Ascorbic acid/iodine and triphenylphosphine/iodine as reducing agents for the As(V)=O group

Article abstract of DOI:10.1080/10426500500272335

The scope of ascorbic acid/iodine and triphenylphosphine/iodine in methanol for the direct reduction of arsenic(V) compounds having the As=O group has been investigated. Ascorbic acid/iodine reduces arsonic acids, diphenylarsinic acid (but not dimethylarsinic acid), and triphenylarsine oxide. The rates of reduction depend on the electronic effects of the ligands bound to arsenic and on the hydrogen-bonding strength of the species, when present. When the As(V) compound has an -NH₂ or an -NH₃⁺ group, the reduction product reacts with a ketonic form of dehydroascorbic acid, giving condensation product(s). Triphenylphosphine/iodine reduced slowly the zwitterionic o-aminophenylarsonic acid but reduced faster the hydrochloric acid salt of the same acid. It reduced dimethylarsinic acid as well because the powerful electron-withdrawing Ph₃P⁺coordinated to As=O seems to outweigh the electronic and hydrogen bonding effects. Copyright Taylor & Francis Group, LLC.

Full text of DOI:10.1080/10426500500272335

This article was downloaded by: [New York University]
On: 06 June 2015, At: 08:09
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK
Phosphorus, Sulfur, and Silicon
and the Related Elements
Publication details, including instructions for
authors and subscription information:
http://www.tandfonline.com/loi/gpss20
Ascorbic Acid/Iodine and
Triphenylphosphine/Iodine as
Reducing Agents for the As(V)
[
dbnd]O Group
a
a
Theodore D. Sideris & Panayiotis V. Ioannou
a
Department of Chemistry , University of Patras ,
Patras, Greece
Published online: 01 Feb 2007.
To cite this article: Theodore D. Sideris & Panayiotis V. Ioannou (2006) Ascorbic Acid/
Iodine and Triphenylphosphine/Iodine as Reducing Agents for the As(V)[dbnd]O Group,
Phosphorus, Sulfur, and Silicon and the Related Elements, 181:5, 1017-1030, DOI:
10.1080/10426500500272335
To link to this article: http://dx.doi.org/10.1080/10426500500272335
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the
information (the “Content”) contained in the publications on our platform.
However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness,
or suitability for any purpose of the Content. Any opinions and views
expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the
Content should not be relied upon and should be independently verified with
primary sources of information. Taylor and Francis shall not be liable for any
losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or

indirectly in connection with, in relation to or arising out of the use of the
Content.
This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan,
sub-licensing, systematic supply, or distribution in any form to anyone is
expressly forbidden. Terms & Conditions of access and use can be found at
http://www.tandfonline.com/page/terms-and-conditions

Phosphorus, Sulfur, and Silicon, 181:1017–1030, 2006
Copyright © Taylor & Francis Group, LLC
ISSN: 1042-6507 print / 1563-5325 online
DOI: 10.1080/10426500500272335
Ascorbic Acid/Iodine and Triphenylphosphine/Iodine
as Reducing Agents for the As(V) O Group
Theodore D. Sideris
Panayiotis V. Ioannou
Department of Chemistry, University of Patras, Patras, Greece
The scope of ascorbic acid/iodine and triphenylphosphine/iodine in methanol for
the direct reduction of arsenic(V) compounds having the As O group has been
investigated. Ascorbic acid/iodine reduces arsonic acids, diphenylarsinic acid (but
not dimethylarsinic acid), and triphenylarsine oxide. The rates of reduction depend
on the electronic effects of the ligands bound to arsenic and on the hydrogen-bonding
strength of the species, when present. When the As(V) compound has an NH or an
2
+
NH group, the reduction product reacts with a ketonic form of dehydroascorbic
3
acid, giving condensation product(s). Triphenylphosphine/iodine reduced slowly
the zwitterionic o-aminophenylarsonic acid but reduced faster the hydrochloric acid
salt of the same acid. It reduced dimethylarsinic acid as well because the powerful
+
electron-withdrawing Ph P coordinated to As O seems to outweigh the electronic
3
and hydrogen bonding effects.
Keywords Arsinic acids; arsonic acids; ascorbic acid; reductions; triphenylarsine oxide;
triphenylphosphine
INTRODUCTION
The reduction of As(V) compounds having the As O group to As(III)
species has analytical, synthetic, and biochemical significance. Thus,
−
arsenic acid can be determined in strongly acidic aqueous media by I
according to Eqs. (1) and (2):
1
(
(
1)
2)
The system ascorbic acid (AA)/iodide/hydrochloric acid has been used
2
for the reduction of arsenic acid in several kinds of natural waters.
Ascorbic acid was used for the “prevention” of the air oxidation of iodide
2
Received April 26, 2005; accepted April 26, 2005.
Address correspondence to Panayiotis V. Ioannou, University of Patras, Department
of Chemistry, Patras, Greece. E-mail: ioannou@chemistry.upatras.gr
1017

