Reductive conversion of phosphoryl P(O) compounds to trivalent organophosphines R3P

Article abstract of DOI:10.1016/j.tetlet.2021.152870

By introducing trimethylsilyl chloride (TMSCl), the pentavalent phosphoryl P(V) compounds such as triphenylphosphine oxides, secondary phosphine oxides etc., were readily converted to the corresponding R₂P(OTMS) intermediates, that can further react efficiently with an electrophile R'X or with a nucleophile R'Li to produce the corresponding trivalent phosphines R₂PR’. Chiral phosphines could also be obtained stereospecifically by this strategy.

Full text of DOI:10.1016/j.tetlet.2021.152870

Tetrahedron Letters 67 (2021) 152870
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Reductive conversion of phosphoryl P(O) compounds to trivalent
organophosphines R3P
Jian-Qiu Zhang a,b, Li-Biao Han a,b,
⇑
a National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki 305-8565, Japan
b Division of Chemistry, Faculty of Pure and Applied Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8571, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
By introducing trimethylsilyl chloride (TMSCl), the pentavalent phosphoryl P(V) compounds such as
triphenylphosphine oxides, secondary phosphine oxides etc., were readily converted to the correspond-
ing R2P(OTMS) intermediates, that can further react efficiently with an electrophile R’X or with a nucle-
ophile R’Li to produce the corresponding trivalent phosphines R2PR’. Chiral phosphines could also be
obtained stereospecifically by this strategy.
Received 26 November 2020
Revised 14 January 2021
Accepted 21 January 2021
Available online 4 February 2021
Ó 2021 Elsevier Ltd. All rights reserved.
Keywords:
Triphenylphosphine oxide
Tertiary phosphines
Chiral phosphines
Reduction
Sodium phosphide
Introduction
more valuable trivalent phosphine Ph2PR that possesses a central
position in organic synthesis and organometallic chemistry [5].
Triphenylphosphine oxide Ph3P(O) is a well-known chemical
waste generated from the Wittig reactions etc. that are widely used
in organic synthesis and industry for the preparation of highly
valuable compounds such as Vitamin A [1]. Tens of thousands tons
of Ph3P(O) are generated every year, and a large portion of them
have to be discarded as useless chemical waste because of the lim-
ited utilities [2]. This situation has been a big concern for more
than half a century. To solve this problem, extensive studies on
Ph3P(O) are carried out world widely [3].
Of course, in the laboratory, Ph2R may be produced from Ph2PCl
and RM (M = Li, MgX) (eq 1). However, chlorophosphines are
expensive and difficult to handle. Beyond the economic advantage,
this shows a new way for converting R3P(O) compounds to triva-
lent phosphines R3P, as the known conventional methods predom-
inantly use a hydride reducing reagent such as LiAlH4 etc. in order
to remove the oxygen on P(O) (eq 1) [3,6].
We have recently reported that, sodium is superior to lithium
and potassium, that can efficiently convert Ph3P(O) and related
organophosphorus compounds easily and efficiently, by cleaving
a Ph-P bond, to other phosphoryl P(O) compounds Ph2P(O)R etc.
(R = H, alkyl and aryl groups), providing a new general and eco-
nomic way for transforming P(V) compounds to other P(V) com-
pounds (Scheme 1A) [4]. During these investigations, we also
noted that, occasionally, a small amount of trivalent phosphine
Ph2PR could be also observed as a side product. This phenomenon
greatly attracted our attention because it unusually indicates that,
from this disposed waste Ph3P(O), we can not only produce the pen-
tavalent phosphoryl compounds Ph2P(O)R, but may also produce the
Herein, we communicate that by introducing TMSCl to the Ph3P
(O)/Na reaction system, diphenyl(trimethylsiloxy)phosphine Ph2P
(OTMS) can be generated that react efficiently further with an elec-
trophile (path a) or with a nucleophile (path b) to produce the
desired Ph2PR (Scheme 1B). The reaction can be carried out one-
pot by sequentially adding the required chemicals starting from
Ph3P(O), thus providing a convenient way for directly converting
Ph3P(O) to Ph2PR. This strategy can be applied to a variety of phos-
phoryl compounds, showing this is a general way for reductively
converting a pentavalent P(V) phosphoryl compound to the corre-
sponding trivalent P(III) organophosphines R3P. It was noted that,
to the best of our knowledge, the present strategies are surpris-
ingly totally unprecedented to date [7].
⇑
Corresponding author at: National Institute of Advanced Industrial Science and
Technology (AIST), Tsukuba, Ibaraki 305-8565, Japan.
E-mail address: libiao-han@aist.go.jp (L.-B. Han).
https://doi.org/10.1016/j.tetlet.2021.152870
0040-4039/Ó 2021 Elsevier Ltd. All rights reserved.

