α,β-Aziridinylphosphonates by lithium amide-induced phosphonyl migration from nitrogen to carbon in terminal aziridines

Article abstract of DOI:10.3762/bjoc.6.110

N-Phosphonate terminal aziridines undergo lithium 2,2,6,6- tetramethylpiperidide-induced N-to C-[1,2]-anionic phosphonyl group migration under experimentally straightforward conditions, to provide a stereocontrolled access to synthetically valuable trans-

Full text of DOI:10.3762/bjoc.6.110

α,β-Aziridinylphosphonates by lithium amide-induced
phosphonyl migration from nitrogen to carbon
in terminal aziridines
David. M. Hodgson* and Zhaoqing Xu
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2010, 6, 978–983.
Department of Chemistry, Chemistry Research Laboratory, University
of Oxford, Mansfield Road, Oxford, OX1 3TA, UK, Fax: +44(1865)
285002
doi:10.3762/bjoc.6.110
Received: 13 August 2010
Accepted: 28 September 2010
Published: 13 October 2010
Email:
David. M. Hodgson* - david.hodgson@chem.ox.ac.uk
Associate Editor: J. Aubé
* Corresponding author
© 2010 Hodgson and Xu; licensee Beilstein-Institut.
License and terms: see end of document.
Keywords:
amino acids; aziridines; lithiation; migration; synthetic methods
Abstract
N-Phosphonate terminal aziridines undergo lithium 2,2,6,6-tetramethylpiperidide-induced N- to C-[1,2]-anionic phosphonyl group
migration under experimentally straightforward conditions, to provide a stereocontrolled access to synthetically valuable trans-α,β-
aziridinylphosphonates. The utility of this chemistry has been demonstrated in the asymmetric synthesis of a β-aminophosphonate.
Introduction
The synthesis of aminophosphonic acids and their derivatives on the generation and subsequent chemistry of α-lithiated
has attracted considerable attention, since the presence of such terminal aziridines [14], we considered whether α,β-
functionality, typically as amino acid surrogates, leads to aziridinylphosphonates 3 could be accessed by α-lithiation of
interesting bioactivity in, for example, antibacterial agents, N-phosphonate terminal aziridines 1, followed by N- to C-[1,2]-
enzyme inhibitors and herbicides [1-3]. α,β-Aziridinylphospho- anionic phosphonyl group migration in lithiated intermediate 2
nates 3, while possessing biological activity themselves, can be (Scheme 1). Here, we present full details of this study [15].
converted into aminophosphonic acids using various ring-
opening processes [4-6]. Aziridinylphosphonates 3 have previ-
ously been prepared by a variety of methods [4-6], of which
ring-closure following either Sharpless aminohydroxylation of
α,β-unsaturated phosphonates [7], or α-halophosphonate add-
ition to sulfinimines [8] constitute notable asymmetric
Scheme 1: Proposed aziridinyl anion induced N- to C-phosphonyl
migration.
approaches, albeit principally leading to β-aryl substituted α,β-
aziridinylphosphonates. Arising from our investigations [9-13]
978

Beilstein J. Org. Chem. 2010, 6, 978–983.
Based-induced migration, involving cleavage of a nitrogen
phosphorus bond and formation of a carbon phosphorus bond,
was first reported over 30 years ago by Hellwinkel and
co-workers (4→5, Scheme 2) [16,17]. More recently, Modro et
al. reported a similar ortho-lithiation followed by [1,3]-shift of a
phosphonyl group (6→7, Scheme 2) [18]; related processes
have been observed with diazaphospholidine oxides [19] and
bicyclic phosphoric triamides [20]. Benzylic lithiation-induced
N- to C-[1,2]-anionic phosphonyl rearrangements (8→9,
Scheme 2) were developed by Hammerschmidt and Hanbauer
[21], and a stereoretentive N- to C-[1,2]-shift in an α-lithiated
pyrrole involving a chiral tert-butyl(phenyl)phosphinyl group
has been reported [22].
