A reactivity study of phenyl and ferrocenyl phosphates within the anionic phospho-Fries rearrangement

Article abstract of DOI:10.1016/j.inoche.2016.07.017

Temperature-dependent anionic phospho-Fries rearrangements of ferrocenyl/phenyl phosphates P(O)(OFc)_n(OPh)₃?_??_n (Fc?=?Fe(η⁵-C₅H₅)(η⁵-C₅H₄); n?=?0,1,2,3) were investigated. Whereas ferrocenyls solely undergo one rearrangement per reaction step, the number of phenyl-based 1,3-O?→?C shifts depends on the temperature. This results in different types of otherwise hardly accessible mixed ferrocenyl/phenyl organophosphorus compounds. Detailed investigations of the triple-rearrangement of triphenyl phosphate (n?=?0) reveals a consecutive formation of its phosphonate and phosphinate prior to the known phosphane oxide.

Full text of DOI:10.1016/j.inoche.2016.07.017

Inorganic Chemistry Communications 72 (2016) 30–32
Contents lists available at ScienceDirect
Inorganic Chemistry Communications
journal homepage: www.elsevier.com/locate/inoche
Short communication
A reactivity study of phenyl and ferrocenyl phosphates within the anionic
☆
phospho-Fries rearrangement
⁎
Marcus Korb, Heinrich Lang
Technische Universität Chemnitz, Faculty of Natural Sciences, Institute of Chemistry, Inorganic Chemistry, D–09107 Chemnitz, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Temperature-dependent anionic phospho-Fries rearrangements of ferrocenyl/phenyl phosphates
P(O)(OFc)n(OPh)3 − n (Fc = Fe(η5-C5H5)(η5-C5H4); n = 0,1,2,3) were investigated. Whereas ferrocenyls solely
undergo one rearrangement per reaction step, the number of phenyl-based 1,3-O → C shifts depends on the tem-
perature. This results in different types of otherwise hardly accessible mixed ferrocenyl/phenyl organophospho-
rus compounds. Detailed investigations of the triple-rearrangement of triphenyl phosphate (n = 0) reveals a
consecutive formation of its phosphonate and phosphinate prior to the known phosphane oxide.
© 2016 Elsevier B.V. All rights reserved.
Received 13 June 2016
Accepted 22 July 2016
Available online 25 July 2016
Organophosphorus compounds are common additives in polymers
to control, for example, their inflammatory properties [1,2]. They also
are of importance in light-emitting diodes [3], due to their easy linking
within polymeric networks [2]. Suitable connecting functionalities, e.g.
the hydroxy group, can be obtained by the reaction of ortho-metallated
species with PX3/P(O)X3 (X = Cl, Br), followed by deprotection to give
the respective triple-ortho-substituted phosphine/-oxide derivatives [4,
5]. More efficiently, the anionic phospho-Fries rearrangement, as a sin-
gle-step 1,3-O → C process, allows the conversion of triaryl phosphates
into tris(2-hydroxyaryl)phosphine oxides in high yields [6,7]. These
species were used for the colorimetric determination of, for example,
Fe3+ ions [4,8]. In general, they can act as binding agents for cations
[9], e.g. Li+ [10] dimethyl tin complexes [11], amino acids [12] and C60
[13]. Ortho-hydroxy phosphine oxides bearing a chiral binol backbone
are exceptional catalysts for the asymmetric addition of ZnEt2 to alde-
hydes, with the alcohols obtained in excellent yield and high ee [14].
The triple ortho-alkoxy-substituted phosphines are widely used for C,C
cross-coupling catalysis by applying axially pure biphenyl phosphanes
in catalytic Heck reactions [15].
a tripodal ligand in transition metal complex chemistry. The anionic
phospho-Fries rearrangement provides a novel and straightforward ac-
cess to 2-hydroxy phosphonates and phosphinates, which will be re-
ported, too.
