Chlorination of Phosphane Selenides

Article abstract of DOI:10.1002/zaac.201400333

Chlorination of the phosphane selenide iPrtBu₂PSe with PhICl₂ leads to the compound iPrtBu₂PSe₂Cl₂, which contains a P-Se-SeCl₂ moiety and is formally a phosphane selenide complex of selenium dichloride. Two polymorphs of the product were identified by X-ray structure analysis. The analogous reaction with tBu₃PSe gave an oil, presumably tBu₃PSe₂Cl₂, from which small quantities of tBu₃PSe₃Cl₂ were obtained and characterized by X-ray structure analysis.

Full text of DOI:10.1002/zaac.201400333

ARTICLE
DOI: 10.1002/zaac.201400333
Chlorination of Phosphane Selenides
Daniel Upmann[a] and Peter G. Jones*[a]
Dedicated to Professor Martin Jansen on the Occasion of His 70th Birthday
Keywords: Phosphane; Selenide; Chloride; Solid state structure
Abstract. Chlorination of the phosphane selenide iPrtBu2PSe with
PhICl2 leads to the compound iPrtBu2PSe2Cl2, which contains a
P–Se–SeCl2 moiety and is formally a phosphane selenide complex of
selenium dichloride. Two polymorphs of the product were identified
by X-ray structure analysis. The analogous reaction with tBu3PSe gave
an oil, presumably tBu3PSe2Cl2, from which small quantities of
tBu3PSe3Cl2 were obtained and characterized by X-ray structure analy-
sis.
Although we have not investigated phosphane tellurides
here, it should also be pointed out that halogen adducts of
triethylphosphane telluride, Et3PTeX2, have been isolated and
structurally characterized for X = Cl, Br, and I; they exhibit
the structure type B and are linked into dimers by Te···X con-
tacts.[6]
Introduction
Phosphane chalcogenides R3P=E (E = S, Se) are known to
form adducts with dihalogen molecules.[1,2] [N.B. For brevity
we denote phosphanes by the general formula R3P, even if the
R groups are not identical]. A considerable number of 1:1 ad-
ducts of phosphane selenides have been isolated because of
their stronger donor abilities. These adducts can have various
forms (Scheme 1), such as a molecular (A and B) or an ionic
form (C and D).[3] For phosphane sulfides only a few examples
of adducts with iodine exist in the literature (form A).[1,4]
Recently we reported the first synthesis of R3PEX+ cations
–
(E = S, Se; X = Cl, Br) as their AuX4 salts, through oxidation
of the corresponding gold(I) complexes with iodobenzene di-
chloride or elemental bromine.[7] The chlorine derivatives rep-
resent the first examples of compounds with a stable P–E–Cl
unit. As a logical consequence of this work we have now fo-
cussed on the direct chlorination of phosphane chalcogenides
to form chlorine adducts.
Results and Discussion
Reacting iodobenzene dichloride with phosphane sulfides in
a 1:1 ratio leads to decomposition of the starting material and
the formation of R3PCl2 and separation of elemental sulfur; the
31P NMR spectra show only the signal for R3PCl2,[8] between δ
= 120 and 130 ppm for the bulky aliphatic R groups employed
herein.[9] Even with three sterically demanding tert-butyl
groups the decomposition upon chlorination is rapid. This
shows the strong tendency of chlorine to attack directly at the
more electropositive phosphorus atom. In case of the selenides
a second product 1, R3PSe–SeCl2, is formed (Scheme 2),
whose stability depends on the nature of the R groups at the
phosphorus atom. For 1a the starting material decomposes
upon chlorination and the 31P NMR spectrum shows only a
singlet at δ = 65.9 ppm for Ph3PCl2. Compound 1b decom-
poses slowly in solution and upon crystallization, but its 31P
NMR signal at δ = 86.5 ppm can be observed. In case of 1c
the compound is stable and can be crystallized from dichloro-
methane/pentane. The product 1d is also stable, but is formed
as orange oil. After a few weeks red single crystals of com-
Scheme 1. Various forms of halogen adducts of phosphane selenides.
–
(X = Br or I; Ix = various polyiodides).[5]
* Prof. Dr. P. G. Jones
Fax: +49-531-5387
E-Mail: p.jones@tu-bs.de
[a] Institut für Anorganische und Analytische Chemie
TU Braunschweig
Hagenring 30
38106 Braunschweig, Germany
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/zaac.201400333 or from the au-
thor.
2776
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2014, 640, (14), 2776–2780

Chlorination of Phosphane Selenides
pound 2 (R3PSe–Se–SeCl2, see below) formed in the oil and the second selenium atom T-shaped. The associated bond
their crystal structure was determined. However, dissolving the lengths and angles (see Figure captions) may be regarded as
crystals for 31P NMR investigations led to decomposition.
“normal”, although Se–Se and Se–Cl bond lengths are very
variable depending on coordination number and the formation
of short contacts leading to polymeric materials.[13] The P–Se
bonds, 2.25–2.26 Å, are appreciably lengthened in comparison
to isolated phosphane selenides (database average value
2.110 Å) and may be compared with 2.27–2.28 Å in a cationic
P+–Se–Se–P+ system[14] and 2.25 Å in cations R3PSeCl+.[7b]
The three molecules of 1c/1cЈ are similar except for some dif-
ferences in Se–Cl bond lengths and, particularly, the P–Se–Se–
Cl torsion angles for 1c, which differ by ca. 15°. The alkyl
groups are arranged such that the isopropyl groups make an
absolute C–P–Se–Se torsion angle of ca. 50°, whereas one tert-
butyl group is antiperiplanar across the same bond. A least-
squares fit for the two molecules of 1c gives a root-mean-
square deviation of 0.13 Å omitting the chlorine atoms, while
a similar fit of 1cЈ to molecule 2 of 1c gives an r.m.s.d. of
0.20 Å including the chlorine atoms.
Scheme 2. Chlorination of phosphane selenides.
The products 1 are to the best of our knowledge the first
neutral molecules containing the linkage –Se–SeCl2 and may
be regarded as adducts LSeCl2 between SeCl2 and phosphane
selenide ligands L. Such compounds are already known for L
= tetramethylthiourea,[10] for which LSeCl2 was synthesized
directly from SeCl2, and for L = dimethylsulfide[11] or various
N-heterocyclic carbenes,[12] for which the adducts were syn-
thesized via reductive processes from SeCl4. We tried to repro-
duce the synthesis of 1c using two different approaches. In the
first, we treated elemental selenium with PhICl2 and then
added iPrtBu2PSe. The mixture was stirred for 10 min, but the
31P NMR spectrum showed mostly a signal for the decomposi-
tion product and only traces of 1c. In the second approach[10]
we used in situ generated SeCl2 and added the phosphane sele-
nide. Soon after adding the phosphane selenide, the formation
of red selenium was observed and the 31P NMR spectrum
showed only the signal for R3PCl2.
Two polymorphs of 1c were obtained. The first crystallizes
in the triclinic space group P1 with two molecules in the asym-
Figure 2. Asymmetric unit of 1cЈ (hydrogen atoms omitted). Selected
molecular dimensions /Å,°: Se2–P1 2.2607(3), Se2–Se1 2.32079(16),
Se1–Cl1 2.4494(3), Se1–Cl2 2.4297(3); Cl1–Se1–Cl2 164.990(12),
P1–Se2–Se1–Cl1 –86.01(1), P1–Se2–Se1–Cl2 95.68(1).
¯
metric unit (Figure 1). The second polymorph (1cЈ) crystallizes
in the monoclinic space group P21/c with one molecule in the
asymmetric unit (Figure 2). In both cases the arrangement
around the selenium atom bonded directly to the phosphorus
atom is bent, with an approximately tetrahedral angle, and at
The SeCl2 adducts of tetramethylthiourea[10] and dimethyl
sulfide[11] display packing patterns that are characterized by
short Se···Cl contacts of 3.276 Å, forming dimers, and
3.085 Å, forming chains, respectively. All contacts between the
heavier atoms of 1c and 1cЈ are appreciably longer than this;
there are also a considerable number of H···X contacts (X = Cl,
Se), the shortest being around 2.8 Å, that might be regarded as
“weak” hydrogen bonds. The packing of 1c shows contacts
between chlorine atoms and selenium (molecule 2) and also
between two selenium atoms (molecule 1), which leads in both
cases to the formation of dimers (Figure 3, Figure 4). In the
packing of 1cЈ the molecules form chains through one selen-
ium-chlorine contact and one H···Cl hydrogen bond (Figure 5).
The crystal structure of 2 reveals the newly formed PSeSe-
SeCl2 unit, which must have arisen following a slow decompo-
sition of 1d. Compound 2 crystallizes in the monoclinic space
group P21/c with one molecule in the asymmetric unit (Fig-
ure 6). The arrangement at the two two-coordinate selenium
atoms is bent and at the third selenium atom T-shaped. Bond
lengths and angles are broadly similar to those of 1c/1cЈ; the
Figure 1. Asymmetric unit of 1c (hydrogen atoms omitted). Ellipsoids
in all diagrams correspond to 50% probability levels. Selected molecu-
lar dimensions /Å,°: Se2–P1 2.2563(6), Se2–Se1 2.3257(3), Se1–Cl1
2.4695(7), Se1–Cl2 2.4002(7), Se4–P2 2.2616(6), Se4–Se3 2.3207(3),
Se3–Cl3 2.4569(6), Se3–Cl4 2.4254(6); Cl1–Se1–Cl2 169.01(2), Cl3–
Se3–Cl4 167.00(2), P1–Se2–Se1–Cl1 –78.04(2), P1–Se2–Se1–Cl2
105.14(2), P2–Se4–Se3–Cl3 93.99(2), P1–Se4–Se3–Cl4 –91.31(2).
Z. Anorg. Allg. Chem. 2014, 2776–2780
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
2777

D. Upmann, P. G. Jones
ARTICLE
(P)Se–Se bond is significantly longer than Se–Se(Cl). The
packing involves the formation of dimers via a selenium-chlo-
rine contact (Figure 7).
Figure 3. Dimers in 1c, molecule 1 (hydrogen atoms omitted). Se-
lected contact distances /Å and angles /°: Se2···Se2A 3.5470(4),
Se2···Se1A 3.8740(3); P1–Se2···Se2A 150.00(2), Se1–Se2···Se2A
(76.82(1).
Figure 7. Dimers in 2 (hydrogen atoms omitted). Selected contact dis-
tance /Å and angle /°: Se2···Cl2A 3.3451(6); Se1–Se2···Cl2A
116.58(1).
Conclusions
Direct halogenation of phosphane sulfides with the chlori-
nating agent PhICl2 is not possible because of the strong ten-
dency of chlorine to attack at the electropositive phosphorus
atom. However, in the case of the selenides some unexpected
adducts R3PSenCl2 (n = 2, 3) are formed.
Figure 4. Dimers in 1c, molecule 2 (hydrogen atoms omitted). Se-
lected contact distances /Å and angles /°: Se4···Se4A 3.5470(4),
Se4···Cl3A 3.6184(6); Se4A···Se4···Cl3A 58.57(1).
Experimental Section
Materials and Measurements: All chemical reagents were commer-
cially available. Solvents were dried and distilled in a nitrogen atmo-
sphere prior to use: dichloromethane (P4O10), pentane and diethyl ether
(SPS-Machine from MBraun). CHN elemental analyses were carried
out with an Elementar Vario MICRO (Germany) elemental analyzer.
1H NMR and 31P NMR spectra were recorded at room temperature
with a Bruker DPX-200 spectrometer.
Figure 5. Chains in 1cЈ (hydrogen atoms omitted). Selected contact
distances /Å and angle /°: Se2···Cl1 3.6102(3), H31A···Cl2 2.78; C31–
H31A···Cl2 175.9.
General Procedure: All reactions were carried out at room tempera-
ture in a nitrogen atmosphere. To a solution of the phosphane chalco-
genide in dichloromethane, PhICl2 was added in a 1:1 ratio and the
mixture was stirred for 30 min at room temperature. The solvent was
evaporated in vacuo and the residue analysed by 31P NMR spec-
troscopy. A signal for the R3PCl2 usually appeared between 120–
130 ppm (in the case of the triphenyl derivative at δ = 65.9 ppm).
Unfortunately only the product 1c was stable enough to be reproduci-
bly synthesized, whereas 2 was only obtained once as a few single
crystals, resulting from decomposition of 1d.
Synthesis of 1c: To a solution of iPrtBu2PSe (230 mg, 0.9 mmol) in
dichloromethane (20 mL), PhICl2 (237 mg, 0.9 mmol) was added and
the reaction mixture was stirred for 30 min. The solvent was evapo-
rated in vacuo and the yellow oily crude product was washed several
times with diethyl ether (4ϫ5 mL). 1c was re-crystallized from dichlo-
romethane and pentane, giving a yellow powder (yield: 168 mg, 48%;
max. 50% yield is possible because of the stoichiometric P:Se imbal-
ance between starting materials and product). Diffusion of pentane into
a dichloromethane solution gave yellow single crystals. 1H NMR
Figure 6. Crystal structure of 2 (hydrogen atoms omitted). Selected
molecular dimensions /Å,°: Se3–P1 2.2453(6), Se3–Se2 2.3581(3),
Se2–Se1 2.3132(3), Se1–Cl1 2.4708(6), Se1–Cl2 2.4168(6); Cl1–Se1–
Cl2 174.29(2), P1–Se3–Se2–Se1 114.24(2), Se3–Se2–Se1–Cl1
59.39(2), Se3–Se2–Se1–Cl2 –118.19(1).
(200 MHz, CDCl3): δ = 4.10–3.86 (m, 1 H, CHCH3), 1.71 (d, JPH
=
16.9 Hz, 18 H, CCH3), 1.70–1.53 (m, 6 H, CHCH3). The signal of the
2778
www.zaac.wiley-vch.de
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2014, 2776–2780

Chlorination of Phosphane Selenides
Table 1. Crystal data and structure refinement for 1c, 1cЈ, and 2.
1c
1cЈ
2
Formula
Temperature /K
C11H25Cl2PSe2
100(2)
C11H25Cl2PSe2
100(2)
C12H27Cl2PSe3
100(2)
Wavelength /Å
0.71073
0.71073
0.71073
Crystal system
Space group
triclinic
P1
monoclinic
P21/c
monoclinic
P21/c
¯
Unit cell dimensions
a /Å
b /Å
c /Å
α /°
9.4388(3)
8.0234(2)
16.1806(3)
12.9830(2)
90
14.3170(4)
8.8361(2)
15.7361(5)
90
12.7374(4)
13.9910(5)
81.746(3)
β /°
86.117(4)
84.547(2)
1654.63(9)
4
1.674
93.578(2)
90
1682.23(6)
4
1.647
4.8
107.982(3)
90
1893.50(9)
4
1.789
6.2
γ /°
V /Å3
Z
Density (calcd.) /Mg·m–3
Absorption coeff. /mm–1
F(000)
4.9
832
832
1000
Crystal size /mm3
Theta range /°
Index range
0.2ϫ0.08ϫ0.02
2.34 to 30.03
–13 Յ h Յ 13,
–17 Յ k Յ 17,
–19 Յ l Յ 19
98.5% to 60°
9537 / 305
1.03
0.18ϫ0.12ϫ0.09
2.52 to 30.90
–11 Յ h Յ 11,
–22 Յ k Յ 23,
–18 Յ l Յ 18
96.2% to 2θmax
5118 / 153
1.08
0.20ϫ0.18ϫ0.05
2.67 to 30.90
–19 Յ h Յ 20,
–12 Յ k Յ 12,
–22 Յ l Յ 22
98.9% to 60°
5665 / 172
1.05
Completeness
Data / parameters
GOOF on F2
Final R indices [I Ͼ 2σ(I)]
R1 = 0.0286,
wR2 = 0.0597
R1 = 0.0389,
wR2 = 0.0630
1.9 and –0.60
R1 = 0.0176,
wR2 = 0.0356
R1 = 0.0222,
wR2 = 0.0368
0.39 and –0.30
R1 = 0.0264,
wR2 = 0.0537
R1 = 0.0375,
wR2 = 0.0572
0.65 and –0.64
R indices (all data)
Largest diff. peak, hole /e·Å–3
one methine proton, split into two septets, was not entirely clear, and
Acknowledgements
there was overlap in the methyl region. 31P NMR (81 MHz, CDCl3):
1
2
δ = 86.4 (s, JPSe = 213.2, JPSe = 168.5 Hz). C11H25PSe2Cl2: calcd. C
31.67, H 6.04%; found: C 31.63, H 6.00%. Not enough material was
obtained for 77Se NMR studies. The second polymorph 1cЈ was ob-
tained by chance on repeating the synthesis under apparently the same
conditions. No attempts were made to isolate a particular polymorph
reproducibly, or to investigate thermodynamic relationships between
the polymorphs; we note however that 1cЈ has a significantly lower
calculated density than 1c and thus may be thermodynamically less
stable.
We are grateful to the referees for constructive criticism.
References
[1] D. C. Apperley, N. Bricklebank, S. L. Burns, D. E. Hibbs, M. B.
Hursthouse, K. M. Abdul Malik, J. Chem. Soc., Dalton Trans.
1998, 1289–1292.
[2] a) R. A. Zingaro, E. A. Meyers, Inorg. Chem. 1962, 1, 771–774;
b) D. J. Williams, K. J. Wynne, Inorg. Chem. 1976, 15, 1449–
1451; c) M. Arca, F. A. Devillanova, A. Garau, F. Isaia, V. Lip-
polis, G. Verani, F. Demartin, Z. Anorg. Allg. Chem. 1998, 624,
745–749.
[3] a) J. Jeske, W.-W. du Mont, P. G. Jones, Chem. Eur. J. 1999, 5,
385–389; b) C. W. Perkins, J. C. Martin, A. J. Arduengo, W. Lau,
A. Alegria, J. K. Kochi, J. Am. Chem. Soc. 1980, 102, 7753–
7759; c) E. Seppälä, F. Ruthe, J. Jeske, W.-W. du Mont, P. G.
Jones, Chem. Commun. 1999, 1471–1472; d) S. M. Godfrey, S. L.
Jackson, C. A. McAuliffe, R. G. Pritchard, J. Chem. Soc., Dalton
Trans. 1998, 4201–4204; e) W.-W. du Mont, A. Martens-
von Salzen, F. Ruthe, E. Seppälä, G. Mugesh, F. A. Devillanova,
V. Lippolis, N. Kuhn, J. Organomet. Chem. 2001, 623, 14–28; f)
W.-W. du Mont, A. Martens-von Salzen, F. Ruthe, E. Seppälä, G.
Mugesh, F. A. Devillanova, V. Lippolis, N. Kuhn, J. Organomet.
Chem. 2001, 628, 280; g) S. M. Godfrey, S. L. Jackson, C. A.
McAuliffe, R. G. Pritchard, J. Chem. Soc., Dalton Trans. 1997,
4499–4502.
X-ray Structure Determination: Data were collected at 100 K with
an Oxford Diffraction Xcalibur E diffractometer using monochromated
Mo-Kα radiation. Structures were refined using the program SHELXL-
97.[15] The structures were refined anisotropically using full-matrix le-
ast-squares on F2. Hydrogen atoms were included using idealized rigid
methyl groups or a riding model for non-methyl H. Crystallographic
data and structure refinement results are summarized in Table 1.
Crystallographic data (excluding structure factors) for the structures in
this paper have been deposited with the Cambridge Crystallographic
Data Centre, CCDC, 12 Union Road, Cambridge CB21EZ, UK.
Copies of the data can be obtained free of charge on quoting the de-
pository numbers CCDC-1014066 (1c), CCDC-1014067 (1cЈ),
and CCDC-1014068 (2) (Fax: +44-1223-336-033; E-Mail:
deposit@ccdc.cam.ac.uk, http://www.ccdc.cam.ac.uk)
[4] N. Bricklebank, S. J. Coles, S. D. Forder, M. B. Hursthouse, A.
Poulton, P. J. Skabara, J. Organomet. Chem. 2005, 690, 328–332.
[5] A referee has pointed out, quite correctly, the discrepancy be-
tween presenting phosphane chalcogenides as R3P=E (historically
Supporting Information (see footnote on the first page of this article):
31P NMR spectra and crystallographic tables.
Z. Anorg. Allg. Chem. 2014, 2776–2780
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
2779

D. Upmann, P. G. Jones
ARTICLE
conditioned) and their derivatives as, for example, R3P–E–X–X.
[10] A. Maaninen, T. Chivers, M. Parvez, J. Pietikäinen, R. S. Lait-
inen, Inorg. Chem. 1999, 38, 4093–4097.
[11] A. Jolleys, W. Levason, G. Reid, Dalton Trans. 2013, 42, 2963–
2972.
[12] J. L. Dutton, R. Tabeshi, M. C. Jennings, A. J. Lough, P. J. Rag-
ogna, Inorg. Chem. 2007, 46, 8594–8602.
The whole concept of double bonds involving elements of the
third and higher periods is indeed the subject of much debate. See,
for example M. S. Schmøkel, S. Cenedese, J. Overgaard, M. R. V.
Jørgensen, Y.-S. Chen, C. Gatti, D. Stalke, B. B. Iversen, Inorg.
Chem. 2012, 51, 8607–8616. We prefer to write the adduct formu-
lae with single bonds (to indicate connectivity) and without ex-
plicit formal charges, which might imply a particular bonding
pattern that may not be strictly valid.
[13] A search of the Cambridge Database (F. H. Allen, Acta Crys-
tallogr., Sect. B 2002, 58, 380–388) gave 153 values in the range
2.25–2.72 Å for bonds between two-coordinate and three-coordi-
nate selenium. The distribution was skewed, with a small number
of very high values, which tended to be observed for cationic
selenourea clusters; the most common values lay in the range
2.35–2.4 Å. Similarly, the search for bonds between three-coordi-
nate selenium and one-coordinate chlorine gave 36 values in the
range 2.04–2.70 Å, with the higher values tending to be observed
[6] J. Konu, T. Chivers, Dalton Trans. 2006, 3941–3946.
[7] a) C. Taouss, P. G. Jones, Dalton Trans. 2011, 40, 11687–11689;
b) D. Upmann, P. G. Jones, Dalton Trans. 2013, 42, 7526–7528.
[8] Compounds R3PCl2 are generally assumed to be ionic, R3PCl+Cl–,
in solution and, in some cases (R = Ph, iPr), this has been con-
firmed in the solid state by X-ray structure determination. For a
more detailed discussion, see: S. M. Godfrey, C. A. McAuliffe,
R. G. Pritchard, J. M. Sheffield, G. M. Thompson, J. Chem. Soc.,
Dalton Trans. 1997, 4823–4827.
2–
in Se2Cl6 anions; there was no obvious preferred range of val-
ues.
[14] F. N. Blanco, L. E. Hagopian, W. R. McNamara, J. A. Golen,
A. L. Rheingold, C. Nataro, Organometallics 2006, 25, 4292–
4300.
[9] Cf. 31P NMR for other R3PCl+Cl–; R = Me, δ = 93.7 ppm and R
= nBu, δ = 106.6 ppm. R. M. Denton, J. An, B. Adeniran, A. J.
Blake, W. Lewis, A. M. Poulton, J. Org. Chem. 2011, 76, 6749–
6767. The δ values increase with the steric bulk of the R groups.
[15] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
Received: July 21, 2014
Published Online: September 30, 2014
2780
www.zaac.wiley-vch.de
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2014, 2776–2780