Benign and high-yielding, large-scale synthesis of diphenylphosphinodithioic acid and related compounds

Article abstract of DOI:10.1021/op300147f

Diphenylphosphinodithioic acid (6b) and its triethyl ammonium salt (6a) were prepared by two new synthetic pathways, each employing cheap and readily available starting materials. These facile one-pot reactions were conducted on a kilogram scale and produ

Full text of DOI:10.1021/op300147f

Article
pubs.acs.org/OPRD
Benign and High-Yielding, Large-Scale Synthesis of
Diphenylphosphinodithioic Acid and Related Compounds
Jochen Wagner,† Michael Ciesielski,∥ Christoph A. Fleckenstein,*,‡ Hartmut Denecke,‡ Florian Garlichs,§
Andreas Ball,§ and Manfred Doering*,∥
†Advanced Materials and Systems Research, GMV/P, ‡Process Research and Chemical Engineering, GCN/F, and §Process Research
and Chemical Engineering, GCN/I, BASF SE, 67056 Ludwigshafen, Germany
∥Fraunhofer LBF, Schloßgartenstraße 6, 64289 Darmstadt, Germany
S
* Supporting Information
ABSTRACT: Diphenylphosphinodithioic acid (6b) and its triethyl ammonium salt (6a) were prepared by two new synthetic
pathways, each employing cheap and readily available starting materials. These facile one-pot reactions were conducted on a
kilogram scale and produced the desired products in high yield and quality, thereby surpassing all previously known routes. The
synthesis of triethyl ammonium salts of 6H-dibenzo[c,e][1,2]oxaphosphinine-6-thiolate 6-sulfide (3a) was also further improved.
INTRODUCTION
RESULTS AND DISCUSSION
Synthesis of Triethyl Ammonium 6H-Dibenzo[c,e]-
■
■
Diphenylphosphinodithioic acid (DPPA, 6b) and its derivatives
are utilized for a plethora of different applications: DPPA is
used as a chelating agent that can bind to different metals such
as copper(I), actinides(III), and metals of the platinum
group.1−5 The difference in the coordination chemistry of
DPPA and its derivatives between lanthanides(III) and
actinides(III) enables these compounds to extract e.g.
Cm(III) and Am(III) from lanthanides(III) in the advanced
processing of spent nuclear fuel.1−3 Metals of the platinum
group can also be solvent extracted through coordination with
6b.4 Multimetal complexes of diphenylphosphinodithioic acid
with copper(I) find application as antioxidation additives and in
biological systems.5 Furthermore, 6b can be used as a corrosion
inhibitor in photochemically cross-linkable coatings,6,7 as a
curing accelerator for epoxy resins,8 and as a cross-linking agent
for rubber polymers.9 Recently, diphenylphosphinodithioic acid
and some derivatives have been reported as flame retardants for
polystyrene foams.10 The mono- and di-sulfur-bridged
derivatives 8a and 8b also find application as UV-stabilizers
in polyolefins.11 Despite the many different applications, no
truly convenient synthesis has been published to the present
day. The majority of preparations utilize a previously reported
procedure.12 While this method represents a reliable and
feasible pathway towards 6b, it suffers from several drawbacks,
including the use of a huge excess of AlCl3 and moderate yields.