Scheme 3: Partial N- to C-migration with N-diphenylphosphinylaziri-
dine 10 [24].
Results and Discussion
So as to examine the migration chemistry outlined in Scheme 1,
access to N-phosphonate terminal aziridines 1 was required
[25]. These can be concisely prepared from alkenes using chem-
istry developed by Zwierzak and co-workers [26-28]. In a one-
flask operation, addition of 1-hexene (13) to Br2NPO(OEt)2
[29] in the presence of UV light [27] and subsequent treatment
with NaH (2 equiv) [26] gave the representative substrate 1a
(57%, Scheme 4) [28].
Scheme 2: Selected previously observed N- to C-phosphorous migra-
tions [17,18,21].
With regard to previous anion-induced N- to C-1,2-shifts in
aziridines, in one isolated example, the N-Boc-aziridine of
styrene was treated with s-BuLi in THF at –98 °C to give a
phenyl-stabilised α-lithiated aziridine, which underwent migra-
tion to give 2-phenyl-2-Boc-aziridine (90%) [23]. Also, our
laboratory has reported the LTMP (lithium 2,2,6,6-tetra-
methylpiperidide)-induced rearrangement of a range of terminal
Scheme 4: Synthesis and rearrangement of aziridine 1a.
N-Boc-aziridines to give trans-aziridinyl esters [12,15] (cf. Initially, we examined organolithiums for their propensity to in-
Scheme 1, with CO2t-Bu instead of PO(OEt)2). However, prior duce deprotonation-migration in N-phosphonate aziridine 1a.
to our studies only one example of a N- to C-[1,2]-anionic Despite it being previously noted by Zwierzak that the reaction
rearrangement of an aziridine involving phosphorus had been of organolithium reagents with such substrates resulted in pre-
observed: lithiation-deuteration of N-diphenylphosphinylaziri- ferential attack at phosphorus giving a complex mixture of
dine 10 gave the anticipated deuterated aziridine 11 (70%), products [30], we did obtain some of the desired rearrangement
along with the rearranged aziridine 12 (25%) [24] (Scheme 3). product 3a (Scheme 4). When aziridine 1a in THF was treated
979

Beilstein J. Org. Chem. 2010, 6, 978–983.
with n-BuLi (1.5 equiv) at −78 °C for 4 h, 46% of rearrange- phonate 3f (entry 6), without potential elimination [34] of the
ment product 3a was obtained (66% based on recovered 1a). primary chloride. From 2-chloroethylamine hydrochloride, the
The stereochemistry of aziridinylphosphonate 3a was assigned simplest N-phosphonate aziridine 1i (Figure 1) could be
as trans on the basis of the small sizes of the vicinal H-H straightforwardly accessed [35]. However, this latter substrate
couplings across the ring in 3a and in the derived N-Boc decomposed under the lithiation conditions. A 2,2-disubstituted
aziridinylphosphonate 14 (3–3.5 Hz), which are diagnostic for N-phosphonate aziridine 1g, available from isobutene and
such systems (the corresponding cis-aziridinylphosphonates are Br2NPO(OEt)2 without UV assistance followed by methanolic
known to have larger 3J values of 6–7 Hz) [31,32]. Generation NaOMe-induced ring-closure [26,27,30], proved viable in the
of trans-stereochemistry was also observed in the corres- lithiation [1,2]-shift chemistry giving aziridinylphosphonate 3g
ponding N-Boc system [12,15] (see earlier discussion), and the (58%, Table 1, entry 7). However, and similarly to the N-Boc
present transformation likely follows a similar reaction aziridine of cyclohexene [12,15], no reaction was observed
pathway: initial trans α-lithiation, which is probably assisted by between LTMP and a 2,3-disubstituted aziridine 1j (Figure 1),
prior complexation of the base with the N-protecting group, fol- presumably due to steric interactions and/or reduced acidity
lowed by intramolecular migration [12] of this group.
impeding lithiation.
