The Reaction of Diethylthiophosphinyl Iodide, Et2P(S)I, with Nucleophiles

Article abstract of DOI:10.1002/zaac.201700143

The compound Me₂AsSI can exist in two different forms, either as dimethylarsinosulfenyl iodide [or (iodothio)dimethylarsane)], Me₂As–S–I (A), or as dimethylthioarsinyl iodide (or dimethylarsinothioic iodide), Me₂As(S)–I (B). To confirm that the structure of the product of the reaction between Bunsen's cacodyl disulfide Me₂As(S)–S–AsMe₂ and iodine is A and not B, the known diethylthiophosphinyl iodide (or diethylphosphinothioic iodide), Et₂P(S)–I (2) was prepared and its hydrolytic stability and reactivity towards a variety of nitrogen, phosphorus(III), arsenic(III), oxygen, and sulfur(II) nucleophiles were studied. The results indicated that only a few reactions of 2 resembled those of A, thus strengthening the proposal that the reaction of Bunsen's cacodyl disulfide with iodine produced A and not B. A series of ³¹P NMR chemical shifts of diethylthiophosphinyl moiety is also reported. Et₂P(S)–DMAP, synthesized and isolated during the presented study, is the ethyl analogue of Me₂P(S)–DMAP, previously described as an important molecule. In our case, Et₂P(S)–DMAP was found to be a good intermediate for the synthesis of phosphoryl or thiophosphoryl derivatives since it was more reactive than 2 towards nucleophiles.

Full text of DOI:10.1002/zaac.201700143

Journal of Inorganic and General Chemistry
DOI: 10.1002/zaac.201700143
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
The Reaction of Diethylthiophosphinyl Iodide, Et2P(S)I, with
Nucleophiles
Gerasimos M. Tsivgoulis,*[a] Golfo G. Kordopati,[a] Dimitris G. Vachliotis,[a] and Panayiotis V. Ioannou[a]
Abstract. The compound Me2AsSI can exist in two different forms, studied. The results indicated that only a few reactions of 2 resembled
either as dimethylarsinosulfenyl iodide [or (iodothio)dimethylarsane)],
Me2As–S–I (A), or as dimethylthioarsinyl iodide (or dimethylarsi- sen’s cacodyl disulfide with iodine produced A and not B. A series of
nothioic iodide), Me2As(S)–I (B). To confirm that the structure of the
31P NMR chemical shifts of diethylthiophosphinyl moiety is also re-
product of the reaction between Bunsen’s cacodyl disulfide Me2As(S)– ported. Et2P(S)–DMAP, synthesized and isolated during the presented
those of A, thus strengthening the proposal that the reaction of Bun-
S–AsMe2 and iodine is A and not B, the known diethylthiophosphinyl
iodide (or diethylphosphinothioic iodide), Et2P(S)–I (2) was prepared
and its hydrolytic stability and reactivity towards a variety of nitrogen,
study, is the ethyl analogue of Me2P(S)–DMAP, previously described
as an important molecule. In our case, Et2P(S)–DMAP was found to be
a good intermediate for the synthesis of phosphoryl or thiophosphoryl
phosphorus(III), arsenic(III), oxygen, and sulfur(II) nucleophiles were derivatives since it was more reactive than 2 towards nucleophiles.
In this work, the stability and electrophilicity of Et2P(S)–I
Introduction
towards nitrogen, phosphorus, arsenic, oxygen, and sulfur nu-
Recently, by reacting Bunsen’s cacodyl disulfide,
cleophiles were studied and compared with those of Me2AsSI.
Me2As(S)–S–AsMe2, with iodine we prepared a compound,
Me2AsSI, which can be formulated as the AsIII compound di-
methylarsinosulfenyl iodide [or (iodothio)dimethylarsane)],
Me2As–S–I (A), or as the isomeric AsV compound dimethyl-
The results support the proposal that the structure of Me2AsSI
can be formulated as Me2As–S–I (A). Moreover, the 31P NMR
thioarsinyl
Me2As(S)–I (B).[1] Theoretical calculations showed that in the
gas phase the arsinyl iodide was more stable by
iodide (or
dimethylarsinothioic
iodide),
B
45.14 kJ·mol–1 than the sulfenyl iodide A. However, its IR
spectrum showed absence of an As=S band at 486 cm–1 and
its chemical properties strongly indicated that the dark red
orange oil was the sulfenyl iodide A. The operation of the
inert pair effect of arsenic can play an important role for this
behavior:
Study of the phosphorus analogues of the arsenic could pro-
vide additional evidence to the above conclusion. To the best
of our knowledge dialkylphosphinosulfenyl halides, R2P–S–X
have not been prepared, whereas the PV dialkylthiophosphinyl
halides, e.g. Et2P(S)–Cl,[2] Et2P(S)–Br,[3,4] and Et2P(S)–I[5]
are known. These compounds can be obtained by the
reaction of halogen with tetraalkyldiphosphine disulfide,
R2P(S)–P(S)R2,[6] the latter obtained from an “anomalous”
Grignard reaction on phosphothio halogenides PSX3.[7,8]
* Associate Prof. Dr. G. M. Tsivgoulis
Fax: +30-2610997118
E-Mail: tsivgoulis@chemistry.upatras.gr
[a] Department of Chemistry
University of Patras
Figure 1. Formulae of compounds 1–11, their 31P NMR (at 162 MHz)
chemical shifts in CDCl3 and, in italics, their Rf values on silica gel
TLCs run in ether/petroleum ether 1:3.