Triferrocenyl phosphate (A) is accessible by the reaction of
ferrocenol FcOH (Fc = Fe(η5-C5H5)(η5-C5H4)) and phosphoryl chloride
in the presence of a base (Scheme 1) [18,19,20]. The yield could be in-
creased from 36 % to 92 % by using BuLi instead of NEt3 as base [16,17,
20]. The investigations of the anionic phospho-Fries reactions for A
have been carried out at various temperature regimes (Scheme 1,
Table SI1). At –70 °C treatment of A with an excess of LDA resulted
in the formation of novel phosphonate A-1 in a yield of 74 %, whereas
17 % of the starting material A remained. At –30 °C the yield of A-1
increased and reached 86 %. However, at higher temperatures A-1
started to decompose.
Phosphonate A-1 undergoes a consecutive anionic phospho-Fries re-
arrangement [1,16,17], resulting in the formation of phosphinate A-2 in
a yield of 86% at 0 °C within 4 h (Scheme 1). The planar chirality of both
ferrocenyls resulted in a diastereomeric excess (de) of 77% with a pre-
dominantly formed configuration of the pair of the meso enantiomers
Rp,Sp,rP and Sp,Rp,sP as it could be confirmed by using single crystal X-
ray diffraction analysis (Fig. 1). Compound A-2 crystallizes in the mono-
clinic space group P21/c in a ratio of 0.9:0.1 of their Rp,Rp/Sp,Sp diastereo-
mers, which is similar to the diastereomeric ratio obtained by 31P{1H}
NMR spectroscopy (0.885:0.115, Experimental section). However,
phosphine oxide A-3 (Scheme 3) could not be obtained, neither by the
reaction of A, A-1 nor A-2 with LDA (Table SI1).
Our recent results on the synthesis of 1,2-P,O ferrocenyl phosphanes
and their use for C,C cross-coupling Suzuki-Miyaura reactions [16] and
diastereomerically enriched multiple-step anionic phospho-Fries rear-
rangements [17], prompted us to investigate the rearrangement of
triferrocenyl phosphate (A), which will be reported herein. The rear-
rangement to a triple-2-functionalized phosphane oxide (A-3) should
give access to an electronically-rich phosphane by reduction [17] or as
The rearrangement of B, in which all ferrocenyls are replaced by
phenyls, has been reported to exclusively result in oxide B-3 within
one reaction step at −78 °C and subsequent warming to 0 °C (Scheme
2) [6,21]. Interestingly, neither phosphonate B-1 nor phosphinate B-2
☆
Dedicated to a dear friend, Professor Dr. Michael Hietschold on the occasion of his 65th
birthday.
⁎
Corresponding author.
E-mail address: heinrich.lang@chemie.tu-chemnitz.de (H. Lang).
http://dx.doi.org/10.1016/j.inoche.2016.07.017
1387-7003/© 2016 Elsevier B.V. All rights reserved.

M. Korb, H. Lang / Inorganic Chemistry Communications 72 (2016) 30–32
31
Scheme 1. Consecutive anionic phospho-Fries rearrangement of triferrocenyl phosphate A
to afford phosphonate A-1 and phosphinate A-2. (i) LDA, thf, −30 °C, 3 h, Me2SO4, 86%
(based on A). (ii) LDA, thf, 0 °C, 3 h, Me2SO4, 86% (based on A-1). a) Obtained by the
reaction of 3 equiv of FcOLi with POCl3 in diethyl ether.
have been isolated so far. The question is, if they were formed during
this reaction or a simultaneous rearrangement from B to B-3 occurs ex-
cluding their formation.
Scheme 2. Anionic phospho-Fries rearrangement of triphenyl phosphate (B) to give B-1–
B-3. (i) thf, LDA, B, 2 h, Me2SO4. a) Yield based on B.
To investigate the reactivity of phosphonate B at low temperatures,
the treatment of this compound with an excess of LDA and the addition
of the electrophile HClaq, were performed at −80 °C (Scheme 2). At this
temperature B was successfully transferred to phosphonate B-1 and
phosphinate B-2, confirming an increased reactivity of B as compared
to A [22]. When the temperature was increased to 0 °C then oxide B-3
was accessible [6,21]. This reactivity behavior of B towards an anionic
phospho-Fries rearrangement can be explained by an increased elec-
tron density at the remaining O-bonded aromatics, requiring higher
temperatures for their lithiation.