Reactions of s-BuLi or t-BuLi with aziridine 1a under the same
conditions as above were less effective, giving a 22% yield of
3a (50% based on recovered 1a) and only traces of 3a (45% 1a
recovered), respectively. In an attempt to induce greater conver-
sion of 1a with n-BuLi, the amount of the latter was increased
to 3 equivalents, however only 25% of 3a was obtained. Given
previous successes with the use of lithium amides to deproton-
ate differently N-protected terminal aziridines [9-14], we moved
Figure 1: Aziridines 1h–j.
on to study such bases with N-phosphonate aziridine 1a. Firstly, LDA has been reported to rapidly (THF, –78 °C, 10 min)
LDA (lithium diisopropylamid) and LiNCy2 were examined. isomerize N-Ph cis-β-phenyl-aziridinylphosphonate to mainly
Although there was no significant increase in yield on using (80:20) the trans-isomer 3 (NPh instead of NH, and R = Ph) via
LDA or LiNCy2, (38% and 15% of 3a, respectively), under the an α-deprotonation pathway [36]. Therefore, although substrate
same conditions as for n-BuLi described above, side-reactions and/or product co-ordination and/or LTMP aggregation
were not observed. These results prompted us to try the stronger phenomena could be invoked to rationalise the beneficial effect
lithium amide, LTMP. Initial studies indicated that with 3 equiv of the excess LTMP on yield, and we viewed α-deprotonation
of LTMP, 3a was obtained in 52% yield (94% based on recov- as more difficult in our likely post-rearrangement intermediate 3
ered 1a). Prolonging the reaction time from 4 to 8 h made no (N–Li instead of N–H, Scheme 1), we sought to rule out the
difference to the yield. However, when the amount of LTMP possibility that LTMP might be being consumed by further
was increased to 4 equiv and to 5 equiv, 3a was obtained in lithiation in the rearranged product. In the event, quenching a
65% and 91% yields, respectively. These latter conditions were migration reaction of N-phosphonate aziridine 1a with CD3OD
then applied to a range of terminal N-phosphonate aziridines 1 did not show any deuterium incorporation. This prompted a
(Table 1).
closer investigation of the reaction using a more rigorously
purified sample of aziridine 1a than which had been used in our
Terminal aziridine 1b, possessing secondary alkyl substitution preliminary studies. This latter work established that direct
on the aziridine, gave the rearranged aziridinylphosphonate 3b application of our earlier N-Boc migration conditions (3 equiv
in 79% yield (Table 1, entry 2). Aziridines 1c and 1d under- LTMP, –78 °C, 90 min) [12] gave complete consumption of the
went migration smoothly without complications arising from starting aziridine 1a, with aziridinylphosphonate 3a being
potential allylic deprotonation [33], intramolecular cyclopropa- isolated in 82% yield (use of 2 equiv LTMP gave 3a in 68%
nation [11,12] or benzylic deprotonation (entries 3 and 4). yield, with 7% recovered 1a) and suggests that 5 equiv of the
Mixed results were obtained when the method was applied to base may not always be needed to effect efficient rearrange-
substrates possessing silyl ether functionality. The distal- ment.
protected aziridine 1e provided stable aziridinylphosphonate 3e
in high yield (entry 5), whereas rapid decomposition was Asymmetric access to an aziridinylphosphonate was also
observed during attempted purification of the product mixture explored to demonstrate further the utility of the above method-
from the proximal-protected substrate 1h (Figure 1). However, ology. As terminal epoxides are readily available as single
the migration reaction could be used to prepare aziridinylphos- enantiomers [37], terminal epoxide-opening with H2NPO(OEt)2
980

Beilstein J. Org. Chem. 2010, 6, 978–983.
Table 1: Aziridinylphosphonate 3 synthesis.