26504 Patras, Greece
Z. Anorg. Allg. Chem. 0000, ᭿,0–0
1
© 0000 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
chemical shifts of a series of Et2P(S)–Y compounds were tabu- chemical shifts of compounds 1–11 (Figure 1), are in agree-
lated (Figure 1).
ment with those expected.[9,10] However, the 31P chemical
shifts of 3 and 9 coincided and in this case the H NMR spec-
1
trum was more informative showing the P(S)–OCH3 doublet
at approx. 3.65 ppm (J approx. 13 Hz). In some cases, follow-
ing the progress of the reactions by 31P NMR was not straight-
forward because the chemical shifts of the products were
slightly shifted by the presence of charged species in the solu-
tion. As the 31P NMR spectra revealed, all the reactions of 2
produced very small amounts of by-products whose nature is
not known, and in the text, we did not cite their values in order
not to overwhelm it with numbers. In all cases, approx. 3–5%
of 3 was produced due to tiny amount of water present in the
dried solvents.
Results and Discussion
Reaction Conditions and Identification of the Products
The electrophilic behavior of 2 was studied by H and 31P
1
NMR spectroscopy. To diminish the risk of hydrolysis of com-
pound 2 to 3 during these studies, part of the experiments were
run directly inside the NMR tubes. For the same reason, in
preparative experiments the volumes of solvents were kept to
minimum and dry glassware was used. Despite these pre-
cautions, approx. 3–5% of 3 and the analogous quantity of the
nucleophile were present in some products and their elemental
analyses (not reported herein) matched to those calculated for
the impure products.
Preparation and Properties of 2
The compounds were identified by TLC, H and 31P NMR
1
Diethylthiophosphinyl iodide (2) was prepared by stirring 1
and an equimolar quantity of I2 at 83 °C instead of 100 °C,[5]
in order to minimize the sublimation of I2, for 2 h. Despite this
precaution in some experiments traces of remaining disulfide
1 were detected by TLC but not by 31P NMR spectroscopy.
However, impurity 1 was not reactive[6] towards the nucleo-
philes studied. The iodide 2 was a dark red oil [while
Me2As–S–I (A) was a red orange oil[1]] and on TLC runs
above 1 with very little hydrolysis, although in moist air it
hydrolyzes.[5] Very small amounts of 2 were produced when 1
and I2 were stirred in dry dichloromethane for 2.5 d. The
mechanism of the reaction probably involves a 4-center transi-
tion state although the bond lengths of 1 (P–P 22.2 pm)[11] do
not exactly match those of Cl–Cl (9.9 pm), Br–Br (11.4 pm),
and I–I (13.3 pm).
(Figure 1). For TLCs, 1 and 3 were co-run as standards, be-
cause most products run in the relatively narrow region of
0.21–0.60. Some compounds, e.g. 9 / 2 and 6 / Ph2NH have
identical Rf values. Compound 2 was not appreciably hy-
drolyzed on TLC.
1
The H NMR spectra showed that the products 3–11 have
the CH2 and CH3 protons always upfield compared to those of
2 (see Figure 2). The anilide 7 and the DMAP derivative 8
showed two signals for the diastereotopic CH2, whereas for
the other compounds only one signal is observed at the region
2.0–2.2 ppm. More useful were the 31P NMR spectra. 31P
In the literature, there are studies on the factors affecting the
position of the P=S band in the IR spectra.[12] For the chlorides
RRЈP(S)–Cl two bands at 775–750 and 626–593 cm–1 are re-
ported,[12] whereas in the case of 2 two bands at 750 and
580 cm–1 were observed (Figure 2).
1
The H NMR spectrum of 2 in CDCl3 (Figure 2) showed
two triplets for –CH3 protons as a result of the splitting by 31P,
whereas two multiplets were seen for the –CH2– protons due
to diastereotopic protons –CHAHB–. The 31P splitting inside
each multiplet could not be directly analyzed.
The singlet in the decoupled 31P resonance of 2 was found
to be at δ = 62.78 ppm in CDCl3 (Figure 1) and at δ =
63.05 ppm when neat.[5] This value is in agreement with the
trend observed for Et2P(S)–Cl (δ
= 108.3 ppm) and
Et2P(S)–Br (δ = 98.3 ppm) as neat liquids.[13]
Stability of Et2P(S)–I (2)
Dialkylthiophosphinyl chlorides, R2P(S)–Cl, are generally
less reactive compared to their oxygen analogues, R2P(O)–Cl,
in reactions with nucleophilic reagents including water.[14a]
However, they can be hydrolyzed with warm water to the acid
Figure 2. (A) IR (neat) spectra of compound 2, (B) 31P spectra of
compound 2, (C) 1H NMR spectra of compound 2. In the 1H NMR
spectrum (D) of the anilide 7 the upfield shifting of the CH2CH3 pro-
tons compared to those of 2 is evident.
Z. Anorg. Allg. Chem. 0000, 0–0
www.zaac.wiley-vch.de
2
© 0000 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
R2P(S)–OH or by alcoholic NaOH to the salt R2P(S)–ONa.[14a] pyridine, or triethylamine. The 31P NMR spectra showed the
The hydrolysis proceeds via the anhydride[15] (see reaction).
presence of 3 and 6 but their chemical shifts showed some
variance. For example, the singlet attributed to 6 was found at
110.77, 108.37, 107.46, and 106.55 ppm, whereas the isolated
6 resonated at δ = 109.08 ppm (Figure 1). Diphenylamine re-
acted to a significant extent with 2 in the presence of triethyl-
amine and 4-dimethylaminopyridine (DMAP) as described in
the Experimental Section, indicating that the reaction involved
the formation and reaction of 8 (see below). It should be noted
1
that in the H NMR spectrum of the isolated 6 containing 6%
of 3 the CH2CH3 protons resonated as multiplets at identical
positions: 2.08 and 1.24 ppm, respectively.
