Selective P-P and P-O-P bond formations through copper-catalyzed aerobic oxidative dehydrogenative couplings of H-phosphonates

Article abstract of DOI:10.1002/anie.201003484

(Chemical Equation Presented) Different copper complexes selectively catalyze the aerobic oxidative coupling of H-phosphonates to afford either hypophosphates and pyrophosphates in high yields with high selectivity (see scheme; tmeda = N, N, N', N'-tetramethylethylenediamine).

Full text of DOI:10.1002/anie.201003484

Communications
DOI: 10.1002/anie.201003484
Coupling Reactions
ꢀ
ꢀ ꢀ
Selective P P and P O P Bond Formations through Copper-
Catalyzed Aerobic Oxidative Dehydrogenative Couplings of
H-Phosphonates
Yongbo Zhou, Shuangfeng Yin, Yuxing Gao, Yufen Zhao, Midori Goto, and Li-Biao Han*
Demands for efficient phosphorus-element bond-forming
reactions to prepare structurally sophisticated phosphorus
compounds that are not readily available by classical methods
have stimulated studies on new phosphorus bond-forming
particularly true for the preparation of hypophosphates 1,
since (RO)2P(O)Cl reacts with NaOP(OR)2 to give a
complicated mixture.[4] Phosphorus compounds having
[6]
ꢀ
ꢀ ꢀ
P(O) P(O) and P(O) O P(O) linkages are of interest as
biologically active substances. The antitumor activity of
hypophosphoric acid analogues 1 was noted,[6c] and a lot of
biologically important compounds, such as ADP and ATP,
reactions by transition-metal catalysis.[1] While catalytic C P
ꢀ
ꢀ
bond-forming reactions through couplings of P H bonds with
ꢀ
ꢀ
organohalides or additions of P H bonds to C C unsaturated
bonds are well known,[1] similar phosphorus–heteroatom
contain P(O) O P(O) linkages.
ꢀ ꢀ
ꢀ
bond-forming reactions (E P formations) are limited
We discovered this new catalytic reaction during a study
on the possible coordination of Z2P(O)H (Z = R or OR) to
copper complexes.[7] Z2P(O)H exists in equilibrium with a
trivalent Z2P(OH), which can coordinate to a variety of
metals.[7c] Very unexpectedly, we found that a slight change of
the substituent Z could lead to the selective generation of
[Eq. (1)].[2,3]
ꢀ ꢀ
ꢀ
both a P O P compound 2’ and a P P compound 1 from a
stoichiometric reaction of Z2P(O)H with copper(II) acetate
(Scheme 1).
ꢀ
ꢀ
Herein we disclose the first copper-catalyzed P P and P
ꢀ
O P bond-forming reactions. By slightly changing the reac-
tion conditions, copper can efficiently catalyze an aerobic
oxidative dehydrogenative coupling of H-phosphonate to
selectively produce hypophosphate 1 and pyrophosphate 2 in
high yields [Eq. (2)].
Scheme 1. Stoichiometric reactions of Cu(OAc)2 with R2P(O)H and
(RO)2P(O)H affording 2’ and 1, respectively.
Traditional methods for the preparation of 1[4] and 2[5] are
through nucleophilic substitution reactions of (RO)2P(O)Cl
with alkali-metal salts such as NaOP(OR)2 and NaO-
P(O)(OR)2, which suffer from drawbacks such as low yields,
tedious procedures, and lack of functionality tolerance. This is
Thus, a mixture of nBu2P(O)H (0.15 mmol) and Cu-
(OAc)2 (0.1 mmol) was dissolved in [D8]THF (0.5 mL) at
room temperature under nitrogen. After 2 h, two new signals
at d = 55.7 ppm and 122.6 ppm were observed by 31P NMR
spectroscopy (the integration ratio of the two signals = 1= ).
2
The signal at d = 55.7 ppm was assigned to nBu2P(O)OH by
comparing with an authentic sample, while the signal at d =
122.6 ppm was assigned to 3a as described below.[8] Complex
3a is extremely air sensitive. On contact with air, 3a
immediately changes from colorless to blue, and phosphinic
acid anhydride 2’ was formed.
