Phosphorochloridates from phosphorohydrazides with ButOCl: Stereospecificity, selectivity and phosphorylium ion intermediates

Article abstract of DOI:10.1016/S0040-4039(01)01329-6

The efficiency with which a six-membered cyclic phosphorochloridate can be obtained from the corresponding hydrazide with retention of configuration at phosphorus, using tert-butyl hypochlorite in tert-butyl alcohol, depends on how stereoelectronic factors influence the fate of an intermediate chloride-phosphorylium ion pair.

Full text of DOI:10.1016/S0040-4039(01)01329-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6749–6752
Phosphorochloridates from phosphorohydrazides with ButOCl:
stereospecificity, selectivity and phosphorylium ion intermediates
Martin J. P. Harger*
Department of Chemistry, University of Leicester, Leicester LE1 7RH, UK
Received 29 May 2001; accepted 20 July 2001
Abstract—The efficiency with which a six-membered cyclic phosphorochloridate can be obtained from the corresponding
hydrazide with retention of configuration at phosphorus, using tert-butyl hypochlorite in tert-butyl alcohol, depends on how
stereoelectronic factors influence the fate of an intermediate chloride–phosphorylium ion pair. © 2001 Elsevier Science Ltd. All
rights reserved.
Phosphoryl transfer reactions play a vital part in biolog-
ical chemistry,1 and phosphorylating agents having a
defined configuration are important in synthesis2,3 and in
mechanistic studies.3,4 Phosphoryl halides are the most
widely used laboratory phosphorylating agents and
a method for changing the configuration at phosphorus
in a controlled and reliable way is of great potential
value.
the crucial retention-of-configuration reaction of the
hydrazide in the second stage.
The phosphorochloridates 2a and 2b (X=Cl) are confor-
mationally mobile (chair–chair inversion) so it is easy for
the PꢀCl bond to assume its preferred orientation
(axial).6 For the related chloridates 3a and 3b (X=Cl),
however, the C-4 phenyl substituent anchors the chair,
with the phenyl group equatorial,7,8 so one of the isomers
cannot have the favoured orientation of the PꢀCl bond.
By comparing the behaviour of these conformationally
mobile and anchored systems we hoped to gain some
indication of the likely generality of the Thenappen−
Wadsworth approach, and some clues as to mechanism
of the hydrazide-to-chloridate conversion.
Thenappen and Wadsworth5 have reported the successful
(and apparently efficient) conversion of a phospho-
rochloridate into its stereoisomer by a two-stage
sequence (Scheme 1) involving the phosphorohydrazide.
The generality of their strategy has never been examined,
however, and nothing is known about the mechanism of
Scheme 1. Conversion of a phosphorochloridate into its stereoisomer.
Keywords: phosphorochloridates; phosphorohydrazides; phosphorylium ions.
* Fax: (0116) 252 3789.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01329-6

6750
M. J. P. Harger / Tetrahedron Letters 42 (2001) 6749–6752
The phosphorohydrazides 2b5 and 3b (X=NHNH2)
were prepared directly from the available chloridates
2a9 and 3a (X=Cl).8,10,11 (CAUTION anhydrous
hydrazine is extremely reactive with oxidising agents
including air.) They were treated with ButOCl and then
H2NNH2 to give (after chromatography) the stereoiso-
meric hydrazides 2a5 and 3a (X=NHNH2). The sym-
metrical hydrazide 1 (X=NHNH2) (no stereoisomers)
was also prepared as a standard for comparison.12
70% and the stereospecificity (de)]88%. In all cases the
other products were the tert-butyl phosphate ester (X=
OBut) (both stereoisomers) and the phosphoric acid
(X=OH). Important stereochemical information is lost
in the dealkylation that produces the acid (Scheme 2)17
so the reactions were also carried out with MeOH as
solvent. The yield of the chloridate was inevitably
reduced but dealkylation was no longer a problem, so
the stereoisomer composition of the methyl ester (X=
OMe) is a true reflection of how the alcohol attacks.
The hydrazides 1–3 all reacted readily with ButOCl in
ButOH (N2 evolution) at room temperature.13 The reac-
tion mixtures were analysed by 31P and 1H NMR
spectroscopy (Table 1) with reference to the spectra of
authentic samples.14–16 In no case was the phospho-
rochloridate (X=Cl) the only product, nor was it
formed with complete retention of configuration, but
with one exception (3b in Table 1) the yield was at least
We suppose that ButOCl converts the hydrazide into
the diazo chloride and that the products are derived
from this.18 If, as the original report seems to imply,5
the chloridate were the only product and its formation
was completely stereospecific a concerted mechanism
might be indicated. As it is, we think a phosphorylium
cation is the product-forming species (Scheme 2), being
Table 1. Reactions of phosphorohydrazides with ButOCl. Yields of products and stereochemistry of formation
Scheme 2. Formation and reactions of a phosphorylium cation.