Anhydrides of this type when pure are usually inert towards
hydrolysis at room temperature.[16] In the absence of a base,
treating 2 with approx. 20 times molar excess of water in acet-
one, revealed (by TLC) that all 2 had reacted and the anhydride
3 plus the acid 4 were produced in 30 min. Traces of 1 were
also detected indicating its non-reactivity towards water. After
2 h at room temperature evaporation of the orange solution
The reaction of 2 with aniline (1:2 molar ratio) in CDCl3 at
room temperature gave at once a white precipitate (PhNH2·HI)
and a colorless supernatant, whose TLC showed traces of 2
that had not reacted, traces of 3 and the product 7, verified by
1
the H and 31P NMR spectra. Thus, aniline being a stronger
1
gave a red oil whose H NMR (CDCl3) showed multiplets at
base (pKb 9.4) than diphenylamine (pKb 13.2) quickly reacted
with 2, while it did not react with Me2As–S–I (A).[1] In a pre-
parative experiment impurities were removed from the product
7 by precipitating it from an ether solution. A yield of 53%
was obtained and its 1H NMR spectrum showed two multiplets
for the diastereotopic CH2 protons (Figure 2).
The reaction of pyridine and DMAP with Me2As–S–I (A)
was complicated because Me2As–S–S–AsMe2 was detected by
1H NMR implying the formation of charge transfer complexes
py·I2 and DMAP·I2.[1] No reaction was observed between the
iodide 2 and pyridine in CDCl3 at room temperature for 20 h.
Only free pyridine and small amount of 3 were detected in the
1H and 31P NMR spectra. Moreover 1 was not detected and
therefore py·I2 was not formed.
The reaction of the iodide 2 and DMAP (1:1 molar ratio) in
either dry dichloromethane or dry ether gave the product 8
contaminated with DMAP·HI that is produced from the partial
hydrolysis of 2 to the anhydride 3. In order to obtain purer 8
we used the stoichiometry described in the Experimental Sec-
tion. Under these conditions, contamination of the product with
DMAP·HI was limited to about 2%. Recrystallization from dry
acetone gave needles not suitable for X-ray analysis.
The P=S stretching vibrations in RRЈP(S)–NϽ compounds
were found in the regions 776–715 and 574–572 cm–1,[12]
while we assign the P=S band at 584, 575, and 611 cm–1 for
6, 7, and 8, respectively.
δ = 1.25 ppm (CH3 protons), 2.01 ppm (CH2 protons of 4) and
2.25 ppm (CH2 protons of 3). Their 31P NMR spectra verified
the assignments (Figure 1), but a singlet signal at δ =
111.61 ppm of medium intensity could not be assigned. The
easier hydrolysis of impure 3 may be due to the presence of
HI, which can protonate the anhydride probably at the oxygen
atom thus facilitating its hydrolysis. Comparing the rate of hy-
drolysis of Me2As–S–I (A) (in CD3OD) by excess water[1]
with that of 2 (in Me2C=O), the former is faster and gave
exclusively the acid Me2As(S)–OH via Me2As(S)–O–
As(S)Me2, whereas 2 gave a mixture of the acid 4 and the
anhydride 3 (predominating) plus an unknown compound.
Repetition of the hydrolysis with 30ϫ excess water, evapo-
1
ration and drying gave a dark red viscous oil, whose H and
31P NMR spectra in CDCl3 revealed that 4 greatly predomi-
nated. By adding pyridine in order to neutralize both the acid
4 and the HI, a white precipitate formed (presumably py·HI).
TLC showed spots at Rf 0.08 and 0.0 due to py·HI and 5. The
1H NMR showed three multiplets at δ = 1.24, 1.96, and
2.17 ppm and protonated pyridine at δ = 7.51, 7.91, and
8.79 ppm, whereas the 31P NMR showed the singlet signal of
the salt 5 at δ = 92.30 ppm, traces of 3, and four small singlet
signals that could not be assigned.
Reaction of Et2P(S)–I (2) with Nitrogen Nucleophiles
DMAP in the presence of Et3N is a powerful catalyst for
acylation and other reactions and its reactivity is due to forma-
tion of intermediates like 8.[17,18] Isolated 1-arylsulfonyl-4-di-
methylaminopyridinium salts have been shown to react with
tyrosine residues in proteins.[19] Also, Me2P(S)–DMAP, the
methyl analogue of 8, has been previously reported as a valu-
able compound.[20] Preliminary experiments with 8 containing
DMAP as an impurity showed that it reacted with diphenyl-
amine (1:1 molar ratio) in CDCl3 to give 6 in 60% yield (by
1H NMR) and in the 31P spectrum two signals were observed:
that of 8 at δ = 108.90 ppm and of 6 at δ = 106.87 ppm. The
upfield movement of the chemical shift of 6 has been men-
The compounds R2P(S)–Cl can react with amines in excess
or an amine plus Et3N or an amine plus pyridine at low tem-
peratures at the beginning, followed by warming and heating
under reflux for approx. 2 h to give amides.[14b] The iodide 2
reacted with H2N–Ph–X (X = Cl, Br) to give the corresponding
p-chloro- and p-bromo anilides in 75–80% yields but the reac-
tion conditions and the 31P NMR spectra were not reported.[5]
We did not try the reaction of 2 with triphenylamine because
triphenylamine is unreactive towards inorganic acids and
Me2As–S–I (A).[1]
While diphenylamine and Me2As–S–I (A) under a 2:1
stoichiometry in CDCl3 did not react,[1] this amine showed tioned above. In this reaction DMAP is not necessarily re-
some reactivity towards 2 in the presence of diphenylamine, quired in stoichiometric quantities. When diphenylamine/2/
Z. Anorg. Allg. Chem. 0000, 0–0
www.zaac.wiley-vch.de
3
© 0000 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
Et3N/DMAP were reacted with a 1:1:1:0.5 molar ratio, com- indicated in the 1H NMR spectra. The 31P NMR spectrum
pound 6 could be isolated.
showed 2, traces of 3, and six additional singlet signals. The
expected (PhS)3As=S is an elusive compound that could not
be prepared from (PhS)3As and octasulfur.[25]
Summarizing, it was found that these two arsenic(III) nu-
cleophiles were virtually inert towards 2, and their behavior
mostly paralleled to that of A.