The increased reactivity of the phenyl-based hydroxyls might affect
the number of possible rearrangements of A and thus, PhOLi (3 and
12 equiv) were added to the reaction of A with LDA at −30 °C (Table
SI1). The yield of phosphonate A-1 decreased from 86% (Scheme 1) to
59% (3 equiv) and 44% (12 equiv), whereas the formation of
phosphinate A-2 was observed with a yield of 16% (3 equiv) and 46%
(12 equiv) as a single diastereomer (de 0.99).
For the synthesis of diferrocenylphenyl phosphate C, FcOLi was
reacted with dichlorophenyl phosphate (Supporting information).
However, treatment of C with 4 equiv of LDA at −75 °C (Scheme 3) pro-
duced only phosphonates C-1 and C-2 in minor yield. Increasing the re-
action temperature to 0 °C resulted in a complete conversion of C giving
C-3. Although an excess of dimethyl sulfate was present, compounds C-
1 and C-3 were obtained in form of C-1a/C-3a (R = H) and as C-1b/C-3b
(R = CH3). As observed for A (Scheme 1), the ferrocenyls prevent the
formation of any type of phosphine oxides, within the anionic
phospho-Fries rearrangement of C (Scheme 3).
To examine the influence between one (C) and two phenyls (D) on
the product-mixture within an anionic phospho-Fries rearrangement,
the treatment of ferrocenyldiphenyl phosphate (D) has been
reinvestigated [16]. Replacing NEt3 by BuLi as the base for the deproton-
ation of ferrocenol, increased the yield of D from 62 [16] to 99%.
Unmethylated D-2 could recently be obtained in a yield of 8% by run-
ning the reaction at −40 °C, due to an unsuccessful methylation by
using iodomethane [17].
To further investigate the difference between the phenyl and
ferrocenyl substituents in A and B, compounds C and D were prepared
bearing both types of aromatics (Schemes 3 and 4).
Carrying out the anionic phospho-Fries rearrangement below −75 °
C reduces the amount of unreacted starting material from 80% for C to
25% for D. The formation of phosphonate D-1 and phosphinate D-3 is
observed. This reveals a positive influence of the two phenyls towards
the lithiation behavior compared to C, where the formation of a
phosphinate has not been observed. At −70 °C compound D was
completely converted and the yields of D-1 and D-3 increased
Scheme 3. Anionic phospho-Fries rearrangement of diferrocenylphenyl phosphate C to
afford phosphonates C-1, C-2 and phosphinate C-3. (i) thf, LDA (4 equiv); C, Me2SO4. a)
Obtained by the reaction of 2 equiv of FcOLi with Cl2P(O)OPh in diethyl ether. b) Yield
based on C. c) 7 equiv of LDA were used.
Fig. 1. ORTEP diagram (50% probability level) of the molecular structure of meso-A-2 with
the atom-numbering scheme. All hydrogen atoms and the co-crystallized racem-A-2
(10%) have been omitted for clarity.

32
M. Korb, H. Lang / Inorganic Chemistry Communications 72 (2016) 30–32
substituents. Thus, adding PhOLi within the reaction of triferrocenyl
phosphate enables a second rearrangement.
Acknowledgment
M.K. thanks the Fonds der Chemischen Industrie for a PhD
fellowship.
Appendix A. Supplementary material
Detailed experimental procedures, Tables SI1–4 and a CIF file giving
additional experimental and crystallographic data as well as spectro-
scopic details for all new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org. Crystallographic data of
A-2 are also available from the Cambridge Crystallographic Database
as file number CCDC 1474108 (A-2). Supplementary data associated
with this article can be found in the online version, at http://dx.doi.
org/10.1016/j.inoche.2016.07.017.
References
[1] C.M. Taylor, A.J. Watson, Curr. Org. Chem. 8 (2004) 623–636.
[2] (a) C.S. Wang, J.Y. Shieh, J. Appl. Polym. Sci. 73 (1999) 353–361;
(b) Y.L. Liu, Y.C. Chiu, C.S. Wu, J. Appl. Polym. Sci. 87 (2003) 404–411;
(c) C.S. Wang, M.C. Lee, Polymer 41 (2000) 3631–3638;
(d) J.Y. Shieh, C.S. Wang, J. Polym. Sci. A 40 (2001) 369–378;
(e) Z. Bai, L. Song, Y. Hu, R.K.K. Yuen, Ind. Eng. Chem. Res. 52 (2013) 12855–12864.
