a decrease of both quantities from 4a to 4b. The same,
at first glance counterintuitive, relationship between bond
length and dissociation energy is known from computa-
tional studies on other diphosphanes,[10] and may reflect the
effect of attractive dispersion forces,[20] which should have
more impact in 4a than in the smaller and less strained 4b.
We conclude thus that 4a is another case that highlights the
importance of dispersion forces in the dissociation reactions
of large molecules,[20] but must, in view of known inherent
difficulties,[13] admit that their realistic depiction remains an
unsolved challenge.
The formation of a significant fraction of phosphanyl
radicals in solution suggests that 4a offers great potential
for addition reactions to multiple bonds. The diphosphan-
ation of alkynes has lately received attention as access to
1,2-bis(phosphanyl)ethenes and may proceed by radical[22]
or dipolar[12,23] pathways.
Figure 2. Crystal structure of 6 (H atoms and solvent molecules
omitted; 50% probability thermal ellipsoids; one of two molecules
with crystallographic Ci symmetry in the unit cell is shown); se-
lected distances [Å] (data for the second molecule in brackets): P1–
C30 1.882(3) [1.879(3)], C30–C30Ј 1.363(6)#1 [1.357(6)#2] (sym-
metry orerators: #1: –x + 1, –y + 2, –z + 1; #2: –x + 2, –y + 1,
–z + 2).
In order to check its behaviour towards electron-poor
alkynes, we treated 4a with dimethyl acetylenedicarboxylate
(DMAD). The reaction consumed 2 equiv. of DMAD to
yield the unusual diphosphane 8 (Scheme 2), which was iso-
lated in good yield and identified by a single-crystal X-ray
diffraction study (Figure 3).
Indeed, 4a reacted cleanly with tolane to give bis(phos-
phane) 6 (Scheme 2), which, according to an X-ray diffrac-
tion study, exhibits crystallographic Ci symmetry and (E)
configuration at the double bond (Figure 2). We assume 6
is formed by a radical pathway as (a) tolane lacks sufficient
electrophilicity to initiate dipolar addition of diphosphanes
like 3,[12] which undergo no detectable bond homolysis up
to 100 °C, and (b) radical-initiated P–P addition to alkynes
yields chiefly (E)-alkenes,[22] which is in accord with the mo-
lecular structure of 6, whereas dipolar addition gives exclu-
Figure 3. Crystal structure of 8 (H atoms omitted; 50% probability
thermal ellipsoids); selected distances [Å]: P1–P2 2.2862(9), P1–N2
1.709(2), N2–C3 1.468(3), C3–C4 1.506(3), C4–C5 1.336(3), C5–
P1 1.813(2), C3–C18 1.509(3), C18–N19 1.261(3), P2–C39 1.817(3),
C39–C38 1.331(3), C37–N36 1.476(3), N36–P2 1.709(2), C37–C52
1.509(3), C52–N53 1.261(3).
NMR studies disclosed that 4a reacts first with one mo-
lecule of DMAD to give spectroscopically detectable 7,
which then takes up a second alkyne to yield 8 as the final
product. The reaction is completed in the time needed to
mix reactants and record the NMR spectrum even at
–20 °C when radical concentration is negligible, and the
product formed under these conditions is identical to that
Scheme 2. Reactions of 4a with alkynes (R = 2,6-iPr2C6H3).
Eur. J. Inorg. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3

Job/Unit: I20633
/KAP1
Date: 05-07-12 16:04:49
Pages: 7
D. Förster, H. Dilger, F. Ehret, M. Nieger, D. Gudat
FULL PAPER
methods. NMR spectra were recorded with Bruker Avance 400 (1H,
400.1 MHz; 13C, 100.5 MHz; 31P, 161.9 MHz) and Avance 250 (1H,
250.1 MHz; 13C, 62.8 MHz; 31P, 101.2 MHz) NMR spectrometers
at 303 K; chemical shifts are referenced to external TMS (1H, 13C)
or 85% H3PO4 (Ξ = 40.480, 747 MHz, 31P). Coupling constants
are given as absolute values. Elemental analyses were determined
with a Perkin–Elmer 24000CHN/O Analyzer. Melting points were
determined in sealed capillaries with a Büchi Point B-545 appara-
tus. EPR spectra were measured with a Bruker EMX X-band spec-
trometer.
obtained at 20 °C when a sizeable amount of radicals is
present. Other than 7, no intermediates were detected, and
the feasibility of any interconversion between 7, 8 and 4a/
5a in the absence of alkyne was ruled out.