Reaction of Et2P(S)–I (2) with Phosphorus Nucleophiles
Triphenylphosphine quickly (1 h) desulfurized Me2As–S–I
(A) giving Me2As–I and Ph3P=S[1] but the conditions for de-
sulfurization of 2 to the readily autoxidized and hydrolyzed
Et2P–I were not stated.[5] The reaction of 2 and Ph3P at room
temperature in CDCl3 under aerobic conditions for 1 h gave a
Reaction of Et2P(S)–I (2) with Oxygen and Sulfur
Nucleophiles
1
yellow solution with a H NMR spectrum featuring 2 that has
mostly not reacted and a broad instead of a sharp Ph3P signal.
In the 31P NMR spectrum Ph3P (δ = –4.64 ppm), Ph3P=S (δ =
44.40 ppm), 2 (δ = 60.83 ppm), and 3 (δ = 109.80 ppm) were
observed. Et2P–I at approx. 77 ppm[5] was not detected and
instead four unknown small singlet signals were present. After
4 d, compounds 2, 3, and Ph3P=S were present, Et2P–I was
absent and eight signals were present, which could not be as-
signed. When two equivalents of Ph3P for 3 d were used, 2
was present and Et2P–I was absent. Evidently Et2P–I reacted
further in a very complicated and, therefore, unknown way.
Methanol or phenol in stoichiometric amounts or excess did
not react with Me2As–S–I (A) in CDCl3 after 2 d at room
temperature. It is of interest, therefore, to compare the reactiv-
ity of MeOH and PhOH towards 2 under the same conditions.
Alcoholysis of R2P(S)–X to give esters R2P(S)–ORЈ(Ar) are
found in the patent literature because of their use as insecti-
cides etc.[26] and are usually prepared from the halide and
alcohol or phenol plus a tertiary base or using alkoxides or
aryloxides.[27]
Overall, Ph3P desulfurized
2 much more slowly than
Dissolving 2 in excess of dry methanol gave an orange-
brownish solution after 30 min stirring at room temperature.
By TLC a spot at Rf 0.55 quite close to that of 2 (Rf 0.51) and
a spot at Rf 0.0 probably due to 4 were found, whereas the
anhydride 3 was not detected. After evaporation and drying in
vacuo the 1H NMR spectrum showed a multiplet at δ =
1.98 ppm due to 4, a multiplet at δ = 2.22 ppm due to 9, a
doublet (δ =3.65 ppm, J = 13.2 Hz) attributed to P(S)–OMe
and a small [20% of P(S)–OMe] doublet at δ = 3.85 ppm (J =
10.4 Hz) tentatively attributed to P(O)–SMe. The 31P NMR
spectrum showed a prominent signal at δ = 109.90 ppm and
four very small signals the one at δ = 99.84 ppm probably
being due to 4 resulted from traces of water in dried methanol.
With stoichiometric amount of methanol in CDCl3 after 20
h at room temperature 62% of 9 was produced. In the 1H NMR
the ϾP(S)–OMe doublet of 9 produced in CDCl3 solvent was
actually more complex: having a doublet at 3.63 and 3.67
(J = 12.8 Hz) and a doublet at 3.64 and 3.67 (J = 13.2 Hz),
whereas 9 produced in MeOH solvent or in CDCl3 in the pres-
ence of pyridine or triethylamine did not show such a splitting.
Adding one equiv. of pyridine, py·HI was precipitated and 9
was produced in 87% yield based on methanol in 30 min as
expected.[27]
Me2As–S–I (A) (1 h)[1] or Bunsen’s cacodyl disulfide,
Me2As(S)–S–AsMe2, (5 min).[21]
Tris(phenylthio)phosphine[22] reacted much more slowly
than triphenylphosphine with Me2As–S–I (A) and the reaction
was quite complicated.[1] Similarly, virtually no reaction with
2 in CDCl3 at room temperature was observed by 31P NMR
after 24 h. Thus, a very strong singlet signal for (PhS)3P at
δ = 133.53 ppm was observed along with signals due to 2, 3,
and (PhS)3P=O [at δ = 56.66 ppm, which was a contaminant
of (PhS)3P]. No signal due to (PhS)3P=S at δ = 91.1 ppm (in
toluene)[13] was detected in our spectrum, although a very
weak signal at δ = 93.61 ppm can be assigned to it.
Summarizing our results, the phosphorus(III) nucleophiles
desulfurized 2 much more slowly than the arsenic(III)
Me2As–S–I (A) or arsenic(V,III) Me2As(S)–S–AsMe2 com-
pounds. The fact that the reaction of Me2As(S)–S–AsMe2 with
Ph3P was instantaneous, while the reaction with Me2As–S–I
(A) was slow may be an indication that the latter is not
Me2As(S)–I (B).
Reaction of Et2P(S)–I (2) with Arsenic Nucleophiles
Triphenylarsine did not react with Me2As–S–I (A), while
In the various runs, where 3 was detected by TLC, the 31P
(PhS)3As reacted very slowly (Ͼ24 h), incompletely and in a NMR spectra showed only one singlet at δ = 109.90 ppm im-
complicated way.[1]
plying that 3 and 9 resonated at the same position. Thus, in
Triphenylarsine did not react at all with 2 in CDCl3, as the contrast to the inert Me2As–S–I (A), the iodide 2 was quickly
1
1
sharp singlet signal at δ = 7.34 ppm in the H NMR spectrum methanolyzed to 9 whose presence was verified by H NMR
revealed, whereas the 31P NMR spectrum after 3 d at room but not with TLC (co-runs with 2) or by 31P NMR (same shift
temperature showed 2, traces of 3, and four unknown species. as 3).
Triphenylarsine sulfide, Ph3As=S, is difficult to prepare[21,23]
and its H NMR spectrum, now recorded, in CDCl3 showed some 3 by 31P NMR) and pyridine (1:1:1 molar ratio) in
Addition of crystalline phenol to a solution of 2 (containing
1
multiplets at δ = 7.50 and 7.74 ppm.