Yet for an industrial application, a cheap and easily scalable
synthesis is crucial. In this work we present two new and
convenient routes for the synthesis of triethyl ammonium
diphenylphosphinodithioate (6a) and diphenylphosphinodi-
thioic acid (DPPA, 6b), respectively. In addition, we present
an improved synthesis of the related compounds 3a (triethyl
ammonium 6H-dibenzo[c,e][1,2]oxaphosphinine-6-thiolate 6-
sulfide) and 3b (6-mercapto-6H-dibenzo[c,e][1,2]-
oxaphosphinine 6-sulfide) which recently became of interest
in the context of phosphinothioyl applications.13
[1,2]oxaphosphinine-6-thiolate 6-Sulfide (3a). Recently,
the synthesis of 3a via sulfidation of 6-chloro-6H-dibenzo[c,e]-
[1,2]oxaphosphinine (1) to form 6-chloro-6H-dibenzo[c,e]-
[1,2]oxaphosphinine 6-sulfide (2) and its subsequent con-
version with elemental sulfur and triethyl amine (route A,
Scheme 1) has been described.14 However, it has now been
Scheme 1. Synthesis of 3a: two-step and one-pot reaction
pathway
found that the initial sulfidation of 1 does not necessarily need
to be carried out in a separate reaction step. Instead, feasibility
of a one-pot reaction was demonstrated which is advantageous
with respect to synthetic effort and reaction time (route B,
Scheme 1). Therefore, 3 equiv of elemental sulfur and 3 equiv
of triethyl amine were placed together with 1 and toluene in a
round-bottomed flask under argon atmosphere and heated for 8
h at solvent reflux temperature. The progress of the reaction
1
was monitored via H- and 31P NMR spectroscopy. NMR
Received: June 6, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/op300147f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 1. Progress of the conversion of 1 (0 h) over the intermediate 2 to 3a after 2, 4, 6, and 8 h, monitored by 31P NMR spectroscopy. 31P NMR
spectra (101 MHz) were measured in CDCl3.
samples were removed from the reaction mixture after 2, 4, 6,
and 8 h. In the course of the reaction, 1 (31P NMR: δ (CDCl3)
= 134.2) is at first fully converted to 2 (31P NMR: δ (CDCl3) =
74.1) before the reaction proceeds to the formation of 3a
(Figure 1). In the process several intermediates are formed
which are also transformed to 2 (31P NMR: δ (CDCl3) = 15.5,
61.9, 62.3). The starting material is consumed within 4 h of
solvent reflux, but complete conversion to 2 is only achieved
after about 5 h. Afterwards the formation of 3a begins, which is
completed within 8 h (31P NMR: δ (CDCl3) = 99.1).
Scheme 3. Proposed reaction scheme for the conversion of 1
to 2 in the presence of sulfur and triethyl amine
To date no detailed investigations of the mechanism of the
above reaction have been carried out. However, it is likely that
hydrogen sulfide is produced in situ by the reaction of triethyl
amine and elemental sulfur and subsequently replaces the
chloride in 2 by nucleophilic substitution. The combination of
amines and sulfur as a source for hydrogen sulfide has been
known for a long time.15−17 General conversion of binary and
tertiary amines with elemental sulfur to thioamides under
release of hydrogen sulfide has been described (Scheme 2).16
From the NMR-data provided in Figure 1, the initial
conversion of 1 to 2 is verified. The proposed conversions of 2
to 3b and of triethylamine to the thioamide via the intermediate
triethylammonium hydrogen sulfide were derived from
reactions that are described in the cited literature. While
these are in line with the here reported conditions and
observations, the occurrence has not yet been validated by
experimental proof.
Scheme 2. Formation of thioamides by reaction of sulfur in
the presence of secondary or tertiary amines
Synthesis of Triethyl Ammonium Diphenylphosphi-
nodithioate (6a) and Diphenylphosphinodithioic Acid
(6b). At present, the Friedel−Crafts approach is still the most
widely used method for the preparation of 6b (Scheme 4).12
This reliable synthesis can be conducted on a laboratory scale
with good reproducibility. However, it has several drawbacks
since it requires an 8-fold (!) molar excess of AlCl3 and two
recrystallization steps to afford DPPA in a moderate overall
yield of 54%. In addition to the huge synthetic effort of this
reaction, the large amount of nonbiodegradeable waste
The formation of ammonium hydrogen sulfide salts
considered to be the major sulfuration agents in the thionation
of olefins has also been reported in this context.17 On the basis
of the stoichiometry of the reaction reported here (with enough
triethyl amine to intermediately catch H2S) and the observation
that no potentially evolved gas left the reaction apparatus, the
formation of ammonium hydrogen sulfide can be considered a
respective key step in the conversion of 2. A proposed reaction
sequence is depicted in Scheme 3:
B
dx.doi.org/10.1021/op300147f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
any drawbacks compared to a smaller scale. Therefore, the
reaction may be deemed applicable for further industrial
employment.