1
018
T. D. Sideris and P. V. Ioannou
3
because it reduces iodine to iodide. The I2 is not only produced by the
−
air oxidation of I but also from the reduction of H3AsO4 according to
Eq. (1). Aiming at the preparation of crystalline dehydro-L-ascorbic acid
4
dimer, AA in methanol has been oxidized by arsenic acid in the pres-
5
ence of a catalytic amount of iodine. In this case the actual reducing
agent of arsenic acid is the hydriodic acid, and the iodine produced [Eq.
(
1)] was reduced by AA. Finally, in a cell, arsenic acid is reduced to ar-
6
senous acid by arsenate reductase (E.C. 1.20.4.1). Here the reduction
involves thiols and not hydriodic acid.
Arsonic acids, RAsO3H2, can be reduced to haloarsines, RAsX2, ar-
seno compounds, (RAs)x, or arsines, RAsH2, depending on the reduc-
7
a
ing agent. The direct reduction of arsonic acids to arsonous acids,
RAs(OH)2, or their dehydrated arsenoso compounds (arsine oxides),
(
RAsO)x, has been achieved in a few cases only (summarized in ref-
erence [8]) because the usual reducing system SO2/I2/conc. HCl gives
the dichloroarsine, RAsCl2, which is then hydrolytically converted into
the arsonous acid or arsenoso compound. The use of concentrated hy-
drochloric acid may be objectionable when an arsonic acid is not stable
in it. We have developed a very mild method for the direct reduction of
arsonic acids to arsenoso compounds using AA, or triphenylphosphine
8
(
TPP), in the presence of a catalytic amount of iodine in methanol,
which can be used for hydrophobic or hydrophilic, respectively, arsenoso
compounds. C As bond fission has, so far, been observed only during
the reduction of 2-arsonohexanoic acid, which gave As2O3 and hexanoic
9
acid.
The reduction of arsinic acids, R2AsO2H, can lead to haloarsines,
R2AsX, to bis(dialkyl)diarsines, R2As-AsR2, or to arsines, R2AsH,
7a
depending on the reducing agent. There are no reports in the
1
0a
literature
on the direct reduction of arsinic acids to their oxides,
R2As-O-AsR2, and the reduction of diphenylarsinic acid by phenylhy-
1
1
drazine gave Ph3As instead of Ph2As-O-AsPh2. The reduction of ar-
sinic acids, e.g., diphenylarsinic acid,12 by SO2/I /conc. HCl followed by
−
1
0a
hydrolysis of the chloroarsines to their oxides also is not common.
Finally, trialkyl- or triarylarsine oxides, R3As O, have been re-
−
duced by SO2/I , SnCl2, or H3PO2 to the corresponding arsines,
R3As.
7
b
Wishing to contribute to the mild reduction of arsenic(V) compounds
having the As O group, we report in this article on the use of the
AA/iodine and TPP/iodine in methanol for the reduction of aromatic ar-
sonic acids having electron-withdrawing and electron-donating groups,
1
–4, to the oxides 5–8, of aromatic and aliphatic arsinic acids 9 and 11
to the oxides 10 and 12 and of triphenylarsine oxide, 13, to tripheny-
larsine, 14.

Reduction of the As(V) O Group
1019
FORMULAE 1–14
RESULTS AND DISCUSSION
Reduction of Aromatic Arsonic Acids
The reduction of o-, m-, and p-nitrophenylarsonic acids 1–3 by AA/iodine
in methanol was smooth. Because we used excess ascorbic acid, the
progress of the reduction was checked by TLC by comparing the rela-
tive intensities of the spots due to AA and of the dehydroascorbic acid
4
8
monomer 15. The lithium hydroxide test did not work well because
of the slow precipitation of the lithium salts of these arsonic acids.
The rates of the reduction of nitrophenylarsonic acids were 1 ≥ 2 ꢀ 3,
and the yields were in the range of 50–90%. The rate of reduction of
phenylarsonic acid was faster than the nitrophenylarsonic acids, and
the yields of phenylarsine oxide, by the water extraction method, were
8
∼
80%.
For the reduction of arsonic acids by HI, we postulated that proto-
nation of the As O (in these and, in fact, in all compounds containing
8
+
the As O group) must precede their reduction. Then the As(V) atom
−
has a vacant site for entry of the soft I . Water elimination, protona-
tion by another HI, and expulsion of I2 will give the As(III) compound.
This hypothesis is in accordance with the rates of reduction of phenyl-,
nitrophenyl-, and aminophenylarsonic acids because the easier the pro-
tonation the faster the reduction will be. In water, the ionization con-
stants of the nitrophenylarsonic acids are13 3 = 2 > 1 (the much lower
∼
7
c
ionization constant of the ortho acid is attributed to the “ortho effect,”
i.e., intramolecular hydrogen bonding). In methanol, these acids will
be less ionized and the protonation of the As O group will be governed
by the position of the electron-withdrawing nitro group on the phenyl
∼
group 1 > 2 = 3. The rates of reduction that we observed closely follow
this order, and therefore, the protonation must be a contributing factor.

