3
14
H. J. FOROUDIAN ET AL.
is preferred for the anion 2, and from HF ab initio
methods for the protonated from 2a. The open-chain
species, 6 and 6a, akin to metaphosphates, would be
formed if elimination of 4-nitrophenoxide ion did no5t
require nucleophilic participation by the peroxo moiety.
Predicted geometries of the anions from calculation at
the 6–31G* level are illustrated for 2 and 6 and O—O
bond lengths of the cyclic intermediates are similar to
those in H O and peroxocomplexes of transition metals,
which are 1.49 and ca 1.47 A, respectively. Predicted
geometries calculated at the 3–21G* level are almost
identical with those shown illustrated for 2 and 6.
Calculations neglect hydration, but its contribution
should be similar for species of like charge. The negative
charge on 6 is localized but that on 2 is delocalized,
which offsets the stronger hydration of cyclic as
compared with extended ions.
CONCLUSIONS
The initially formed peroxophosphate from the reaction
À
of HO2 with PPC decomposes by reaction with H O /
2
2
À
HO2 rather than by eliminating phenoxide ion, but the
peroxophosphate from reaction of TNPP eliminates 4-
nitrophenoxide ion in competition with decomposition by
À
reaction with H O /HO . Quantum mechanical calcula-
2
2
2
tions indicate that elimination of aryloxide ion from the
initially formed peroxophosphate ester generates a labile
cyclic, rather than an open-chain, peroxophosphate.
2
2
13
˚
Acknowledgements
This work was supported by CONACYT Grant 25183-E
and DGAPA-UNAM IN214998, and the US Army Office
of Research, and H.J.F. thanks the Materials Research
Laboratory (NSF; DMR 9362716) for support. We
gratefully acknowledge this support.
The postulated intermediate 2 can break down by
À
attack of H O or OH on phosphorus generating the
2
strained cyclic species, 7 and 7a, but they are disfavored
relative to the open-chain peroxophosphates, 8 and 8a
REFERENCES
(
Table 2), which form inorganic phosphate by reaction
6
with H O . Therefore, we conclude that 2 ring opens by
attack of H O OH on phosphorus rather than by addition
2
2
1. (a) Larsson L. Acta Chem. Scand. 1958; 12: 723–730; (b) Epstein
J, Demek MM, Rosenblatt DH. J. Org. Chem. 1956; 21: 796–797.
À
2
2
. (a) Yang Y-C. Chem. Ind. (London) 1995; 334–337; (b) Yang Y-C,
Berg FJ, Szafraniec LL, Beaudry WT, Bunton CA, Kumar A. J.
Chem. Soc., Perkin Trans. 2 1997; 607–613.
giving 7 and 7a, followed by ring opening. Our
calculations relate to intermediates, not transition states,
but in fast reactions conversion barriers are low and
changes in geometries small. On the assumption that the
semi-empirical are less reliable than the ab initio
treatments, it appears that the general conclusions do
not depend on the charge of the intermediates, but it is
reasonable to assume that elimination of nitrophenoxide
ion from a peroxophosphate involves an anionic inter-
mediate (Scheme 1). The various methods of structure
3. (a) Jencks WP. Catalysis in Chemistry and Enzymology. McGraw-
Hill: New York, 1969; Chapt. 2; (b) Edwards JO, Pearson RG. J.
Am. Chem. Soc. 1962; 84: 16–24; Hoz S, Buncel E. Isr. J. Chem.
1
985; 26: 313–319.
4. (a) Yang Y-C, Baker JA, Ward JR. Chem. Rev. 1992; 98: 1729–
1
743; (b) Yang Y-C. Acco. Chem. Res., 1999; 33: 109–115.
5
. Mejia-Radillo Y, Yatsimirsky A, Foroudian HJ, Gillitt ND, Bunton
CA. J. Phys. Org. Chem. 2000; 13: 505–510.
6. Creaser II, Edwards JO. Top. Phosphorus Chem. 1972; 7: 379–432.
7
8
9
. (a) Hudson RF, Moss G. J. Chem. Soc. 1962; 3599–3604; (b)
Hudson RF, Moss G. J. Chem. Soc. 1964; 1040–1045.
. Jennings RN, Capomacchia AC. Anal. Chim. Acta 1989; 227: 37–
48.
11,14
simulations have been critically analyzed,
and the
semi-empirical AM1 and PM3 methods appear to be less
reliable than the HF ab initio methods.
. Kirby AJ. Adv. Phys. Org. Chem. 1980; 17: 183–278.
1
0. Everett AJ, Minkoff GJ. Trans. Faraday Soc. 1953; 49: 410–414.
1. Hehre WJ, Yu J, Klunzinger PE, Lou L. A Brief Guide to
Molecular Mechanics and Quantum Mechanical Calculations.
Wavefunction: Irvine, CA, 1998.
2. (a) Evans MG, Uri N. Trans. Faraday Soc. 1949; 45: 224–230; (b)
Muhammad SS, Rao TN. J. Chem. Soc. 1957; 1077–1078.
Predicted dimensions of the hypothetical peroxo inter-
mediates 2 and 6 are shown on the structures. We expect the
peroxo intermediates to be extensively deprotonated in our
reaction conditions, and the cyclic peroxophosphate 2 is
predicted to be symmetrical. These structures are from 6–
1
1
13. (a) Dickman MM, Pope MT. Chem. Rev. 1994; 94: 569–584; (b)
Cotton FA, Wilkinson G. Advanced Inorganic Chemistry (3rd
edn). Wiley: New York, 1972; Chapt. 25-C.
3
1G* calculations, but essentially identical structures were
obtained from 3–21G* calculations.
1
4. Levine IN. Quantum Chemistry (5th edn). Prentice Hall: Upper
Saddle River, NJ, 2000; Chapt. 15, 16.
Copyright 2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 310–314