The 'valence tautomers' of o-iodosobenzoic acid: The case of 4-pentyl-2-iodosobenzoic acid

Article abstract of DOI:10.1039/a802902d

Contrary to a previous report, only the closed (iodoxolone) form of 4-pentyl-2-iodosobenzoic acid (3a) can be isolated; the previously assigned open (iodoso) form (3b) is actually 4-pentanoyl-2-iodobenzoic acid.

Full text of DOI:10.1039/a802902d

View Article Online / Journal Homepage / Table of Contents for this issue
The ‘valence tautomers’ of o-iodosobenzoic acid: the case of
4-pentyl-2-iodosobenzoic acid
Robert A. Moss,*† Saketh Vijayaraghavan and T. J. Emge
Department of Chemistry, Rutgers, The State University of New Jersey, New Brunswick, New Jersey 08903, USA
Contrary to a previous report, only the closed (iodoxolone)
iii
C5H11
CH2OH
C5H11
CO2H
form of 4-pentyl-2-iodosobenzoic acid (3a) can be isolated;
the previously assigned open (iodoso) form (3b) is actually
4-pentanoyl-2-iodobenzoic acid.
R
I
6
iv
Y = '3b'
o-Iodosobenzoic acid (IBA, 1) has long been known1 to exist in
the cyclic 1-hydroxy-1,2-benziodoxol-3(1H)-one form (2).2,3
X-Ray crystal structure4 and theoretical studies5 agree that the
cyclic form is the better representation, although the internal
I–O bond is longer than a ‘normal’ single I–O bond, indicating
significant ‘open’ character.4b,c,5
4 R = H
5 R = I
i, ii
X = '3a'
Scheme 1 Reagents and conditions: i, BuLi, Me2NCH2CH2NMe2; ii, I2; iii,
KMnO4–H2O, C6H6, cat. Bu4P+Cl2; iv, 30% H2O2, Ac2O, 40 °C, 20 h
A valence tautomeric representation of IBA (1 " 2) suggests
carbonyl bands for X at 1710 cm21 and Y at 1650 cm21. The
higher frequency CNO band of X was considered indicative of
a ‘lactone’ structure as in 3a.7 However, it is known2 that the
carbonyl band of IBA, in its iodoxolone form, is at 1633 cm21
(Nujol), so that the reported IR band of X at 1710 cm21 is
inconsistent with structure 3a; it is Y (1650 cm21) that is more
likely to merit this assignment.
that both 1 and 2 can exist as separate entities. Such
O
CO2H
O
I
I
O
We repeated the synthetic sequence of Scheme 1.7 In
particular, the permanganate oxidation of 5, after chromatog-
raphy of the product on silica gel (hexanes–EtOAc–HOAc,
79:20:1 to 28:70:2) afforded 6 (48%, mp 67.5–68.5 °C, lit.7
68.0–69.0 °C) and X (10%, mp 115–116 °C, lit.7 115–116
°C).
OH
1
2
representations have often appeared,6 even if they were not
intended to imply the simultaneous presence of interconverting
open (‘iodoso’) and closed (iodoxolone) forms. Thus, in 1990,
Panetta et al. reported the separate isolation of the iodoxolone
and iodoso valence tautomers of 4-propyl- as well as 4-pentyl-
2-iodosobenzoic acids.7 These extraordinary results have been
reiterated in a recent authoritative review,3b so that it becomes
imperative to verify them, particularly because of the im-
portance of IBA and its analogues as decontamination agents
for toxic phosphonates and phosphates.4b,c,5–8
We have now reinvestigated the case of the 4-pentyl
‘tautomers’ (3a and 3b), and report here that the previously
described7 compounds were misassigned; in fact, only a single
4-pentyl-2-iodosobenzoic acid can be isolated, and it is best
represented as the ‘closed’ iodoxolone compound, 3a.
Our sample of X displayed the same mp, an experimentally
comparable elemental analysis, and a similar IR CNO band
(1707 cm21) relative to those reported7 for ‘3a’. Nevertheless,
several observations indicated that X was not 3a. (1) The NMR
spectrum of X‡ revealed only four sets of alkyl protons rather
than the anticipated five. (2) The pentyl benzylic resonance of 5
(a triplet at d 2.55) was missing in X, while a more deshielded
triplet appeared at d 3.1. (3) The IR spectrum (KBr) of X
revealed two intense CNO absorptions at 1707 and 1686 cm21
.
