Isomerization of Ferrocenyl Phosphinites to Phosphane-oxides and retro-Phospha-Brook Rearrangement

Article abstract of DOI:10.1016/j.jorganchem.2021.121801

Ferrocene-containing phosphorus derivatives are important as ligands in transition-metal catalysis. We have investigated the synthesis of ferrocene-based phosphanes from corresponding phosphinites via retro-phospha-Brook rearrangement. Ferrocenyl phosphin

Full text of DOI:10.1016/j.jorganchem.2021.121801

Journal of Organometallic Chemistry 941 (2021) 121801
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Isomerization of Ferrocenyl Phosphinites to Phosphane-oxides and
retro-Phospha-Brook Rearrangement
Ambroz Almássy, Andrea Hejtmánková, Denisa Vargová, Radovan Šebesta∗
Department of Organic Chemistry, Faculty of Natural Sciences, Comenius University in Bratislava Mlynská dolina, Ilkovicˇova 6, 842 15 Bratislava, Slovakia
a r t i c l e i n f o
a b s t r a c t
Article history:
Ferrocene-containing phosphorus derivatives are important as ligands in transition-metal catalysis. We
have investigated the synthesis of ferrocene-based phosphanes from corresponding phosphinites via
retro-phospha-Brook rearrangement. Ferrocenyl phosphinites isomerize to thermodynamically more sta-
ble phosphine oxides. Interestingly, retro-phospha-Brook rearrangement proceeds on the adjacent phenyl
ring with the assistance of the ferrocenyl moiety. DFT calculations provided insight into the reaction
mechanisms of these transformations. Isomerization probably proceeds intramolecularly with the par-
ticipation of developing charge stabilization with irons´ d-orbitals. BH3 complexation seems to favor the
phosphinite structure as was evidenced by 31 P-11 B scalar coupling. Retro-phospha-Brook rearrangement
also likely proceeds as intramolecular migration of the phosphanyl group from oxygen to carbon, from
both BH3 complexes and the uncomplexed starting material.
Received 29 January 2021
Revised 8 March 2021
Accepted 26 March 2021
Available online 1 April 2021
© 2021 Elsevier B.V. All rights reserved.
1. Introduction
investigate the possibility of performing a related retro-phospha-
Brook rearrangement on ferrocenyl derivatives. The successful ap-
Ferrocene derivatives are invaluable as materials, pharmaceu-
ticals, and catalysts [1]. Chiral ferrocene compounds serve as
highly efficient catalysts for many transformations [2]. Phosphorus-
containing ligands have found the widespread use in metal-
catalyzed reactions. Chiral ferrocene phosphanes are a privileged
ligand class for many transformations in asymmetric catalysis, in-
cluding chemical industry applications [3–11]. Diphosphanes such
as Josiphos, [12] Taniaphos, [13–15] or Walphos [16] are proto-
typical catalysts. Rich coordination chemistry of ferrocenyl phos-
phorus compounds is in the center of interest of many chemists
as well [17–20]. Recently, ferrocene-based nitrogen and phospho-
rus derivatives became useful as nucleophilic organocatalysts, thus
sidestepping the need for transition metal catalysts [21]. Given
this interest, a variety of preparative methods were developed for
synthesis and transformations of ferrocenyl phosphorus derivatives
[22].
plication of retro-phospha-Brook rearrangement would constitute a
new strategy to obtain valuable ferrocenyl phosphorus derivatives.
Brook rearrangement is typically a migration of a silyl group be-
tween carbon and oxygen [24–26]. The transfer of the silyl group
from oxygen to carbon is termed retro-Brook rearrangement. Brook
rearrangement has been instrumental also in asymmetric synthe-
sis, [27] and natural product synthesis [28]. Besides silicon, other
heteroatoms can promote rearrangements as well. Arguably, the
transfer of phosphorus groups is one of the most relevant from
a synthetic point of view as it leads to useful organophosphorus
derivatives. Phospha-Brook rearrangement was described in sev-
eral instances where a carbanion was generated at a suitable dis-
tance from the phosphorus group [29–36]. Recently, Terada and co-
workers showed that [1,2]-phospha-Brook rearrangement gener-
ated α-oxygenated propargyl anions that engaged in a Mannich or
aldol type additions, which ultimately provided access to pyrroles
and furan derivatives [37,38]. This type of Brook rearrangement
was also utilized in the formation of allylic alcohols from α,β-
epoxyketones [39]. Carbanions generated by [1,2]-phospha-Brook
rearrangement were employed in aromatic nucleophilic substi-
tutions of fluoroarenes [40]. Base-catalyzed phospha-Brook rear-
rangement of isatine and diethyl phosphite afforded intermediate
electrophilic species that underwent a Friedel−Crafts alkylation. In
this way, α-aryl oxindoles were obtained in a one-pot procedure
[41]. However, phospha-Brook rearrangement has never been used
for the synthesis of ferrocenyl phosphorus derivatives.
Recently, we have investigated the retro-Brook rearrangement
of ferrocenyl silyl ethers as a means to obtain new ferrocenyl sil-
icon derivatives [23]. Bromine-lithium exchange or ortho-lithiation
produced suitably positioned carbanions that acted as nucleophilic
centers and initiated an attack on the silyl group in the retro-Brook
rearrangement. Motivated by these results, we have decided to
∗
Corresponding auhtor:
E-mail address: radovan.sebesta@uniba.sk (R. Šebesta).
https://doi.org/10.1016/j.jorganchem.2021.121801
0022-328X/© 2021 Elsevier B.V. All rights reserved.

A. Almássy, A. Hejtmánková, D. Vargová et al.
Journal of Organometallic Chemistry 941 (2021) 121801
In this context, we decided to study the utilization of Br-Li
exchange-initiated retro-phospha-Brook rearrangement as a possi-
ble means for the synthesis of ferrocenyl phosphane derivatives.