Entry
1
Aziridine 1
Aziridinylphosphonate 3
Yield (%)
91
2
3
4
5
6
7
79
79
87
95
87
58
followed by ring-closure was initially considered as a potential (15) using ammonia, which gave β-amino alcohol 16 in good
asymmetric route to N-phosphonate terminal aziridines 1. yield (Scheme 5) [40]. By analogy with a preparation of
However, 1,2-epoxyhexane could not be successfully ring- N-diphenylphosphinyl aziridines [41], subsequent one-pot N-
opened with H2NPO(OEt)2 under a variety of conditions (NaH, and O-phosphonylation followed by NaH-induced ring-closure
KH, or NaH/DMPU in THF; (i-PrO)4Ti or BF3•Et2O in gave (S)-aziridine 1k (52%). Lithiation–rearrangement of (S)-
CH2Cl2; aminolytic kinetic resolution [12,38]), and use of aziridine 1k provided aziridinylphosphonate (–)-3k in excellent
(Me3)SiHNPO(OEt)2 [39] also proved ineffective. A conveni- yield (89%), and without any degradation of enantiopurity
ent asymmetric access was eventually developed, starting with (>99% ee, as determined by chiral HPLC analysis of the
ring-opening of commercially available (R)-1,2-epoxybutane benzoyl derivative 17).
981

Beilstein J. Org. Chem. 2010, 6, 978–983.
Conclusion
In summary, lithiation-induced phosphonyl migration from
nitrogen to carbon in terminal aziridines 1 effects simultaneous
N-deprotection and accesses synthetically valuble N–H trans-
aziridinylphosphonates 3. The process is experimentally
straightforward to carry out and occurs in an efficient and
completely stereocontrolled fashion. The utility of the chem-
istry is further highlighted by a subsequent conversion into an
enantiopure β-aminophosphonate 18, demonstrating an entry to
a biologically important β-amino acid mimic class.
Supporting Information
Full preparative details of all compounds are reported,
together with their spectroscopic data.
Scheme 5: Synthesis and rearrangement of aziridine (S)-1k.
Supporting Information File 1
Experimental and analytical data
Hydrogenolytic ring-opening of aziridinylphosphonates
provides an attractive entry to α- or β-aminophosphonates [6].
α,β-Aziridinylphosphonates bearing a β-aryl group undergo
cleavage at the formally benzylic C–N bond, leading to
α-aminophosphonates. With β-alkyl groups, regioselectivity is
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-6-110-S1.pdf]
influenced by the presence or absence of an N-substituent. N-Ts Acknowledgements
Cis-β-alkyl-substituted aziridinylphosphonates give We thank the Royal Society for an International Incoming
β-aminophosphonates [32], whereas both N-Boc cis- and trans- Fellowship (to Z. X.), P. G. Humphreys for some additional
β-alkyl-substituted aziridinylphosphonates lead to experimental results, C. Pousset (Solvay) for hydrogenolysis
α-aminophosphonates. For example, hydrogenolysis of N-Boc details and the EPSRC National Mass Spectrometry Service
trans-aziridinylphosphonate 14 (Bn instead of the C4H9 group) Centre for mass spectra.
using 10% Pd/C (H2 (1 atm), EtOH, 12 h) has been reported to
References
1. Kukhar, V. P.; Hudson, H. P., Eds. Aminophosphonic and
give the corresponding N-Boc α-aminophosphonate (63%) [31].
In the absence of an N-substituent, cis-β-alkyl-substituted
Aminophosphinic Acids: Chemistry and Biological Activity; Wiley:
aziridinylphosphonates give β-aminophosphonates [32] and we
Chichester, 2000.
observed that trans-β-alkyl-substituted aziridinylphosphonate
2. Palacios, F.; Alonso, C.; de los Santos, J. M. Chem. Rev. 2005, 105,
(–)-3k underwent completely regioselective hydrogenolysis
899–932. doi:10.1021/cr040672y
under transfer hydrogenation conditions [31] to produce
3. Orsini, F.; Sello, G.; Sisti, M. Curr. Med. Chem. 2010, 17, 264–289.
β-aminophosphonate (+)-18 [32,42] in 68% yield with >99% ee
(determined by chiral HPLC analysis of the benzoyl derivative
19, Scheme 6).