Jian-Qiu Zhang and Li-Biao Han
Tetrahedron Letters 67 (2021) 152870
Scheme 1. Conversion of Ph3P(O) using sodium to Ph2P(O)R and Ph2PR. (A): P(V) to
P(V); (B): P(V) to P(III).
Results and discussion
Scheme 2. Reduction of Ph3P(O) to Ph2PNa.
We have found that the trivalent dibenzophospholes can be
generated by a mild reduction with sodium via sequential reac-
tions starting from Ph3P(O) (eq 2) [3b]. Therefore, we initially
thought that Ph2PONa might also be similarly reduced to Ph2PNa
Table 1
Reaction conditions optimization.a
by sodium (SD: sodium dispersed in paraffin with
lm-scale
particles). However, disappointedly, contrary to our expectation,
we found that Ph2PONa could hardly be reduced to Ph2PNa even
under heating at an elevated temperature for a long time in the
presence of a large excess amount of SD.
run
SD
TMSCl
Yield of 3a/3a’
1
2
3
4
5
4 equiv
6 equiv
8 equiv
4 equiv
4 equiv
2 equiv
2 equiv
2 equiv
1 equiv
3 equiv
72%/22%
62%/17%
69%/15%
33%/67%
42%/39%
a
Reaction condition: Ph3P(O) 1a, (0.25 mmol), THF (2.0 mL), TMSCl (0.25–
0.75 mmol), SD (1.0–2.0 mmol), rt, 4 h. After filtration, n-BuBr (0.5 mmol), rt, 0.5 h.
Since Na alone is not able to reduce Ph2PONa to Ph2PNa, we
then added an additive to the reaction system that might act as
an ‘‘activator” to ‘‘activate” Ph2PONa so that it could be reduced
to Ph2PNa. Many such ‘‘activators” were then tested, and finally
we noted that when TMSCl was introduced to the reaction system,
the corresponding Ph2Pn-Bu was detected (eq 3)! Thus, to a mix-
ture of Ph3P(O) (0.25 mmol) and TMSCl (0.50 mmol) in THF
(2 mL) was added SD (0.50 mmol) at room temperature. The color
of the solution gradually changed from grey to black and then
orange, after stirring the mixture for 4 h. The precipitates were
removed by centrifugation and 31P NMR spectroscopy showed that
Ph2PNa (dp = À19.6 ppm) was generated (Scheme 2). Then n-BuBr
(0.5 mmol) was added to the transparent solution and Ph2Pn-Bu
(dp = -15.8 ppm) was obtained in 57% yield based on Ph3P(O) used
as estimated by 31P NMR spectra (Fig. S1). The product Ph2P(O)n-
Bu (dp = 28.7 ppm) via Ph2P(ONa) (dp = 90.3 ppm) was also gener-
ated in 19% yield. Then, efforts were devoted to improving the yield
of Ph2Pn-Bu (Table 1). However, only 72% yield of 3a could be
achieved (run 1).
Yields refer to 31P NMR yields based on 1a used.
Having verified that Ph2P(OTMS) was the real active intermedi-
ate for the conversion of Ph3P(O) to Ph2Pn-Bu, we can conduct a
one-pot reaction starting from Ph3P(O) more efficiently without
isolating Ph2P(OTMS). Thus, as shown in eq 5, Ph3P(O) reacted with
sodium to generate Ph2P(ONa) first. To this mixture was then
added TMSCl to in situ give Ph2P(OTMS). Sodium was then added
to generate Ph2PNa which was trapped by n-BuBr to give Ph2Pn-
Bu in 90% yield!
Carefully analyzing the reaction paths for the formation of Ph2-
PNa in eq 3, we realized that Ph2POTMS perhaps was the reactive
intermediate because Ph2P(ONa) generated from Ph3P(O) with
sodium could be readily trapped by TMSCl. This speculation was
correct as confirmed by a separate experiment by employing the
isolated Ph2P(OTMS), and Ph2Pn-Bu could be obtained in a quanti-
tative yield (eq 4)!
2