1
020
T. D. Sideris and P. V. Ioannou
−
The next step is the addition of I in the positively charged tetracoordi-
+
nated As(V) . The nitro groups at the ortho, meta, and para positions,
through their canonical structures, can and do affect the rate of the I
−
+
entry into –As(OH) , so the rates of reduction are 1 ≥ 2 ꢀ 3.
3
When an arsonic acid is reduced, the product can be either the ar-
sonous acid, R As(OH)2, or its dehydrated form: arsenoso compound
(
arsine oxide), (RAsO)x. The reduction of nitrophenylarsonic acids by
−
SO2/I /conc. HCl gave, after hydrolysis of the dichloroarsine, both the
arsonous acids and the oxides.10b The AA/iodine reduction of the nitro-
phenylarsonic acids 1–3 in methanol most likely gave in solution the
arsonous acids, Ar-As(OH)2, but the products isolated were the arsine
oxides 5–7 as evidenced by their colors,1
4,15
elemental analyses and
IR spectra in KBr. The As O As asymmetric and symmetric stretch-
ing vibrations of arsine oxides, (ArAsO)x, are expected in the region of
−1
7
50–700 cm based on the strong bands of (PhAsO)x found at 742 and
−
1 8
721 cm .
FORMULAE 15 and 16
We isolated two oxides 5, a yellow (50%) and a pale yellow (95%) ox-
ide, free of water and 1 (by elemental analyses and IR spectra: absence
of As O stretching16 at 910 cm ). Both decomposed above 230 C (the
melting points of oxides have little value as a criterion of purity for they
can be mixtures 17 of cyclic, oligomeric, and polymeric (RAsO)x), but they
had different IR (KBr) spectra. The yellow had one strong, somewhat
−1
◦
14
−1
broad band at 732 cm , while the pale yellow oxide had three sharp
bands at 732 (m), 798 (mw), and 636 (s). These bands were also seen in
Nujol. The bands at 732 and 798 were also found in 1 being assigned
1
6
to ω(CH)arom. and to asymmetric As-OH vibrations. At the moment, the
different color and IR spectra of the two oxides, 5, cannot be explained.
−
The reduction of 2 with SO2/I /conc. HCl followed by alkaline hydrol-
ysis of the dichloroarsine gave the arsonous acid, Ar As(OH)2, based
on the elemental analysis.14 Neither the exact melting point nor color
◦
18
was reported. A melting point of 147–151 C was stated by Hiratuka.
Later, Morgan and colleagues1 found that the alkaline hydrolysis of
5
the dichloroarsine gave, by autoxidation-reduction, the acid 2 and the

Reduction of the As(V) O Group
1021
ꢁ
yellow 3,3 -diarsonoazoxybenzene, and therefore, they reduced 2 in boil-
−
ing 5 N H2SO4 with SO2/I and obtained the oxide 6 as a white solid
after recrystallizations from ethanol and acetone in 72% yield with a
◦
melting point of 184.5–187.5 C. The reduction of 2 with AA/iodine in
methanol gave the very pale yellow oxide 6 based on the elemental
analysis. The IR (KBr) spectra of various preparations of 6 consistently
−
1
showed a strong, somewhat broad band at 732 cm covering the sharp,
−
1
medium 732 cm band of 2, but their melting points were neither con-
◦
sistent nor sharp: They started darkening from ∼70 C, then formed a
◦
red granule which, decomposed (swelled) at 107, 140, 168, or 185 C.
Also, our oxides 6 were very soluble in MeOH and Me2CO and mod-
erately soluble in CHCl3. All these data most likely indicate that the
oxides 6 we isolated were mixtures of various oligomeric and polymeric
entities.
−
10b
The reduction of 3 with SO2/I /conc. HCl
gave the yellow arson-
ous acid, Ar As(OH)2, after the alkaline hydrolysis of the intermediate
◦
18
14
dichloroarsine, with a melting point of 234 C. Doak and colleagues,
however, under apparently the same conditions as previously men-
tioned, obtained the white oxide 7. The reduction of 3 with AA/iodine
gave, by elemental analysis, the oxide 7 (45–55%) as an off-white pow-
◦
◦
−
der, which changed color at 236 C and decomposed at 244 C. Its IR
1
(
KBr) spectrum showed a very strong, narrow band at 742 cm (which
contained the ω(CH)ar. band ) flanked by two strong bands at 782 and
16
−
1
7
12 cm .
The reduction of 4 in its zwitterionic form by stoichiometric amounts
of either AA or TPP in the presence of ∼10 mol % I2 was a slow pro-
cess (∼48 h for the consumption of AA or TPP by TLC analyses).
8
−
+
Evidently, the protonation of the AsO3H group to –As (OH)3 or
+
+
As (OH)2(OP Ph3) was difficult. When the zwitterionic 4 was con-
verted into its hydrochloride, the reductions were fast: 1 h for AA and
<
3 h for TPP using 5 mol % I2. Thus, the electron-withdrawing ortho
+
NH group facilitated the reduction –AsO3H2 → –As(OH)2 in a man-
3
ner analogous to ortho –NO2 group.
The reduction of the zwitterionic o-aminophenylarsonic acid 4 or its
hydrochloride by AA/iodine gave a small amount of 15 implying that a
small amount of 8 was formed. A less hydrophilic compound running
just above 15 was isolated chromatographically. Its IR spectrum showed
−
1
−1
a γ -lactone (at 1792 cm ) and an O As O band (at 756 cm ), and
1
the complex H NMR spectrum showed both aromatic (6.5–8.0 ppm)
and aliphatic (3.5–5.0 ppm) hydrogens and very small amounts of ethyl
acetate and diethyl ether. Most likely, it is a condensation product of
4,5
cyclo-8 and one of the forms1 of dehydroascorbic acid 16 probably form-
ing a Schiff base. Pecherer 9 reported that dehydroascorbic acid dimer
4

