A Stable Primary Phosphane Oxide and Its Heavier Congeners

Article abstract of DOI:10.1002/chem.202003702

(Ferrocenylmethyl)phosphane (1) oxidation with hydrogen peroxide, elemental sulfur and grey selenium produced (ferrocenylmethyl)phosphane oxide 1O, sulfide 1S and selenide 1Se, respectively, as the first isolable primary phosphane chalcogenides lacking steric protection. At elevated temperatures, compound 1O disproportionated into 1 and (ferrocenylmethyl)phosphinic acid. In reactions with [(η⁶-mes)RuCl₂]₂, 1O underwent tautomerization into a phosphane complex [(η⁶-mes)RuCl₂{FcCH₂PH(OH)-κP}], whereas 1S and 1Se lost their P-bound chalcogen atoms, giving rise to the phosphane complex [(η⁶-mes)RuCl₂(FcCH₂PH₂-κP)] (Fc=ferrocenyl, mes=mesitylene). No tautomerization was observed in the reaction of 1O with B(C₆F₅)₃, which instead produced a Lewis pair FcCH₂P(O)H₂-B(C₆F₅)₃. Phosphane oxide 1O added to C=O bonds of aldehydes and ketones and even to cumulenes PhNCE (E=O and S). However, both PH hydrogens were only employed in the reactions with aldehydes and cyanates.

Full text of DOI:10.1002/chem.202003702

Accepted Article
Accepted Article
Title: A stable primary phosphane oxide and its heavier congeners
Authors: Filip Horký, Ivana Císařová, and Petr Stepnicka
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202003702
Link to VoR: https://doi.org/10.1002/chem.202003702
01/2020

10.1002/chem.202003702
Chemistry - A European Journal
COMMUNICATION
A stable primary phosphane oxide and its heavier congeners
Filip Horký, Ivana Císařová, and Petr Štěpnička*
[a]
F. Horký, Dr. I. Císařová, Prof. Dr. P. Štěpnička
Department of Inorganic Chemistry
Faculty of Science, Charles University
Hlavova 2030, 128 40 Prague, Czech Republic
E-mail: petr.stepnicka@natur.cuni.cz
Supporting information for this article is given via a link at the end of the document.
Abstract: (Ferrocenylmethyl)phosphane (1) oxidation with hydrogen
peroxide, elemental sulfur and grey selenium produced
(ferrocenylmethyl)phosphane oxide 1O, sulfide 1S and selenide 1Se,
respectively, as the first isolable primary phosphane chalcogenides
lacking steric protection. At elevated temperatures, compound 1O
disproportionated into 1 and (ferrocenylmethyl)phosphinic acid. In
reactions with [(η6-mes)RuCl2]2, 1O underwent tautomerization into a
phosphane complex [(η6-mes)RuCl2{FcCH2PH(OH)-κP}], whereas
1S and 1Se lost their P-bound chalcogen atoms, giving rise to the
phosphane complex [(η6-mes)RuCl2(FcCH2PH2-κP)] (Fc = ferrocenyl,
mes = mesitylene). No tautomerization was observed in the reaction
Compound 1O could be stored under ambient conditions but,
upon heating to 100°C in a solution, it disproportionated into 1
and (ferrocenylmethyl)phosphinic acid (2) (see Equation 1):
2FcCH2P(O)H2 (1O) → 1 + FcCH2P(O)OH (2)
Eq. 1
Acid 2 also formed when 1 was treated with an excess of
MCPBA (2 equiv.); similar over-oxidation with H2O2 was only
noted over extended reaction times.
PH2
P(E)H2
of 1O with B(C6F5)3, which instead produced
a Lewis pair
=
=
or
for E O: H2O2 MCPBA
reagent
FcCH2P(O)H2−B(C6F5)3. Phosphane oxide 1O added to C=O bonds
of aldehydes and ketones and even to cumulenes PhNCE (E = O
and S). However, both PH hydrogens were only employed in the
reactions with aldehydes and cyanates.