Jian-Qiu Zhang and Li-Biao Han
Tetrahedron Letters 67 (2021) 152870
Scheme 3. Efficient P-OTMS bond cleavage of R2P(OTMS) by sodium generating R2PNa.
nucleophilic substitution reaction, remarkably demonstrating that
Ph2P(OTMS) can act like ‘‘an ambiguous reagent” that both an elec-
trophile and an nucleophile can be used for converting Ph2P(OTMS)
to Ph2PR (Scheme 1)! For example, Ph2P(OTMS) in THF rapidly
reacted with n-BuLi at 0 °C to produce Ph2Pn-Bu in 96% yield
(Scheme 4). In addition, MeLi and PhLi could also be used as effi-
cient nucleophiles to react with Ph2P(OTMS), to give the substitu-
tion product Ph2PMe and Ph3P in 96% and 97% yield, respectively.
Very importantly, this substitution reaction can be used for the
preparation of the highly valuable chiral phosphines that are quite
difficult to prepare by other methods despite its high value. We
have developed an efficient way for the preparation of chiral-P
(O) compounds [8]. As demonstrated in Scheme 5, we first con-
firmed that a chiral-P(O)H compound (Rp)-1e (s, dp = 25.1 ppm)
can readily be changed to the corresponding (Rp)-(-)MenPhPOTMS
2e (s, dp = 105.9 ppm) (B). No epimerization took place at all, and
the stereochemistry retained at the phosphorus center! Treatment
As could be readily expected, by using this strategy, other sec-
ondary phosphine oxides R2P(O)H could be also readily reductively
converted to the corresponding trivalent phosphines (Scheme 3).
A further study on the reactivity of Ph2P(OTMS) fantastically
revealed that the OTMS unit can be replaced easily by RLi via a
Scheme 4. Efficient substitution reactions of Ph2P(OTMS) with RLi generating
Ph2PR.
Scheme 5. Preparation of chiral-phosphines via the reaction of (Rp)-(-)MenPhP(O)H with SD or n-BuLi monitored by 31P NMR.
3