1
022
T. D. Sideris and P. V. Ioannou
+
or 15 reacted with –NH2/–NH containing compounds like amides and
3
amino acids etc, giving colored products, a behavior resembling that of
ninhydrin.
The reaction of the hydrochloride of 4 and TPP/I2 gave the protonated
, which was separated from TPP oxide and excess TPP by extraction in
8
water. After neutralization, 8 precipitated as a gum, which on prolonged
cooling, was transformed to an off-white solid. Its melting point was
1
4
not sharp, resembling the behavior of the meta-isomer but not of the
◦
para-isomer. The latter had a sharp melting point (56–58 C) but the
2
0
elemental analysis and its empirical formula cited are strange. The
IR spectrum of 8 showed a very strong band at 750 cm and strong
−
1
−1
bands at 721, 690, and 629 cm .
Summarizing, the rates of reduction of aromatic arsonic acids by
AA/iodine and TPP/iodine depend on the nature and the position of
the substituent on the aromatic ring. The ease of the protonation of
the As O oxygen and the electronic effects of the aromatic ligand that
+
affect the degree of the positive character of arsenic(V) in –As (OH)3
+
+
−
+
or –As (OH)2(OP Ph3) determine the rate of I addition to As(V) and
+
therefore its reduction. An arsonic acid having an –NH2 or –NH group
3
is reduced by both AA/I2 and TPP/I2, but with AA, condensation be-
+
tween the –NH2/– NH group and dehydroascorbic acid takes place.
3
Reduction of Arsinic Acids and Triphenylarsine Oxide
The reduction of diphenylarsinic acid 9/AA/iodine with a 1:1:0.15 molar
ratio in nonde-aerated methanol consumed all AA (TLC analysis) but
various amounts of 9 were recovered. With a molar ratio of 1:1.5:0.15,
all AA had reacted in 16 h, but 18% oxide 10 and 51% acid 9 were iso-
lated. However, with a molar ratio of 1:1.5:0.15 in de-aerated methanol,
5
5–60% oxide 10 and only ∼10% acid 9 were isolated. Finally, with a
molar ratio of 1:1.5:0.3 in methanol under reflux for 4 h, ∼70% oxide
0 and 8% acid 9 were isolated. These results indicate that dipheny-
1
8
larsinic acid, 9, is reduced by the mild reducing system AA/iodine
much slower than the phenylarsonic acid. Because pure oxide 10 in
8
3
2
methanol is not appreciably oxidized by air , it seemed that the con-
sumption of the AA by air was catalyzed by the oxide. However, control
experiments (AA alone, AA + 5% wt/wt 9, and AA + 5% wt/wt 10, in
methanol at r.t.) showed that 15 was not formed after 48 h of stirring
in air. These results leave the precursor of 10, Ph2As-OH, as a prob-
able catalyst for the oxidation of AA and regenation of 9 as shown in
Scheme 1.
The binding of dioxygen to Ph2As-OH converts it to a peroxyarsinic
acid, Ph2AsO3H. Peracids, e.g., peroxyphenylarsonic acid, PhAsO4H2,

Reduction of the As(V) O Group
1023
SCHEME 1
have been postulated as intermediates in the arsonic-acid catalyzed
2
1
epoxidation of olefins by hydrogen peroxide. Then the peracid can
react with either AA or HI or both, giving the observed results.
The system 11/AA/iodine (1:1:0.15 molar ratio) in nonde-aerated
methanol did not produce any 15, and therefore, the oxide 12 was not
produced. With a 1:1:0.3 molar ratio, the reduction of cacodylic acid was
very slow, consuming ∼50% of the AA after 24 h stirring at r.t. (TLC
analysis).
8
The stronger reducing system TPP/iodine reduced both 9 (1:1:0.05
molar ratio) in ∼1 h and 11 (1:1:0.05 molar ratio) in 3 h as evidenced
by the consumption of TPP (TLC analysis). No attempts were made to
isolate the oxide 10 from the lipophilic TPP oxide, while we were unable
to separate the oxide 12 from TPP oxide.
The moderate reactivity of diphenylarsinic acid, 9, and the unreac-
tivity of dimethylarsinic acid (cacodylic acid), 11, towards AA/iodine can