doi:10.2174/092986710790149729
4. Moonen, K.; Laureyn, I.; Stevens, C. V. Chem. Rev. 2004, 104,
6177–6216. doi:10.1021/cr030451c
5. Zhou, P.; Chen, B.-C.; Davis, F. A. Asymmetric Syntheses with
Aziridinecarboxylate and Aziridinephosphonate Building Blocks. In
Aziridines and Epoxides in Organic Synthesis; Yudin, A. K., Ed.;
Wiley-VCH: Weinheim, 2006; pp 73–115. doi:10.1002/3527607862.ch3
6. Singh, G. S.; D'hooghe, M.; De Kimpe, N. Chem. Rev. 2007, 107,
2080–2135. doi:10.1021/cr0680033
7. Thomas, A. A.; Sharpless, K. B. J. Org. Chem. 1999, 64, 8379–8385.
doi:10.1021/jo990060r
8. Davis, F. A.; Wu, Y.; Yan, H.; McCoull, W.; Prasad, K. R. J. Org. Chem.
2003, 68, 2410–2419. doi:10.1021/jo020707z
9. Hodgson, D. M.; Humphreys, P. G.; Ward, J. G. Org. Lett. 2005, 7,
1153–1156. doi:10.1021/ol050060f
10.Hodgson, D. M.; Miles, S. M. Angew. Chem., Int. Ed. 2006, 45,
935–938. doi:10.1002/anie.200503303
Scheme 6: Hydrogenolysis of aziridinylphosphonate (–)-3k.
982

Beilstein J. Org. Chem. 2010, 6, 978–983.
11.Hodgson, D. M.; Humphreys, P. G.; Ward, J. G. Org. Lett. 2006, 8,
995–998. doi:10.1021/ol060101n
35.Osowska-Pacewicka, K.; Zwierzak, A. Synthesis 1996, 333–335.
doi:10.1055/s-1996-4213
12.Hodgson, D. M.; Humphreys, P. G.; Miles, S. M.; Brierley, C. A. J.;
Ward, J. G. J. Org. Chem. 2007, 72, 10009–10021.
doi:10.1021/jo701901t
36.Coutrot, P.; Elgadi, A.; Grison, C. Heterocycles 1989, 28, 1179–1192.
doi:10.3987/COM-88-S147
37.Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.;
Hansen, K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N.
J. Am. Chem. Soc. 2002, 124, 1307–1315. doi:10.1021/ja016737l
38.Bartoli, G.; Bosco, M.; Carlone, A.; Locatelli, M.; Melchiorre, P.;
Sambri, L. Org. Lett. 2004, 6, 3973–3975. doi:10.1021/ol048322l
39.Zwierzak, A. Synthesis 1982, 920–922. doi:10.1055/s-1982-29996
40.Wenghoefer, H. M.; Lee, K. T.; Srivastava, R. M.; Srimannarayana, G.;
Clapp, L. B. J. Heterocycl. Chem. 1970, 7, 1407–1411.
doi:10.1002/jhet.5570070631
13.Hodgson, D. M.; Humphreys, P. G.; Hughes, S. P. Pure Appl. Chem.
2007, 79, 269–279. doi:10.1351/pac200779020269
14.Florio, S.; Luisi, R. Chem. Rev. 2010, 110, 5128–5157.
doi:10.1021/cr100032b
15.Hodgson, D. M.; Humphreys, P. G.; Xu, Z.; Ward, J. G.
Angew. Chem., Int. Ed. 2007, 46, 2245–2248.
doi:10.1002/anie.200604920
16.Hellwinkel, D.; Hofmann, G.; Lämmerzahl, F. Tetrahedron Lett. 1977,
18, 3241–3244. doi:10.1016/S0040-4039(01)83207-X
17.Hellwinkel, D.; Lämmerzahl, F.; Hofmann, G. Chem. Ber. 1983, 116,
3375–3405. doi:10.1002/cber.19831161014
41.Osborn, H. M. I.; Cantrill, A. A.; Sweeney, J. B.; Howson, W.