[3] (a) Y.M. Sun, C.S. Wang, Polymer 42 (2001) 1035–1045;
(b) Y.M. Sun, A.Y.C. Hung, C.T. Wang, J. Appl. Polym. Sci. 85 (2002) 2367–2376.
[4] M.J. Holdoway, J.L. Willans, Anal. Chim. Acta 18 (1958) 376–380.
[5] F.C. Krebs, P.S. Larsen, J. Larsen, C.S. Jacobsen, C. Boutton, N. Thorup, J. Am. Chem. Soc.
119 (1997) 1208–1216.
Scheme 4. Anionic phospho-Fries rearrangement of ferrocenyldiphenyl phosphate D to
give phosphonates D-1 and D-2, phosphinate D-3 as well as phosphine oxide D-4. (i)
thf, LDA (8 equiv), D, 2 h, Me2SO4 (12 equiv). a) Obtained by the reaction of 0.5 equiv of
FcOLi with ClP(O)(OPh)2 in diethyl ether. b) Yield based on D.
accordingly. Additionally, the formation of phosphinate D-2 is observed.
Phosphine oxide D-4 could be isolated at 0 °C in a yield of 25%
(Scheme 4).
[6] (a) B. Dhawan, D. Redmore, Synth. Commun. 17 (1987) 465–468;
(b) J. Heinicke, I. Böhle, A. Tzschach, J. Organomet. Chem. 317 (1986) 11–21;
(c) B. Dhawan, D. Redmore, J. Organomet. Chem. 49 (1984) 4018–4021;
(d) S. Li, G. Wang, Phosphorus Sulfur Silicon Relat. Elem. 61 (1991) 119–129;
(e) K.P. Jayasundera, A.J. Watson, C.M. Taylor, Tetrahedron Lett. 46 (2005)
4311–4313;
Characterization and differentiation between the different families
of organophosphorus compounds is based on 13C{1H} and 31P{1H}
NMR spectroscopy. The 31P{1H} NMR downfield shift increases by re-
placing oxygens with carbon atoms [16,17]. Thus, phosphates (−15.3
(A) to −17.8 (B) ppm) and phosphonates (11.5 to 18.8 ppm) could
be distinguished, whereby the electron-rich phosphonates A-1 and C-
2 are shifted to lower field (D-1, 18.8 ppm; C-2, 18.3 ppm) as compared
with their phenyl derivatives (C-1b, 11.5 ppm to C-1a, 15.8 ppm). Addi-
tionally, in phosphinates D-3b (26.1 ppm) – D-3a (44.1 ppm) and phos-
phine oxides D-4b (38.2 ppm), D-4a (49.3 ppm) and B-3 (51.1 ppm) the
presence of an electron-donating hydroxyl groups increases the chemi-
cal shift as compared to their methoxy derivatives (Supporting informa-
tion). Further criteria are the decreasing 1JC,P coupling constants with
significant smaller values for P–CPhenyl (phosphonates B-1, C-1 and D-
1: 184–190 Hz; phosphinates B-2, C-3 and D-2,3: 134–147 Hz; oxides
B-3 and D-4b: 107 and 110 Hz) as compared to P–CFc (phosphonates
A-1 and C-2: 220 and 222 Hz; phosphinates A-2, C-3 and D-3: 162.5–
183.9 Hz; oxide D-4: 120 ppm).
(f) L.S. Melvin, Tetrahedron Lett. 22 (1981) 3375–3376.
[7] B. Dhawan, D. Redmore, J. Organomet. Chem. 56 (1991) 833–835.
[8] K.R. Dunbar, S.C. Haefner, A. Quillevéré, Polyhedron 9 (1990) 1695–1702.
[9] I.Y. Kudryavtsev, T.V. Baulina, M.P. Pasechinik, R.R. Aisin, S.V. Matveev, P.V.
Petrovskii, E.E. Nifant'ev, Russ. Chem. Bull., Int. Ed. 62 (2013) 1086–1090.