1
024
T. D. Sideris and P. V. Ioannou
be explained as follows. This reducing system generates HI, which can
form an adduct with arsinic acids, L2AsO2H·HI. Such an adduct has
been postulated in the extraction of HI by di-n-octylarsinic acid and
2
2
the adduct Me2AsO2H·HCl has been prepared. In our case, the ac-
tual structure of the adduct under a limited amount of HI in methanol
(
which has a moderate dielectric constant) may be dimeric and not
2
2
monomeric (Scheme 2).
SCHEME 2
Evidence of the hydrogen-bonded species shown in Scheme 2 comes
from the observations that arsinic acids are strongly hydrogen bonded
2
3
1
7, attempts to crystallize the oxide 10 in air and gave, after ox-
2
4
idative hydrolysis, diphenylarsinic acid, 9, and its dimer 17, the
sulfide Ph2 As-S-AsPh2, likewise, gave a hydrogen bonded dimer
2
5
(
Ph2AsO2H)2·2Ph2As(S)OH, and attempts at the preparation of the
triethylammonium salts of dimethyl- and diphenylarsinic acids in
methanol gave after evaporation and drying the free acids and not
their salts (this work). Assuming that protonation of the As O group
has taken place (Scheme 2) there are two factors that will determine
the outcome of the reduction. The first factor is the addition of I to
the tetracoordinated As(V) , which will be favored by the two (weakly)
−
+
electron-withdrawing phenyl groups, compared to two (weakly) elec-
+
trondonating methyl groups because they will make the As(V) more
−
positive and willing to accept the soft I . The second factor is the ex-
pulsion of a H2O molecule from the hydrogen-bonded species 19 to give
L2As(O)I en route to L2As(O)I2,8 which may be difficult unless the
hydrogen bonding breaks down, e.g., by refluxing, as was found.
The high reactivity of 9 and the low reactivity of 11 towards the
,26
8
stronger reducing system TPP/iodine can be explained as follows. With

Reduction of the As(V) O Group
1025
+
8
this system, “protonation” is achieved by Ph3P -OMe as shown in
Eq. (3):
(3)
+
The Ph3P -group in 20 and 21, being more electron-withdrawing than
the –H, probably disrupts the hydrogen-bonding system analogous to 18
and 19 in Scheme 2 and by outweighing the electron-donating ability of
the two methyl groups in cacodylic acid allows the formation of 21. The
+
−
expulsion of Ph3P O gives the L2As (OH)I·I or L2As(OH)I2, which is
easily converted into oxides 10 and 12.
Triphenylarsine oxide, 13, in methanol was reduced by AA/iodine to
triphenylarsine, 14, which precipitated out in 45–50% yields after a 2-
hour reaction. No effort was made to optimize the reaction. A probable
mechanism for this reduction is shown in Scheme 3. The Ph3AsO·HI
7
d
adduct may exist as tetra- and pentacoordinated species 22 and 23.
The intramolecular dehydration of 23 should give 24, which on further
7
e
reaction with HI gives 23 and 25. The latter is a well known compound
7
e
and by the expulsion of I2 or via disproportionation to Ph3As and
+
−
Ph3AsI and I gives the product Ph3As, 14.
3
SCHEME 3
To summarize, the direct reduction of arsinic acids by HI, gener-
ated by AA and catalytic amount of I2, involves two key steps: the
−
protonation of the As O group and entry of the soft Lewis base I to
+
the tetra-coordinated As(V) . The feasibility of both steps depends on
the electronic effects of the alkyl or aryl groups bound to the As O
group. Because arsinic acids have the tendency to dimerize in the
solid state and probably in solvents with moderate to low dielectric

1
026
T. D. Sideris and P. V. Ioannou
constants, another step should be considered: that of the expulsion
of water from a hydrogen-bonded >As(OH)2I. The powerful electron-
+
withdrawing nature of the Ph3P group seems to outweigh these elec-
tronic and hydrogen bonding effects in the reduction of arsinic acids
by TPP/iodine in methanol. Triphenylarsine oxide, having three weak
electron-withdrawing groups, was easily reduced by AA/iodine.
EXPERIMENTAL
Dimethylarsinic acid (cacodylic acid) was from Serva, while dipheny-
larsinic acid was prepared in a ∼60% yield from triphenylar-
sine oxide (Aldrich) and sodium hydroxide according to Horner
and colleagues,27 m.p. 173 C [lit.
◦
27
174 C]. 2-Nitrophenylarsonic
and 2-aminophenylarsonic acids were from Aldrich, while 3-
◦
◦
28
◦
nitrophenylarsonic acid (m.p. 177–180 C dec.; lit. 182 C) and 4-
nitrophenylarsonic acid (m.p. 300 C dec.; lit. 300 C dec.) were pre-
◦
28
◦
2
8
pared according to Ruddy and colleagues. Ascorbic acid was from
Merck. Silica gel 60 H for TLC was from Merck, while silica gel Si60
for column chromatography was from Serva. AR grade solvents and so-
lutions were de-aerated by boiling, flushing with nitrogen, stoppering,
and cooling to r.t. TLC was run on microslides. The spots were made
visible by iodine vapors and in the case of ascorbic acid and 15 by spray-
ing with 35% sulfuric acid and charring. IR spectra were obtained on
a Perkin-Elmer model 16PC FT-IR spectrometer. Elemental analyses
were obtained through the Centre of Instrumental Analyses, Univer-
sity of Patras, Patras, Greece.
Reductions with Ascorbic Acid/Iodine
2
-Nitrophenylarsonic Acid, 1
To a yellow solution of 2-nitrophenylarsonic acid, 1, (247 mg, 1 mmol)
and ascorbic acid (246 mg, 1.5 mmol) in methanol (4 mL), iodine (13
mg, 0.05 mmol) was added and stirred at r.t. for 90 min. TLC (ethyl
acetate) showed the presence of ascorbic acid (Rf 0.38 streaks) and 15
(
Rf 0.61). The concentration and addition of water (4 mL) precipitated
the product 5 as a yellow solid. Centrifugation and washing with water
(
5
2
3 × 1 mL) and methanol (1 × 2 mL) gave the 2-nitrophenylarsenoxide,
◦
, (185 mg, 87%) as a very pale yellow powder. M.p. 240–245 C shrinks,
◦
47 C dec. It is insoluble in MeOH, Me2CO, and CHCl3. Calculated for
C6H4NO3As (Mr 213.02): C, 33.83; H, 1.89; N, 6.58%; found: C, 33.62;
H, 1.99; N, 6.26%. IR (KBr): 3404 w, broad, 3062 w, 1596 w, 1564 w, 1525
vs, 1454 w, 1338 vs, 1308 m, 1106 m, 856 mw, 796 s, 732 s, 706 m, 634
vs, 590 m.

Reduction of the As(V) O Group
1027
The reduction of 1 (1 mmol) as previously mentioned gave after evap-
orating and drying in vacuo an orange foam. The addition of chloro-
form (15 mL) was followed by the extraction of 15 and AA with wa-
ter (1 × 10, 3 × 5 mL). An off-white solid was noticed at the interface,
which was discarded. The yellow chloroform layer was dried (Na2SO4),
filtered, evaporated, and dried to give the product, 5 (111 mg, 52%) as
a yellow powder, insoluble in CHCl3, Me2CO, and MeOH. M.p.: from
◦
◦
◦
1
2
35 C starts darkening, at 225 C forms a dark red granule, and from
30 C melts slowly. Calculated for C6H4NO3As (Mr 213.02): C, 33.83;
H, 1.89; N, 6.58%; found: C, 33.67; H, 1.92; N 6.37%. IR (KBr): 3428 w,
broad, 1596 w, 1520 vs, 1338 vs, 1308 w, 1104 m, 856 w, 792 m, 734 vs,
650 w, 550 w.
3
-Nitrophenylarsonic Acid, 2
Reduced at 1 mmol scale and worked up after 4 h, as in the case of 1.
Concentration of the methanolic solution gave a dark red oil which was
extracted with water to give 170 mg, 63%, of a very pale yellow powder.
◦
◦
M.p: from 66 C shrinks and darkens, at 82 C forms an orange gran-
◦
◦
15
ule which from 92 C swells and until 114 C decomposes (lit. 184.5–
87.5 C for the recrystallized solid from ethanol and acetone 6). It is
◦
1
very soluble in MeOH, soluble in Me2CO, moderately soluble in CHCl3,
and insoluble in Et2O, H2O. Calculated for C6H4NO3As (Mr 213.02): C,
3
3
3.83; H, 1.89; N, 6.58%; found: C, 33.63; H, 1.97; N, 6.82%. IR (KBr):
444 w, broad, 3076 w, 2843 vw, 1600 w, 1522 vs, 1346 vs, 1274 w, 1084
w, 1014 w, 868 w, 810 m, 732 vs, 674 m, 648 m, 566 w.
4
-Nitrophenylarsonic Acid, 3
Reduced at 1 mmol scale as in the case of 3-nitrophenylarsonic acid.
After 24 h work up gave the oxide 7 as an off white powder (55% yield).
◦
◦
M.p. 236 C shrinks and turns red, 244 C dec. (swells). It is insoluble
in MeOH, Me2CO, and CHCl3. Calculated for C6H4NO3As (Mr 213.02):
C, 33.83; H, 1.89; N 6.58%; found: C, 33.98; H, 1.93; N 6.15%. IR (KBr):
3
444 mw, broad, 1596 w, 1508 vs, 1348 vs, 1312 w, 1280 w, 1104 w, 1072
w, 1012 w, 850 m, 782 s, 742 vs, 712 s, 686 m, 536 w.
Diphenylarsinic Acid, 9
To a de-aerated solution of diphenylarsinic acid, 9 (262 mg, 1 mmol)
and ascorbic acid (264 mg, 1.5 mmol) in methanol (5 mL), iodine (38
mg, 15 mol %) was added, and the solution was stirred at r.t. for 24 h.
Evaporation gave a yellowish oil, which was dissolved in chloroform (20
mL), washed with water (4 × 5 mL), dried (Na2SO4), and evaporated to
give a semisolid (290 mg; expected oxide 10: 237 mg). Extraction with
boiling petroleum ether (2 × 5 mL) left impure diphenylarsinic acid, 9,

1
028
T. D. Sideris and P. V. Ioannou
◦
(
28 mg, m.p. 168–170 C; 11% recovery). The extracts were concentrated
◦
to ∼1.5 mL and cooled at −20 C overnight to give the oxide 10 (139 mg,
◦
29
◦
12
◦
5
9%) as a white solid, m.p. 91–93 C (lit. 92.5– 93.5 C; lit. 91–92 C).
IR (KBr): 3454 w, 3052 vw, 1576 vw, 1476 w, 1432 m, 1074 w, 1022 w,
9
96 w, 736 vs, 694 vs, 550 mw.
In nonde-aerated methanol using 15 mol % iodine, ascorbic acid was
completely consumed in 3 days. Yield of 10: 56%; recovered 9: 10%.
Triphenylarsine Oxide, 13
In a centrifuge tube, triphenylarsine oxide, 13, (161 mg, 0.5 mmol)
and ascorbic acid (132 mg, 0.75 mmol) were dissolved in de-aerated
methanol (2 mL), and iodine (20 mg, 15 mol %) was added. Stirring at
r.t. for 2 h gave the solid product (sometimes an oil was obtained, but
seeding gave the solid product). TLC (AcOEt) of the supernatant showed
that most ascorbic acid had reacted. Centrifugation and washing with
water (1 × 1 mL) and methanol (2 × 0.5 mL) gave the product 14 (70
◦
30
◦
31
◦
mg, 46%), m.p. 58–61 C (lit. 61 C; lit. 59–60 C), having the same
IR (KBr) with an authentic sample of 14.
Reductions with Triphenylphosphine/Iodine
Hydrochloric Acid Salt of 2-Aminophenylarsonic Acid, 4
2
-aminophenylarsonic acid, 4, (217 mg, 1 mmol) dissolved in
methanol (3 mL) was acidified to pH ∼2 with concentrated aqueous
hydrochloric acid, evaporated, and dried in vacuum over phosphorus
pentoxide to give a beige-white solid (272 mg; expected 253 mg). The
solid was dissolved in methanol (5 mL), triphenylphosphine (334 mg,
1
.2 mmol) and iodine (13 mg, 5 mol %) were added and stirred at r.t.
for 8 h. TLC (Et2O) showed the excess Ph3P (Rf ∼1), Ph3P O (Rf 0.15),
and spots at Rf 0.08 and 0.0. Evaporation gave a light orange gummy
oil, which was treated with warm water (2 × 3 mL), cooled to r.t., and
filtered (Pasteur pipette plugged with cotton). The insoluble solid was
3
88 mg (expected 350 mg Ph3P and Ph3P O). To the clear aqueous fil-
trate concentrated aqueous ammonia was added with stirring till pH 8.
The sticky white gum, which precipitated, was transformed to a hard
◦
solid on cooling at +4 C overnight. It was broken with a spatula and
centrifuged, and the powder washed with water (1 × 1 mL), centrifuged,
and dried over phosphorus pentoxide to give the oxide 8 (110 mg, 60%)
as a white powder. It is slightly soluble in CHCl3, moderately soluble in
◦
Me2CO, soluble in MeOH, DMSO, and insoluble in Et2O. M.p. at 69 C
forms a granule and at >90 C swells. Calculated for C6H6NOAs (Mr
◦
1
83.04): C, 39.37; H, 3.31; N 7.65%; found: C, 39.15; H, 3.39; N 7.41%.
IR (KBr): 3410 m, 3324 m, 3205 m, 3055 w, 1618 ms, 1586 ms, 1476 ms,