Herein, we describe the results of our experimental and DFT inves-
tigation of phosphinite - phosphane-oxide isomerization and retro-
phospha-Brook rearrangement of selected ferrocenyl derivatives.
the corresponding BH3-complex of phosphinite 3. In addition to
this product, 40% of 2-bromobenzylferrocene (6) and 15% of un-
complexed 3 were obtained (Scheme 2b). 2-Bromobenzylferrocene
probably resulted from the secondary reduction of the alcohol with
borane.
Experimental evidence in favor of BH3-complexed phosphinite
derivative was obtained by 31P NMR. The scalar nuclear spin-spin
coupling between 31P and 11 B nuclei is an indication of a dative
bond in BH3-complexed phosphanes [43]. Indeed, such coupling,
2. Results and Discussion
1
with a constant JPB = 42 Hz, was observed in BH3-complexed
We have started our investigation by a synthesis of ferrocenyl
alcohol 2 using literature procedures [15]. Racemic alcohol 2 was
obtained in 71% yields by reduction of the corresponding ferro-
cenyl ketone 1 (Scheme 1). This ferrocenyl ketone was obtained by
Friedel-Crafts acylation of ferrocene. With racemic ferrocenyl alco-
hol 2 in our hands, we have tried to synthesize the corresponding
phosphinite 3 by a reaction with chlorodiphenylphosphine under
various conditions. The reactions with Ph2PCl in toluene, with Et3N
or BuLi as base were not successful. The best reaction conditions
comprised Ph2PCl, 4-dimethylaminopyridine (DMAP), and Et3N in
dichloromethane (DCM). Interestingly, the spectral characteristics,
in particular NMR data, of the obtained compound matched more
closely with corresponding phosphane-oxide 4 than with expected
phosphinite 3 (Scheme 1).
compound 3. This result hints at a partly covalent bond between
phosphorus and boron.
As an alternative, Arbusov-type reaction can be considered on
phosphinite 3 involving intermolecular nucleophilic attack of phos-
phorus on the αC. We think that it is unlikely because such an
event should produce a mixture of 3 and 4 that we did not ob-
serve. Furthermore, these kind or reactions usually require higher
temperature and SN2 reaction at the αC next to the ferrocene moi-
ety is also unlikely.
Phospha-Brook rearrangement proceeded on BH3-complexed
phosphinite BH3.3 as well, however alcohol 5 was obtained only
in 11% yield along with 17% of a debrominated compound 7.
Compound
7 was likely formed via protodelithiation from an
organolithium intermediate, which was formed via Br-Li exchange
(Scheme 2b).
NMR spectroscopic characteristics of compound
4 resemble
those of similar ferrocenyl phosphane-oxides obtained by a differ-
ent method [42]. We have calculated the chemical shifts for phos-
phinite 3 and phosphane-oxide 4 at the ωB97X-D/6-31G∗ level.
DFT calculated chemical shifts of phoshane-oxide 4 are closer to
the experimental ones. Key NMR characteristics are gathered in
Table 1.
To gain insight into the details of the phosphinite – phosphane-
oxide isomerization, we have carried out DFT calculations. Opti-
mized structures were obtained using long-range corrected hybrid
density ωB97X-D functional [44]. This functional performs well
for both main group elements as well as for organometallic com-
pounds. Geometry optimizations were performed with LACVP basis
set, which combines Hay and Wadts´ LANL2DZ effective core basis
set for Fe, [45] and 6-31G∗ basis set for all other atoms. Energies
were refined at ωB97X-D/6-311+G∗∗ level using C-PCM scheme to
account for solvent effects (dichloromethane). [46].
Compound rac-4 afforded base-induced phospha-Brook re-
arrangement when treated with three equivalents of nBuLi
(Scheme 2a). The corresponding alcohol 5 was obtained in 26% iso-
lated yield. This result is surprising as phospha-Brook rearrange-
ment is expected to proceed via a phosphinite. In another attempt
to obtain ferrocenyl phosphinites, we have tried to stabilize the
resulting phosphinite via coordination with BH3. Alcohol 2 was
treated with chlorodiphenylphosphine in the presence of DMAP
and triethylamine and then with dimethylsulfide-borane complex.
Even though in only low isolated yield (25%), we have obtained
DFT calculations suggest that isomerization of phosphinite 3 to
the corresponding phosphane-oxide 4 proceeds via a cyclic tran-
sition state (Scheme 3a). Developing a positive charge on the α-
carbon might be stabilized by donor-acceptor orbital interaction
with d-electrons of iron. Similar orbital interactions were described
in related compounds by computational methods as well as X-ray
crystallography [47–50].
DFT calculations (ωB97X-D/LACVP//ωB97X-D/6-311+G∗∗(DCM))
suggest that phosphane-oxide form 4-H is 84.5 kJ/mol more sta-
ble than the model phosphinite 3-H. The difference is even more
pronounced in brominated phosphane-oxide 4-Br, which is even
more stable (106.2 kJ/mol) than the corresponding phosphinite 3-
Br (Scheme 3b).
Table 1
Comparison of selected NMR characteristics of compound
phosphane-oxides.
4 with similar
Compound
δP [ppm] δC(αC)
1JPC [Hz] δH [ppm] 2JPH [Hz]
[ppm]
rac-4 (experimental) 30.9
rac-4 (DFT)a
rac-3 (DFT)a
45.9
43.8
79.4
35.6
46.2
63.0
5.12
4.15
6.03
3.39
3.37
9.2
The natural charge on the α-carbon in the transition state
TS-isom-H is +0.170 (ωB97X-D/6-311+G∗∗ (DCM)). In comparison,
α-carbon in the corresponding carbocation has a natural charge
+0.017. To visualize the stabilization of the transition state of
the isomerization via Fe-CH interaction, we have used intrinsic
bond orbitals (IBO) [51,52]. Scheme 3c shows IBOs for the Fe-CH
126.7
220.0
33.9
FcCH(POPh2)-Me
FcCH(POPh2)-iPr
66.4
66.6
15.8
15.0
32.3
a
NMR chemical shifts were calculated at ωB97X-D/6-31G∗ level using ωB97X-
D/6-31G∗ optimized geometries; stated chemical shifts are weighted arithmetic
mean of chemical shifts of most populated conformers, see SI for details.
Scheme 1. Synthesis of starting material for phospha-Brook rearrangement.
2

A. Almássy, A. Hejtmánková, D. Vargová et al.
Journal of Organometallic Chemistry 941 (2021) 121801
Scheme 2. Experimental results in the phospha-Brook rearrangement.
interaction in the transition state of isomerization and in the
corresponding carbocation for comparison.
that in the presence of BuLi or on the carbanionic intermediate,
the phosphinite/phosphane-oxide isomerization is faster and some
amount of phosphinite is present. This would then allow retro-
phospha-Brook rearrangement to proceed.
DFT calculation (ωB97X-D/LACVP//ωB97X-D/6-311+G∗∗(DCM))
of the BH3 phosphinite and phosphane-oxide showed that the dif-
ference in stability is higher than in uncomplexed species. BH3-
complexed phosphane oxide is 50.9 kJ/mol more stable than BH3-
complexed phosphinite (Fig. 1).
3. Conclusions
To understand retro-phospha-Brook rearrangement on both free
phosphinite 3 as well as BH3-protected phosphinite BH3.3, DFT
calculations were conducted. We have again employed long-range
corrected hybrid density ωB97X-D functional with LACVP basis set
for geometrical optimizations. Energies were refined at ωB97X-
D/6-311+G∗∗ level with C-PCM scheme to account for solvent ef-
fects (tetrahydrofuran). To better model organolithium species in-
volved in this reaction, we have used explicit solvent coordina-
tion with two molecules of THF complexed to lithium. As a start-
ing compound for retro-phospha-Brook rearrangement, the organo-
lithium intermediate Int-PhLi was modeled. Intramolecular trans-
position of the diphenylphosphanyl group from oxygen to car-
bon occurs via a transition state TS-Brook with an activation bar-
rier 45.7 kJ/mol. This process is accompanied by the simultaneous
transposition of lithium from carbon to oxygen. The retro-phospha-
Brook rearrangement’s primary product is a lithium alkoxide Int-
OLi, which is upon aqueous work-up transformed into product 5
(Scheme 4).
Experiments and DFT calculations showed that ferrocenyl
phosphinites rearrange to more stable phosphine-oxides. Borane-
protection seems to stabilize these phosphinites and the pres-
ence of scalar coupling between 31P and 11 B supports this notion.
nBuLi initiated retro-phospha-Brook rearrangement can be realized
both with uncomplexed as well as BH3-complexed starting mate-
rials. Interestingly, this suggests that rearrangement can be accom-
plished also from a phosphine-oxide. Despite the relatively low iso-
lated yields of rearrangement products, this study presents an ex-
ample of an unusual reaction mechanism worth further investiga-
tion.
4. Experimental
All experiments were performed under an argon atmosphere,
using standard Schlenk techniques. Alcohol 2 (CAS: 1277-49-2) was
prepared according to procedures in the literature [15,23]. Other
chemicals were purchased from commercial suppliers (Sigma-
Aldrich, Alfa-Aesar, Acros, TCI Chemicals) and used without fur-
ther purification. Solvents were dried (DCM, and Et3N with CaH2,
THF with Na/benzophenone) and distilled. Thin-layer chromatogra-
phy was performed on Merck TLC-plates silica gel 60, F-254, or on
Al2O3 TLC-plates. Compounds were visualized by irradiation with
UV light. Products were separated by flash chromatography using
Merck silica gel 60 (0.040-0.063 mm). NMR spectra were recorded
on Varian NMR System 300 and 600 (300 or 600 MHz for 1H, 151
MHz for 13C, and 121 or 243 MHz for 31P). Chemical shifts (δ) are
given in ppm relative to tetramethylsilane. High-resolution mass
spectra (HRMS) were recorded using HESI heated electrospray ion-
ization (HESI).
DFT calculations (ωB97X-D/LACVP//ωB97X-D/6-311+G∗∗ (THF))
showed that coordinated borane likely accelerates the rearrange-
ment pathway. Activation barrier (32.0 kJ/mol) is slightly lower
in comparison with the reaction of uncoordinated substrate
(Fig. 2).
The reason why retro-phospha-Brook rearrangement via BH3-
complexed starting material affords products in lower yield is
probably because of side reactions between BuLi and BH3. An in-
teresting question remains, why retro-phospha-Brook rearrange-
ment proceeds even if DFT calculations and spectroscopic stud-
ies of 4 suggest that the starting material is in the form of
phosphane-oxide from which it is difficult to imagine a success-
ful retro-phospha-Brook rearrangement. One hypothesis may be
3

A. Almássy, A. Hejtmánková, D. Vargová et al.
Journal of Organometallic Chemistry 941 (2021) 121801
Scheme 3. a) Phosphinite-phosphane-oxide isomerization; b) DFT (ωB97X-D/6-31G∗//ωB97X-D/6-311+G∗∗(DCM)) structures and relative Gibbs free energies; lengths of
˚
formed and broken bonds in transition states are in A; molecules visualized with CYLview; [53] c) intrinsic bond orbitals for transition state of the isomerization (left)
and the corresponding ferrocenyl(phenyl)methylium cation (right); generated with IboView [54].
Fig. 1. BH3-complexes of phosphinite and phosphane-oxide calculated at (ωB97X-D/LACVP//ωB97X-D/6-311+G∗∗(DCM)) level with relative Gibbs free energies.
4

A. Almássy, A. Hejtmánková, D. Vargová et al.
Journal of Organometallic Chemistry 941 (2021) 121801
Scheme 4. a) Reaction scheme for the retro-phospha-Brook rearrangement; b) reaction energy profile with relative Gibbs free energies obtained at ωB97X-D/LACVP//ωB97X-
D/6-311+G∗∗(THF) level; lengths of formed and broken bonds in the transition state are in A.
˚
5H, Cp), 3.74 (m, 1H, Fc) ppm; 13C NMR (151 MHz, CDCl3): δ 138.1
4.1. 1-(2-Bromobenzyl)phenyl diphenylphosphinite (rac-3)
(Ph), 132.6 (Ph), 131.6 (d, J = 14.8 Hz, Ph), 131.6 (d, J = 8.8 Hz, Ph),
Into a solution of alcohol 2 (0.27 mmol, 100 mg) and DMAP
131.2 (Ph), 131.1 (Ph), 128.8 (Ph), 128.3 (d, J = 12.0 Hz, Ph),128.2
(0.014 mmol, 1.7 mg) in anhydrous DCM, anhydrous Et3N (1.89
(d, J = 12.2 Hz, Ph),127.7 (Ph), 125.7 (d, J = 8.9 Hz, Ph), 84.1 (Fc),
mmol, 0.26 mL) was added. The reaction mixture was cooled to
70.9 (Fc), 68.8 (Fc), 68.7 (Cp), 68.2 (Fc), 66.7 (Fc), 45.9 (d, J = 63.0
Hz, C-O) ppm; 31P NMR (121 MHz, CDCl3): δ 30.9 ppm.
0°C and PPh2Cl (0.27 mmol, 50 μL) was added. After 1h, the re-
action mixture did not contain any starting alcohol 2. The reac-
tion mixture was filtered through Celite® and crude product puri-
fied by column chromatography (SiO2, hexanes/AcOEt, 6:1). Prod-
uct rac-3 (31.5 mg, 21%) was obtained as a yellow crystalline solid.
Mp 187°C (decomp.); Rf = 0.43 (hexanes/AcOEt, 1:1); HRMS: (m/z):
[M]+ calc. for C29H24BrFeOP, 554.0098, found 554.0095; IR (ATR):
ʋ 3052, 2915, 2651, 1719, 1466, 1435, 1259, 1177, 1149, 1105, 1025,
4.2. (2-(Diphenylphosphanyl)phenyl)(ferrocenyl)methanol (rac-5)
Into a solution of compound rac-3 (0.09 mmol, 50 mg) in anhy-
drous THF (9.4 mL) was at -60°C slowly added nBuLi (1,6 M in hex-
ane, 0.27 mmol, 0.16 mL). The reaction mixture was stirred while
the temperature increases slowly to 14°C and then cold water was
added (8 mL). Aqueous phase was diluted with brine and ex-
tracted with Et2O. Combined organic phases were dried over anhy-
drous Na2SO4. The product was isolated by column chromatogra-
phy (SiO2, hexanes/AcOEt 5:6). Alcohol rac-5 (11 mg, 26%) was ob-
tained as a yellow crystalline solid. Mp 205°C (decomp.); Rf = 0.29
927, 811, 791, 747, 693, 645, 609, 542, 523, 504 cm−1 1H NMR (300
.
MHz, CDCl3): δ 8.18 (dt, J = 7.9; 1.7 Hz, 1H, Ph), 7.74-7.59 (m, 4H,
Ph), 7.50-7.27 (m, 8H, Ph), 7.07 (ddd, J = 8.9; 2.6; 1.3 Hz, 1H, Ph),
5.12 (d, J = 9.2 Hz, 1H, HC-O), 4.44 (d, J = 1.2 Hz, 1H, Fc), 4.14 (dd,
J = 3.7; 2.5 Hz, 1H, Fc), 3.89 (td, J = 2.4; 1,2 Hz, 1H, Fc), 3.75 (s,
5

A. Almássy, A. Hejtmánková, D. Vargová et al.
Journal of Organometallic Chemistry 941 (2021) 121801
Fig. 2. DFT calculated structures for retro-phospha-Brook rearrangement under BH3-complexation with relative Gibbs free energies; lengths of formed and broken bonds in
˚
the transition state are in A.
(hexanes/AcOEt 5:6); IR (ATR): ʋ 3297 (br s, OH), 2921, 2852, 1908,
1719, 1465, 1435, 1379, 1260, 1180, 1102, 1075, 1039, 1019, 867, 801,
;
751, 716 cm−1 1H NMR (300 MHz, CDCl3): δ 7.79-7.45 (m, 10H,
4H, Ph), 7.44-7.40 (m, 2H, Ph), 7.37-7.30 (m, 2H, Ph), 7.28-7.24 (m,
2H, Ph), 7.11 (t, J = 7.6 Hz, 1H, Ph), 5.47 (d, J = 16.1 Hz, 1H, HC-
O), 4.26 (s, 1H, Fc), 4.13 (s, 1H, Fc), 3.93 (s, 1H, Fc), 3.81 (s, 5H,
Cp), 3.58 (s, 1H, Fc), 0.6-1.3 (m, 3H, BH3) ppm. 13C NMR (151 MHz,
CDCl3): δ 137.6 (d, J = 1.6 Hz, Ph), 134.1 (d, J = 8.5 Hz, Ph), 132.9
(d, J = 8.2 Hz, Ph), 132.9 (Ph), 131.6 (d, J = 3.0 Hz, Ph), 131.6 (Ph),
131.0 (d, J = 2.3 Hz, Ph), 129.0 (Ph), 128.5 (d, J = 9.8 Hz, Ph), 128.5
(d, J = 52.0 Hz, Ph), 128.3 (d, J = 9.6 Hz, Ph), 127.7 (d, J = 52.7 Hz,
Ph), 127.4 (Ph), 126.1 (d, J = 6.9 Hz, Ph), 84.4 (Fc), 70.6 (d, J = 2.0
Hz, C-O), 68.8 (Cp), 68.6 (Fc), 68.4 (Fc), 87.3 (Fc), 43.9 (d, J = 28.2
Hz, C-O) ppm. 31P NMR (243 MHz, CDCl3): 27.2 (br d, JPB = 42 Hz)
ppm.
Ph), 7.40 (ddd, J = 7.6; 1.5; 1.5 Hz, 1H, Ph), 7.28-7.23 (m, 1H, Ph),
7.16 (dddd, J = 7.5; 7.5; 2.6; 1.3 Hz, 1H, Ph), 6.97 (ddd, J = 14.1; 7.6;
1.4 Hz, 1H, Ph), 6.07 (d, J = 4.1 Hz, 1H, HC-O), 5.41 (d, J = 4.1 Hz,
1H, HO), 4.61-4.57 (m, 1H, Fc), 4.13-4.09 (m, 1H, Fc), 4.02 (s, 5H,
Cp), 3.99-3.95 (m, 1H, Fc), 3.58-3.55 (m, 1H, Fc) ppm; 13C NMR
(151 MHz, CDCl3): δ 150.6 (d, J = 7.6 Hz, Ph), 133.2 (d, J = 102.8
Hz, Ph), 133.0 (d, J = 13.0 Hz, Ph), 132.6 (d, J = 2.5 Hz, Ph), 132.3
(d, J = 9.8 Hz, Ph), 132.2 (d, J = 2.8 Hz, Ph), 132.2 (d, J = 2.9 Hz,
Ph), 131.7 (d, J = 10.0 Hz, Ph), 131.6 (d, J = 106.3 Hz, Ph), 130.3 (d,
J = 101.3 Hz, Ph), 129.3 (d, J = 9.5 Hz, Ph), 128.8 (d, J = 12.2 Hz,
Ph), 128.6 (d, J = 12.4 Hz, Ph), 126.7 (d, J = 13.0 Hz, Ph), 89.9 (Fc),
70.0 (d, J = 5.9 Hz, C-O), 68.9 (Cp), 67.6 (Fc), 67.1 (Fc), 67.0 (Fc),
66.8 (Fc) ppm; 31P NMR (243 MHz, CDCl3): δ 34.1 ppm.
4.4. Procedure for preparation of rac-5 from BH3.rac-3
Compound BH3.rac-3 (0.10 mmol, 56 mg) was dissolved in an-
hydrous THF (3 mL) and the resulting solution was cooled to -78°C.
Into this solution, nBuLi was added (1.6 M in hexane, 0.12 mmol,
70 μL). The reaction mixture was stirred for 1.5 h at -78°C and then
the temperature was allowed to slowly rise to r.t. (18 h). The reac-
tion was quenched with cold H2O (3 mL). Aqueous phase was ex-
tracted with AcOEt and combined organic extracts were dried over
anhydrous Na2SO4. After filtration, the mixture was concentrated
under reduced pressure and product was isolated by column chro-
matography (SiO2, hexanes/AcOEt 10:1). Product rac-5 (5 mg, 11%)
was obtained as a yellow crystalline solid along with debrominated
compound 7 (8 mg, 17%).
4.3. 4-(((Diphenylphosphanyl)oxy)(2-(diphenylphosphanyl)
phenyl)methyl)ferrocene-borane complex (BH3.rac-3)
Alcohol rac-2 (0.27 mmol, 100 mg) and DMAP (0.014 mmol, 1.7
mg) were dissolved in anhydrous DCM. Into this solution, anhy-
drous Et3N (1.89 mmol, 0.26 mL) was added dropwise and then
PPh2Cl (0.27 mmol, 0.05 mL) was added slowly at 0°C. After the
complete conversion of starting alcohol 3 (approx. 1 h), BH3.SMe2
(0.27 mmol, 0.03 mL) was added at 0°C. The mixture was stirred
for 30 min and then the solvent was evaporated, and the products
were separated by column chromatography (SiO2, hexanes/AcOEt,
6:1). The product BH3.rac-3 (38 mg, 25%) was obtained as an or-
ange crystalline solid. In addition, compound rac-3 (22 mg, 15%)
and a 2-bromobenzylferrocene (6, 38 mg, 40%) [55] were also iso-
lated. Rf 0.56 (hexanes/AcOEt, 6:1). HRMS: (m/z): [M - BH3]+ calc.
for C29H24BrFeOP, 554.0098, found 554.0097. IR (ATR): ʋ 3055,
2961, 2923, 2853, 2651, 1891, 1718, 1467, 1432, 1307, 1260, 1190,
1102, 1056, 1020, 948, 925, 831, 813, 743, 713, 688, 670, 630, 585,
4.5. DFT calculations
DFT calculations were performed using ωB97X-D functional,
[44] as implemented in Spartan 18 program package [56]. For
geometrical optimizations, LACVP basis set (a combination of 6-
31G∗ for H, C, N, O, Li, P, and LANL2DZ basis set for Fe) was
used. Energies were refined at ωB97X-D/6-311+G∗∗ level. Sol-
vent effects were evaluated within the context of self-consistent
reaction field theory using the polarizable continuum model
556, 520, 482, 439 cm−1
;
1H NMR (600 MHz, CDCl3): δ 7.83 (d,
J = 7.8 Hz, 1H, Ph), 7.70 (dd, J = 9.1; 7.6 Hz, 2H, Ph), 7.58-7.49 (m,
6

A. Almássy, A. Hejtmánková, D. Vargová et al.
Journal of Organometallic Chemistry 941 (2021) 121801
(PCM)[46] and dichloromethane (ε = 8.93), and tetrahydrofuran
(ε = 7.52) as solvents. Conformational searches were performed
for all starting compound and products using molecular mechanics.
Lowest energy conformers were gradually reoptimized at HF/3-21G
level and then at ωB97X-D/LACVP level.
[18] J.-C. Hierso, R. Smaliy, R. Amardeil, P. Meunier, New concepts in multiden-
tate ligand chemistry: effects of multidentarity on catalytic and spectroscopic
properties of ferrocenyl polyphosphines, Chem. Soc. Rev. 36 (2007) 1754–1769,
doi:10.1039/B701207C.
ꢂ
[19] H. Lang, Organometallic π-tweezers and 1,1 -bis(diphenylphosphanyl)ferrocene
as bidentate chelating ligands for the synthesis of heteromultimetallic com-
pounds, Polyhedron 139 (2018) 50–62, doi:10.1016/j.poly.2017.09.020.
[20] P. Šteˇpnicˇka, Coordination and catalytic chemistry of phosphinoferrocene car-
boxamides, Coord. Chem. Rev. 353 (2017) 223–246, doi:10.1016/j.ccr.2017.10.
003.
Declaration of Competing Interest
[21] J.-C. Zhu, D.-X. Cui, Y.-D. Li, R. Jiang, W.-P. Chen, P.-A. Wang, Ferrocene as
a Privileged Framework for Chiral Organocatalysts, ChemCatChem 10 (2018)
907–919, doi:10.1002/cctc.201701362.
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.
[22] I.R. Butler, The Simple Synthesis of Ferrocene Ligands from a Practitioner’s
Perspective, Eur. J. Inorg. Chem. 2012 (2012) 4387–4406, doi:10.1002/ejic.
201200540.
[23] V. Bariak, A. Malastová, A. Almássy, R. Šebesta, Retro-Brook Rearrangement of
Ferrocene-Derived Silyl Ethers, Chem. Eur. J. 21 (2015) 13445–13453, doi:10.
1002/chem.201501711.
Acknowledgements
[24] A.G. Brook, Molecular rearrangements of organosilicon compounds, Acc. Chem.
Res. 7 (1974) 77–84, doi:10.1021/ar50075a003.
[25] P. Jankowski, P. Raubo, J. Wicha, Tandem Transformations Initiated by the Mi-
gration of a Silyl Group, in: Some New Synthetic Applications of Silyloxiranes,
Synlett, 1994, pp. 985–992, doi:10.1055/s-1994- 23070.
We thank for financial support from Slovak Grant Agency VEGA,
grant no. VEGA 1/0595/17. This publication is the result of the
project implementation: 262401200025 supported by the Research
& Development Operational Programme funded by the ERDF.
[26] W.H. Moser, The Brook rearrangement in tandem bond formation strategies,
Tetrahedron 57 (2001) 2065–2084, doi:10.1016/S0040-4020(00)01089-9.
[27] N. Lee, C.-H. Tan, D. Leow, Asymmetric Brook Rearrangement, Asian J. Org.
Chem. 8 (2019) 25–31, doi:10.1002/ajoc.201800585.
Supplementary materials
[28] Y. Deng, A.B. Smith, Evolution of Anion Relay Chemistry: Construction of Ar-
chitecturally Complex Natural Products, Acc. Chem. Res. 53 (2020) 988–1000,
doi:10.1021/acs.accounts.0c00076.
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.jorganchem.2021.
121801.
˙
˙
[29] A.S. Demir, Ö. Reis, A.Ç. Igˇdir, I. Esiringü, S. Eymur, Generation of Acyl Anion
Equivalents from Acylphosphonates via Phosphonate−Phosphate Rearrange-
ment: A Highly Practical Method for Cross-Benzoin Reaction, J. Org. Chem. 70
(2005) 10584–10587, doi:10.1021/jo051811u.
References
[30] M. Hayashi, S. Nakamura, Catalytic Enantioselective Protonation of α-
Oxygenated Ester Enolates Prepared through Phospha-Brook Rearrangement,
Angew. Chem. Int. Ed. 50 (2011) 2249–2252, doi:10.1002/anie.201007568.
[31] M.T. Corbett, D. Uraguchi, T. Ooi, J.S. Johnson, Base-Catalyzed Direct Aldoliza-
tion of α-Alkyl-α-Hydroxy Trialkyl Phosphonoacetates, Angew. Chem. Int. Ed.
51 (2012) 4685–4689, doi:10.1002/anie.201200559.
[1] P. Štepnicˇka, Ferrocenes: Ligands, Materials and Biomolecules, Wiley, Chich-
ester, 2008.
[2] L.-X. Dai, X.-L. Hou, Chiral Ferrocenes in Asymmetric Catalysis, Wiley-VCH,
Weinheim, 2010.
[3] M. Drusan, R. Šebesta, Enantioselective C-C and C-heteroatom bond form-
ing reactions using chiral ferrocene catalysts, Tetrahedron 70 (2014) 759–786,
doi:10.1016/j.tet.2013.11.012.
[4] Š. Toma, J. Csizmadiova, M. Mecˇiarová, R. Šebesta, Ferrocene phosphane-
heteroatom/carbon bidentate ligands in asymmetric catalysis, Dalton Trans 43
(2014) 16557–16579, doi:10.1039/C4DT01784F.
[32] A. Kondoh, M. Terada, Brønsted Base Catalyzed [2,3]-Wittig/Phospha-Brook
Tandem Rearrangement Sequence, Org. Lett. 15 (2013) 4568–4571, doi:10.1021/
ol402144q.
[33] A. Kondoh, T. Aoki, M. Terada, Intramolecular Cyclization of Alkynyl α-
Ketoanilide Utilizing [1,2]-Phospha-Brook Rearrangement Catalyzed by Phos-
phazene Base, Org. Lett. 16 (2014) 3528–3531, doi:10.1021/ol501479t.
[34] A. Kondoh, T. Aoki, M. Terada, Synthesis of Phenanthrene Derivatives by In-
tramolecular Cyclization Utilizing the [1,2]-Phospha-Brook Rearrangement Cat-
alyzed by a Brønsted Base, Chem. Eur. J. 21 (2015) 12577–12580, doi:10.1002/
chem.201501377.
[35] A. Kondoh, M. Terada, Bronsted base-catalyzed three-component coupling re-
action of [small alpha]-ketoesters, imines, and diethyl phosphite utilizing
[1,2]-phospha-Brook rearrangement, Org. Biomol. Chem. 14 (2016) 4704–4711,
doi:10.1039/C6OB00739B.
[5] F.L. Lam, F.Y. Kwong, A.S.C. Chan, Recent developments on chiral P,S-type lig-
ands and their applications in asymmetric catalysis, Chem. Commun. 46 (2010)
4649–4667, doi:10.1039/C000207K.
[6] R.G. Arrayas, J. Adrio, J.C. Carretero, Recent Applications of Chiral Ferrocene
Ligands in Asymmetric Catalysis, Angew. Chem. Int. Ed. 45 (2006) 7674–7715,
doi:10.1002/anie.200602482.
[7] T.J. Colacot, A Concise Update on the Applications of Chiral Ferrocenyl Phos-
phines in Homogeneous Catalysis Leading to Organic Synthesis, Chem. Rev. 103
(2003) 3101–3118, doi:10.1021/cr000427o.
[36] A. Kondoh, K. Koda, Y. Kamata, M. Terada, Synthesis of Indolizine Derivatives
Utilizing [1,2]-Phospha-Brook Rearrangement/Cycloisomerization Sequence,
Chem. Lett. 46 (2017) 1020–1023, doi:10.1246/cl.170377.
[8] H.U. Blaser, W. Brieden, B. Pugin, F. Spindler, M. Studer, A. Togni, Solvias
Josiphos ligands: from discovery to technical applications, Top. Catal. 19 (2002)
3–16, doi:10.1023/A:1013832630565.
[37] A. Kondoh, A. Iino, S. Ishikawa, T. Aoki, M. Terada, Efficient Synthesis of Poly-
substituted Pyrroles Based on [3+2] Cycloaddition Strategy Utilizing [1,2]-
Phospha-Brook Rearrangement under Brønsted Base Catalysis, Chem. Eur. J. 24
(2018) 15246–15253, doi:10.1002/chem.201803809.
[9] P. Barbaro, C. Bianchini, G. Giambastiani, S.L. Parisel, Progress in stereoselective
catalysis by metal complexes with chiral ferrocenyl phosphines, Coord. Chem.
Rev. 248 (2004) 2131–2150, doi:10.1016/j.ccr.2004.03.022.
[10] K. Yoshida, R. Yasue, Planar-Chiral Ferrocene-Based N-Heterocyclic Carbene Lig-
ands, Chem. Eur. J. 24 (2018) 18575–18586, doi:10.1002/chem.201803903.
[11] N. Dwadnia, J. Roger, N. Pirio, H. Cattey, J.-C. Hierso, Input of P, N-(phosphanyl,
amino)-ferrocene hybrid derivatives in late transition metals catalysis, Coord.
Chem. Rev. 355 (2018) 74–100, doi:10.1016/j.ccr.2017.07.015.
[38] A. Kondoh, K. Aita, S. Ishikawa, M. Terada, Synthesis of Tetrasubstituted Fu-
rans through One-Pot Formal [3 + 2] Cycloaddition Utilizing [1,2]-Phospha-
Brook Rearrangement, Org. Lett. 22 (2020) 2105–2110, doi:10.1021/acs.orglett.
0c00619.
[39] A. Kondoh, N. Tasato, T. Aoki, M. Terada, Brønsted Base-Catalyzed Transfor-
mation of α,β-Epoxyketones Utilizing [1,2]-Phospha-Brook Rearrangement for
the Synthesis of Allylic Alcohols Having a Tetrasubstituted Alkene Moiety, Org.
Lett. 22 (2020) 5170–5175, doi:10.1021/acs.orglett.0c01765.
[12] A. Togni, C. Breutel, A. Schnyder, F. Spindler, H. Landert, A. Tijani, A Novel Eas-
ily Accessible Chiral Ferrocenyldiphosphine for Highly Enantioselective Hydro-
genation, Allylic Alkylation, and Hydroboration Reactions, J. Am. Chem. Soc.
116 (1994) 4062–4066, doi:10.1021/ja00088a047.
[40] A. Kondoh, T. Aoki, M. Terada, Organocatalytic Arylation of α-Ketoesters Based
on Umpolung Strategy: Phosphazene-Catalyzed SNAr Reaction Utilizing [1,2]-
Phospha-Brook Rearrangement, Chem. Eur. J. 24 (2018) 13110–13113, doi:10.
1002/chem.201803218.
[13] J.W. Bats, A. Doppiu, A. Rivas Nass, A.S.K. Hashmi, Sp)-1-Diphenylphosphanyl-
2-{(S)-[2-(diphenylphosphanyl)phenyl]hydroxymethyl}ferrocene, Acta Cryst.
64 (2008) m1585, doi:10.1107/S1600536808038294.
E
[14] J.W. Bats, A. Rivas Nass, A. Doppiu, R. Karch, A.S.K. Hashmi, Rp)-
1-{(R)-(Dimethylamino)[2-(diphenylphosphanyl)phenyl]methyl}-2-
[41] B.V. Rokade, P.J. Guiry, Synthesis of α-Aryl Oxindoles by Friedel–Crafts Alkyla-
tion of Arenes, J. Org. Chem. 85 (2020) 6172–6180, doi:10.1021/acs.joc.0c00370.
[42] L. Ling, J. Hu, Y. Huo, H. Zhang, Synthesis of ferrocene-based phosphine lig-
ands via Cu-catalyzed reductive coupling of ferrocenyl ketone-derived tosylhy-
drazones and H-phosphorus oxides, Tetrahedron 73 (2017) 86–97, doi:10.1016/
j.tet.2016.11.054.
(diphenylphosphanyl)ferrocene chloroform solvate, Acta Cryst.
m1639, doi:10.1107/S160053680803955X.
E
64 (2008)
[15] T. Ireland, K. Tappe, G. Grossheimann, P. Knochel, Synthesis of
Class of Chiral 1,5-Diphosphanylferrocene Ligands and Their Use in Enan-
tioselective Hydrogenation, Chem. Eur. J. (2002) 843–852, doi:10.1002/
a
New
8
[43] R.W. Rudolph, C.W. Schultz, Phosphorus-31-boron-11 coupling constant as a
qualitative measure of dative bond strength, J. Am. Chem. Soc. 93 (1971) 6821–
6822, doi:10.1021/ja00754a020.
1521-3765(20020215)8:4ꢀ843::AID-CHEM843ꢁ3.0.CO;2-9.
[16] T. Sturm, W. Weissensteiner, F. Spindler, A Novel Class of Ferrocenyl-Aryl-Based
Diphosphine Ligands for Rh- and Ru-Catalysed Enantioselective Hydrogenation,
Adv. Synth. Catal. 345 (2003) 160–164, doi:10.1002/adsc.200390003.
[17] P. Šteˇpnicˇka, Phosphino-carboxamides: the inconspicuous gems, Chem. Soc.
Rev. 41 (2012) 4273–4305, doi:10.1039/C2CS00001F.
[44] J.-D. Chai, M. Head-Gordon, Long-range corrected hybrid density functionals
with damped atom–atom dispersion corrections, Phys. Chem. Chem. Phys. 10
(2008) 6615–6620, doi:10.1039/B810189B.
7

A. Almássy, A. Hejtmánková, D. Vargová et al.
Journal of Organometallic Chemistry 941 (2021) 121801
[45] P.J. Hay, W.R. Wadt, Ab initio effective core potentials for molecular calcula-
tions. Potentials for the transition metal atoms Sc to Hg, J. Chem. Phys. 82
(1985) 270–283, doi:10.1063/1.448799.
[46] V. Barone, M. Cossi, Quantum Calculation of Molecular Energies and Energy
Gradients in Solution by a Conductor Solvent Model, J. Phys. Chem. A 102
(1998) 1995–2001, doi:10.1021/jp9716997.
[47] R. Šebesta, M. Mecˇiarová, É. Molnár, J. Csizmadiová, P. Fodran, O. Onomura,
Š. Toma, Enantioselective reductions of [m] ferrocenophanones, J. Organomet.
Chem. 693 (2008) 3131–3134, doi:10.1016/j.jorganchem.2008.07.004.
[48] A.Z. Kreindlin, F.M. Dolgushin, A.I. Yanovsky, Z.A. Kerzina, P.V. Petro-
vskii, M.I. Rybinskaya, Synthesis, crystal and molecular structure of
[50] P. Liptau, T. Seki, G. Kehr, A. Abele, R. Frohlich, G. Erker, S. Grimme, Formation
of a Chelate Bis(phosphino)[3]ferrocenophane Ligand and Its Use in Palladium-
Catalyzed Alternating CO/Ethene Copolymerization, Organometallics 22 (2003)
2226–2232, doi:10.1021/om030076e.
[51] G. Knizia, Intrinsic Atomic Orbitals: An Unbiased Bridge between Quantum
Theory and Chemical Concepts, J. Chem. Theory Comput. 9 (2013) 4834–4843,
doi:10.1021/ct400687b.
[52] G. Knizia, J.E.M.N. Klein, Electron Flow in Reaction Mechanisms—Revealed from
First Principles, Angew. Chem. Int. Ed. 54 (2015) 5518–5522, doi:10.1002/anie.
201410637.
[53] CYLview, 1.0b, Université de Sherbrooke, 2009, http://www.cylview.org.
[54] IboView v20150427, http://www.iboview.org, 2015,
[55] A. Almássy, A. Škvorcová, B. Horváth, F. Bilcˇík, V. Bariak, E. Rakovský,
R. Šebesta, Explanation of Different Regioselectivities in the ortho-Lithiation
of Ferrocenyl(phenyl)methanamines, Eur. J. Org. Chem. (2013) 111–116, doi:10.
1002/ejoc.201201325.
[{C5Me5FeC5Me4CH2}+B{C6H3(CF3)2}4−], the first example of
a struc-
turally characterized primary ferrocenylcarbocation, J. Organomet. Chem. 616
(2000) 106–111, doi:10.1016/s0022-328x(00)00566-0.
[49] K. Müther, P. Hrobárik, V. Hrobáriková, M. Kaupp, M. Oestreich, The Fam-
ily of Ferrocene-Stabilized Silylium Ions: Synthesis, 29Si NMR Characterization,
Lewis Acidity, Substituent Scrambling, and Quantum-Chemical Analyses, Chem.
Eur. J. 19 (2013) 16579–16594, doi:10.1002/chem.201302885.
[56] Spartan’18, Wavefunction, Inc., Irvine, CA.
8