for E S: elemental sulfur
Fe
Fe
=
for E Se:
selenium
grey
1
1E
Scheme 1. Synthesis of P-chalcogenides 1E
Despite unrelenting research efforts, the chemistry of
organophosphorus compounds still poses major challenges. For
instance, although the long-sought air-stable and isolable
primary phosphanes without substantial steric protection have
been already reported,[1,2] the corresponding primary phosphane
chalcogenides continue to elude chemists.[ 3,4] To the best of our
knowledge, primary phosphane oxides have only been
generated or postulated in reaction mixtures so far,[5] with the
sole exception of the highly unstable n-octylphosphane oxide,[6]
and their stable congeners with heavier chalcogen atoms have
never been reported at all.[ 7] Yet, considering the remarkable
The synthesis of the corresponding phosphane sulfide 1S was
rather difficult due to the thermal sensitivity of 1S (decomposition
above 50°C) and to its tendency to enter into follow-up reactions,
producing mixtures of polythionated compounds. When using 10
equiv. of sulfur (in benzene at 50°C for 1 day), the reaction
mixture contained 8% of the starting phosphane, 61% of 1S and
31% of other unidentified products. From this mixture,
nevertheless, 1S could be isolated in a 30% yield as a yellow
solid, which was stable under inert atmosphere but gradually
decomposed when exposed to air. Thionation of 1 or 1O with
Lawesson reagent proceeded similarly. In contrast, 1Se was
prepared smoothly as the sole product of the reaction between 1
and grey selenium (1.5 equiv. of Se in toluene at 50°C
overnight) and isolated in a 93% yield as a yellow, extremely
malodorous solid. Under inert atmosphere, the compound could
be stored practically indefinitely, albeit gradually converting into
1, 2 and red selenium when exposed to air and moisture. Unlike
its lighter congeners, 1Se did not decompose when heated to
100°C for 2 days (in CH2Cl2/hexane under argon), but failed to
provide crystalline samples. Our attempts at chalcogenation of 1
with elemental tellurium under otherwise similar conditions failed,
only producing some 2.
influence of the ferrocenylmethyl group on the overall chemical
8 ]
stability[
of the primary phosphane FcCH2PH2 (1; Fc =
ferrocenyl),[
we hypothesized that the corresponding
9
]
phosphane oxide, FcCH2P(O)H2 (1O), could be stabilized and
even isolated. Accordingly, in this contribution, we describe the
isolation of the bench-stable primary phosphane oxide 1O, the
analogous phosphane sulfide (1S) and the first primary
phosphane selenide (1Se).
For this purpose, the starting phosphane 1 was synthesized by
an alternative, three-step procedure from ferrocene
carboxaldehyde (see the Supporting Information). Subsequent
oxidation of 1 with hydrogen peroxide (10 equiv. of H2O2,
methanol/CH2Cl2 at 0°C for less than 10 min, 80% yield;
Scheme 1) or with 3-chloroperoxybenzoic acid (MCPBA; 1 equiv.
in dichloromethane at 0°C; 90% yield) cleanly and swiftly
produced the phosphane oxide 1O. This compound was isolated
by a conventional chromatography as an air-stable, yellow solid
with a characteristic though not intense smell of phosphane.
Compounds 1O, 1S and 1Se are reluctant to crystallize, and
their formulation was thus mostly based on spectroscopic
methods. In their 31P NMR spectra, they displayed triplets of
triplets due to interactions of the phosphorus atom with two
directly bonded hydrogen atoms and with the protons of the
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202003702
Chemistry - A European Journal
COMMUNICATION
2
adjacent CH2 group. In proton-decoupled spectra, these signals
collapsed into singlets (for parameters, see Table S5). The
presence of PH hydrogens was also indicated by 1H NMR
spectra, showing a doublet of triplets resulting from their
interaction with 31P and with the methylene protons. The spectra
of 1Se further contained characteristic 77Se satellites (Figure 1).
The 1JPSe coupling constant (733 Hz) of 1Se was larger than that
of the tertiary phosphane selenide FcCH2P(Se)Ph2 (728 Hz),
hence suggesting that the basicity of 1 is lower than that of
FcCH2PPh2.[10]
355 Hz, JPH = 8 Hz)]. Both complexes were unequivocally
authenticated by X-ray diffraction analysis (see the Supporting
Information).
In a reaction with ZnI2, 1O produced a phosphine oxide complex,
compound 5 (Scheme 3). Although the solution NMR spectra
indicated that 1O was coordinated to Zn(II) ions, they also
suggested fluxional nature of the species formed in this reaction,
corresponding with the rather weak (predominantly electrostatic)
interaction of the O-donor ligand 1O with the hard Lewis acid
Zn(II).