[*] Dr. Y. B. Zhou, Dr. M. Goto, Dr. L.-B. Han
National Institute of Advanced Industrial Science and Technology
(AIST), Tsukuba, Ibaraki, 305-8565 (Japan)
Fax : (+81)29-861-6344
E-mail: libiao-han@aist.go.jp
Prof. Dr. S. F. Yin
College of Chemistry and Chemical Engineering, Hunan University,
410082, Changsha (China)
Although crystals suitable for X-ray analysis were not
obtained from 3a, by changing the substituent R from nBu to
Ph(CH2)4, air- and moisture-sensitive colorless crystals of 3b[9]
suitable for X-ray analysis were obtained in 56% yield
(Figure 1). Compound 3b is a copper(I) carboxylate complex
Y. X. Gao, Prof. Dr. Y. F. Zhao
Department of Chemistry and Key Laboratory for Chemical Biology of
Fujian Province, Xiamen University, Xiamen, 361005 (China)
having an interesting tetranuclear structure.[10] A novel R2P
ꢀ
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201003484.
ꢀ
O PR2 unit, formally generated by the dehydration of two
6852
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6852 –6855

Angewandte
Chemie
Stimulated by the above results of the stoichiometric
reactions of Z2P(O)H with copper(II) acetate, we thought
that a catalytic reaction to selectively generate 1 and 2 should
be possible. This indeed proved to be the case. After an
extensive screening of the reaction conditions, we found that
the oxidative dehydrogenative coupling of (RO)2P(O)H does
take place efficiently with a catalytic amount of copper under
ꢀ
air. Moreover, it can produce both the P P hypophosphate 1a
ꢀ ꢀ
and the P O P pyrophosphate 2a highly selectively by
changing the reaction conditions only slightly (Table 1).[13]
Table 1: Copper-catalyzed aerobic oxidative dehydrocoupling of
(iPrO)2P(O)H leading to hypophosphate 1a and pyrophosphate 2a.[a]
Run
[Cu] [mol%]
Amine [mol%]
Yield (1a/2a) [%][b]
1
Cu(OAc)2 (2)
CuCl (5)
Et3N[c]
Et3N (20)
98/trace (32/1)[d]
17/6
9/15
16/42
2/44
11/49
71/28
97/2
2/97
0/99
2[e]
3[e]
4[e]
5[e]
6[e]
7[e]
8[g]
9[e,h]
10[e]
iPr2NH (20)
(tBuNHCH2)2 (10)
2,2’-bipyridyl (10)
tmeda (10)
teeda[f] (10)
teeda (20)
Figure 1. Top: Molecular structure of 3b at 20% probability (hydrogen
atoms omitted for clarity). Bottom: View of core with substituent R
omitted. Selected bond lengths [ꢀ] and angles [8]: Cu1–Cu2 2.6809(6),
Cu1–P1 2.1467(8), Cu1–O5 2.004(2), Cu1–O3 2.1291(18), Cu1–O3’
2.1643(18), Cu2–O2 1.983(2), Cu2–O4 2.014(2), Cu2–P2 2.1438(9),
O1–P2 1.639(2), O1–P1 1.649(2), O3-Cu1 2.1643(18); P1-Cu1-Cu2
85.42(3), P2-Cu2-Cu1 97.28(3), P1-O1-P2 123.35(12), O1-P1-Cu1
118.37(8), O1-P2-Cu2 108.57(8), O2-Cu2-Cu1 90.85(5), O4-Cu2-Cu1
84.75(6), O5-Cu1-Cu2 81.41(6), O3-Cu1-Cu2 153.99(5), O3’-Cu1-Cu2
75.00(5), P1-Cu1-Cu2 85.42(3), O3-Cu1-O3’ 80.14(7), Cu1-O3-Cu1’
99.86(7).