Jian-Qiu Zhang and Li-Biao Han
Tetrahedron Letters 67 (2021) 152870
of the (Rp)-2e with 2.5 equivalents of SD at room temperature and
subsequent quenching with n-BuBr resulted in the racemic mix-
Acknowledgments
ture of (Sp,Rp)-(-)MenPhPn-Bu (3h/3h’) (d, dp
=
À19.8/–
This study was partially supported by a joint-research of
National Institute of Advanced Industrial Science and Technology
(AIST) with Katayama Chemical Industries Co., Ltd. We thank Pro-
fessor Chang-Qiu Zhao for a sample of (Rp)-1e.
23.1 ppm), albeit with high yield (eq 6). Secondly, and excitingly,
the substitution with RLi took place stereospecifically. Thus, to
the POTMS solution was dropwise added 1.2 equivalents of n-BuLi
at 0 °C and stirring for 1 h, only one signal appeared at À19.8 ppm
as confirmed by 31P NMR, indicating that (Sp)-(-)MenPhPn-Bu 3h
was generated stereospecifically in a high yield (eq 7). Protection
with BH3·THF afforded the air-stable P-B compound 3h’’ (d, 20.5
ppm, JB-P = 64.8 Hz) that was characterized by 1H and 13C NMR
(SI) [9a].
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tetlet.2021.152870.
Using the same strategy, MeLi could also efficiently react with
(Rp)-2e with inversion of configurations at phosphorus to give
the stereospecific product (Sp)(-)MenPhPMe in 88% yield (eq 8)
[9b].
References
[1] (a) H. Pommer, Angew. Chem. 72 (1960) 811–819;
(b) M.M. Heravi, M. Ghanbarian, V. Zadsirjan, B.A. Jani, Monatsh. Chem. (2019)
1–43.
[2] H.A. van Kalkeren, A.L. Blom, F.P.J.T. Rutjes, M.A.J. Huijbregts, Green. Chem. 15
(2013) 1255–1263.
[3] (a) D. Herault, D.H. Nguyen, D. Nuel, G. Buono, Chem. Soc. Rev. 44 (2015) 2508–
2528;
(b) J.Q. Zhang, J. Ye, T. Huang, H. Shinohara, H. Fujino, Han L. - B. Commun.
Chem. 3 (2020) 1–9;
(b) C. Tejo, J.H. Pang, D.Y. Ong, M. Oi, M. Uchiyama, R. Takita, S. Chiba, Chem.
Commun. 54 (2018) 1782–1785.
[4] J.Q. Zhang, E. Ikawa, H. Fujino, Y. Naganawa, Y. Nakajima, L.-B. Han, J. Org. Chem.
85 (2020) 14166–14173.
[5] (a) H. Guo, Y.C. Fan, Z. Sun, Y. Wu, O. Kwon, Chem. Rev. 118 (2018) 10049–
10293;
Conclusions
(b) D.J. Durand, N. Fey, Chem. Rev. 119 (2019) 6561–6594.
[6] T. Kovács, A. Urbanics, F. Csatlós, J. Binder, A. Falk, F. Uhlig, G. Keglevich, Curr.
Org. Synth. 13 (2016) 148–153.
[7] For TMSCl-promoted reduction of Ph3P(O) to Ph3P, see: (a) Kuroboshi, M.; Kita,
T.; Aono, A.; Katagiri, T.; Kikuchi, S.; Yamane, S.; Kawakubo, H.; Tanaka, H.
Tetrahedron Lett. 2015, 56, 918–920. (b) Tanaka, H.; Kuroboshi, M.; Kawakubo,
H.; Yano, T.; Kobayashi, K. Synthesis 2011, 2011, 4091–4098. (c) Kawakubo, H.;
Tanaka, H.; Yano, T.; Kobayashi, K.; Kamenoue, S.; Kuroboshi, M., Synlett 2011,
2011, 582–584. Kovács, T.; Urbanics, A.; Csatlós, F.; Binder, J.; Falk, A.; Uhlig, F.;
Keglevich, G. Curr. Org. Synth. 2016, 13, 148–153.
In summary, we have developed a TMSCl/Na reduction strategy
for the conversion of pentavalent phosphoryl compounds to triva-
lent phosphines. This method completed our on-going studies on
the utilization of Ph3P(O), showing that from Ph3P(O) not only
Ph2P(O)R but also Ph2PR could be produced via the key Ph2P
(OTMS) intermediate. In addition, the highly valuable chiral-phos-
phines can also be easily generated stereospecifically from the cor-
responding phosphine oxide by using this new strategy.
[8] T. Chen, L.-B. Han, Synlett 26 (2015) 1153–1163.
[9] (a) J.-J. Ye, S.-Z. Nie, J.-P. Wang, J.-H. Wen, Y. Zhang, M.-R. Qiu, C.-Q. Zhao, Org.
Lett. 19 (2017) 5384–5538;
(b) A.B. Shortt, L.J. Durham, H.S. Mosher, J. Org. Chem. 48 (1983) 3125–3126.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
4