The Supplementary Material for ref. 7 reports this band at 1685
cm21; it can be assigned to an aromatic CO2H. The former band
was assigned7 to the carbonyl of 3a, but the ‘lactone’ carbonyl
group of (e.g.) 2 is known to absorb at 1633 cm21 (Nujol).2 (4)
Compound X was kinetically inactive toward p-nitrophenyl
diphenyl phosphate (PNPDPP) in aqueous micellar cetyl-
trimethylammonium chloride (CTACl) at pH 8 (see Table 1),
whereas authentic benziodoxolones (e.g. 2) rapidly cleave
PNPDPP under these conditions.8a,b (5) Additionally, X did not
oxidize iodide to iodine, a common property of iodoso-
benzoates.8b
O
CO2H
O
I
C5H11
I
O
C5H11
3a
OH
3b
The origin of the misassignments in ref. 7 lies in the synthetic
sequence, which we summarize in Scheme 1. 4-Pentylbenzyl
alcohol (4) was first regiospecifically iodinated to 5.7,9
Accordingly, an X-ray crystal structure determination was
carried out for X,§ revealing it to be not an iodoso compound at
all, but 4-pentanoyl-2-iodobenzoic acid (7) (Fig. 1). Clearly, the
KMnO4 oxidation of 5 to 6 must have been accompanied by
overoxidation12 at the benzylic position of the pentyl chain,
affording (both) ketone 7 (and iodoterphthalic acid). Structure 7
immediately accounts for the spectral characteristics of X
itemized above in points (1)–(3),¶ and, of course, 7 should also
be inactive in the hydrolysis of PNPDPP or the oxidation of
iodide [points (4) and (5)].
In the following and key step, iodo alcohol 5 was oxidized to
4-pentyl-2-iodobenzoic acid (6) using phase transfer catalysis in
a KMnO4–water–benzene system.7,10 This reaction led not only
to the desired 6, in 63% yield, but also to a second,
chromatographically-separated product, X (25%, mp 115–116
°C), assigned7 as 4-pentyl-2-iodosobenzoic acid in its cyclic
(iodoxolone) form, 3a. A separate H2O2/Ac2O oxidation7,11 of
6 afforded 4-pentyl-2-iodosobenzoic acid Y (mp 188.5–189.5
°C) assigned7 as the open (iodoso) valence tautomer, 3b.
The assignments7 of X and Y rest on acceptable elemental
analyses for C12H15IO3, suggestive of isomerism, as well as IR
Compound Y, which we obtained from the peroxide
oxidation of 6 (Scheme 1) had a mp identical to the compound
previously obtained,7 and is actually 4-pentyl-2-iodosobenzoic
Chem. Commun., 1998
1559

View Article Online
hydrophobic catalyst into the oily interior of the micro-
emulsion.7 However, we find 3a to be quite reactive toward
PNPDPP under these conditions (Table 1, entry 6); indeed, it is
actually ~ 3.5 times more reactive than IBA (entry 5),
paralleling the results in micellar CTACl (see above, and entries
3 and 4). Note (Table 1) that both IBA and 3a are less reactive
toward PNPDPP in the microemulsion than in micellar CTACl,
an expected consequence of lessened mutual catalyst/substrate
concentration in the microemulsion.14
In conclusion, only one 4-pentyl-2-iodosobenzoic acid can be
isolated, and it is best represented as iodoxolone 3a. A similar
situation is likely to hold for 4-propyl-2-iodosobenzoic acid7 as
well.
We are grateful to the US Army Research Office for financial
support.
Fig. 1 ORTEP diagram of X (4-pentanoyl-2-iodobenzoic acid, 7)
acid, best represented as 3a (not 3b7). Thus, Y (3a) gave both an
appropriate elemental analysis [C, 43.l; H, 4.53; I, 38.0%) and
NMR spectrum.∑ The IR (KBr) spectrum of 3a displayed its
CNO band at 1602 cm21, considerably lower than the reported7
1650 cm21. However, benziodoxolone carbonyl bands are very
sensitive to conditions of their determination; the CNO absorp-
tion of 2 has been variously reported at 1633,2 1612,6b and
16056b cm21. Additionally, 3a showed the expected2 (I)OH
absorptions at 2928 and 2444 cm21. A standard iodometric
titration13 of 3a gave 93% of INO oxidative activity.