CDCl3 gave a brown precipitate and a yellow supernatant. By
1
Triphenyltrithioarsenite (or tris(phenylthio)arsine)[24] in both H and 31P NMR spectra, 2 had not totally reacted after
CDCl3 reacted slowly over a period of 4 d with 2 at room 24 h. However, the esterification was over in approx. 10 min
temperature as the change of the pattern of the aromatic signals in the presence of triethylamine. Work up gave 96% of the
Z. Anorg. Allg. Chem. 0000, 0–0
www.zaac.wiley-vch.de
4
© 0000 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
addition was complete stirring was continued for 40 min or until all
ester 10 as an oil with 94% purity, the impurities being phenol
and 3.
the magnesium turnings have reacted and the suspension was cooled to
0–5 °C in an argon atmosphere (solution A). Thiophosphoryl chloride
(1.00 mL, 1.672 g, 9.87 mmol) diluted by 9 mL dry tetrahydrofuran
(freshly distilled over sodium) was added to the previously prepared
Grignard solution A during a 25 min period and under vigorous stir-
ring. The reaction temperature was kept at 0–5 °C and during the
course of the addition a thick grey precipitate was formed. After com-
pletion of the addition, the reaction mixture was refluxed for additional
2 h. Subsequently, the suspension was cooled to 0 °C and acidified
Thiophenol and Me2As–S–I (A) in CDCl3 virtually did not
react after 24 h at room temperature.[1] The bromide Et2P(S)–
Br reacted with the salt Et2P(S)–S–Na+·2H2O giving the
thioanhydride Et2P(S)–S–P(S)Et2.[16]
Thiophenol added to a solution of 2 in CDCl3 did not react
at room temperature after 22 h. Its 31P NMR spectrum showed
signals of 2 and traces of 3 and of other compounds but not
11 at 82–83 ppm. After addition of an equivalent amount of with 1 m hydrochloric acid. After a few minutes of stirring, the organic
layer was separated and the aqueous phase was extracted with Et2O
(2ϫ10 mL). The combined organic phases were dried with Na2SO4
and concentrated, giving a clear light green oil. The crude oily product
was dissolved in the minimum amount of hot methanol (2.5 mL) and
then, while still hot, water (0.9 mL) was added to the point where the
solution turned to permanently turbid. Cooling at 0 °C gave 0.4836 g
(40%) of white needles, m.p.: 75.7–76.2 °C (lit. m.p.: 75–76 °C;[29]
77–78 °C[7]). IR (KBr): ν˜ = 2974 m, 2934 m, 2900 w, 2874 w, 1456
m, 1404 m, 1378 s, 1228 w, 1036 s, 1022 s, 996 m, 890 w, 772 vs,
pyridine the product 11 appeared, its amount increased after
24 h at room temperature but 2 had not totally reacted as the
31P NMR spectra revealed. However, in the presence of trieth-
ylamine, 2 gave 11 in less than 1 h that was isolated in 93%
yield with 94% purity. In the IR (neat) a very strong band
was observed at 588 cm–1 which should be attributed to P=S
stretching that compares well with the reported[12] range of
617–595 cm–1 for RRЈP(S)–SH compounds.
1
738 s, 684 vs, 548 vs, 434 vs cm–1 (see also reference[12]). H NMR
3
(400 MHz, CDCl3, TMS): δ = 1.29 (dt, 12 H, JCH3,P = 20.0,
3JCH3,CH2 = 7.6 Hz, CH3), 2.13–2.31 (m, 8 H, CH2) ppm. 1H NMR
Conclusions
3
(400 MHz, [D6]DMSO, TMS): δ = 1.16 (dt, 12 H, JCH3,P = 20.0,
The following summary of results supports the hypothesis
of our previous publication[1] that the structure of Me2AsSI is
Me2As–S–I (A) and not Me2As(S)–I (B).
3JCH3,CH2 = 7.6 Hz, CH3), 2.09–2.86 (m, 8 H, CH2) ppm. 31P NMR
(162 MHz, CDCl3):
δ
=
52.04 ppm. 31P NMR (162 MHz,
[D6]DMSO): δ = 51.51 ppm (Figure 1).
The hydrolysis of 2 was slower than that of A giving the
anhydride 3 and some acid 4. In the presence of pyridine, 3
was hydrolyzed too. While diphenylamine and aniline were
inert towards A, they reacted with 2, the former slowly and
the latter quickly. Although pyridine and DMAP reacted in a
complicated way with A, pyridine and 2 did not react, while
with DMAP the compound 8 precipitated out at once. Tri-
phenylphosphine desulfurized A quickly and 2 very slowly,
whereas (PhS)3P reacted very slowly and in a complicated way
with A but it was unreactive towards 2. The arsenic(III) nu-
cleophiles, Ph3As and (PhS)3As, were virtually unreactive
towards both A and 2. Methanol and phenol did not react with
A, whereas they reacted with 2 both in the absence and in the
presence of a base. Finally, thiophenol was inert towards A
and 2 but it reacted with 2 in the presence of a base.
Preparation of Diethylthiophosphinyl Iodide (2): Equimolar quanti-
ties of 1 and iodine in an oven-dried centrifuge tube were stirred at
83 °C (oil bath temperature) for 2 h. TLC of the dark red oil showed
the product 2 and occasionally traces of the starting 1. IR (neat): ν˜ =
2972 m, 2932 m, 2874 m-w, 1452 s, 1396 m, 1378 m, 1236 w,
1048 m, 1024 m, 1010 m, 930 w, 772 vs, 750 s, 716 s, 692 m, 580 vs
cm–1 (Figure 2) (see also reference[12]). 1H NMR (400 MHz CDCl3,
TMS): δ = 1.33 (dt, 12 H, JCH3,P = 24.4, JCH3,CH2 = 7.2 Hz, CH3),
2.34–2.45 (m, 4 H, CHAHB), 2.50–2.62 (m, 4 H, CHAHB) ppm (Fig-
ure 2). 31P NMR (162 MHz, CDCl3): δ = 62.78 ppm (Figure 2).