Tetrahedron Lett. 1994, 35, 3159–3162.
doi:10.1016/S0040-4039(00)76856-0
18.Jardine, A. M.; Vather, S. M.; Modro, T. A. J. Org. Chem. 1988, 53,
3983–3985. doi:10.1021/jo00252a019
42.Yuan, C.; Xu, C.; Zhang, Y. Tetrahedron 2003, 59, 6095–6102.
doi:10.1016/S0040-4020(03)00995-5
19.Legrand, O.; Brunel, J. M.; Buono, G. Angew. Chem., Int. Ed. 1999, 38,
1479–1483.
doi:10.1002/(SICI)1521-3773(19990517)38:10<1479::AID-ANIE1479>3
.0.CO;2-G
License and Terms
20.He, Z.; Modro, T. A. Synthesis 2000, 565–570.
doi:10.1055/s-2000-6368
This is an Open Access article under the terms of the
Creative Commons Attribution License
21.Hammerschmidt, F.; Hanbauer, M. J. Org. Chem. 2000, 65,
6121–6131. doi:10.1021/jo000585f
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
22.Au-Yeung, T.-L.; Chan, K.-Y.; Haynes, R. K.; Williams, I. D.;
Yeung, L. L. Tetrahedron Lett. 2001, 42, 457–460.
doi:10.1016/S0040-4039(00)01952-3
23.Capriati, V.; Florio, S.; Luisi, R.; Musio, B. Org. Lett. 2005, 7,
3749–3752. doi:10.1021/ol051412l
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
24.Luisi, R.; Capriati, V.; Florio, S.; Di Cunto, P.; Musio, B. Tetrahedron
2005, 61, 3251–3260. doi:10.1016/j.tet.2005.01.045
25.Sweeney, J. B. Aziridines. In Science of Synthesis: Houben-Weyl
Methods of Molecular Transformations; Enders, D., Ed.; Thieme:
Stuttgart, 2008; Vol. 40a, pp 643–772.
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
26.Zwierzak, A.; Zawadzki, S. Synthesis 1972, 416–417.
doi:10.1055/s-1972-21890
doi:10.3762/bjoc.6.110
27.Zawadzki, S.; Zwierzak, A. Tetrahedron 1981, 37, 2675–2681.
doi:10.1016/S0040-4020(01)98974-4
28.Osowska-Pacewicka, K.; Zwierzak, A. J. Prakt. Chem. 1986, 328,
441–444. doi:10.1002/prac.19863280321
29.Zawadzki, S.; Zwierzak, A. Tetrahedron 1973, 29, 315–320.
doi:10.1016/S0040-4020(01)93296-X
30.Gajda, T.; Napieraj, A.; Osowska-Pacewicka, K.; Zawadzki, S.;
Zwierzak, A. Tetrahedron 1997, 53, 4935–4946.
doi:10.1016/S0040-4020(97)00188-9
31.Pousset, C.; Larchevêque, M. Tetrahedron Lett. 2002, 43, 5257–5260.
doi:10.1016/S0040-4039(02)01035-3
32.Palacios, F.; Aparicio, D.; Ochoa de Retana, A. M.;
de los Santos, J. M.; Gil, J. I.; López de Munain, R.
Tetrahedron: Asymmetry 2003, 14, 689–700.
doi:10.1016/S0957-4166(03)00089-2
33.Mordini, A.; Peruzzi, D.; Russo, F.; Valacchi, M.; Reginato, G.;
Brandi, A. Tetrahedron 2005, 61, 3349–3360.
doi:10.1016/j.tet.2005.01.102
34.Hodgson, D. M.; Bray, C. D.; Kindon, N. D.; Reynolds, N. J.;
Coote, S. J.; Um, J. M.; Houk, K. N. J. Org. Chem. 2009, 74,
1019–1028. doi:10.1021/jo802016t
983