Based on these findings we deduce that primary adduct
7 is not formed in a radical process but arises from the
direct reaction of the alkyne with diphosphane 4a. Adopt-
ing a widely accepted concept for reactions of DMAD with
phosphanes,[23,24] the unprecedented opening of the N-het-
erocyclic ring of 4a is rationalized by assuming that nucleo-
philic attack of one phosphane unit at the triple bond pro-
duces a zwitterionic intermediate, which then rearranges to
7 (Scheme 3). The reaction of another DMAD molecule
with the second phosphane unit completes the formation of
8.
4a: Mg turnings (0.11 g, 4.5 mmol) and a few crystals of I2 were
added to a solution of 2-chloro-1,3-bis(dipp)-1,3-dihydro-1,3,2-di-
azaphosphol (2.0 g, 4.5 mmol) in tetrahydrofuran (THF, 10 mL).
The mixture was stirred for 12 h. The volatiles were removed in
vacuo, the residue was suspended in hexane (15 mL) and filtered
through Celite. The filtrate was concentrated and stored at –20 °C
to give dark green crystals (1.3 g, 72%, m.p. 157 °C). 31P CP-MAS
NMR: δ = 147.8, 151.7 ppm; |1JPP| = 365 Hz. 31P{1H} NMR ([D8]-
1
THF, 303 K): δ = 152.8 (s) ppm. H NMR ([D8]THF, 203 K): δ =
7.33 (m, 4 H, C6H3), 7.22 (m, 2 H, C6H3), 6.88–7.00 (m, 6 H,
3
C6H3), 6.35 (m, 2 H, NCH), 6.00 (d, JHH = 2 Hz, 2 H, NCH),
3
3
4.30 (d, JHH = 6 Hz, 2 H, CHMe2), 3.55 (d, JHH = 6 Hz, 2 H,
CHMe2), 3.04 (m, 2 H, CHMe2), 2.99 (m, 2 H, CHMe2), 1.50 (d,
3JHH = 6 Hz, 6 H, CH3), 1.31 (d, 3JHH = 6 Hz, 6 H, CH3), 1.22 (d,
3JHH = 6 Hz, 6 H, CH3), 1.15 (d, 3JHH = 6 Hz, 6 H, CH3), 1.11 (d,
3JHH = 6 Hz, 6 H, CH3), 1.07 (d, 3JHH = 6 Hz, 6 H, CH3), 0.42 (d,
Scheme 3. Proposed mechanism for ring metathesis in the reaction
of 4a with DMAD (R = 2,6-iPr2C6H3).
3
3JHH = 6 Hz, 6 H, CH3), 0.27 (d, JHH = 6 Hz, 6 H, CH3) ppm.
C52H72N4P2 (805.12): calcd. C 76.62, H 8.90, N 6.87; found C
75.90, H 8.67, N 6.78 (the rather large deviation between found
and calculated values owes to the high oxidative sensitivity of the
sample).