CuCl (10)
CuCl (2)
CuBr2 (1)
tmeda (200)
tmeda (10)
[a] (iPrO)2P(O)H (1 mmol), copper compound, and amine under air for
20 h, unless otherwise noted. [b] Determined by GC. [c] 0.2 mL Et3N was
used; reaction time 4 h. [d] 50 mL of water was added. [e] Conducted in
acetone (1 mL). [f] teeda=Et2NCH2CH2NEt2. [g] Neat, 1.5 h. [h] 3 h.
Thus,
a mixture of (iPrO)2P(O)H and Cu(OAc)2
R2P(O)H units, chelates a pair of copper(I) atoms, which are
further linked into a dimer by intermolecular copper–oxygen
interactions. The Cu1–Cu2 length (2.6809(6) ꢀ) is signifi-
cantly shorter than the sum of their van der Waals radii
(2.80 ꢀ),[11] indicating the existence of a CuI–CuI interaction.
Surprisingly, a different product was obtained when H-
phosphonate (RO)2P(O)H was employed as the substrate.
Thus, when R2P(O)H was replaced by (RO)2P(O)H, a similar
copper complex could not be observed at all. Instead, the
exclusive formation of 1 took place [Eq. (3)]. For example,
[(iPrO)2P(O)]2 from dehydrogenative coupling (24 h, 18%
yield) was observed upon mixing a stoichiometric mixture of
(iPrO)2P(O)H with Cu(OAc)2 in THF. Compound
[(iPrO)2P(O)]2 was obtained quantitatively upon further
heating at 708C for 4 h [Eq. (3)].[12] Under similar stoichio-
metric reaction conditions, other H-phosphonates also effi-
ciently reacted with Cu(OAc)2 to give the corresponding
dehydrogenetive coupling products 1 in high yields (yields of
isolated products: R = Et 90%; R = nBu 92%; R = nC12H25
ꢀ
(2 mol%) in Et3N under air was stirred at room temperature
ꢀ
for 4 h to afford the P P coupling product 1a almost
ꢀ ꢀ
quantitatively (less than 0.1% yield of the P O P coupling
product 2a, Table 1, run 1). The use of Et3N in the reaction is
essential for this reaction. In the absence of Et3N, only a trace
amount of the coupling products could be obtained. This
coupling reaction can be carried out in solvents such as
acetone, THF, and EtOAc to give 1a in high yields. However,
since the copper catalyst was hydrolyzed to the catalytically
inactive Cu(OH)2 (see below), addition of water lowered the
yields of the products. For example, when 50 mL of H2O was
added, only 32% yield of 1a was obtained after 20 h (run 1).
With the exception of Cu(OH)2, which hardly catalyzes the
reaction, a variety of CuII or CuI compounds can serve as good
catalysts for this coupling reaction to give high yields of the
products. It was noted that both the efficiency of the catalysts
and the selectivity of the products are dramatically affected
by the amine used. Thus, a screening on amines using CuCl
(Table 1, runs 2–9) revealed that bidentate amines are
suitable compounds for the copper-catalyzed reactions. More-
over, while N,N,N’,N’-tetramethylethylenediamine (tmeda)
prefers the formation of 2a (Table 1, run 6), a bulky
91%; R = iPr 93%; R = CH2Ph 94%;
R
=
CH2C-
ꢀ
(Me2)CH2 90%).
N,N,N’,N’-tetraethylethylenediamine (teeda) leads to
a
reverse in the selectivity to the products favoring the
formation of 1a (Table 1, run 7). Further efforts on the
optimization of the reactions[13] established the best condi-
Angew. Chem. Int. Ed. 2010, 49, 6852 –6855
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6853

Communications
tions for the selective generation of both 1a and 2a. Thus, 1a
In summary, a copper-catalyzed oxidative dehydrocou-
ꢀ
ꢀ ꢀ
can be produced in 97% yield with greater than 97%
selectivity by using a catalyst CuCl/teeda (Table 1, run 8),
while 2a can be obtained in 97% yield with greater than 97%
selectivity by replacing tmeda with teeda (Table 1, run 9), or
by using CuBr2/tmeda as the catalyst to produce a quantita-
tive yield of 2a (Table 1, run 10).[13]
pling of (RO)2P(O)H selectively affording P P and P O P
compounds was revealed for the first time. We believe that
the present work not only provides an efficient and clean
method for the selective preparation of hypophosphate 1 and
pyrophosphate 2, but also is highly informative for the design
of new copper-catalyzed oxidative coupling reactions.[16]
Further studies on the clarification of the reaction mechanism
and applications to other substrates are underway.
This catalytic dehydrogenative coupling reaction can be
applied to other H-phosphonates to produce the correspond-
ꢀ
ꢀ ꢀ
ing P P compounds 1 and the P O P compounds 2 in high
yields with high selectivity (Table 2). An analogue of a
nucleotide that shows unique biological activity (run 7)[14]
Received: June 8, 2010
Published online: August 16, 2010
ꢀ
ꢀ
could also be used to give the corresponding P P and P
Keywords: copper · dehydrogenation · homogeneous catalysis ·
phosphorus
ꢀ
O P products in high yields, demonstrating clearly the high
.
potential of the current method to allow access to highly
functionalized targets.
[1] a) L.-B. Han, M. Tanaka, Chem. Commun. 1999, 395; b) F.
Alonso, I. P. Beletskaya, M. Yus, Chem. Rev. 2004, 104, 3079;
c) D. Prim, J. Campagne, D. Joseph, B. Andrioletti, Tetrahedron
2002, 58, 2041; d) D. S. Glueck, Synlett 2007, 2627; e) A. Schwan,
Chem. Soc. Rev. 2004, 33, 218; f) L. Coudray, J.-L. Montchamp,
Eur. J. Org. Chem. 2008, 3601; g) O. Delacroix, A. C. Gaumont,
Curr. Org. Chem. 2005, 9, 1851.
[2] a) S. Greenberg, D. W. Stephan, Chem. Soc. Rev. 2008, 37, 1482;
b) R. J. Less, R. L. Melen, V. Naseri, D. S. Wright, Chem.
Commun. 2009, 4929; c) D. W. Stephan, Angew. Chem. 2000,
112, 322; Angew. Chem. Int. Ed. 2000, 39, 314; d) R. Waterman,
Dalton Trans. 2009, 18; e) A. R. McWilliams, H. Dorn, I.
Manners, Top. Curr. Chem. 2002, 220, 141.
[3] A few transition-metal complex mediated dehydrocoupling
reactions with R2PH and R2PH·BH3 under an oxygen-free
inert atmosphere are known. Selected examples: a) T. L. Clark,
J. M. Rodezno, S. B. Clendenning, S. Aouba, P. M. Brodersen,
A. J. Lough, H. E. Ruda, I. Manners, Chem. Eur. J. 2005, 11,
4526; b) H. Dorn, R. A. Singh, J. A. Massey, A. J. Lough, I.
Manners, Angew. Chem. 1999, 111, 3540; Angew. Chem. Int. Ed.
1999, 38, 3321; c) H. Dorn, R. A. Singh, J. A. Massey, J. M.
Nelson, C. A. Jaska, A. J. Lough, I. Manners, J. Am. Chem. Soc.
2000, 122, 6669; d) R. H. Shu, L. J. Hao, J. F. Harrod, H. G. Woo,
E. Samuel, J. Am. Chem. Soc. 1998, 120, 12988; e) A. J. Roering,
S. N. MacMillan, J. M. Tanski, R. Waterman, Inorg. Chem. 2007,
46, 6855; f) J. D. Masuda, A. J. Hoskin, T. W. Graham, C. Beddie,
M. C. Fermin, N. Etkin, D. W. Stephan, Chem. Eur. J. 2006, 12,
8696; g) V. P. W. Bohm, M. Brookhart, Angew. Chem. 2001, 113,
4832; Angew. Chem. Int. Ed. 2001, 40, 4694; h) L.-B. Han, T. D.
Tilley, J. Am. Chem. Soc. 2006, 128, 13698; i) R. Waterman,
Organometallics 2007, 26, 2492. For metal-catalyzed dehydro-
coupling reactions of other heteroatom compounds, see also:
j) D. Miles, J. Ward, D. A. Foucher, Macromolecules 2009, 42,
9199; k) T. J. Clark, K. Lee, I. Manners, Chem. Eur. J. 2006, 12,
8634; l) F. Gauvin, J. F. Harrod, H. G. Woo, Adv. Organomet.
Chem. 1998, 42, 363.
Table 2: Copper-catalyzed selective aerobic oxidative couplings of
H-phosphonates affording hypophosphates 1 and pyrophosphates 2.[a]
Run 1[a,b]
Yield 2[d,e]
Yield
[%][c]
[%][c]
1
2
3
4
6
93[a]
80[b]
85[b]
84[a]
83[b]
99[d]
99[d]
99[d]
98[e]
90[e]
7
84[b]
96[e]
8
78[b]
91[e]
[a] Cu(OAc)2 (2 mol%), Et3N (0.2 mL), (RO)2P(O)H (1 mmol), 4 h. [b] CuCl
(10 mol%), teeda (30 mol%), acetone (0.5 mL), (RO)2P(O)H (1 mmol),
1.5–6 h. [c] Yields of isolated products. [d] CuBr2 (1 mol%), tmeda
(10 mol%), acetone (1 mL), (RO)2P(O)H (1 mmol), 6 h. [e] CuBr2
(2 mol%), tmeda (15 mol%), THF (1 mL), (RO)2P(O)H (1 mmol), 8–24 h.
By referring to the known copper-mediated coupling
ꢀ
reactions in the literature, the formation of the P P com-
pound 1 was rationalized to take place by the coupling of
phosphoryl radical (RO)2(O)P, which was generated by an
electron-transfer (ET) process of (RO)2P(O)H to CuII.[15,16] In
contrast, the reaction via intermediate 3 should be one of the
[4] a) R. W. Kinas, A. Okruszek, W. J. Stec, Tetrahedron Lett. 2002,
43, 7875; b) W. J. Stec, A. Zwierzak, J. Michalski, Bull. Acad.
Pol. Sci. Ser. Sci. Chim. 1970, 18, 23; c) J. E. Nycz, R. Musiol,
Heteroat. Chem. 2006, 17, 310; d) M. Z. Baudler, Naturforsch.
1953, 8, 326.
[5] a) N. Suzuki, T. Kawai, S. Inoue, N. Sano, Y. Izawa, Bull. Chem.
Soc. Jpn. 1980, 53, 1421; b) Y. Butsugan, K. Nagai, F. Nagaya, H.
Tabuchi, K. Yamada, S. Araki, Bull. Chem. Soc. Jpn. 1988, 61,
1707; c) Z. M. Jaszay, I. Petnehazy, L. Toke, Heteroat. Chem.
2004, 15, 447; d) W. S. Wadsworth, J. Org. Chem. 1987, 52, 1748;
e) A. D. F. Toy, J. Am. Chem. Soc. 1948, 70, 3882.
ꢀ ꢀ
possible ways for the formation of the P O P compound 2.
Although detailed mechanistic aspects remained to be
clarified, it was rationalized that a bulky amine could retard
the intermolecule reaction pathway for the formation of the
ꢀ ꢀ
ꢀ
P O P compound 2, and favor the formation of the P P
compound 1 by couplings of phosphoryl radicals.
6854 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6852 –6855

Angewandte
Chemie
[6] a) A. V. Mukovnya, V. L. Tunitskaya, A. L. Khandazhinskaya,
N. A. Golubeva, N. F. Zakirova, A. V. Ivanov, M. K. Kukhanova,
S. N. Kochetkov, Biochemistry 2008, 73, 660; b) M. K. Kukha-
nova, N. F. Zakirova, A. V. Ivanov, L. A. Alexandrova, M. V.
Jasco, A. R. Khomutov, Biochem. Biophys. Res. Commun. 2005,
338, 1335; c) M. G. Rose, M. P. Farrell, J. C. Schmitz, Clin.
Colorectal Cancer 2002, 1, 220.