Notes and References
† E-mail: moss@rutchem.rutgers.edu
‡ dH[(CD3)2CO, 200 MHz] 0.92 (t, J 8, 3H), 1.4 (sext., J 8, 2H), 1.7 (pent.,
J 8, 2H), 3.1 (t, J 8, 2H), 7.9 (d, J 8, 1H), 8.1 (AB dd, J 8, 1.6, 1H), 8.5 (d,
J 1.6, 1H).
§ Crystal data for 7: C12H13O3I, M = 332.12, colorless rods, 0.06 3 0.54
Most importantly, 3a was very reactive toward PNPDPP. Its
kinetic properties were assessed from a rate constant–[surfac-
tant] profile for the cleavage of PNPDPP in micellar CTACl;8b
conditions and results appear in Table 1. Not only is Y (3a)
highly reactive toward PNPDPP, where X (7) is inactive (entry
2), but 3a affords an acceleration of 1460 relative to micellar
CTACl alone (entry 1), 4.6 times greater than the acceleration
provided by the parent IBA (2) (entry 3). This reactivity
advantage is an expected consequence of the hydrophobic
pentyl group of 3a, which affords better binding of 3a to the
micellar phase in which the phosphorolytic reaction oc-
curs.8b,c
3 0.60 mm, monoclinic, space group P21/n, a
= 4.2397(11), b =
29.477(5), c = 10.222(2) Å, b = 101.92(2)° U = 1249.9(5) Å3, Z = 4, Dc
= 1.765 g cm23, m(Mo-Ka) = 25.52 cm21, F(000) = 648. The 2340 total
data were collected at 20 °C using graphite monochromatized Mo-Ka
radiation (l = 0.71073 Å), and converged at R1 = 0.0507, wR2 = 0.0889
for all 2157 unique data. CCDC 182/910.
¶ Note too that the calculated elemental analysis of 7, C12H13IO3 [C, 43.4;
H, 3.94; I, 38.2%] is nearly within accepted limits of the calculated analysis
for 3, C12H15IO3 [C, 43.1; H, 4.52; I, 38.0%].
∑ d([2H6]DMSO, 200 MHz) 0.83 (t, J 8, 3H), 1.3 (m, 2CH2, 4H), 1.62 (pent.,
J 8, 2H), 2.75 (t, J 8, 2H), 7.5 (d, J 8, 1H), 7.6 (s, 1H), 7.9 (d, J 8, 2H), 7.95
(s, 1H, OH exchangeable with D2O).
1 V. Meyer and W. Wachter, Chem. Ber., 1892, 25, 2632; C. Wilgerodt,
Die Organische Verbindungen mit Mehrwertigen Jod, Enke, Stuttgart,
1914, p. 134; D.E. Banks, Chem. Rev., 1966, 66, 243 (see p. 255).
2 G. P. Baker, F. G. Mann, N. Sheppard and A. J. Tetlow, J. Chem. Soc.,
1965, 3721.
3 (a) G. F. Koser, in The Chemistry of Functional Groups, Suppl. D, ed.
S. Patai and Z. Rappoport, Wiley, New York, 1983, pp. 721f; (b) P. J.
Stang and V. V. Zhdankin, Chem. Rev., 1996, 96, 1123; (c) A.
Varvoglis, The Organic Chemistry of Polycoordinated Iodine, VCH,
New York, 1992, p. 168f.
4 (a) E. Shefter and W. Wolf, J. Pharm. Sci., 1965, 54, 104; (b) A. R.
Katritzky, G. P. Savage, G. J. Palenik, K. Qian, Z. Zhang and H. D.
Durst, J. Chem. Soc., Perkin Trans. 2, 1990, 1657; (c) R. A. Moss, K.
Bracken and T. J. Emge, J. Org. Chem., 1995, 60, 7739.
5 R. A. Moss, B. Wilk, K. Krogh-Jespersen, J. T. Blair and J. D.
Westbrook, J. Am. Chem. Soc., 1989, 111, 250; R. A. Moss, B. Wilk, K.
Krogh-Jespersen and J. D. Westbrook, J. Am. Chem. Soc., 1989, 111,
6729.