[10] (a) T. Sone, A. Yamaguchi, S. Matsunaga, M. Shibasaki, Molecules 17 (2012)
1617–1634;
(b) N. Yamagiwa, Y. Abiko, M. Sugita, J. Tian, S. Matsunaga, M. Shibasaki, Tetrahe-
dron Asymmetry 17 (2006) 566–573.
[11] M. Wada, T. Fujii, S. Iijima, S. Hayase, T. Erabi, J. Organomet. Chem. 445 (1993)
65–70.
[12] A.H. Alberts, K. Timmer, J.G. Noltes, A.L. Spek, J. Am. Chem. Soc. 101 (1979)
3375–3376.
[13] M. Yamamura, T. Saito, T. Nabeshima, J. Am. Chem. Soc. 136 (2014) 14299–14306.
[14] (a) M. Hatano, T. Mixamoto, K. Ishihara, J. Org. Chem. 71 (2006) 6474–6484;
(b) M. Hatano, K. Ihsihara, Chem. Rec. 8 (2008) 143–155.
[15] E. Gorobets, G.-R. Sun, B.M.M. Wheatley, M. Parvez, B.A. Keay, Tetrahedron Lett. 45
(2004) 3597–3601.
In conclusion, the synthesis of a series of ferrocenyl/phenyl phos-
phates of type P(O)(OFc)n(OPh)3 – n (Fc = Fe(η5-C5H5) (η5-C5H4);
n = 0, 1, 2, 3) was investigated in anionic phospho-Fries rearrange-
ments [1]. Triferrocenyl phosphate (A, n = 3) undergoes step-wise re-
arrangements to the respective phospho- and phospinate. In contrary,
the product mixture derived from the triphenyl derivative B (n = 0)
solely depends on the reaction temperature. Furthermore, the investi-
gations of B reveal that the literature-stated reaction temperature of
−78 °C does not result in the formation of the respective phosphine
oxide B-3. Moreover, stirring at 0–25 °C is essential. Our investigations
provide a new straightforward access to mixed phosphonates and
phosphinates, derived from ferrocenyl/phenyl phosphates within one
reaction step by applying the anionic phospho-Fries rearrangement.
The number of rearrangements could be controlled by varying the reac-
tion temperature, revealing an increased reactivity for phenyl
[16] M. Korb, D. Schaarschmidt, H. Lang, Organometallics 33 (2014) 2099–2108.
[17] M. Korb, H. Lang, Organometallics 33 (2014) 6643–6659.
[18] Y. Kudryavtsev, T.V. Baulina, V.N. Khrustalev, P.V. Petrovskii, M.P. Pasechnik, E.E.
Nifant'ev, Dokl. Chem. 448 (2013) 55–60.
[19] M. Herberhold, A. Hofmann, W. Milius, F.F. de Biani, P. Zanello, Inorg. Chim. Acta 273
(1998) 24–30.
[20] M. Herberhold, A. Hofmann, W. Milius, J. Organomet. Chem. 555 (1998) 187–200.
[21] Examples for this temperature regime are reported in references 1, 6, 7, 10, 19.
[22] For similar compounds, derived from the addition of phosphites to quinones, see:
reference 2e, 3a and
(a) B. Xiong, R. Shen, M. Goto, S.-F. Yin, L.-B. Han, Chem. Eur. J. 18 (2012)
16902–16910;
(b) P. Müller, Y. Bykov, O. Walter, M. Döring, Heteroat. Chem. 23 (2012) 383–394;
(c) I. Neda, H.-J. Plinta, A. Fischer, P.G. Jones, R. Schmutzler, J. Fluor. Chem. 71 (1995)
65–74;
(d) O.M. Demchuk, B. Yoruk, T. Blackburn, V. Snieckus, Synlett (2006) 2908–2913;
(e) N.G. Frolova, E.D. Savin, E.I. Goryunov, K.A. Lysenko, Y.V. Nelyubina, P.V.
Petrovskii, E.E. Nifant'ev, Dokl. Chem. 430 (2010) 18–23;
(f) J.M. Brown, S. Woodward, J. Organomet. Chem. 56 (1991) 6803–6809.