Conclusions
Our study of the reversible dissociation of 4a enabled the
first experimental evaluation of the energetics of P–P bond
homolysis in a sterically strained diphosphane. The reaction
6: Diphenylacetylene (55 mg, 0.31 mmol) was added to a solution
of 4a (250 mg, 0.31 mmol) in anhydrous THF (8 mL). The solution
remains endergonic, in spite of the electronic stabilization was stirred at 50 °C for 12 h, allowed to cool to room temp., and
the solvents were evaporated to dryness. The residue was washed
twice with Et2O (5 mL) and recrystallized at –20 °C from THF
of radical 5a and steric crowding in 4a, but the dissociation
enthalpy (79 kJmol–1) is much lower than the average P–
P bond enthalpy of 214 kJmol–1 or gas-phase dissociation
energies of small diphosphanes R2P–PR2 (R = H, Et, F, Cl,
I; DP–P = 239–368 kJmol–1[25]). The existence of a substan-
tial energetic barrier to dissociation as well as its spectro-
scopic and structural properties suggest that 4a resembles
more closely a conventional – although sterically strained –
diphosphane with a localized P–P bond rather than a labile
dimer formed by the π*–π* interaction between two radi-
cals, for which a more delocalized bonding situation would
be expected. Computational studies suggest that dispersion
forces contribute to the P–P bonding but show also that a
realistic model of the reaction energetics is still out of reach.
Reactions of the equilibrium mixture of 4a/5a with alkynes
proceed either by a radical pathway, or by direct interaction
of the alkyne with intact 4a to result in the unprecedented
opening of the N-heterocycle. The mechanism is controlled
by the substituent pattern in the alkyne, and the observed
dichotomy indicates that radicals do not necessarily react
faster than their closed-shell, neutral dimers.
(200 mg, 65%, m.p. 199 °C). 31P{1H} NMR (THF, 253 K): δ =
3
117.0 (s) ppm. 1H NMR ([D8]THF, 253 K): δ = 7.29 (d, JHH
=
7 Hz, 4 H, C6H3), 6.96–7.01 (m, 10 H, C6H5), 6.84 (broad, 4 H,
C6H3), 6.63 (3JHH = 8 Hz, 4 H, C6H3), 5.70 (s, 4 H, NCH), 3.27
3
(m, 4 H, CHMe2), 2.67 (m, 4 H, CHMe2), 0.98 (d, JHH = 7 Hz,
3
3
12 H, CH3), 0.85 (t, JHH = 8 Hz, 24 H, CH3), 0.21 (d, JHH
=
6 Hz, 12 H, CH3) ppm. C66H82N4P2 (993.33): calcd. C 79.80, H
8.32, N 5.64; found C 79.37, H 8.23, N 5.72.
8: DMAD (87 mg, 0.62 mmol) was added to a solution of 4a
(250 mg, 0.31 mmol) in anhydrous THF (8 mL). The solution was
stirred at room temp. for 1 h and concentrated under reduced pres-
sure. The residue was washed with hexane (2ϫ 5 mL) and recrys-
tallized at –20 °C from Et2O (177 mg, 52%, m.p. 174 °C). 31P{1H}
NMR (THF, 303 K): δ = 112.8 (s) ppm. 1H NMR ([D8]THF,
3
253 K): δ = 6.85–7.25 (12 H, C6H3), 6.07 (m, JHH = 6 Hz, 2 H,
NCH), 6.05 (m, 3JHH = 6 Hz, 2 H, NCH), 4.00 (s, 6 H, OMe), 3.86
(s, 6 H, OMe), 3.36 (m, 4 H, CHMe2), 3.25 (m, 2 H, CHMe2), 3.07
3
3
(m, 2 H, CHMe2), 1.41 (d, JHH = 6 Hz, 6 H, CH3), 1.33 (d, JHH
3
= 6 Hz, 6 H, CH3), 0.86 (d, JHH = 6 Hz, 18 H, CH3), 0.82 (d,
3JHH = 6 Hz, 12 H, CH3), 0.76 (d, JHH = 6 Hz, 6 H, CH3) ppm.
3
C64H84N4O8P2 (1099.32): calcd. C 69.92, H 7.70, N 5.10; found C
69.32, H 7.86, N 4.92.
7: Compound 7 was spectroscopically observed as the main prod-
uct of the reaction of 4a with 1 equiv. of DMAD at –20 °C along
with some 8 and unreacted 4a. 31P{1H} NMR (THF, 253 K): δ =
Experimental Section
General: All manipulations were carried out under dry argon using
standard vacuum-line techniques. Solvents were dried by standard
1
1
155.2 (d, JPP = 371.4 Hz), 137.0 (d, JPP = 371.4 Hz) ppm.