3
3
Stability of Et2P(S)–I (2): The hydrolytic stability of 2 was studied
by reacting it with excess water in acetone. In other experiments pyr-
idine was added to neutralize both the produced acid 4 and HI. The
produced salt 5 was identified mainly by 31P NMR spectroscopy.
Reaction of Et2P(S)–I (2) with Various Nucleophiles: To a solution
of the iodide 2 (approx. 0.2 mmol) in CDCl3 (approx. 0.6 mL) the
nucleophile was added, stirred at room temperature for 5 min, and
transferred to an NMR tube. The reaction was followed by TLC (Et2O/
Experimental Section
1
petroleum ether 1:3), H and 31P NMR spectroscopy.
Materials and Methods: The sources and purity of the nucleophiles
[amines, phosphorus(III), arsenic(III), oxygen, and sulfur(II) com-
pounds] are described in a previous publication[28]. The other materials
and methods have been described.[1] All solvents were dried with 4 Å
molecular sieves and all experiments with moisture-sensitive com-
pounds were run in a dry argon atmosphere.
Reaction of Et2P(S)–I (2) with Diphenylamine: To a solution of 2
(0.4 mmol) in CDCl3 (approx. 0.6 mL) was added diphenylamine
(0.38 mmol) and dry triethylamine (0.6 mmol) and the light orange
solution stirred at room temperature for 20 h. TLC showed a blue spot
(Rf 0.55) (due to Ph2NH) and traces of 3 and the 31P NMR spectrum
Preparation of 1,1,2,2-Tetraethyldiphosphane 1,2-Disulfide (1): showed 2 and small singlets at δ = 107.46 ppm (due to 6) and 108.94
Compound 1 was prepared by a modification of the literature (due to 3). Addition of 25% 4-dimethylaminopyridine (DMAP) gave
method.[29] Magnesium turnings (0.7915 g, 32.57 mmol) placed in a
25 mL two-naked round-bottomed flask connected to a reflux con-
denser and a dropping funnel in an argon atmosphere. A few crystals
a brown solution that after 2 d the 31P NMR spectrum showed dimin-
ished 2 and two main singlets at 106.79 and 108.87 ppm due to 6 and
3, respectively. Addition of 25% more DMAP, after 3 d gave a dark
(approx. 0.05 mg) of iodine and dry diethyl ether (6 mL) were added brown solution showed by 31P NMR very little 2, a strong singlet at δ
giving a clear pale orange supernatant. After 10 min, ethyl bromide = 106.55 ppm due to 6 and a medium one at δ = 108.82 ppm that was
(2.43 mL, 3.549 g, 32.57 mmol) was added dropwise at a rate suf-
ficient to maintain reflux (approx. 20 min addition period). When the
assigned to 3 and probably 8. Evaporation of CDCl3 and addition of
ether (5 mL) gave a brown solid and a yellowish supernatant that was
Z. Anorg. Allg. Chem. 0000, 0–0
www.zaac.wiley-vch.de
5
© 0000 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
filtered (Pasteur pipette plugged with cotton and silica gel). Evapora-
NMR spectroscopy. Evaporation of the CDCl3 and addition of ether
tion and drying in vacuo gave a light orange semi-solid (96 mg) con-
taining 94% 6, 3% 3, and 3% Ph2NH (by H NMR), soluble in Et2O
(6 mL) precipitated Et3N·HI (84 mg, expected 91 mg) and the superna-
tant filtered (Pasteur pipette plugged with cotton and celite). Evapora-
1
and insoluble in petroleum ether. TLC gave a brown spot for 6 and tion and drying the filtrate gave 10 as a nearly colorless oil (83 mg,
Ph2NH (Rf 0.55) and a yellow spot for 3. IR (neat): ν˜ = 3042 m, 3020
m, 2974 m, 2936 m, 2910 w, 2878 w, 1592 vs, 1510 vs, 1494 vs, 1458
97%), soluble in Et2O and CHCl3. The product contained 2% of 3
and 4% phenol (by H NMR). IR (neat): ν˜ = 2975 m, 2937 m, 2879
1
s, 1418 m, 1380 w, 1244 m, 1174 w, 1156 w, 1084 w, 1028 w, 994 w, m, 1591 s, 1490 vs, 1473 m, 1458 m, 1405 m, 1261 w, 1205 vs, 1162
908 s, 878 m, 786 m, 746 vs, 690 s, 584 w cm–1. 1H NMR (400 MHz, m, 1070 m, 1034 s, 1023 s, 910 vs, 891 vs, 784 vs, 764 vs, 732 s, 691
3
3
CDCl3, TMS): δ = 1.24 (dt, 6 H, JCH3,P = 21.6, JCH3,CH2 = 7.2 Hz,
vs cm–1. 1H NMR (400 MHz, CDCl3, TMS): δ = 1.28 (dt, 6 H,
3
4
3
CH3), 2.15–2.25 (m, 4 H, CH2), 6.91 (td, 2 H, Jmp = 7.6, Jop
=
3JCH3,P = 20.8, JCH3,CH2 = 7.2 Hz, CH3), 2.09–2.85 (m, 4 H, CH2),
3
4
0.8 Hz, p-C6H5), 7.09 (dd, 4 H, Jom = 8.4, Jop = 0.8 Hz, o-C6H5), 7.15–7.18 (m, 3 H, p-C6H5 and o-C6H5), 7.29–7.35 (apparent triplet, 2
7.25 (apparent triplet, 4 H, Jmp = 7.6 Hz, m-C6H5). 31P NMR H, m-C6H5). 31P NMR (162 MHz, CDCl3): δ = 109.59 ppm (Figure 1).
3
(162 MHz, CDCl3): δ = 109.09 ppm (Figure 1).