[7] a) R. Peng, M. Li, D. Li, Coord. Chem. Rev. 2010, 254, 1; b) A.
Grodzicki, I. Lakomska, P. Piszczek, I. Szymanska, E. Szlyk,
Coord. Chem. Rev. 2005, 249, 2232; c) D. M. Roundhill, R. P.
Sperline, W. B. Beaulieu, Coord. Chem. Rev. 1978, 26, 263.
[8] The reaction took place as: 3R2P(O)H + 2Cu(OAc)2!3 +
R2P(O)OH + 2AcOH. However, the exact pathway for the
formation of 3 is not clear.
[9] CCDC 777312 contains the supplementary crystallographic 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.
[10] a) A. S. Filatov, O. Hietsoi, Y. Sevryugina, N. N. Gerasimchuk,
M. A. Petrukhina, Inorg. Chem. 2010, 49, 1626; b) Y. Sevryugina,
O. Hietsoi, M. A. Petrukhina, Chem. Commun. 2007, 3853.
[11] a) S. Sculfort, P. Croizat, A. Messaoudi, M. Benard, M. M.
Rohmer, R. Welter, P. Braunstein, Angew. Chem. 2009, 121,
9843; Angew. Chem. Int. Ed. 2009, 48, 9663; b) A. Bondi, J. Phys.
Chem. 1964, 68, 441.
acetic acid (97% yield measured by NMR spectroscopy), which
accounts for the stoichiometry of the reaction: 2(RO)2P(O)H +
2Cu(OAc)2!1 + 2Cu(OAc) + 2AcOH.
[13] For details of the experiments, see the Supporting Information.
A separate experiment using isolated 1a showed that compound
1a is stable under the reaction conditions for the formation of
2a, indicating that 2a was not formed by further oxidation of 1a.
[14] a) A. G. Pokrovsky, T. R. Pronayeva, N. V. Fedyuk, E. A.
Shirokova, A. L. Khandazhinskaya, N. B. Tarusova, I. L. Kar-
penko, A. A. Krayevsky, Nucleosides Nucleotides Nucleic Acids
2001, 20, 767; b) X. B. Sun, J. X. Kang, Y. F. Zhao, Chem.
Commun. 2002, 2414.
[15] a) T. Punniyamurthy, L. Rout, Coord. Chem. Rev. 2008, 252, 134;
b) T. Pintauer, K. Matyjaszewski, Chem. Soc. Rev. 2008, 37, 1087;
c) J. Iqbal, B. Bhatia, N. K. Nayyar, Chem. Rev. 1994, 94, 519;
d) T. Pintauer, K. Matyjaszewski, Coord. Chem. Rev. 2005, 249,
1155.
[16] Cu-catalyzed oxidative couplings are of high interest, but their
mechanisms are not well understood: a) K. Kamata, S. Yama-
guchi, M. Kotani, K. Yamaguchi, N. Mizuno, Angew. Chem.
2008, 120, 2441; Angew. Chem. Int. Ed. 2008, 47, 2407; b) P.
Siemsen, R. C. Livingston, F. Diederich, Angew. Chem. Int. Ed.
2000, 39, 2633; c) Y. Wei, H. Q. Zhao, J. Kan, W. P. Su, M. C.
Hong, J. Am. Chem. Soc. 2010, 132, 2522; d) C. Zhang, N. Jiao, J.
Am. Chem. Soc. 2010, 132, 28; e) T. Hamada, X. Ye, S. S. Stahl, J.
Am. Chem. Soc. 2008, 130, 833; f) Y. X. Gao, G. Wang, L. Chen,
P. X. Xu, Y. F. Zhao, Y. B. Zhou, L.-B. Han, J. Am. Chem. Soc.
2009, 131, 7956.
[12] No possible intermediates of (iPrO)2P(O)H-ligated Cu species
could be detected by 31P NMR spectroscopy. Other products are
CuOAc as white precipitate (95% yield of isolated product) and
Angew. Chem. Int. Ed. 2010, 49, 6852 –6855
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6855