4
www.eurjic.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 0000, 0–0

Job/Unit: I20633
/KAP1
Date: 05-07-12 16:04:49
Pages: 7
Properties of a Sterically Encumbered Diphosphane
X-ray Analysis: Crystal structures were determined with a Nonius
Kappa-CCD diffractometer at T = 100(2) K, using Mo-Kα radia-
tion and SHELX97 for structure solution and refinement (full-ma-
trix least squares refined against F2). The positions of hydrogen
atoms were refined with a riding model. Semi-empirical absorption
corrections were applied for 6 and 8. CCDC-880738 (for 4a),
-880759 (for 6) and -880718 (for 8) contain the supplementary crys-
tallographic 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.
Academy of Finland (M. N.), and the COST action CM 0802 (pho-
scinet) is gratefully acknowledged. We thank the bw-grid project[28]
for the computational resources and Dr. W. Frey for measurement
of XRD data sets.
[1] H. Bock, K. Ruppert, C. Näther, Z. Havlas, H.-F. Herrmann,
C. Arad, I. Göbel, A. John, J. Meuret, S. Nick, A. Rauschen-
bach, W. Seitz, T. Vaupel, B. Solouki, Angew. Chem. 1992, 104,
564–595; Angew. Chem. Int. Ed. Engl. 1992, 31, 550–581.
[2] S. L. Hinchley, H. E. Robertson, A. Parkin, D. W. H. Rankin,
G. Tekautz, K. Hassler, Dalton Trans. 2004, 759–766.
[3] N. Wiberg, H. Schuster, A. Simon, K. Peters, Angew. Chem.
1986, 98, 100–101; Angew. Chem. Int. Ed. Engl. 1986, 25, 79–
80.
[4] a) S. L. Hinchley, C. A. Morrison, D. W. H. Rankin, C. L. B.
Macdonald, R. J. Wiacek, A. H. Cowley, M. F. Lappert, G.
Gundersen, J. A. C. Clyburne, P. P. Power, Chem. Commun.
2000, 2045–2046; b) S. L. Hinchley, C. A. Morrison, D. W. H.
Rankin, C. L. B. Macdonald, R. J. Wiacek, A. Voigt, A. H.
Cowley, M. F. Lappert, G. Gundersen, J. A. C. Clyburne, P. P.
Power, J. Am. Chem. Soc. 2001, 123, 9045–9053.
[5] J.-P. Bezombes, K. B. Borisenko, P. B. Hitchcock, M. F. Lap-
pert, J. E. Nycz, D. W. H. Rankin, H. E. Robertson, Dalton
Trans. 2004, 1980–1988.
[6] A. J. C. Wilson, Internat. Tables for Crystallography, vol. C,
Kluwer Academic Publishers, London, 1992.
[7] M. J. S. Gynane, A. Hudson, M. F. Lappert, P. P. Power, H.
Goldwhite, J. Chem. Soc., Dalton Trans. 1980, 2428–2433.
[8] J.-P. Bezombes, K. B. Borisenko, P. B. Hitchcock, M. F. Lap-
pert, J. E. Nycz, Dalton Trans. 2004, 499–501.
[9] S. L. Masters, D. A. Grassie, H. E. Robertson, M. Hoelbling,
K. Hassler, Chem. Commun. 2007, 2618–2620.
4a: C52H72N4P2; M = 815.08; crystal size 0.30ϫ0.25ϫ0.10 mm;
¯
triclinic, space group P1, a = 10.2824(3), b = 12.2901(5), c =
20.8313(9) Å, α = 80.845(2)°, β = 88.824(3)°, γ = 66.237(2)°, V =
2375.85(16) Å3, Z = 2, ρ = 1.139 Mgm–3, μ = 0.13 mm–1, F(000) =
884, θmax = 27.0°; 19051 reflections/10264 independent reflections
(Rint = 0.075), 520 parameters, 4 restraints, R1 = 0.079 [IϾ2σ(I)],
wR2 = 0.191 (all data), S = 1.06, largest diff. peak/hole 0.477/
–0.590 eÅ–3.