M. J. P. Harger / Tetrahedron Letters 42 (2001) 6749–6752
6751
Scheme 3. Fates of chloride−phosphorylium ion pairs.
same (axial) ester. Both the yield and the stereochemi-
cal integrity are, therefore, low for the chloridate but
high for the ester (3b in Table 1). In the case of the ion
pair A from the axial hydrazide, formation of the
chloridate (retention; ax-Cl) is more favourable (faster)
and competition from ester formation (inversion; eq-
OR) less severe. There is also less leakage, and that
which does occur results in axial ester rather than the
unstable equatorial chloridate. So now the yield and the
stereochemical integrity are high for the chloridate but
low for the ester (3a in Table 1).
trapped competitively by chloride ion and the alcohol
solvent (ROH).19 The competition depends on the bulk
of the alcohol (Table 1), but not as much as it would
for a normal associative substitution reaction [SN2(P)]
at a four-coordinate phosphoryl centre.20 This itself
may be seen as evidence for a reactive and sterically
accessible three-coordinate phosphoryl intermediate.21
A free and symmetrically solvated phosphorylium ion
cannot be the principal product-forming species how-
ever, as there is little stereochemical congruence of the
products from stereoisomeric hydrazides. Also, the
competing chloride and alcohol nucleophiles form
products with contrasting stereochemistries (retention
and predominant inversion, respectively). The phospho-
rylium ion will, in fact, be generated alongside a chlo-
ride ion and it is the behaviour of this ion pair that
determines the outcome of the reaction. Stereoelec-
tronic factors seem to be of major importance, influenc-
ing the composition of the product (chloridate versus
ester) as well as the stereochemistry of its formation
(retention versus inversion).
As a method for reversing the configuration of a phos-
phorochloridate the hydrazine–hypochlorite protocol
may not be as efficient as was thought,5 but that is
hardly surprising given the (apparent) intermediacy of a
high energy phosphorylium ion in the second stage. It
should still prove very useful for systems in which the
stereoelectronic requirements of the reaction are not
opposed by powerful conformational constraints.22
References
The PꢀX bond of the product (X=Cl or OR) prefers to
be axial, especially when X=Cl (anomeric effect).6,8
With the conformationally mobile system 2, it makes
little difference whether the ClCH2 or CH3 group at
C-5 is axial, so whatever the geometry of the product
(cis or trans) the PꢀX bond can easily be placed axial.
The isomeric hydrazides 2a and 2b (X=NHNH2),
therefore, behave in much the same way as one another
and as the symmetrical substrate 1. But with the con-
formationally anchored system 3 the PꢀX bond can be
axial only when X is trans to the equatorial C-4 phenyl
group. In the case of the ion pair B (Scheme 3) from the
equatorial hydrazide formation of the less stable equa-
torial chloridate (retention) is relatively unfavourable
(slow), so competition from ester formation (inversion;
ax-OR) is severe. Also, there is considerable leakage to
the alternative ion pair A and to the symmetrically-sol-
vated phosphorylium ion (ROH in place of Cl− in B);
the former collapses mainly to the other (axial) chlori-
date while the latter stereoselectively forms more of the
1. Corbridge, D. E. C. Phosphorus: An Outline of its Chem-
istry, Biochemistry and Technology; Elsevier: Amsterdam,
1995.
2. Kolodiazhnyi, O. I. Tetrahedron: Asymmetry 1998, 9,
1279–1332.
3. Edmundson, R. S. In The Chemistry of Organophosphorus
Compounds; Hartley, F. R., Ed.; Wiley: Chichester, 1996;
Vol. 4, Chapter 6.
4. (a) Emsley, J.; Hall, D. The Chemistry of Phosphorus;
Harper & Row: London, 1976; (b) Thatcher, G. R. J.;
Kluger, R. Adv. Phys. Org. Chem. 1989, 25, 99–265; (c)
Hall, C. R.; Inch, T. D. Tetrahedron 1980, 36, 2059–2095.
5. Thenappan, A.; Wadsworth, W. S. Tetrahedron Lett.
1986, 27, 1755–1756. This communication contains little
experimental detail and a full account has not been
published.
6. Bentrude, W. G. In Conformational Behaviour of Six-
Membered Rings; Juaristi, E., Ed.; VCH Publishers: New
York, 1995, Chapter 7.

6752
M. J. P. Harger / Tetrahedron Letters 42 (2001) 6749–6752
15. (a) For 2-X-1,3,2-dioxaphosphorinane 2-oxides the 31P
7. Leusen, F. J. J.; Bruins Slot, H.; Noordik, J. H.; van der
NMR chemical shift lP is smaller (higher field) when X is
axial (PꢁO equatorial). Gallagher, M. J. In Phosphorus-
31 NMR Spectroscopy in Stereochemical Analysis;
Verkade, J. G.; Quin, L. D., Eds.; VCH Publishers:
Deerfield Beach, 1987; Chapter 9; (b) For X=Cl, lP
(CDCl3): 1, −2.9; 2a, −2.9; 2b, −2.8; 3a, −2.25; 3b, 4.0.
For X=OBut, lP (CDCl3): 1, −12.6; 2a, −13.3; 2b, −10.8;
3a, −12.3; 3b, −6.3. For X=OMe, lP (CDCl3): 1, −6.6;
2a, −7.1; 2b, −5.5; 3a, −6.2; 3b, −2.4. For X=OH, lP
(CDCl3): 1, −4.25; 2, −4.45; 3, −3.2.
Haest, A. D.; Wynberg, H.; Bruggink, A. Recl. Trav.
Chim., Pays-Bas 1991, 110, 13–18.
8. Hulst, R.; Visser, J. M.; Koen de Vries, N.; Zijilstra, R.
W. J.; Kooijman, H.; Smeets, W.; Spek, A. L.; Feringa,
B. L. J. Am. Chem. Soc. 2000, 122, 3135–3150.
9. Wadsworth, W. S.; Larsen, S.; Horten, H. L. J. Org.
Chem. 1973, 38, 256–263.
10. Cullis, P. M.; Fawcett, J.; Griffith, G. A.; Harger, M. J.
P.; Lee, M. J. Am. Chem. Soc. 2001, 123, 4147–4154.
11. The chloridate was added in portions to anhydrous
H2NNH2 (2.5 equiv.) in MeCN (N2 atmosphere; safety
screen). Hydrazine hydrochloride was removed by filtra-
tion or washing with water and the hydrazide product
was purified by crystallisation. [Anhydrous hydrazine is
commercially available (Aldrich) but we preferred to
prepare it in small amounts (1–2 g) as required from the
hydrate: Day, A. C.; Whiting, M. C. Org. Synth. Coll.
1988, 6, 10–12.]
12. Phosphorohydrazides (X=NHNH2): 1, mp 148–150°C,
lP (CDCl3) 3.05; 2a, mp 159–161°C (lit. 177–178°C), lP
1.4; 2b, mp 156–158°C (lit. 156°C), lP 4.15; 3a, mp
211–214°C, lP 0.35; 3b, mp 208–210°C, lP 5.35. In the 1H
NMR spectra (CDCl3) the NH group gave a doublet (JPH
25–35 Hz) at lH 4.3–4.8. The new compounds 1, 3a and
3b gave satisfactory results in elemental analysis and/or
accurate mass measurement.
1
16. For the compounds 2 the H NMR chemical shifts lH of
the ClCH2 and CH3 groups on C-5 are larger (lower field)
when they are axial: Wadsworth, W. S. J. Org. Chem.
1987, 52, 1748–1753.
17. tert-Butyl phosphate esters are very easily dealkylated
with acid: Perich, J. W.; Johns, R. B. Synthesis 1988,
142–144 and references cited therein.
18. For a similar supposition, see: Carpino, L. A. J. Am.
Chem. Soc. 1957, 79, 96–98 (reaction of RCONHNH2
with Cl2 leading to RCOCl).
19. Phosphorylium ions probably play no part in normal
nucleophilic substitution reactions of phosphoryl com-
pounds. Quin, L. D. A Guide to Organophosphorus Chem-
istry; Wiley–Interscience: New York, 2000; p. 28. In the
present reactions the liberation of N2 could provide the
driving force for their formation.
20. Treatment of the hydrazide 1 (X=NHNH2) with ButOCl
in a 1:1 mixture of MeOH and ButOH gave products
derived from the two alcohols in a 2.2:1 ratio. By con-
trast Ag+-catalysed solvolysis of the chloridate 1 (X=Cl)
[SN2(P)] gave]96.5% of the methyl ester.
13. ButOCl (2.5 equiv.) was added to a suspension or solu-
tion of the phosphorohydrazide (22.5 mmol) in ButOH or
MeOH (0.45 ml). Reaction mixtures were examined
directly by 31P NMR spectroscopy (relative amounts of
products) and after replacement of the alcohol solvent
with CDCl3, by 1H and 31P NMR (identities of products).
14. The phosphorochloridates (X=Cl) 2b (Ref. 9) and 3b
(Ref. 10) were only available as mixtures with their
stereoisomers. The phosphoric acids (X=OH) were pre-
pared by hydrolysis of the chloridates, and the phosphate
esters (X=OMe or OBut) (mixtures of stereoisomers) by
treatment of the acids with diazomethane or the chlori-
dates with KOBut.
21. Cf.: Ramirez, F.; Marecek, J. F. J. Am. Chem. Soc. 1979,
101, 1460–1465.
22. (a) Most potential alternatives can be discounted because
the chloridate product would be generated under condi-
tions in which it is configurationally unstable; (b) In
extreme cases even the first stage (hydrazide formation)
may not be completely stereospecific (inversion); thus 3b
(X=Cl) reacts with H2NNH2 with partial retention of
configuration.