Reduction of the As(V) O Group
1029
1438 vs, 1308 m, 1158 w, 1116 m, 896 w, 865 w, 750 vs, 722 s, 690 s, 629
ms, 540 w, 506 w.
REFERENCES
[
1] A. I. Vogel, Textbook of Quantitative Inorganic Analysis, Revised by J. Bassett,
R. C. Denney, G. H. Jeffery, and J. Mendham (Longman, London, 1979), 4th ed.,
Ch. 10, p. 383.
[
2] B. J. A. Haring, W. van Delft, and C. M. Bom, Fresenius Z. Anal. Chem., 310, 217
(
1982).
[
3] R. W. Herbert, E. L. Hirst, E. G. V. Percival, R. J. W. Reynolds, and F. Smith,
J. Chem. Soc., 1270 (1933).
[
[
4] E. K. Koliou and P. V. Ioannou, Carbohydr. Res., 340, 315 (2005).
5] G. M. Tsivgoulis, P. A. Afroudakis, and P. V. Ioannou, J. Inorg. Biochem., 98, 649
(
2004).
[
6] P. Martin, S. DeMel, J. Shi, T. Gladysheva, D. L. Gatti, B. P. Rosen, and B. F. P.
Edwards, Structure, 9, 1071 (2001).
[
7] G. O. Doak and L. D. Freedman, Organometallic Compounds of Arsenic, Antimony,
and Bismuth (Wiley, New York, 1970), (a) Ch. II, pp. 32–36; (b) Ch. V, pp. 187–188;
(
c) Ch. II, pp. 26–32; (d) Ch. V, pp. 209–213; (e) Ch. V, pp. 201–207.
[
8] P. V. Ioannou, Appl. Organometal. Chem., 14, 261 (2000).
[
9] P. V. Ioannou and M. G. Siskos, Appl. Organometal. Chem., 15, 511 (2001).
[
10] M. Dub, Organometallic Compounds: Compounds of Arsenic, Antimony and Bis-
muth (Springer-Verlag, Berlin, 1968), 2nd ed., Vol. 3. (a) pp. 212–218; (b) pp. 219–
250.
[
11] H. Wieland, Justus Liebigs Ann. Chem., 431, 30 (1923).
[
12] K. Takahashi and T. Ueda, Dai Nippon Pharmaceutical Co., U. S. 2,701,812 (Feb. 8,
1955); Chem. Abs., 50, 1907b (1956).
[13] D. Pressman and D. H. Brown, J. Am. Chem. Soc., 65, 540 (1943).
[14] G. O. Doak, H. Eagle, and H. G. Steinman, J. Am. Chem. Soc., 62, 168 (1940).
[15] J. F. Morgan, E. J. Cragoe, Jr., R. J. Andres, B. Elpern, R. F. Coles, J. Lawhead, R. L.
Clark, E. B. Hatlelid, F. H. Kahler, H. W. Paxton, C. K. Banks, and C. S. Hamilton,
J. Am. Chem. Soc., 69, 930 (1947).
[
16] F. D. Yambushev, V. P. Kovyrzina, R. R. Shagidullin, L. A. Gorchakova, B. G. Fedotov,
Yu. A. Bespalov, M. D. Nikiforov, G. I. Kokorev, and F. G. Khalitov, J. Gen. Chem.
USSR, 51, 1919 (1981).
[
[
[
[
[
[
[
[
17] W. Steinkopf, S. Schmidt, and H. Penz, J. Prakt. Chem., 141, 301 (1934).
18] K. Hiratuka, J. Chem. Soc. Japan, 1060 (1937); Chem. Abs., 33, 1578 (1939).
19] B. Pecherer, J. Am. Chem. Soc. , 73, 3827 (1951).
20] K. J. Stevenson, G. Hale, and R. N. Perham, Biochemistry, 17, 2189 (1978).
21] S. E. Jacobson, F. Mares, and P. M. Zambri, J. Am. Chem. Soc., 101, 6946 (1979).
22] A. M. Olivares and K. J. Irgolic, J. Inorg. Nucl. Chem., 34, 1399 (1972).
23] M. R. Smith, R. A. Zingaro, and E. A. Meyers, J. Organometal. Chem., 20, 105 (1969).
24] L. H. Doerrer, J. C. Green, M. L. H. Green, I. Haiduc, C. N. Jardine, S. I. Pascu,
L. Silaghi-Dumitrescu, and D. J. Watkin, J. Chem. Soc., Dalton Trans., 3347
(
2000).
[
25] L. Silaghi-Dumitrescu, M. N. Gibbons, I. Silaghi-Dumitrescu, J. Zukerman-
Schpector, I. Haiduc and D. B. Sowerby, J. Organometal. Chem., 517, 101 (1996).
26] K. J. Irgolic, R. A. Zingaro and L. J. Edmonson, Jr., Phosphorus, 5, 183 (1975).
27] L. Horner, H. Hoffman, and H. G. Wippel, Chem. Ber., 91, 64 (1958).
[
[

1
030
T. D. Sideris and P. V. Ioannou
[28] A. W. Ruddy, E. B. Starkey, and W. H. Hartung, J. Am. Chem. Soc., 64, 828 (1942).
[29] F. F. Blicke and F. D. Smith, J. Am. Chem. Soc, 51, 1558 (1929).
[30] R. C. Weast, CRC Handbook of Chemistry and Physics (CRC Press, Boca Raton,
Florida, 1985), 65th ed.
[
[
31] R. A. Zingaro, R. E. McGlothin, and R. M. Hedges, Trans. Faraday Soc., 59, 798
(
1963).
32] T. D. Sideris and P. V. Ioannolu, Phosphorus, Sulfur, and Silicon, 181, 4, p. 751–762
2006).
(