6: C70H90N4OP2; M = 1065.40; crystal size 0.22ϫ0.17ϫ0.12 mm;
¯
triclinic, space group P1, a = 12.2960(11), b = 12.4740(11), c =
21.8851(18) Å, α = 73.465(4)°, β = 74.135(5)°, γ = 77.549(5)°, V =
3060.6(5) Å3, Z = 2, ρ = 1.156 Mgm–3, μ = 0.12 mm–1, F(000) =
1152, θmax = 26.4°; 62159 reflections/12458 independent reflections
(Rint
= 0.048), 694 parameters, 420 restraints, R1 = 0.077
[IϾ2σ(I)], wR2 = 0.222 (all data), S = 1.03, largest diff. peak/hole
1.069/–0.744 eÅ–3.
8: C64H84N4O8P2; M = 1099.29; crystal size 0.21ϫ0.19ϫ0.15 mm;
monoclinic, space group P21/c, a = 24.9956(19), b = 11.7832(9), c
= 22.9093(15) Å, β = 115.500(3)°, V = 6090.1(8) Å3, Z = 4, ρ =
1.199 Mgm–3, μ = 0.128 mm–1, F(000) = 2360, θmax = 25.0°; 70971
reflections/10719 independent reflections (Rint = 0.098), 703 param-
eters, R1 = 0.051 [IϾ2σ(I)], wR2 = 0.107 (all data), S = 1.00,
largest diff. peak/hole 0.354/–0.251 eÅ–3.
[10] a) K. B. Borisenko, D. W. H. Rankin, Inorg. Chem. 2003, 42,
7129–36; b) K. B. Borisenko, D. W. H. Rankin, Dalton Trans.
2004, 2382–2387.
[11] S. L. Hinchley, H. E. Robertson, K. B. Borisenko, A. R.
Turner, B. F. Johnston, D. W. H. Rankin, M. Ahmadian, J. N.
Jones, A. H. Cowley, Dalton Trans. 2004, 2469–2476.
[12] S. Burck, D. Gudat, M. Nieger, Angew. Chem. 2004, 116, 4905–
4908; Angew. Chem. Int. Ed. 2004, 43, 4801–4804.
Calculations: DFT calculations were carried out with the
Gaussian 09[16] package using three different methods – Becke’s
three-parameter exchange functional[26] with the Lee–Yang–Parr
correlation energy functional (B3LYP)[27] with 6-31g(d) basis sets,
the ωB97X-D functional by Head–Gordon,[18] and finally Truhlar’s
M06-2X functional[19] – in combination with both 6-31g* or cc-
pVDZ basis sets. Attempts to use larger basis sets (i.e. including
diffuse functions or of triple-zeta quality) failed due to numerical
or computation time problems. Numerical integrations were per-
formed on an ultrafine grid. The molecular structures of 4a,b and
5a,b were first optimized using the smaller (6-31g*) basis set and
then re-optimized with the larger basis set (cc-pVDZ). Harmonic
frequencies and zero-point energies (ZPE) of the optimized gas-
phase structures were calculated at the same levels and showed all
molecular geometries to present local minima (only positive eigen-
values of the Hessian matrix) on the potential energy surface. Lists
of computed energies and molecular coordinates are given in
Table 1 and the Supporting Information.
[13] R. Edge, R. J. Less, E. J. L. McInnes, K. Müther, V. Naseri,
J. M. Rawson, D. S. Wright, Chem. Commun. 2009, 1691–1693.
[14] A. Haghverdi, PhD Thesis, University of Bonn, 2000.
[15] G. R. Eaton, S. S. Eaton, D. P. Barr, R. T. Weber in Quantita-
tive EPR – A practitioner’s guide, Springer, Wien, 2010.
[16] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-
erov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Ad-
amo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-
tin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, Re-
vision B01, Gaussian, Inc., Wallingford, 2009.
Supporting Information (see footnote on the first page of this arti-
cle): Plots of the molecular structure of 4a in the crystal, VT-NMR
spectroscopic data for 4a, details of the EPR measurements, lists
of computed thermochemical data for the dissociation reaction of
4a,b i 2 5a,b, computed absolute energies and atomic coordinates
of 4a,b and 5a,b.
[17] Problems of B3LYP in describing large molecules were recently
reviewed: I. Y. Zhang, J. Wu, X. Xu, Chem. Commun. 2010, 46,
3057–3070.
[18] J.-D. Chai, M. Head-Gordon, Phys. Chem. Chem. Phys. 2008,
Acknowledgments
10, 6615–6620.
[19] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215–241.
[20] The necessity to consider dispersion corrections in realistic
models of the dissociation energetics of large molecules has
Financial support by the Deutsche Forschungsgemeinschaft, the
Deutscher Akademischer Austauschdienst (DAAD) (D. F.), the
Eur. J. Inorg. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5

Job/Unit: I20633
/KAP1
Date: 05-07-12 16:04:49
Pages: 7
D. Förster, H. Dilger, F. Ehret, M. Nieger, D. Gudat
FULL PAPER
recently been pointed out: S. Grimme, P. R. Schreiner, Angew.
Chem. 2011, 123, 12849–12853; Angew. Chem. Int. Ed. 2011,
50, 12639–12642.
Phys. Chem. 1963, 67, 124; c) F. E. Saalfeld, H. J. Svec, Inorg.
Chem. 1964, 3, 1442–1443; d) N. N. Grishin, G. M. Bogolyu-
bov, A. A. Petrov, J. Gen. Chem. USSR 1968, 38, 2595; e) C. S.
Dean, A. Finch, P. J. Gardner, D. W. Payling, J. Chem. Soc.
Faraday Trans. 1973, 1921–1925.
[21] R. Robiette, V. K. Aggarwal, J. N. Harvey, J. Am. Chem. Soc.
2007, 129, 15513–15525.
[22] A. Sato, H. Yorimitsu, K. Oshima, Angew. Chem. 2005, 117,
1722–1724; Angew. Chem. Int. Ed. 2005, 44, 1694–1696.
[23] D. L. Dodds, M. F. Haddow, A. G. Orpen, P. G. Pringle, G.
Woodward, Organometallics 2006, 25, 5937–5945.
[24] a) X. Lu, C. Zhang, Z. Xu, Acc. Chem. Res. 2001, 34, 535–
544; b) Z. Duan, J. Zhang, R. Tian, L. Bai, F. Mathey, Organo-
metallics 2008, 27, 5169–5171.
[26] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[27] B. J. Lynch, P. L. Fast, M. Harris, D. G. Truhlar, J. Phys. Chem.
A 2000, 104, 4811–4815.
[28] bwGRiD (http://www.bw-grid.de), member of the German D-
Grid initiative, funded by the Federal Ministry for Education
and Research and the Ministry for Science, Research and Arts,
Baden-Württemberg.
[25] a) A. Finch, A. Hameed, P. J. Gardner, N. Paul, J. Chem. Soc.
C 1969, 391; b) A. A. Sandoval, H. C. Moser, R. W. Kiser, J.
Received: June 11, 2012
Published Online: ᭿
6
www.eurjic.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 0000, 0–0

Job/Unit: I20633
/KAP1
Date: 05-07-12 16:04:49
Pages: 7
Properties of a Sterically Encumbered Diphosphane
P–P Bond Reactivity
The tendency of sterically encumbered di-
phosphanes to dissociate at room tempera-
ture into phosphanyl radicals is explained
by a low P–P bond enthalpy of 79 kJmol–1.
Still, reactions with alkynes do not neces-
sarily proceed as radical additions to the
triple bond but may proceed by direct at-
tack of the diphosphane at the alkyne,
yielding unexpected ring metathesis prod-
ucts with intact P–P bonds.
D. Förster, H. Dilger, F. Ehret, M. Nieger,
D. Gudat* ......................................... 1–7
An Insight into the Reversible Dissociation
and Chemical Reactivity of a Sterically En-
cumbered Diphosphane
Keywords: Phosphanes / Radicals / Di-
merization / Bond energy / Reaction mech-
anisms
Eur. J. Inorg. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
7