Table 1 Rate constants for the cleavage of PNPDPPa
Entry
Catalyst
ky/1024 s21
krel
1
2
3
4
5g
6g
Noneb
X (7)
2
2.05c
2.00
1.00
0.98
640d
312
1460
8.9
30.9
Y (3a)e
2
3000f
18.3
63.3
Y (3a)
a For background, see ref. 8(b). Conditions for entries 1–4: [CTACl] = 1.0
3 1023 m, [PNPDPP] = 1.0 3 1025 m, [catalyst] = 1.0 3 1024 m, pH 8,
0.02 m phosphate buffer, m = 0.08 (NaCl), 25 °C. Rate constants were
determined by monitoring the time dependent absorbance of the released
p-nitrophenylate ion at 400 nm. b CTACl alone. c Given as 1.8 3 1024 s21
in ref. 6(a). d Ref. 8(b). e [PNPDPP] = 3.0 3 1025 m, [Y] = 3.0 3 1024
f
g
m. Stopped-flow determination. Microemulsion conditions:7 8% (w/w)
CTABr, 8% N-methylpyrrolidinone, 4% toluene, 80% 0.03 m aqueous
Na2B4O7·10H2O buffer, pH 9.4, 25 °C; [PNPDPP] = 3 3 1025 m, [catalyst]
= 3 3 1024 m.
6 (a) R. A. Moss, S. Chatterjee and B. Wilk, J. Org. Chem., 1986, 51,
4303; (b) A. R. Katritzky, G. P. Savage, J. K. Gallos, and H. D. Durst,
J. Chem. Soc., Perkin Trans. 2, 1990, 1515.
7 C. A. Panetta, S. M. Garlick, H. D. Durst, F. R. Longo and J. R. Ward,
J. Org. Chem., 1990, 55, 5202.
8 (a) R. A. Moss, K. Alwis and G. O. Bizzigotti, J. Am. Chem. Soc., 1983,
105, 681; (b) R. A. Moss, K. W. Alwis and J.-S. Shin, J. Am. Chem. Soc.,
1984, 106, 2651; (c) R. A. Moss, K. Y. Kim and S. Swarup, J. Am.
Chem. Soc., 1986, 108, 788; (d) P. S. Hammond, J. S. Forster, C. N.
Lieske and H. D. Durst, J. Am. Chem. Soc., 1989, 111, 7860.
9 N. Meyer and D. Seebach, Angew. Chem., Int. Ed. Engl., 1978, 17,
521.
10 A. W. Herriott and D. Picker, Tetrahedron Lett., 1974, 16, 1511.
11 A. R. Katritzky, B. L. Duell, H. D. Durst and B. L. Knier, J. Org. Chem.,
1988, 53, 3972.
12 H-J. Schmidt and H.J. Schäfer, Angew. Chem., Int. Ed. Engl., 1979, 18,
68.
13 H. J. Lucas and E. R. Kennedy, Org. Synth., 1955, Coll. Vol. 3, 482.
14 (a) R. A. Moss, R. Fujiyama, H. Zhang, Y.-C. Chung and K. McSorley,
Langmuir, 1993, 9, 2902; (b) R. Mackay, F. R. Longo, B. L. Knier and
H. D. Durst, J. Phys. Chem., 1987, 91, 861.
Although we could not obtain crystals of 3a suitable for
X-ray analysis, its closed, ‘lactone’ structure follows from the
IR spectrum,2 and from its kinetic properties toward PNPDPP
(which link 3a to other phosphorolytically reactive iodoso-
benzoates for which the closed structure has been estab-
lished).4–6,8
True
‘iodoso’
compounds,
such
as
m-iodosobenzoic acid, show little esterolytic reactivity.8a
Additionally, we determined the pKa of 3a as 6.8 from a pH–rate
constant profile4c,5,8b for the cleavage of PNPDPP by 3a in 0.02
m micellar CTACl and 0.02 m phosphate buffer over the pH
range 5.35–7.68.
A pKa ~ 7 is appropriate for an
o-iodosobenzoate in the iodoxolone form.2,8
Finally, 3a was reported to be 477 times less reactive than
IBA
itself
toward
PNPDPP
in
a
CTABr–
N-methylpyrollidinone–toluene–aqueous borate microemul-
sion, a phenomenon attributed to incorporation of the more
Received in Corvallis, OR, USA, 17th April 1998; 8/02902D
1560
Chem. Commun., 1998