Reaction of Et2P(S)–I (2) with PhSH-Et3N: Reacting 2, thiophenol,
Reaction of Et2P(S)–I (2) with Aniline: To a solution of 2 (0.4 mmol)
in dry dichloromethane (0.5 mL), aniline (0.8 mmol) was added and
the reaction was stirred at room temperature for 1 h. Centrifugation
and washing with diethyl ether (2ϫ2 mL) gave PhNH2·HI (75 mg,
expected 88 mg) and a supernatant that was evaporated and dried in
vacuo. The impure product 7 (86 mg) contained by 1H NMR 4% anil-
ine and traces of 3 by 31P NMR spectroscopy. The solid was dissolved
in diethyl ether (5 mL), filtered (Pasteur pipette plugged with cotton)
and the filtrate concentrated to approx. 1 mL to give the product 7
(47 mg, 53%), soluble in CH2Cl2, CHCl3, Me2C=O and Et2O. M.p.
89–90 °C. C10H16NPS (Mr 213.27): calcd. C 56.31, H 7.56, N 6.57, S
15.03%; found C 55.98, H 7.58, N 6.50, S 16.40%. IR (KBr): ν˜ =
3234 vs, 3046 w, 3018 w, 2976 m, 2936 w, 2917 m, 2878 w, 1603 s,
1498 s, 1488 s, 1449 m, 1411 m, 1401 m, 1304 m, 1293 s, 1239 m,
1182 w, 1156 w, 1084 m, 1039 m, 1029 s, 923 vs, 895 w, 796 s, 763
and dry triethylamine (1:1:1 molar ratio) in CDCl3, the reaction was
over in less than 1 h. Working up as in the case of phenol the product
11 was obtained as an oil (86 mg, 93%) with 94% purity containing
2% PhSH and 5% 3, and it was soluble in Et2O and CHCl3. IR (neat):
ν˜ = 3056 w, 2972 s, 2934 m, 2904 w, 2876 m, 1576 w, 1472 s, 1454
s, 1438 s, 1404 m, 1376 m, 1068 m, 1024 s, 1010 m, 1000 m, 914 m,
772 vs, 748 vs, 714 s, 698 s, 690 vs, 588 vs cm–1. 1H NMR (400 MHz,
CDCl3, TMS): δ = 1.30 (dt, 6 H, JCH3,P = 21.2, JCH3,CH2 = 7.6 Hz,
CH3), 1.92- 2.10 (m, 4 H, CH2), 7.35–7.45 (m, 3 H, p-C6H5 and
o-C6H5), 7.52–7.56 (m, 2 H, m-C6H5). 31P NMR (162 MHz, CDCl3):
δ = 82.27 ppm (Figure 1).
3
3
Keywords: Arsenic; Diethylthiophospinyl iodide; Diethyl-
phosphinothioic iodide; Main group elements; Phosphorus;
vs, 739 vs, 713 m, 696 s, 684 vs, 617 m, 575 s, 508 m, 434 m cm–1. Tetraethyldiphosphine disulfide
3
1H NMR (400 MHz, CDCl3, TMS): δ = 1.23 (dt, 6 H, JCH3,P = 20,
3JCH3,CH2 = 7.6 Hz, CH3), 1.98–2.10 (m, 2 H, CHAHB), 2.12–2.25 (m,
3
2 H, CHAHB), 4.40 (s broad, 1 H, NH), 6.97 (d, 2 H, Jom = 8.4 Hz,
o-C6H5), 7.00 (t, 1 H, 3Jmp = 7.4 Hz, p-C6H5), 7.26 (apparent triplet, 2
H, m-C6H5). 31P NMR (162 MHz, CDCl3): δ = 74.12 ppm (Figure 1).
References
[1] P. V. Ioannou, D. G. Vachliotis, A. Chrissanthopoulos, Z. Anorg.
Allg. Chem. 2015, 641, 1340–1346.
Reaction of Et2P(S)–I (2) with 4-Dimethylaminopyridine (DMAP):
In an oven-dried centrifuge tube containing 1,1,2,2-tetraethyldiphos-
phane 1,2-disulfide (1) (160 mg, 0.66 mmol) and flushed with argon,
iodine (98 mg, 0.386 mmol) was added, immersed into an oil bath
(83 °C), and the light orange oil was stirred for 2 h. After cooling to
room temperature, 4-dimethylaminopyridine (DMAP) (59 mg,
0.486 mmol) was added followed by dry dichloromethane (1 mL) and
the yellow-orange solution stirred at room temperature for 3 h. Ad-
dition of dry ether (7 mL) and stirring for 30 min precipitated the prod-
uct, which was centrifuged and washed with diethyl ether (2ϫ3 mL).
The product 8 (174 mg, 97% on DMAP, 98% purity) was a white
solid insoluble in Et2O, moderately soluble in Me2C=O and soluble in
CH2Cl2 and CHCl3. Thermal behavior: at 150 °C melts to an opales-
cent light orange oil that clears at approx. 168 °C. C11H20IN2PS (Mr
370.22): calcd. C 35.68, H 5.45, N 7.57, S 8.66%; found C 35.58, H
4.92, N 7.98, S 8.97%. IR (KBr): ν˜ = 3066 w, 3031 w, 2972 w, 2928
w, 2909 w, 2882 w, 1650 vs, 1635 vs, 1576 s, 1559 s, 1399 m, 1312
m, 1273 w, 1220 m, 1051 vs, 1019 s, 838 m, 806 w, 786 s, 739 m,
718 w, 710 w, 668 w, 611 s, 511 w cm–1. 1H NMR (400 MHz, CDCl3,
[2] K. A. Pollart, H. J. Harwood, J. Org. Chem. 1962, 27, 4444–4447.
[3] H. J. Harwood, K. A. Pollart, J. Org. Chem. 1963, 28, 3430–3433.
[4] W. Kuchen, H. Buchwald, Angew. Chem. 1959, 71, 162.
