Chemistry of Thermally Generated Transient Phosphanoxyl Complexes

Article abstract of DOI:10.1021/acs.organomet.7b00347

Investigations on the reactivity of the transiently formed phosphanoxyl complex [(CO)₅W(Ph₂PO^a€￠)], thermally generated from [(CO)₅W(Ph₂PO)] in toluene, is presented. Apart from self-reactions, trapping of this radical complex was achieved using group 14 hydrides Ph₃EH (E = Si, Ge, Sn), leading to new phosphane complexes possessing a P-O-EPh₃ bonding motif and the corresponding TEMP-H as byproduct. Reaction pathways, derived from DFT calculations, clearly revealed the intermediacy of various open-shell complexes; EPR measurements showed the presence of radicals, but unfortunately interpretation was not achieved.

Full text of DOI:10.1021/acs.organomet.7b00347

Article
pubs.acs.org/Organometallics
Chemistry of Thermally Generated Transient Phosphanoxyl
Complexes
Tobias Heurich,† Zheng-Wang Qu,*,‡ Gregor Schnakenburg,† Yaser NejatyJahromy,§ Olav Schiemann,§
Stefan Grimme,‡ and Rainer Streubel*,†
†Institut fur Anorganische Chemie der Rheinischen Friedrich-Wilhelms-Universitat Bonn, Gerhard-Domagk-Straße 1, 53121 Bonn,
̈
̈
Germany
‡Mulliken Center for Theoretical Chemistry der Rheinischen Friedrich-Wilhelms-Universitat Bonn, Beringstraße 4, 53115 Bonn,
̈
Germany
§Institut fur Physikalische und Theoretische Chemie der Rheinischen Friedrich-Wilhelms-Universitat Bonn, Wegelerstraße 12, 53115
̈
̈
Bonn, Germany
S
* Supporting Information
ABSTRACT: Investigations on the reactivity of the tran-
siently formed phosphanoxyl complex [(CO)5W(Ph2PO•)],
thermally generated from [(CO)5W(Ph2PO-TEMP)] in
toluene, is presented. Apart from self-reactions, trapping of
this radical complex was achieved using group 14 hydrides
Ph3EH (E = Si, Ge, Sn), leading to new phosphane complexes
possessing a P−O−EPh3 bonding motif and the corresponding
TEMP-H as byproduct. Reaction pathways, derived from DFT
calculations, clearly revealed the intermediacy of various open-
shell complexes; EPR measurements showed the presence of
radicals, but unfortunately interpretation was not achieved.
derivatives formed at low temperature have been reported
recently. One example led, after homolytic O−N bond cleavage
and silyl migration, to an O-silyl phosphinate derivative.4 In a
more recent study, we showed that a PIII derivative of II (Z =
lone pair, R = Ph) can be obtained at low temperature via the
reaction of diphenylphosphane with TEMPO,5 which also
suffered from homolytic O−N bond cleavage to yield the
corresponding PV derivative (Z = O) as the final product.6
The knowledge of transition-metal phosphane complexes
having the P-TEMPO ligand II (Z = MLn) is also very scarce,
and only a few examples have been described. The first report
was on a stable AuI complex,7 shortly followed by an unstable
W0 complex.8 In the case of the latter, hints were obtained that
radicals are involved in synthesis and decomposition which
seemed to occur through O−N bond homolysis, thus leading to
the transient (and at that time unknown) phosphanoxyl
complex III (Z = MLn). Very recently, we were able to get
access to a somewhat more stable complex II (Z = W(CO)5, R
= Ph)9 that enabled the first reaction studies. Again, a
homolytic O−N bond cleavage led to the transient
phosphanoxyl complex III that was trapped with the
triphenylmethyl (trityl) radical to give complex IV. Further-
more, the use of this transient phosphanoxyl complex as a
radical initiator for styrene polymerization was demonstrated9
INTRODUCTION
■
Among the commonly used nitroxides,1 TEMPO (2,2,6,6-
tetramethyl-piperidin-1-oxyl) plays an important role and is
largely used in polymer chemistry, especially in living free
radical polymerization.2 Despite this huge interest and
numerous reports on main-group-element derivatives of
TEMPO (I)3 (Scheme 1), only a few examples of P-nitroxyl-
substituted phosphanes and their complexes have been
reported so far.
All known compounds of type II show the interesting feature
of having a labile O−N bond that tends to undergo homolytic
bond cleavage, yielding a radical of type III.
To date, there are no examples of PIII derivatives II (Z = lone
pair) that are stable at ambient temperature, but transient
Scheme 1. Main-Group-Element Derivatives of TEMPO (I),
Their Related Phosphorus Derivatives (II; R = Organic
Substituent, Z = Electron Pair or MLn for a Transition-Metal
Complex), Radical Species (III) Derived from O−N Bond
Cleavages in II, and a Phosphane Complex (IV) Trapping
Product of III
Received: May 4, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00347
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and DFT calculations provided information on the reaction
mechanism.
Herein, investigations on the decomposition of a tungsten(0)
complex having the ligand Ph2P-OTEMP in the absence of
trapping reagents are reported as well as trapping of the
transient phosphanoxyl complex using group 14 hydrides
Ph3EH (E = Si, Ge, Sn). The reaction pathways have been
studied via DFT calculations.
RESULTS AND DISCUSSION
■
First, the thermal stability of tungsten complex 19 in toluene
solutions was investigated. Complex 1, which was stable at
ambient temperature, started to decompose at elevated
temperature, as monitored by 31P{1H} NMR spectroscopy: at
50 °C two resonances of products appeared high field of 1 but
the conversion was quite slow. The same result with respect to
selectivity and product ratio could be obtained by heating
complex 1 for 3 h to 80 °C, leading to a much faster conversion
into the two main products 210 and 3 (ratio 1:1) (Scheme 2).
Figure 1. Molecular structure of complex 2 (50% probability level;
hydrogen atoms omitted for clarity). Selected bond lengths (in Å) and
angles (in deg): W1−P1 2.4823(12), W2−P2 2.4920(13) C1−P1
1.811(4), C7−P1 1.810(4), O1−P1 1.652(4), C18−P2 1.805(4),
C24−P2 1.818(4), O1−P2 1.636(4); P1−O1−P2 144.04(16), W1−
P1−C1 120.19(16), W1−P1−C7 110.61(15), W1−P1−O1
109.95(10), O1−P1−C1 107.09(19), O1−P1−C7 102.6(2), C1−
P1−C7 104.9(2), W2−P2−C18 118.68(16), W2−P2−C24
112.69(15), W2−P2−O1 110.74(11), O1−P2−C18 107.0(2), O1−
P2−C24 112.69(15), C18−P2−C24 103.6(2).
Scheme 2. Thermal Decomposition of Complex 1 in
Toluene
This reaction pointed strongly to the intermediacy of transient
phosphanoxyl complex 4, generated via O−N bond cleavage in
complex 1. After column chromatography, 2 and 3 were
obtained as beige powders and both compositions were
confirmed by X-ray analysis.
Complex 2 showed a resonance in the 31P{1H} NMR
spectrum at 128.0 ppm as part of a superposition of A2, ABX,
and AA′XX′ spin systems (A, A′, B = 31P, X = 183W); viable
coupling constants were derived from gNMR11a and the
WinDNMR11b simulation program (see the Supporting
Information). The molecular structure of 2 is shown in Figure
1.
Figure 2. Molecular structure of complex 3 (50% probability level;
hydrogen atoms omitted for clarity). Selected bond lengths (in Å) and
angles (in deg): W−P1 2.4894(8), C1−P1 1.812(3), C7−P1 1.818(3),
O1−P1 1.657(2), O2−P2 1.473(2), O1−P2 1.605(2), C13−P2
1.799(3), C19−P2 1.790(3); O1−P1−W 118.55(8), P1−O1−P2
127.40(14), O2−P2−O1 113.03(13), W−P1−C1 111.42(10), W−
P1−C7 119.08(11), O2−P2−C13 112.37(14), O2−P2−C19
115.28(14), O1−P−C1 97.06(13), C1−P1−C7 105.16(15), C13−
P2−O1 106.60(13), C19−P2−O1 100.90(13), C13−P2−C19
107.73(15).
Complex 3 displayed overlapping AB and ABX spin systems
(A, B = 31P, X = 183W) in the 31P{1H} NMR spectrum, showing
1
two doublets at 126.3 ppm (2JP,P = 38.6 Hz and JW,P = 296.2
Hz) and at 27.5 ppm, the latter without a tungsten-phosphorus
coupling. Figure 2 shows the molecular structure of complex 3,
obtained from single-crystal X-ray diffraction studies. The mass
spectrometry of 3 did not show the expected molecular ion
peak (m/z 726.0) but instead the loss of one carbonyl ligand at
the tungsten center (m/z 698.1). A LIFDI mass spectrometric
measurement also showed the same result. However, NMR
spectra and also elemental analysis prove the compound to be
as shown in Figure 2 (and Scheme 2).
The thermal decomposition pathway of tungsten complex 1,
involving only neutral closed-shell and radical species, were
calculated at the TPSS-D3/def2-QZVP+COSMO-RS-
(toluene)//TPSS-D3/def2-TZVP+COSMO(toluene) level of
theory (see the Supporting Information for a fuller description
of computational details).12 Briefly, all structures were
optimized at the TPSS-D3/def2-TZVP level using the
COSMO solvation model for toluene, followed by vibrational
frequency analysis to confirm the nature of obtained stationary
points (true minima and transition structures) and to provide
thermal free energy corrections. Spin densities for open-shell
radicals were computed according to a Mulliken population
analysis. Final free energies (298 K, 1 atm) were computed by
combining the TPSS-D3/def2-QZVP electronic energies, the
COSMO-RS solvation free energies in toluene, and the
aforementioned thermal free energy corrections and are used
throughout our discussion.
The computed reaction paths for the decomposition of
complex 1 via radical intermediates are depicted in Scheme 3.
Upon moderate heating in toluene, the homolytic O−N
cleavage of complex 1 is only 14.8 kcal/mol endergonic to form
phosphanoxyl radical 4 and is thus feasible, while P−O cleavage
to form phosphanyl radical 513 and direct W(CO)5 or CO
B
DOI: 10.1021/acs.organomet.7b00347
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 3. DFT Predicted Decomposition Reaction Paths of
Tungsten Complex 1
Scheme 4. Reaction of Complex 1 in Toluene with Different
Group 14 Hydrides
a
According to the 31P{1H} NMR spectroscopic monitoring
the reaction starts slowly at room temperature to give
selectively complex 6a in 63% conversion (according to
31P{1H} NMR integration) after 19 days. Gentle heating to
50 °C led to a faster and full conversion within 36 h, and the
product was obtained as an orange oily liquid after extraction
with n-pentane and washing at −50 °C. The formed amine
1
(TEMP-H) could be detected in the reaction mixture by H
NMR spectroscopy. Complex 6a showed a resonance in the
31P{1H} NMR spectrum at 101.5 ppm with a direct tungsten-
phosphorus coupling of 278.5 Hz and 2J couplings with the tin
2
2
nuclei: i.e. J(117Sn,31P) and J(119Sn,31P) values of 145.9 and
152.1 Hz, respectively. The 119Sn{1H} NMR spectrum
displayed a doublet at 142.2 ppm.
a
The free energy changes and barriers (ΔG and ΔG* in kcal/mol) are
shown above the arrows for each reaction step, respectively.
To obtain a trapping product that might possess a higher
tendency to crystallize, the tin substituents were changed to
phenyl. Heating a toluene solution of complex 1 to 50 °C with
triphenylstannane led selectively to complex 6b (Scheme 4).
The reaction mixture of 6b appeared as a brownish solution,
from which, after solvent removal in vacuo, a brown oily
compound was obtained. In the reaction mixture the amine
(TEMP-H) could be detected by 1H NMR spectroscopy. After
column chromatography at low temperatures a beige powder
was obtained. The 31P{1H} NMR spectrum of 6b showed a
elimination are highly endergonic and thus forbidden. The
resultant radicals TEMP (2,2,6,6-tetramethylpiperidinyl) and 4
(spin densities on central W, P and O: 0.32, 0.12 and 0.32 e,
respectively) are unable to form stable homodimers via
covalent N−N and O−O bonds or abstract a W(CO)5 unit
from complex 1. On the other hand, the radical 4 may easily
convert into its slightly less stable, P-centered radical form 4′
(spin densities on central W, P and O: 0.21, 0.31 and 0.22 e,
respectively) via a W(CO)5 shift. Without additional trapping
reagent, the exergonic O−W and O−P radical couplings
between two radicals 4 and between radicals 4 and 4′ may form
the complexes 2O and 3W, respectively. Note that complex 2O
has two relatively loose and weak W···O and W···P bonds
bound by only 19.4 and 10.9 kcal/mol, respectively. The
sterically less hindered terminal oxygen of 2O and W(CO)5
moiety of 3W can now easily be removed by the reactive radical
TEMP, eventually leading to the observed products 2 and 3,
respectively. Alternative complex 2 formation via O··P attack
between radical 4 and complex 1 followed by TEMPO release
is however kinetically less likely over an estimated barrier of
about 24 kcal/mol.
signal at 107.2 ppm with tungsten and tin satellites (1JW,P
=
2
2
281.3 Hz, J(117Sn,31P) = 157.8 Hz, J(119Sn,31P) = 165.0 Hz).
A doublet in the 119Sn{1H} NMR spectrum showed a
resonance at −81.5 ppm. The identity could be confirmed by
X-ray single-crystal structure determination, which is shown in
Figure 3.
To get further experimental evidence for the existence of the
transient phosphanoxyl complex 4 (or other radicals), EPR
measurements were performed by heating a solution of 1 in
toluene-d8 and following the decomposition. Unfortunately, we
obtained only broad and complex spectra, and interpretations
were not achieved (see the Supporting Information for more
information), but the presence of radicals could be confirmed.
Which role the latter play in the decomposition cannot be
judged fully at this stage.
To make use of the transient formation, thermal reactions of
P-nitroxyl complex 1 with various group 14 hydrides R3EH (E
= Si, Ge, Sn) were investigated. First, we examined the case of
tri-n-butylstannane on the basis of the ability of the transiently
formed tri-n-butylstannyl14 radical to act as an effective trap for
(other) radical species (Scheme 4).
Figure 3. Molecular structure of complex 6b (50% probability level;
hydrogen atoms omitted for clarity). Selected bond lengths (in Å) and
angles (in deg): W−P 2.4981(5), C19−P 1.821(2), C25−P 1.832(2),
O1−P 1.5822(13), O1−Sn 2.0279(13); P−O1−Sn 139.98(8), W−P−
C19 111.97(7), W−P−C25 119.86(6), W−P−O1 111.59(6), O1−P−
C19 103.18(8), O1−P−C25 106.56(8), C19−P−C25 102.04(9).
C
DOI: 10.1021/acs.organomet.7b00347
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
In the same manner as for the tin derivatives, the reaction
with the germanium derivative Ph3GeH was performed as well.
Heating a mixture of complex 1 and triphenylgermane in
toluene to 50 °C led to very good selectivity (95% according to
31P{1H} NMR integration) to complex 6c (Scheme 4). Here,
the amine could be proven to be formed during the reaction by
1H NMR spectroscopy, as in the reactions before. From the
brown reaction mixture a colorless powder could be isolated by
low-temperature column chromatography. A signal at 110.7
ppm in the 31P{1H} NMR spectrum with tungsten satellites
(1JW,P = 287.4 Hz) could be assigned to the proposed product,
from which the identity could be confirmed by an X-ray single-
crystal structure determination, shown in Figure S1 in the
Supporting Information.
large bond angles. Consistent with such a subtle difference
between C and its heavier analogues, slightly decreasing P−W
bond lengths are also observed in the order from Sn over Ge to
Si except for E = C.
Further extensive DFT calculations were also performed to
understand the observed selectivity and reactivity trends for the
trapping of phosphanoxyl complex 4 using the group 14
hydrides Ph3EH (E = Si, Ge, Sn). The predicted reaction paths
are compared in Scheme 5 for the tin and silicon derivatives.
Scheme 5. DFT Predicted Reaction Pathways for the
Reaction between Tungsten Complex 1 and Group 14
a
Hydrides Ph3EH (E = Si, Sn)
The case of 1 and triphenylsilane (see Scheme 4) showed a
different reaction behavior, and here the reaction with 1 equiv
of the silane led only to a conversion of 68% (31P{1H} NMR
integration) of the desired product after heating to 50 °C. In
addition to residual starting material, complexes 2 and 3 (about
10% each) were detected as well. Therefore, and to obtain a
higher selectivity, 2 equiv of the silane was used, which yielded
1
a 91% conversion to complex 6d. A H NMR spectrum of the
reaction solution showed the existence of the amine that is
formed in the reaction as well. After low-temperature column
chromatography a colorless powder was obtained showing a
resonance signal in the 31P{1H} NMR spectrum at 112.1 ppm
with tungsten satellites (1JW,P = 293.1 Hz). In the 29Si{1H}
NMR spectrum a doublet with a resonance signal at −12.7 ppm
2
appeared showing a J phosphorus silicon coupling of 19.0 Hz.
The composition of 6d could be confirmed by an X-ray single-
crystal structure determination, which is shown in Figure S3 in
the Supporting Information.
As all these products 6b−d are related to the lighter
homologue complex IV, obtained via trapping the phospha-
noxyl complex 4 with the trityl radical,9 a comparison of some
bond lengths and angles with the group 14 elements from
carbon to tin is presented in Table 1. It is clearly seen that the
a
The free energy changes and barriers (ΔG and ΔG* in kcal/mol) are
shown above the arrows for each reaction step, respectively, with red
italic numbers for E = Si.
Table 1. Comparison of Bond Lengths (in Å) and Bond
Angles (in deg) of Complexes 6b−d (E = Sn, Ge, Si) and
Their Lighter Homologue IV (E = C)
Initiated by the O−N bond cleavage of complex 1, the
resultant radicals 4 and TEMP turn out to be reactive with
respect to hydrogen atom abstraction from the E−H bond of
hydrides Ph3EH (E = Si, Sn). Both H abstractions from
Ph3SnH by 4 and TEMP are exergonic over a rather low barrier
to form the transient radical Ph3Sn•, with the latter being
kinetically less but thermodynamically more favorable. The
temporarily formed complex 4H ((CO)5W(Ph2POH)) can be
easily removed by another TEMP radical via H abstraction
from the O−H site to regenerate 4, which is −7.0 kcal/mol
exergonic and almost barrierless. The transient radical Ph3Sn•
can be easily coupled with radicals 4 and TEMP to form
complexes 6b and TEMP-SnPh3, respectively, with the latter
being easily removed by another radical 4 via radical
replacement at the tin center to form complex 6b. The overall
reaction is thus initiated by the O−N bond cleavage of complex
1, followed by formally selective H abstraction by TEMP from
Ph3EH and radical coupling between transient 4 and Ph3Sn•. A
very similar mechanism is also found for the reaction between
complex 1 and Ph3SiH, except for more difficult H abstractions
from the stronger Si−H bond that eventually leads to a 6.3
kcal/mol higher overall barrier at the rate-limiting H abstraction
no. (E)
d(O−E)
d(P−O)
d(P−W)
∠(P−O−E)
6b (Sn)
6c (Ge)
6d (Si)
IV (C)
2.0279(13)
1.809(3)
1.662(2)
1.460(3)
1.5822(13)
1.600(3)
2.4981(5)
2.4945(11)
2.4866(9)
2.5282(7)
139.98(8)
140.67(17)
146.27(17)
131.96(16)
1.609(2)
1.6282(18)
O−E (E = Sn, Ge, Si, C) bond lengths evidently decrease in the
order of decreasing atomic radii: Sn > Ge > Si > C. The
adjacent P−O bond lengths slightly but consistently increase,
likely due to increasing steric hindrance from the bulky EPh3
group from carbon to tin. Though the P−O−E bond angles are
increased in the order Sn < Ge < Si, the smallest angle of 132°
is found for E = C. DFT calculations showed that this is very
likely due to the lack of suitable empty d valence orbitals that
are present for heavier group 14 elements to accept π-electron
donation from central oxygen 2p lone pairs. Consistent with
the largest P−O−Si bond angle, the best O(2p) → E(nd) π-
electron donation is expected for E = Si due to both the short
O−Si bond length and low lying Si 3d orbital; due to similar
O(2p) → P(3d) π-electron donation, the P−O−E moieties
show some conjugated π-system nature, as evidenced by their
D
DOI: 10.1021/acs.organomet.7b00347
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Hz, ipso-CPh of [W]PPh), 196.8 (dSat, 2JP,C = 7.8 Hz, 1JW,C = 126.0 Hz,
step. This fact is consistent with the observed reactivity trend
for group 14 hydrides as radical trapping reagents.
2
1
cis-CO), 199.5 (dSat, JP,C = 28.2 Hz, JW,C = 138.7 Hz, trans-CO).
31P{1H} NMR (202.48 MHz, C6D6): δ 27.5 (dSat, 2JP,P = 38.6 Hz, 1JP,C
2
1
= 137.5 Hz, P(O)), 126.3 (dSat, JP,P = 38.6 Hz, JW,P = 296.2 Hz,
[W]P). IR (ν(CO), neat): ν
/cm−1 2073 (m), 1993 (w), 1917 (vs),
CONCLUSION
■
̃
It was shown that P-nitroxyl phosphane complexes undergo
homolytic bond cleavage upon gentle heating. The transient
phosphanoxyl complex, formed preferentially via O−N bond
cleavage, can be trapped effectively with group 14 hydrides,
giving the corresponding product complexes having a P−O−E
motif (E = Si, Ge, Sn). In the absence of any trapping reagent
two complexes are formed in a 1:1 ratio which are (mainly)
derived from phosphanyl and phosphanoxyl complexes and/or
rearranged species, which was analyzed via state of the art DFT
calculations. Furthermore, insights into the bond-breaking and
-forming processes of different transient radical species were
obtained. EPR measurements also showed that open-shell
molecules are formed during the thermolysis, but a satisfactory
assignment of their role was not achieved.
1233 (s, ν(O−P(O))), 893 (vs, ν(P−O−P)). MS: calcd for
C29H20O7P2184W, m/z 726.0; found (EI, 70 eV, 184W), m/z (%)
698.1 [M − CO]•+ (1); LIFDI-MS in toluene, m/z (%) 698.1 [M −
CO]•+ (100), 92.1 [toluene]+• (98). Anal. Calcd for C29H20O7P2W: C,
47.96; H, 2.78. Found: C, 48.02; H, 2.74.
Complex 6a. Complex 1 (665 mg, 1 mmol) was dissolved in
toluene (5 mL), and tri-n-butylstannane (292 mg,1 mmol) was added.
The solution was then heated to 50 °C for 36 h, and volatiles were
removed under reduced pressure (3 × 10−2 mbar) from the yellow
solution. The oily dark orange residue was extracted with n-pentane
and recrystallized from n-pentane at −50 °C to yield a light orange oil
1
(at room temperature). Yield: 312 mg (0.38 mmol, 38%). H NMR
(300.13 MHz, CDCl3): δ 0.88 (t, 9 H, CH3), 0.99−1.12 (m, 6 H, Sn-
CH2),), 1.19−1.34 (m, 6 H, CH2-CH3), 1.40−1.55 (m, 6 H, CH2-
CH2-CH2), 7.37−7.49 (m, 6 H, m- and p-HPh), 7.50−7.64 (m, 4 H, o-
HPh). 13C{1H} NMR (75.48 MHz, CDCl3): δ 13.7 (s, CH3), 17.6 (sSat,
1JSn,C = 334.2 Hz, 1JSn,C = 349.7 Hz, Sn-CH2), 27.1 (s, CH2-CH3), 27.7
EXPERIMENTAL SECTION
■
(s, CH2-CH2-CH2), 128.3 (d, 3JP,C = 9.4 Hz, m-CPh), 130.1 (d, 4JP,C
=
All manipulations involving air- and moisture-sensitive compounds
were carried out under an atmosphere of purified argon by using
standard Schlenk-line techniques or in a glovebox. Solvents were dried
with appropriate drying agents and degassed before use. The 1H,
13C{1H}, and 31P{1H} NMR spectroscopic data (δ in ppm) were
recorded unless otherwise noted at 25 °C on a Bruker DMX 300 or a
Bruker AV III 500 MHz Prodigy NMR spectrometer. The standard for
1H, 13C, and 29Si NMR is tetramethylsilane (Me4Si), for 31P NMR 85%
phosphorus acid (H3PO4), and for 119Sn NMR tetramethylstannane
(Me4Sn). Mass spectra were recorded on a MAT 95 XL Thermo
Finnigan spectrometer and a MAT 90 Thermo Finnigan sector
instrument equipped with a LIFDI ion source (Linden CMS)
(selected data given). Infrared spectra were recorded on a Thermo
Nicolet 380 FT-IR or a Bruker Alpha Diamond ATR FTIR
spectrometer (selected data given). Melting points were determined
using a Buchi Type S apparatus with samples sealed in capillaries
under argon and are uncorrected. Elemental analyses were performed
using an ElementarVarioEL instrument.
1.9 Hz, p-CPh), 130.4 (d, 2JP,C = 13.7 Hz, o-CPh), 144.8 (d, 1JP,C = 37.0
2
1
Hz, ipso-CPh), 198.1 (dSat, JP,C = 8.5 Hz, JW,C = 126.0 Hz, cis-CO),
201.3 (d, JP,C = 22.0 Hz, trans-CO). 31P{1H} NMR (121.51 MHz,
2
1
2
2
CDCl3): δ 101.5 (sSat, JW,P = 278.5 Hz, JSn,P = 145.91 Hz, JS2n,P
=
152.1 Hz). 119Sn{1H} NMR (111.87 MHz, CDCl3): δ 142.2 (d, JSn,P
= 152.1 Hz). IR (ν(CO), neat): ν
̃
/cm−1 2066 (m), 1977 (w), 1900
(vs). MS: calcd for C29H37O6P118Sn184W, m/z 814.0; found (EI, 70 eV,
118Sn, 184W), m/z (%) 814.1 [M]•+ (2).
Complex 6b. Complex 1 (665 mg, 1 mmol) was dissolved in
toluene (5 mL), and triphenylstannane (352 mg, 0.26 mL, 1 mmol)
was added. The solution was then heated to 50 °C for 36 h, and
volatiles were removed under reduced pressure (3 × 10−2 mbar) from
the brown solution. Column chromatography was performed at −20
°C (Al2O3, h = 3 cm, ø = 2 cm; eluent, petroleum ether/CH2Cl2 1/1),
yielding the product as a beige powder after evaporation of the solvent
and recrystallization from toluene and n-pentane at lower temperature.
1
Mp: 107 °C. Yield: 325 mg (0.37 mmol, 37%). H NMR (500.13
MHz, C6D6): δ 6.86−6.96 (m, 6 H, m- and p-HPh of PPh), 7.05−7.14
Complexes 2 and 3. Complex 1 (665 mg, 1 mmol) was dissolved
in toluene (5 mL) and the solution heated to 80 °C for 3 h. Volatiles
were removed under reduced pressure (3 × 10−2 mbar) from the dark
brown solution. The residue was purified by column chromatography
at −20 °C (SiO2, h = 6 cm, ø = 4 cm; eluent petroleum ether/Et2O 4/
1). Evaporation of the first fraction gave 2 as a beige solid after
washing with Et2O and n-pentane. Mp: 189 °C. Yield: 140 mg (0.14
mmol, 28%). 1H NMR (300.13 MHz, CDCl3): δ 7.31−7.51 (m, 20 H,
o-, m- and p-HPh). 13C{1H} NMR (75.48 MHz, CDCl3): δ 128.6 (“t”,
3
(m, 9 H, m- and p-HPh of SnPh), 7.33−7.49 (mSat, JSn,H = 52.9 Hz,
3JSn,H = 64.7 Hz, 6 H, o-HPh of SnPh) 7.49−7.58 (m, 4 H, o-HPh of
PPh). 13C{1H} NMR (125.77 MHz, C6D6): δ 128.4 (d, 3JP,C = 9.6 Hz,
m-CPh of PPh), 129.4 (sSat, 3JSn,C = 61.9 Hz, 3JSn,C = 64.7 Hz, m-CPh of
SnPh), 130.3 (d, 4JP,C = 1.8 Hz, p-CPh of PPh), 130.8 (sSat, 4JSn,C = 13.3
2
Hz, p-CPh of SnPh), 130.8 (d, JP,C = 13.7 Hz, o-CPh of PPh), 136.7
2
2
(sSat, JSn,C = 47.0 Hz, JSn,C = 48.7 Hz, o-CPh of SnPh), 137.6 (sSat,
1JSn,C = 611.2 Hz, 1JSn,C = 639.2 Hz, ipso-CPh of SnPh), 144.41(d, JP,C
1
2
1
2
3JP,C = 4.9 Hz, m-CPh), 131.1 (“t”, JP,C = 7.1 Hz, o-CPh), 131.4 (s, p-
= 37.1 Hz, ipso-CPh of PPh), 198.3 (dSat, JP,C = 8.4 Hz, JW,C = 125.8
Hz, cis-CO), 201.3 (dSat, 2JP,C = 22.6 Hz, 1JW,C = 139.8 Hz, trans-CO).
CPh), 139.0 (“t”, 1JP,C = 20.4 Hz, ipso-CPh), 196.5 (“t”Sat, 2JP,C = 7.6 Hz,
31P{1H} NMR (202.48 MHz, C6D6): δ 107.2 (sSat, JW,P = 281.3 Hz,
1
2
1JW,C = 126.3 Hz cis-CO), 221.0 (“t”, JP,C = 14.3 Hz, trans-CO).
2JSn,P = 157.8 Hz, JSn,P = 165.0 Hz). 119Sn{1H} inverse gated NMR
2
31P{1H} NMR (121.51 MHz, CDCl3): δ 128.0 (sSat, 1JW,P = 301.0 Hz,,
2
1
3
2JP,P = 64.7 Hz, JW,P = 1.4 Hz). IR (ν(CO), neat): ν
̃
/cm−1 2072 (s),
(186.50 MHz, C6D6): δ −81.5 (dSat, JSn,P = 165.0 Hz, JSn,C = 639.2
Hz). IR (ν(CO), neat): ν
̃
/cm−1 2067 (m), 1974 (s), 1907 (vs). MS:
1996 (w), 1904 (s), 857 (s, ν(P−O−P)). MS: calcd for
C34H20O11P2184W2, m/z 1034.2; found (EI, 70 eV, 184W) m/z (%)
1034.1 [M]•+ (27). Anal. Calcd for C34H20O11P2W2: C, 39.49; H, 1.95.
Found: C, 39.45; H, 1.92.
calcd for C35H25O6P118Sn184W, m/z 874.0; found (EI, 70 eV, 118Sn,
184W), m/z (%) 874.0 [M]•+ (11). Anal. Calcd for C35H25O6PSnW: C,
48.04; H, 2.88. Found: C, 48.00; H, 2.92.
Complex 6c. Complex 1 (665 mg, 1 mmol) was dissolved in
toluene (5 mL), and triphenylgermane (305 mg, 1 mmol) was added.
The solution was then heated to 50 °C for 41 h, and volatiles were
removed under reduced pressure (3 × 10−2 mbar) from the brown
solution. The oily brown residue was purified by column
chromatography at −20 °C (Al2O3, h = 3 cm, ø = 2 cm; eluent,
petrol eumether/CH2Cl2 1/0.2). Evaporation of the first fraction gave
a beige powder that was recrystallized from toluene and n-pentane at
lower temperature. Mp: 146 °C. Yield: 300 mg (0.36 mmol, 36%). 1H
NMR (500.13 MHz, C6D6): δ 6.87−6.94 (m, 6 H, m- and p-HPh of
PPh), 7.03−7.12 (m, 9 H, m- and p-HPh of GePh), 7.44−7.51 (m, 6 H,
Evaporation of the third fraction gave 3 as a beige solid after
washing with Et2O and n-pentane. Mp: 128 °C. Yield: 80 mg (0.11
mmol, 22%). H NMR (500.13 MHz, C6D6): δ 6.90−6.95 (m, 10 H,
1
m- and p-HPh of [W]PPh and m-HPh of (O)PPh), 6.95−7.00 (m, 2 H,
p-HPh of (O)PPh), 7.66−7.72 (m, 4 H, o-HPh of [W]PPh), 7.72−7.78
(m, 4 H, o-HPh of (O)PPh). 13C{1H} NMR (125.77 MHz, C6D6): δ
128.4 (d, 3JP,C = 10.3 Hz, m-CPh of [W]PPh), 128.7 (d, 3JP,C = 13.5 Hz,
m-CPh of (O)PPh), 131.4 (d, 2JP,C = 10.6 Hz, o-CPh of (O)PPh), 131.5
2
(d, 4JP,C = 1.9 Hz, p-CPh of [W]PPh), 132.0 (d, JP,C = 14.8 Hz, o-CPh
of [W]PPh), 132.1 (d, 4JP,C = 2.9 Hz, p-CPh of (O)PPh), 133.2 (d, 1JP,C
= 137.5 Hz, ipso-CPh of (O)PPh), 138.4 (dd, 1JP,C = 41.0 Hz, 3JP,C = 2.0
E
DOI: 10.1021/acs.organomet.7b00347
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
o-HPh of GePh), 7.51−7.57 (m, 4 H, o-HPh of PPh). 13C{1H} NMR
Notes
(125.77 MHz, C6D6): δ 127.9 (d, 3JP,C = 9.6 Hz, m-CPh of PPh), 128.5
The authors declare no competing financial interest.
4
(s, m-CPh of GePh), 130.1 (d, JP,C = 1.8 Hz, p-CPh of PPh), 130.4 (s,
2
p-CPh of GePh), 130.8 (d, JP,C = 13.9 Hz, o-CPh of PPh), 133.9 (s,
ACKNOWLEDGMENTS
ipso-CPh of GePh), 134.4 (s, o-CPh of GePh), 142.1 (d, 1JP,C = 38.2 Hz,
ipso-CPh of PPh), 197.6 (dSat, 2JP,C = 8.2 Hz, 1JW,C = 125.9 Hz, cis-CO),
■
We are grateful to the Deutsche Forschungsgemeinschaft (SFB
813 “Chemistry at Spin Centers”) for financial support. G.S.
thanks Prof. Dr. A. C. Filippou for support. We dedicate this
work to Prof. G. Bertrand on the occasion of his 65th birthday.
2
1
200.1 (dSat, JP,C = 24.0 Hz, JW,C = 138.9 Hz, trans-CO). 31P{1H}
1
NMR (202.48 MHz, C6D6): δ 110.7 (sSat, JW,P = 287.4 Hz). IR
(ν(CO), neat): ν
̃
/cm−1 2070 (s), 1917 (vs). MS: calcd for
C35H2572GeO6P184W, m/z 828.0; found (EI, 70 eV, 72Ge, 184W), m/
z (%) 827.9 [M]•+ (13). Anal. Calcd for C35H25GeO6PW: C, 50.71; H,
3.04. Found: C, 50.75; H, 3.06.
REFERENCES
■
(1) For a few selected reviews on nitroxides, see: (a) Vogler, T.;
Studer, A. Synthesis 2008, 2008, 1979−1993. (b) Ciriminna, R.;
Pagliaro, M. Org. Process Res. Dev. 2010, 14, 245−251. (c) Tebben, L.;
Studer, A. Angew. Chem., Int. Ed. 2011, 50, 5034−5068; Angew. Chem.
2011, 123, 5138−5174 and references cited therein.
(2) (a) Braunecker, W. A.; Matyjaszewski, K. Prog. Polym. Sci. 2007,
32, 93−146. (b) Shipp, D. A. Polym. Rev. 2011, 51, 99−103.
(c) Sciannamea
Complex 6d. Complex 1 (665 mg, 1 mmol) was dissolved in
toluene (5 mL), and triphenylsilane (520 mg, 2 mmol) was added.
The solution was then heated to 50 °C for 46 h, and volatiles were
removed under reduced pressure (3 × 10−2 mbar) from the brown
solution. The brown oily residue was further purified by column
chromatography at −20 °C (SiO2, h = 2 cm, ø = 2 cm; eluent,
petroleum ether/CH2Cl2 1/0.5). Evaporation of the first fraction gave
a beige powder that was further recrystallized from toluene and n-
pentane at lower temperature. Mp: 159 °C. Yield: 350 mg (0.45 mmol,
45%). 1H NMR (500.13 MHz, C6D6): δ 6.88−6.95 (m, 6 H, m- and p-
HPh of PPh), 7.07−7.13 (m, 6 H, m- and p-HPh of SiPh), 7.13−7.19
(m, 3 H, p-HPh of SiPh), 7.52−7.68 (m, 10 H, o-HPh of SiPh and PPh).
13C{1H} NMR (125.77 MHz, C6D6): δ 128.4 (s, m-CPh of SiPh), 128.5
(d, 3JP,C = 10.0 Hz, m-CPh of PPh), 130.8 (s, p-CPh of SiPh), 130.9 (d,
4JP,C = 1.8 Hz, p-CPh of PPh), 131.0 (d, 2JP,C = 14.6 Hz, o-CPh of PPh),
́ ́ ̂
, V.; Jerome, R.; Detrembleur, C. Chem. Rev. 2008, 108,
1104−1126.
(3) For examples of TEMPO bound to main-group-element
derivatives, see the following. (a) For B derivatives, see: Armbrecht,
M.; Maringgele, W.; Meller, A.; Noltemeyer, M.; Sheldrick, G. M. Z.
Naturforsch., B: J. Chem. Sci. 1985, 40, 1113−1122. (b) Ollivier, C.;
Huard, R. C.; Renaud, P. Synlett 1999, 1999, 807−809. (c) For Si
derivatives, see: Losehand, U.; Mitzel, N. W.; Rankin, D. W. H. J.
Chem. Soc., Dalton Trans. 1999, 4291−4297. (d) For Li, Na, and Mg
derivatives, see: Forbes, G. C.; Kennedy, A. R.; Mulvey, R. E.; Rodger,
P. J. A. Chem. Commun. 2001, 1400−1401. (e) For Sn, Ge, and Si
derivatives, see: Lucarini, M.; Marchesi, E.; Pedulli, G. F.;
Chatgilialoglu, C. J. Org. Chem. 1998, 63, 1687−1693. (f) For Al
and Ga derivatives, see: Jones, C.; Rose, R. P. New J. Chem. 2007, 31,
1484−1487. (g) For Sb and Bi derivatives, see: Ishida, S.; Hirakawa,
F.; Furukawa, K.; Yoza, K.; Iwamoto, T. Angew. Chem., Int. Ed. 2014,
53, 11172−11176; Angew. Chem. 2014, 126, 11354−11358.
(4) Ishida, S.; Hirakawa, F.; Iwamoto, T. Chem. Lett. 2015, 44, 94−
96.
1
133.4 (sSat, JC,Si = 83.1 Hz, ipso-CPh of SiPh), 135.9 (s, o-CPh of
GePh), 141.6 (d, 1JP,C = 41.5 Hz, ipso-CPh of PPh), 197.3 (dSat, 2JP,C
8.0 Hz, 1JW,C = 125.8 Hz, cis-CO), 199.5 (dSat, 2JP,C = 25.5 Hz, 1JW,C
=
=
137.9 Hz, trans-CO). 29Si{1H} DEPT20 NMR (99.36 MHz, C6D6): δ
2
1
−12.7 (dSat, JP,Si = 19.0 Hz, JC,Si = 83.1 Hz). 31P{1H} NMR (202.48
1
MHz, C6D6): δ 112.1 (sSat, JW,P = 293.1 Hz). IR (ν(CO), neat): ν
̃
/
cm−1 2073 (m), 1983 (w), 1921 (vs). MS: calcd for C35H25O6PSi184W,
m/z 784.1; found (EI, 70 eV, 184W), m/z (%) 784.1 [M]•+ (13). Anal.
Calcd for C35H25O6PSiW: C, 53.59; H, 3.21. Found: C, 53.60; H, 3.29.
ASSOCIATED CONTENT
■
(5) The reaction of Ph2PH with TEMPO has already been described,
and the P-OTEMP substituted PIII compound was proposed as an
intermediate, but without further evidence: Baker, R. J.; Hashem, E.
Helv. Chim. Acta 2010, 93, 1081−1085.
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00347.
̌
́
(6) Heurich, T.; Qu, Z.-W.; Nozinovic, S.; Schnakenburg, G.;
Matsuoka, H.; Grimme, S.; Schiemann, O.; Streubel, R. Chem. - Eur. J.
2016, 22, 10102−10110.
Crystallographic information for complexes 2, 3, and
6b−d, NMR simulation of complex 2, NMR spectra of
the obtained compounds, information on the EPR
results, and details of the theoretical calculations
(PDF)
Cartesian coordinates of the calculated molecules,
transition states and intermediates(PDF)
(7) Siu, P. W.; Serin, S. C.; Krummenacher, I.; Hey, T. W.; Gates, D.
P. Angew. Chem., Int. Ed. 2013, 52, 6967−6970; Angew. Chem. 2013,
125, 7105−7108.
(8) (a) Nesterov, V.; Schwieger, S.; Schnakenburg, G.; Grimme, S.;
Streubel, R. Organometallics 2012, 31, 3457−3459. (b) Nesterov, V.;
Qu, Z.-W.; Schnakenburg, G.; Grimme, S.; Streubel, R. Chem.
Commun. 2014, 50, 12508−12511.
(9) Heurich, T.; Nesterov, V.; Schnakenburg, G.; Qu, Z.-W.;
Grimme, S.; Hazin, K.; Gates, D. P.; Engeser, M.; Streubel, R.
Angew. Chem., Int. Ed. 2016, 55, 14439−14443. Angew. Chem. 2016,
128, 14654−14658
(10) Complex 2 was obtained first via a different route, but no crystal
structure was described: Zeiher, C.; Mohyla, J.; Lorenz, I.-P.; Hiller, W.
J. Organomet. Chem. 1985, 286, 159−170.
(11) (a) Budzeelar, P. H. gNMR, Version 5.0.6.0; IvorySoft, 2006.
(b) Reich, H. J. J. Chem. Educ. 1995, 72, 1086.
Accession Codes
CCDC 1545939−1545942 and 974093 contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge via www.ccdc.cam.ac.uk/data_request/
cif, or by emailing data_request@ccdc.cam.ac.uk, or by
contacting The Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(12) (a) Tao, J. M.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E.
Phys. Rev. Lett. 2003, 91, 146401. (b) Grimme, S.; Antony, J.; Ehrlich,
S.; Krieg, H. J. Chem. Phys. 2010, 132, 154104. (c) Grimme, S.;
Goerigk, L. J. Comput. Chem. 2011, 32, 1456−1465. (d) Schaefer, A.;
Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100, 5829−5835.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail for Z.-W.Q.: qu@thch.uni-bonn.de.
*E-mail for R.S.: r.streubel@uni-bonn.de.
(e) Weigend, F.; Haser, M.; Patzelt, H.; Ahlrichs, R. Chem. Phys. Lett.
̈
ORCID
1998, 294, 143−152. (f) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem.
Phys. 2005, 7, 3297−3305. (g) Weigend, F.; Ahlrichs, R. Phys. Chem.
Chem. Phys. 2006, 8, 1057−1065. (h) Eichkorn, K.; Weigend, F.;
Stefan Grimme: 0000-0002-5844-4371
Rainer Streubel: 0000-0001-5661-8502
F
DOI: 10.1021/acs.organomet.7b00347
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Treutler, O.; Ahlrichs, R. Theor. Chem. Acc. 1997, 97, 119−124.
(i) Deglmann, P.; May, K.; Furche, F.; Ahlrichs, R. Chem. Phys. Lett.
2004, 384, 103−107. (j) TURBOMOLE V7.0 2015, a development of
the University of Karlsruhe and Forschungszentrum Karlsruhe GmbH,
1989−2007, TURBOMOLE GmbH, since 2007; available from
http://www.turbomole.com. (k) Klamt, A.; Schuurmann, G. J. Chem.
̈
̈
Soc., Perkin Trans. 2 1993, 799−805. (l) Grimme, S. Chem. - Eur. J.
2012, 18, 9955−9964. (m) Eckert, F.; Klamt, A. COSMOtherm,
Version C3.0, Release 14.01; COSMOlogic GmbH & Co. KG,
Leverkusen, Germany, 2013. (n) Klamt, A. J. Phys. Chem. 1995, 99,
2224−2235. (o) Eckert, F.; Klamt, A. AIChE J. 2002, 48, 369−385.
(13) For complex 5, formed via X-ray radiolysis from
(CO)5W(PPh3) and identified by EPR in the solid state, see:
Ndiaye, B.; Bhat, S.; Jouaiti, A.; Berclaz, T.; Bernardinelli, G.; Geoffroy,
M. J. Phys. Chem. A 2006, 110, 9736−9742.
(14) RajanBabu, T. V.; Bulman Page, P. C.; Buckley, B. R. e-EROS
Encyclopedia of Reagents for Organic Synthesis, 2004; Tri-n-butylstan-
nane.
G
DOI: 10.1021/acs.organomet.7b00347
Organometallics XXXX, XXX, XXX−XXX