Fe
O
P
H
I
1O
(2 equiv.)
CH2Cl2
H
ZnI2
Zn
H
I
O
P
H
Fe
5*
Scheme 3. Reaction of 1O with ZnI2 (left) and the crystal structure of complex
5 (right; for details, see the Supporting Information)
Already during the initial experiments, we noted that phosphane
oxide 1O adds across the C=O bond of the solvent when
performing the oxidation of 1 by H2O2 in acetone. In this respect,
1O is more reactive than its parent phosphane. This result led us
to investigate addition reactions of 1O in more detail (Scheme 4).
Figure 1. 31P NMR spectrum of 1Se (162 MHz, CDCl3)
R
F5 3
)
O
B(C6
P
P
H
OH
O
H
Fe
6a*
Fe
RCHO
R
F5 3
)
B(C6
Notably, no signs of phosphinous acid tautomer were observed
in the NMR spectra of 1O recorded in common deuterated
solvents (even after adding 1,8-bis(dimethylamino)naphthalene
as a proton scavenger). Nevertheless, a facile tautomerization[11]
occurred during the reaction of 1O with [(η6-mesitylene)RuCl(µ-
Cl)]2, producing complex 3 as the sole product (Scheme 2).[12]
Unexpectedly, similar reactions with 1S and 1Se took an
different course, with the elimination of the P-bound chalcogen
atom and the formation of complex 4, which was alternatively
obtained by conventional cleavage of the Ru(II) precursor with a
stoichiometric amount of phosphane 1.
OH
=
=
Me)
Ph)
9
(R
(R
P(O)H2
6b
Fe
R
E
OH
1O
P
P
PhNCE
R
2CO
R
NHPh
O
Fe
O
Fe
H
E
NHPh
=
=
=
=
8a*
8b*
7a*
7b
O)
S)
Me)
Ph)
(E
(E
(R
(R
Scheme 4. Reactions of 1O (asterisk indicates that the crystal structure was
determined; for details, see the Supporting Information)
Ru
Thus, in a clean reaction of compound 1O with acetaldehyde
and benzaldehyde, the respective two-fold addition products 6a
and 6b[7a] were formed as a mixture of three diastereoisomers in
Ru
Ru
,
,
1O
1 1S 1Se
Cl
Cl
Cl
Cl
P
HO
P
H
Cl
Cl
CH2Cl2
CH2Cl2
Cl
Cl
H
H
Ru
Fe
Fe
a
statistical 1:2:1 ratio. These isomers differed in the
configuration at the two stereogenic carbon centers and at the
phosphorus atom (for detailed discussion and crystal
3*
4*
a
Scheme 2. Reactions of [(η6-mesitylene)RuCl(µ-Cl)]2 with 1E (the asterisk
indicates that the crystal structure was determined; see the Supporting
Information)
structures, see the Supporting Information). Conversely, the
reaction with ketones only led to the monoaddition products 7,
even when using an excess of the carbonyl compound.[7a] Of
these compounds, only 7a resulting from 1O and acetone could
be isolated in pure form. The more sterically congested
compound 7b was in equilibrium with the starting materials,[13]
which precluded its isolation by chromatography or
crystallization because 1O and Ph2CO were partly regenerated
during the purification. Lastly, in the reaction of the phosphane
The 31P NMR signatures of compounds 3 and 4 markedly
differed in the magnitude of the upfield-shift and fine structure
according to the substitution at the phosphorus atom [3: δP 115.0
1
2
1
(ddt, JPH = 110 Hz, JPH = 15 and 6 Hz), 4: δP −24.0 (tt, JPH
=
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202003702
Chemistry - A European Journal
COMMUNICATION
oxide 1O with phenyl isocyanate[ 14 ] and isothiocyanate, the
respective phosphabiurets 8a and 8b were formed in good
yields.[15] In contrast, when treated with neat 4-ethynyltoluene
(80°C/overnight), 1O disproportionated into a mixture of 1 and 2
(vide supra) rather than adding to the alkyne triple bond, as
expected for secondary phosphane oxides.[16]
[2]
For a sterically shielded, primary phosphane oxide and sulfide, see: M.
Yoshifuji, K. Shibayama, K. Toyota, N. Inamoto, Tetrahedron Lett. 1983,
24, 4227-4228.
[3]
[4]
a) L. D. Quin, A Guide to Organophosphorus Chemistry, Wiley-
Interscience, New York, 2000, pp. 95-113.
By contrast, the more stable secondary phosphane oxides found
attractive applications as ligands for catalysis: a) L. Ackermann,
Synthesis 2000, 1557-1571; b) T. M. Shaikh, C.-M. Weng, F.-E. Hong,
Coord. Chem. Rev. 2012, 256, 771-803; c) A. Gallen, A. Riera, X.
Verdauger, A. Grabulosa, Catal. Sci. Technol. 2019, 9, 5504-5561.
a) S. A. Buckler, M. Epstein, Tetrahedron 1962, 18, 1211-1219; b) B.
Pellerin, P. Guenot, J.-M. Denis, Tetrahedron Lett. 1987, 28, 5811-
5814; c) M. Scherer, D. Stein, F. Breher, J. Geier, H. Schönberg, H.
Grützmacher, Z. Anorg. Allg. Chem. 2005, 631, 2770-2774; d) E. V.
Gorbachuk, E. K. Badeeva. S. A. Katsyuba, P. O. Pavlov, K. R.
Khayarov, O. G. Sinyashin, D. G. Yakhvarov, Phosphorous, Sulfur
Silicon Relat. Elem. 2016, 191, 1480-1481; e) M. A. Gunawan, O.
Moncea, D. Poinsot, M. Keskes, B. Domenichini, O. Heintz, R.
Chassagnon, F. Herbst, R. M. K. Carlson, J. E. P. Dahl, A. A. Fokin, P.
R. Schreiner, J.-C. Hierso, Adv. Funct. Mater. 2018, art no. 1705786; f)
O. Moncea, D. Poinsot, A. A. Fokin, P. R. Schreiner, J.-C. Hierso,
ChemPlusChem 2018, 13, 2915-2922; g) GB Patent 912269 (1960).
S. A. Buckler, M. Epstein, Tetrahedron 1962, 18, 1221-1230.
a) F. Uhlig, E. Herrmann, D. Schädler, G. Ohms, G. Grossmann, S.
Besser, R. Herbst-Irmer, Z. Anorg. Allg. Chem. 1993, 619, 1962-1970;
b) S. Stalbauer; A. Torvisco; F. Uhlig, Phosphorus, Sulfur, Silicon, Relat.
Elem. 2014, 189, 1084-1093.
Considering the possibility of tautomerizaton, we also
investigated the interaction of 1O with borane and
tris(pentafluorophenyl)borane. Similarly to diphenylphosphane
oxide,[ 17] when mixing 1O with B(C6F5)3, a bipolar Lewis pair
FcCH2P(+)(O)H2−B(–)(C6F5)3 (9) formed instead of the
“conventional” Lewis adduct FcCH2PH(OH)·B(C6F5)3. Although 9
readily decomposed and could not be crystallized, its formulation
was unambiguously established from the spectroscopic data.
Specifically, the signal observed in the 31P{1H} NMR spectrum of
9 was observed at a position close to that of complex 5 (δP 17.8)
and was split into a septet by a scalar interaction with six fluorine
atoms of B(C6F5)3 in ortho positions. The non-decoupled 31P
NMR and 1H NMR spectra confirmed the presence of two
equivalent hydrogen atoms at phosphorus, whereas the position
of the 11B NMR resonance (δB 0.7) corresponded with that
reported for the similar adduct resulting from Ph2P(O)H.[17a] In
contrast, BH3 (1 equiv. of THF solution) addition to 1O produced
a complex mixture. 31P NMR analysis allowed us to identify the
parent phosphane 1 as the dominant product, in addition to
unreacted 1O, adducts 1·BH3 and FcCH2PH(OH)·BH3 and some
unidentified minor byproducts (see the Supporting Information).
In summary, we reported the synthesis of the first bench-stable
phosphane oxide. Although this compound can be conveniently
synthesized, purified by chromatography and stored under
ambient conditions without precautions, it retains the reactivity
typical of phosphane oxides. Together with the corresponding
phosphane sulfide and selenide, these compounds are the first
isolable primary phosphane chalcogenides with a conventional
albeit somewhat exotic “organic” substituent at the phosphorus
atom.[ 18 ] With their parent compound 1, they form a unique
family of homologous compounds with a high potential for
comparative studies. The first results from such studies are also
presented here. Although no signs of phosphinous acid tautomer
were detected in solutions of 1O, the phosphane oxide
undergoes facile tautomerization upon coordination to a soft
metal ion but binds to relatively harder metal ions in its native P-
oxide form. This P-oxide form also seems to be the active
species in additions of 1O to various unsaturated systems.
[5]
[6]
[7]
[8]
[9]
B. Stewart, A. Harriman, L. J. Higham, Organometallics 2011, 30, 5338-
5343.
N. J. Goodwin, W. Henderson, B. K. Nicholson, Chem. Commun. 1997,
31-32.
[10] a) C. A. Tolman, Chem. Rev. 1977, 77, 313-348; b) U. Beckmann, D.
Süslüyan, P. C. Kunz, Phosphorus, Sulfur, Silicon 2011, 186, 2061-
2070.
[11] B. G. Janesko, H. C. Fisher, M. J. Brindle, J.-L. Montchamp, J. Org.
Chem. 2015, 80, 10025-10032; b) D. Vincze, P. Ábrányi-Balogh, P.
Bagi, G. Kegevich, Molecules 2019, 24, art. no. 3859. For a similar
reaction of in situ generated H3P=O, see: c) D. Yakhvarov, M. Caporali,
L. Gonsalvi, S. Latypov, V. Mirabello, I. Rizvanov, O. Sinyashin, P.
Stoppioni, M. Peruzzini, Angew. Chem. Int. Ed. 2011, 50, 5370-5373.
[12] a) R. Krafczyk, H. Thönnessen, P. G. Jones, R. Schmutzler, J. Fluorine
Chem. 1997, 83, 159-166; b) S. M. M. Knapp, T. J. Sherbow, R. B.
Yelle, J. J. Juliette, D. R. Tyler, Organometallics 2013, 32, 3744-3752;
c) L. V.Graux, M. Giorgi, G. Buono, H. Clavier, Dalton Trans. 2016, 45,
6491-6502.
[13] M. Y. Dvorko, T. E. Glotova, I. A. Ushakov, N. K. Gusarova, Russ. J.
Org. Chem. 2010, 46, 485-490.
[14] The reaction of 1O with PhNCO required the presence of a base
catalyst (5 mol.% dabco) and stepwise addition of the isocyanate to
avoid the formation of a 1:1 addition product. See ref. [7a].
[15] D. H. M. W. Thewissen, H. P. M. M. Ambrosius, Rec. Trav. Chim. Pays-
Bas 1980, 99, 344-345 and references cited therein.
Acknowledgements
[16] A. C. Gaumont, M. Gulea in Science of Synthesis, Vol. 33 (Ed.: G. A.
Molander), Thieme, Stuttgart, 2006, pp. 701-710.
We acknowledge the support from the Grant Agency of Charles
University (project no. 920119) and from the Charles University
Research Centre program (project UNCE/SCI/014).
[17] a) R. Kather, E. Lork, J. Beckmann, Eur. J. Inorg. Chem. 2017, 2595-
2599. For examples of B(C6F5)3 adducts with tertiary phosphanes, see:
b) M. A. Beckett, D. S. Brassington, M. E. Light, M. B. Hursthouse, J.
Chem. Soc., Dalton Trans. 2001, 1768-1772; c) E. Regulska, S. Christ,
J. Zimmermann, F. Rominger, G. Hernandez-Sosa, C. Romero-Nieto,
Dalton Trans. 2019, 48, 12803-12807.
Keywords: Primary phosphane chalcogenides • Phosphanes •
Metallocenes • Synthesis design • Structure elucidation
[18] For a complete series of phosphanylborane chalcogenides, see: C.
Marquardt, O. Hegen, T. Kahoun, M. Scheer, Chem. Eur. J. 23, 2017,
4397-4404.
[1]
a) M. Brynda, Coord. Chem. Rev. 2005, 249, 2013-2034; b) J. T.
Fleming, L. J. Higham, Coord. Chem. Rev. 2015, 297-298, 127-145.
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202003702
Chemistry - A European Journal
COMMUNICATION
Entry for the Table of Contents
Remarkably stable (ferrocenylmethyl)phosphane oxide was synthesized as the first isolable primary phosphane oxide. Reactivity of
this compound and the corresponding phosphane sulfide and selenide, which were also prepared, was studied toward diverse metal
precursors and toward unsaturated organic substrates.
Institute Twitter username: @UniKarlova
4
This article is protected by copyright. All rights reserved.