[5] N. G. Feshchenco, N. D. Gomelya, A. G. Matyusha, J. Gen.
Chem. USSR 1982, 52, 202–203.
[6] D. E. C. Corbridge, Phosphorus: An Outline of its Chemistry, Bio-
chemistry and Technology, Elsevier, Amsterdam, 1978, p. 436.
[7] H. Niebergall, B. Langenfeld, Chem. Ber. 1962, 95, 64–76.
[8] H. Reinhardt, D. Bianchi, D. Mölle, Chem. Ber. 1957, 90, 1656–
1660.
[9] N. Muller, P. C. Lauterbur, J. Coldenson, J. Am. Chem. Soc. 1956,
78, 3557–3561.
[10] J. R. Van Wazer, C. F. Callis, J. N. Shoolery, R. C. Jones, J. Am.
Chem. Soc. 1956, 78, 5715–5726.
[11] S. N. Duta, M. M. Woolfson, Acta Crystallogr. 1961, 14, 178–
185.
[12] R. A. Chittenden, L. C. Thomas, Spectrochim. Acta 1964, 20,
1679–1696.
[13] K. Moedritzer, L. Maier, L. C. Groenweghe, J. Chem. Eng. Data
1962, 7, 307–310.
[14] a) P. C. Crofts, Phosphinic Acid and Derivatives in Organic Phos-
phorus Compounds (Eds.: G. M. Kosolapoff, L. Maier), Wiley-
Interscience, New York, 1973, vol. 6, ch.14, p. 38; b) P. C. Crofts,
Phosphinic Acid and Derivatives in Organic Phosphorus Com-
pounds (Eds.: G. M. Kosolapoff, L. Maier), Wiley-Interscience,
New York, 1973, vol. 6, ch.14, p. 41–42.
3
3
TMS): δ = 1.12 (dt, 6 H, JCH3,P = 22.4, JCH3,CH2 = 7.2 Hz, CH3),
2.55–2.66 (m, 2 H, CHAHB), 3.01–3.14 (m, 2 H, CHAHB), 3.39 (s, 6
H, NCH3), 7.05 (d, 2 H, 3J = 7.6 Hz, Me2NCCH), 9.31 [t, 3J = 7.6 Hz,
2 H, 3JH,P = 7.2 Hz, Et2P(S)NCH] ppm. 31P NMR (162 MHz, CDCl3):
δ = 108.79 ppm (Figure 1).
[15] M. Fild, R. Schmutzler, Phosphonyl-(Thiono-, Seleno-) and Phos-
phinyl-(Thiono-, Seleno-) Halides and Pseudohalides in Organic
Phosphorus Compounds (Eds.: G. M. Kosolapoff, L. Maier),
Wiley-Interscience, New York, 1972, vol. 4, ch. 9, p. 180.
Reaction of Et2P(S)–I (2) with PhOH-Et3N: The reaction of equi-
molar quantities of 2, crystalline phenol, and dry triethylamine
(0.4 mmol each) in CDCl3 was over in less than 10 min by TLC and
Z. Anorg. Allg. Chem. 0000, 0–0
www.zaac.wiley-vch.de
6
© 0000 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
[16] W. Kuchen, K. Strolenberg, H. Buchwald, Chem. Ber. 1962, 95,
[24] S. V. Serves, Y. C. Charalambidis, D. N. Sotiropoulos, P. V. Ioan-
nou, Phosphorus Sulfur Silicon 1995, 105, 109–116.
[25] M. N. Haikou, G. M. Tsivgoulis, P. V. Ioannou, Z. Anorg. Allg.
Chem. 2004, 630, 1084–1089.
[26] G. Schrader, W. Lorenz, R. Coelln, H. Schloer, Phosphinic Esters,
U. S. 3,232,830, 1966 to Bayer AG; G. Schrader, W. Lorenz, R.
Coelln, H. Schloer, Chem. Abstr. 1966, 64, 15923.
[27] L. Maier, Helv. Chim. Acta 1964, 47, 1448–1459.
[28] P. V. Ioannou, D. G. Vachliotis, Z. Anorg. Allg. Chem. 2015, 641,
710–714.
1703–1710.
[17] G. Höfle, W. Steglish, H. Vorbrüggen, Angew. Chem. 1978, 90,
602–615; Angew. Chem. Int. Ed. Engl. 1978, 17, 569–583.
[18] E. F. V. Scriven, Chem. Soc. Rev. 1983, 12, 129–161.
[19] E. Guibe-Jampel, M. Wakselman, D. Raulais, J. Chem. Soc.,
Chem. Commun. 1980, 993–994.
[20] J. J. Weigand, N. Burford, D. Mahnke, A. Decken, Inorg. Chem.
2007, 46, 7689–7691.
[21] P. V. Ioannou, D. G. Vachliotis, T. D. Sideris, Z. Anorg. Allg.
Chem. 2007, 633, 2077–2084.
[22] G. M. Tsivgoulis, P. V. Ioannou, Main Group Chem. 2011, 10,
51–62.
[29] J. P. Christen, L. M. van der Linde, F. N. Hoose, Rec. Trav. Chim.
Pay. Bas 1959, 78, 161–172.
[23] T. D. Sideris, G. M. Tsivgoulis, D. G. Vachliotis, P. V. Ioannou,
Main Group Chem. 2009, 8, 163–176.
Received: May 4, 2017
Published Online: ᭿
Z. Anorg. Allg. Chem. 0000, 0–0
www.zaac.wiley-vch.de
7
© 0000 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
G. M. Tsivgoulis,* G. G. Kordopati, D. G. Vachliotis,
P. V. Ioannou ..................................................................... 1–8
The Reaction of Diethylthiophosphinyl Iodide, Et2P(S)I, with
Nucleophiles
Z. Anorg. Allg. Chem. 0000, 0–0
www.zaac.wiley-vch.de
8
© 0000 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim