The first flights of a molecular shuttle transporting elements: Easy one-pot formation of organic cyclic arsanes, stibanes and bismutanes

Article abstract of DOI:10.1002/chem.200802007

A study was conducted to demonstrate the synthesis of new arsenic-, antimony-, and bismuth-donor reagents. The study also demonstrated the use of these reagents, to perform one-pot synthesis of cyclic arsanes, stibanes, and bismutanes. The one-pot synthes

Full text of DOI:10.1002/chem.200802007

COMMUNICATION
DOI: 10.1002/chem.200802007
The First Flights of a Molecular Shuttle Transporting Elements: Easy
One-pot Formation of Organic Cyclic Arsanes, Stibanes and Bismutanes
Graziano Baccolini,* Carla Boga,* Marzia Mazzacurati, and Gabriele Micheletti[a]
The increasing role of organic derivatives of arsenic, anti-
mony, and bismuth in many fields of chemistry is due to
their versatility as ligands,[1] their use in materials science[2]
and their applications in the pharmacological field.[3] De-
spite this, few methods of synthesis of triorganoderivatives
of these elements have been reported to date and they all
involve multi-step procedures giving products in very low
overall yields.[4] Moreover, the toxicity of these derivatives
requires that their preparation must be carried out with
strong attention and adequate precautions in their handling.
For these reasons, to find synthetic one-pot procedures that
permit the manipulation of these compounds to be mini-
mised represents an interesting challenge.
Scheme 1. One-pot synthesis of tertiary phosphanes.
Here we report a very simple synthesis of new arsenic-,
antimony- and bismuth-donor reagents and their use in the
one-pot synthesis of cyclic arsanes, stibanes and bismutanes,
respectively, through a procedure in which the simultaneous
tion of the reagent 1a can be accomplished by simple addi-
tion of one equivalent amount of PCl3 to the crude reaction
mixture.[8]
À
À
À
formation of three C As, C Sb or C Bi bonds is achieved.
Recently, we reported[5] (Scheme 1) a new synthesis of ter-
tiary cyclic phosphanes 2 and 3 in 60–70% yields[5a,b] by ad-
Taking these findings into consideration, together with the
fact that as As, Sb and Bi, belong to the same group of P
they could show similar behaviour, we decided to try a
transport process for these elements, in order to find a new
protocol for obtaining cyclic organoarsanes, organostibanes
and organobismutanes, which, to date, have been synthes-
ised with multi-step procedures and in very low overall
yields.[9]
For this purpose, firstly we tried the synthesis of new het-
erocycles 1b–d (Scheme 2), by reaction between compound
4 and AsCl3, SbCl3, and BiCl3, respectively.[10] With this reac-
tion benzothiaphospharsole (1b), benzothiaphosphastibole
(1c), and benzothiaphosphabismole (1d; Scheme 2) were
isolated in almost quantitative yields (95–98%) and charac-
terised.
dition of a bisACHTUNGTRENNUNG(Grignard) reagent and a Grignard reagent to
an unusual phosphorus-donor reagent, the benzothiadi-
phosphole 1a,[6] called by us a “butterfly reagent”[7] owing
to its particular folded structure. At the end of the reaction
the starting reagent 1a was quantitatively regenerated by
addition of one equivalent amount of PCl3 to the end-prod-
uct 4, which is the remainder of 1a obtained after quenching
the reaction mixture with water (Scheme 1). More recently,
we found that tertiary phosphanes are formed spontaneously
in the reaction mixture, and that the quantitative regenera-
[a] Prof. G. Baccolini, Dr. C. Boga, Dr. M. Mazzacurati,
Dr. G. Micheletti
It is worth noting that mass spectra of 1a–d showed the
same peak (m/z=243), originating by the loss from the mo-
lecular ion of the fragment PS, AsS, SbS and BiS, respective-
ly. At the same time, the 31P NMR spectra showed a regular
decrease of the chemical shifts on going from 1b (d=
78.6 ppm) to 1c (d=52.7 ppm), to 1d (d=35.0 ppm) in
[D8]THF, as a result both of the decrease in electronegativi-
Department of Organic Chemistry “A. Mangini”
ALMA MATER STUDIORUM—Universitꢀ di Bologna
Viale Risorgimento, 4 40136 Bologna (Italy)
Fax : (+39)051 2093654
E-mail: baccolin@ms.fci.unibo.it
boga@ms.fci.unibo.it
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802007.
Chem. Eur. J. 2009, 15, 597 – 599
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
597

of bismuth have been often reported[12] to be very stable or
substituted by ionic species).[1,13]
However, it has to be pointed out that our methodology
allows us to obtain derivatives 2a–d and 3a–d in overall
yields higher with respect to those reported so far, with a
noticeable improvement in terms of safety of the whole pro-
cess thanks to the possibility to carry out it in a one-pot pro-
cess.
In addition, the yield of the re-formation of compound 1
from 5 is nearly quantitative, and the cycle can be repeated
more times (i.e., after three cycles, 2a was isolated in
amount corresponding to a mean of 70% yields each cycle),
thus avoiding the step-by-step separation of reaction prod-
ucts, which could imply a decrease of yields since com-
pounds 1a–d and 5 (as well as its neutral form 4) are stable
in solution but, once isolated, became air-sensitive and tend
to form the corresponding oxides.
The efficiency of the one-pot one-step synthesis of cyclic
derivatives of P, As, Sb and Bi can be explained by the inter-
vention of penta- and hexacoordinated intermediates, such
as A and B in Scheme 3, shown for the case of phosphorus
derivatives as an example.
Scheme 2. Synthesis and reactions of element-donor reagents 1a–d.
ty and of the concomitant increase in metallic character
down to the group. Furthermore, we observed a decreasing
solubility of these new heterocycles moving toward the
heaviest elements that parallels the growing tendency to the
self-assembling in dimeric or cluster-like structures.[11]
Once 1b–d were obtained, we focused our efforts on their
use as starting reagents to realise a one-pot synthesis of
cyclic arsanes, stibanes and bismutanes. Taking in consider-
ation that tertiary phosphanes are produced from 1a directly
in the reaction mixture,[8] we wanted to try to carry out a
“catalytic process of transport of an element” aimed at the
continuous formation of cyclic tertiary organoderivatives of
P, As, Sb and Bi without isolation of any intermediate and
with the “in situ” re-formation of starting “butterfly” re-
agents, as shown in Scheme 2. The “transporter system” is
formed by two molecules. The first is the reagent 1, which
donates the atom that lies between the two sulfur atoms to
Grignard
reagents
(BrMg
(CH2)
G
and
CH3MgBr) forming cyclic tertiary organoderivatives 2 or 3.
The second molecule, compound 5—the magnesium salt
precursor of 4—obtained after spontaneous expulsion of the
cyclic product as residue of the “butterfly reagent”, can be
easily re-transformed “in situ” into the starting reagent 1 by
simple addition of ECl3 (E=P, As, Sb, Bi). This double abili-
ty of compound 5, either as acceptor of element when it
reacts with ECl3 and as donor of element in its “activate
form” 1, means that it can be considered as a catalytic carri-
er of the element, recovered at the end of the process.
With this method, we obtained compounds 2a–d and 3a–
d in good to satisfactory yields. In particular, phosphanes
were obtained in 70–80% yields, the new benzothiaphos-
pharsole 1b also acts as efficient arsenic-donor molecule,
giving cyclic arsanes in 60–70% yields, while cyclic stibanes
and bismutanes were obtained in 50–55 and 25–30% yields,
respectively. This gradual decrease of the yields of cyclic
products on going from starting reagent 1a to 1d could be
explained by considering that the heavier elements Sb and
Bi in Group 15 possess a reactivity different from that of the
preceding elements, probably due to the formation of hyper-
valent species of Sb and Bi that make difficult the reaction
with Grignard reagents (penta- and hexacoordinated species
Scheme 3. Hypervalent phosphorus intermediates involved in the forma-
tion of tertiary cyclic phosphanes.
The formation of these species, detected by 31P NMR
spectroscopy in the case of formation of phosphanes,[5b,14] is
highly favoured because the rigid “butterfly”[6] structure of
1a, with its bicyclic condensed system around the phospho-
rus atom, strongly favours the formation of hypervalent in-
termediates,[15] each ring being a factor of great stability, re-
ducing the overcrowding.
In summary, we have realised the synthesis of new hetero-
À
À
À
cycles 1b–d bearing P As, P Sb and P Bi bonds that can
act, in turn, as carriers of arsenic, antimony, and bismuth to
give, in good yields, cyclic organoarsanes, organostibanes
and organobismutanes. The starting phosphorus-, arsenic-,
antimony- and bismuth-donor reagents can be re-formed in
situ by simple addition to the crude reaction mixture of
PCl3, AsCl3, SbCl3 or BiCl3, respectively, without isolation of
598
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 597 – 599

Heterocyclic Compounds
COMMUNICATION
[2] a) G. G. Briand, N. Burford, Chem. Rev. 1999, 99, 2601–2657;
b) W. R. Cullen, K. J. Reimer, Chem. Rev. 1989, 89, 713–764; Y.-Z.
Huang, Acc. Chem. Res. 1992, 25, 182–187; c) Inorganic Materials
(Eds.: D. W. Bruce, D. O. Hare), Wiley, New York, 1992; d) Chem-
ical Vapor Deposition: Principles and Applications (Eds.: M.-L.
Hitchman, K.-F. Jensen), Academic, London, 1993.
intermediates, thus permitting to consider the whole process
as a catalytic cycle.
Experimental Section
[3] a) J. Mishra, A. Saxena, S. Sing, Curr. Med. Chem. 2007, 14, 1153–
1169; b) E. Guillamet, A. Creus, J. Ponti, E. Sabbioni, S. Fortaner,
R. Marcos, Mutagenesis 2004, 19, 129–135; c) T. Kotani, D. Nagai,
K. Asahi, H. Suzuki, F. Yamao, N. Kataoka, T. Yagura, Antimicrob.
Agents Chemother. 2005, 49, 2729–2734; d) I. Mancini, G. Guella,
M. Frostin, E. Hnawia, D. Laurent, C. Debitus, F. Pietra, Chem. Eur.
J. 2006, 12, 8989–8994.
[4] a) M. D. Smith, Product Class 1: Arsenic Compounds, Science of
Synthesis Vol. 4 (Ed.: I. Fleming), Thieme, Stuttgart, 2002, pp. 13–
51; b) J. W. Burton, Product Class 2: Antimony Compounds, Science
of Synthesis Vol. 4 (Ed.: I. Fleming), Thieme, Stuttgart, pp. 53–57;
c) H Suzuki, T. Ikegami, Product Class 3: Bismuth Compounds, Sci-
ence of Synthesis Vol. 4 (Ed.: I. Fleming), Thieme, Stuttgart, pp. 77–
116.
[5] a) G. Baccolini, C. Boga, U. Negri, Synlett 2000, 1685–1687; b) G.
Baccolini, C. Boga, R. A. Buscaroli, Eur. J. Org. Chem. 2001, 3421–
3424; c) G. Baccolini; C. Boga, M. Galeotti, Angew. Chem. 2004,
116, 3120–3122; Angew. Chem. Int. Ed. 2004, 43, 3058–3060.
[6] a) G. Baccolini, E. Mezzina, P. E. Todesco, E. Foresti, J. Chem. Soc.
Chem. Commun., 1988, 304–305; b) G. Baccolini, M. Beghelli, C.
Boga, Heteroatom. Chem. 1997, 8, 551–556.
Preparation of compounds 1b–d:[10] AsCl3 (or SbCl3, or BiCl3, 1.0 mmol)
was added, at room temperature, to a solution of compound 4 (0.278 g,
1.0 mmol) in dichloromethane (30 mL) under argon atmosphere. After
about 20 m. the solvent was removed giving quantitatively the corre-
sponding compound 1.
Preparation of compound 5:
A solution of a Grignard reagent
(3.0 mmol) in THF was added, at 258C, to a solution of 1a (1.0 mmol) in
anhydrous THF (35 mL) under argon. The reaction was monitored by
31P NMR spectrometry: when the signals of reagent 1a disappeared, and
those of the tertiary phosphane and of 5 appeared, the solvent was re-
moved under vacuum to about 1/5 of the starting volume, and the result-
ing suspension was filtered under argon. The white salt 5 was washed
once with anhydrous THF (1–2 mL) and the solvent was removed under
argon atmosphere.
Preparation of compounds 2a–d and 3a–d: ECl3 (2.0 mmol, E=P, As,
Sb, Bi) was added to a suspension of 5 (2.0 mmol) in THF (20–30 mL).
After about 20 min the formation of compound 1 (1a, 1b, 1c or 1d) was
complete. Then, a solution containing BrMgACTHNUTGRNEG(UN CH2)nMgBr (n=1 or 2,
2.0 mmol) and CH3MgBr (2.0 mmol) in THF was quickly added, at room
temperature. The reaction course was monitored by GC-MS and
31P NMR spectroscopy. After about 40 min. the signal of the starting re-
agent 1 disappeared, and concomitantly appeared those of 5 and 2 (n=1,
or 3, n=2). At this stage, an equivalent amount of ECl3 was added to the
crude reaction mixture and the re-formation of corresponding reagent 1
was detected. The cycle was repeated at least three times; at the end
compound 5 was recovered by filtration, and cyclic compound was isolat-
ed from the solution by removal of the solvent followed by distillation.
More details and other procedures are given in the Supporting Informa-
tion, together with chemicophysical data of all synthesised compounds.
[7] G. Baccolini, C. Boga, M. Mazzacurati, J. Org. Chem. 2005, 70,
4774–4777.
[8] a) G. Baccolini, C. Boga. M. Mazzacurati, Eur. J. Org. Chem. 2007,
4529–4534.
[9] a) F. G. Mann, The Heterocyclic Derivatives of Phosphorus, Arsenic,
Antimony and Bismuth, Wiley-Interscience: New York, 1970;
b) R. E. Atkinson, Heterocyclic Rings containing Arsenic, Antimomy
or Bismuth, Comprehensive Heterocyclic Chemistry. Vol. 1 (Eds.:
A. R. Katritzky, C. W. Rees), Pergamon, Oxford, 1984, pp. 539–561;
c) K. C. Caster, Arsoles, Stiboles, and Bismoles, Comprehensive Het-
erocyclic Chemistry II. Vol. 2 (Eds.: A. R. Katritzky, C. W. Rees,
E. F. Scriven, C. W. Bird), Pergamon, Oxford, 1996, pp. 891–895.
[10] Compounds 1a–d can be also obtained by reaction between 5 and
ECl3, see the Supporting Information.
[11] a) C. Jones, Coord. Chem. Rev. 2001, 215, 151–169; b) J. W. Wie-
landt, N. L. Kilah, A. C. Willis, S. B. Wild, Chem. Commun. 2006,
3679–3680.
[12] a) R. Luckenbach, Dynamic Stereochemistry of Pentacoordinated
Phosphorus and Related Elements, Thieme, Stuttgart, 1973; b) G.
Wittig, D. Hellwinkel, Angew. Chem. 1962, 74, 76; Angew. Chem.
Int. Ed. Engl. 1962, 1, 53; c) D. Hellwinkel, M. Bach, Justus Liebigs
Ann. Chem. 1968, 720, 198.
Acknowledgements
Work supported by Alma Mater Studiorum—Universitꢀ di Bologna
(RFO funds) and MIUR (PRIN Project 2007)
Keywords: antimony
· arsenic · bismuth · phosphorus
heterocycles · pnictides
[13] a) Chemistry of Hypervalent Compounds (Ed.: K. Akiba), Wiley-
VCH, Weinheim, 1999; b) Organobismuth Chemistry (Eds.: H.
Suzuki, Y. Matano), Elsevier, Amsterdam, 2001.
[1] a) W. Levason, G. Reid, Coord. Chem. Rev. 2006, 250, 2565–2594,
and references therein; b) H. Werner, Angew. Chem. 2004, 116,
956–972; Angew. Chem. Int. Ed. 2004, 43, 938–954; c) N. A. Comp-
ton, R. J. Errington, C. Nicholas, Adv. Organomet. Chem. 1990, 31,
91–182; d) C. A. McAuliffe, W. Levason, Phosphine, Arsine and Sti-
bine Complexes of the Transition Elements, Elsevier, Amsterdam,
1978; e) J. L. Wardell, Arsenic Antimony, and Bismuth in Compre-
hensive Organometallic Chemistry II, Vol. 2 (Eds.: E. W. Abel,
F. G. A. Stone, G. Wilkinson, A. G. Davies), Pergamon, Oxford,
1995.
[14] G. Baccolini, C. Boga. M. Mazzacurati, Tetrahedron 2007, 63,
12595–12600.
[15] For reviews on hypervalent phosphorus see: a) R. R. Holmes, Penta-
coordinated Phosphorus Structure and Spectroscopy, ACS Mono-
graph 175, Vol. 1–2, American Chemical Society, Washington, 1980;
b) R. R. Holmes, Acc. Chem. Res. 1998, 31, 535–542; c) C. Y. Wong,
D. K. Kennepohl, R. G. Cavell, Chem. Rev. 1996, 96, 1917–1951.
Received: October 31, 2008
Published online: December 4, 2008
Chem. Eur. J. 2009, 15, 597 – 599
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
599