Scheme 4. Traditional synthesis of DPPA
While the triethyl ammonium salt can be obtained in high
yield and is often adequate for subsequent chemistry, the
release and isolation of the free acid 6b goes along with a
decrease in yield. Further on, there is still room for
improvement of atom efficiency for this reaction route.
Therefore, a second new route to 6b was explored. A first
approach was the direct conversion of 5 to 6b by treatment
with hydrogen sulfide (Scheme 6).
produced is a drawback that prevents this reaction from
application on an industrial scale.
In an earlier approach a Grignard reagent and phosphorous
pentasulfide were employed, but the product was achieved in
only 24% yield.18 An obvious approach through the sulfidation
of diphenylphosphine has been published.19 Better yields than
those previously reported are available via this method (91%
crude green oil compared to 80% before recrystallization). This
amount is reduced by recrystallization steps to obtain the pure
white and crystalline acid 6b. However, the applicability of this
reaction is primarily less attractive due to the high price of
diphenylphosphine.
Scheme 6. Direct conversion of 5 to 6b by treatment with
hydrogen sulfide
The lack of a convenient and easily scalable synthesis for 6a
and 6b prompted a renewed effort in this area. Similar to the
synthesis of 3a the triethyl ammonium salt of DPPA (6a) can
be obtained by the reaction of 4 with 3 equiv of elemental
sulfur and 3 equiv of triethyl amine (Scheme 5). This
However, in toluene at atmospheric pressure and room
temperature, no conversion was obtained. Even when
stoichiometric amounts of pyridine were added, no reaction
occurred. High yields were achieved when pyridine was applied
as solvent (Scheme 7).
Scheme 5. Synthesis of compounds 6a and 6b: two-step and
one-pot reaction pathway
Scheme 7. Reaction scheme for the conversion of 5 to 6d by
treatment with hydrogen sulfide in pyridine; 6b can be
obtained from 6d in a separate reaction step
The conversion of 5 to 6d under these conditions is fast and
nearly quantitative. The reaction mixture was then concentrated
by solvent evaporation under reduced pressure and diluted with
water and aqueous hydrochloric acid solution, and the product
6b was extracted with MTBE in high yield (95%) and good
purity (98%, determined by 31P NMR). While this route
represents a scalable and reproducible process, the direct
conversion of 5 to 6b was not feasible. The large amount of
hydrochloric acid needed for the neutralization of 6d and
remaining excess pyridine is a drawback for this synthetic route.
Toxicity and limitation in transportation of hydrogen sulfide
reflect further disadvantages of the applicability of this process.
Since the synthesis of 6b in just one reaction step failed, the
next approach was the use of sodium hydrogen sulfide as
reagent for the conversion of 5 to 6b.
Remarkably, simple treatment of 5 with technical grade
sodium hydrogen sulfide, utilizing water as a reaction solvent
affords sodium dithiodiphenylphosphinate (6c) in nearly
quantitative yield (Scheme 8). Subsequent acidification with
hydrochloric acid readily affords 6b in very high yield (>98%)
and with excellent purity. The acidification starting from the
sodium salt 6c proceeds more smoothly compared to the same
reaction starting from the triethyl ammonium salt 6a. This
synthetic access to 6b is superior to all other routes reported.
represents a new high-yielding route to the ammonia salt, 6a,
and to the free dithiophosphinic acid, 6b, respectively, after
further conversion by treatment with hydrochloric acid.
Apart from small changes in the reaction times, reaction
conditions were the same as for the synthesis of 3a. Again, it is
possible to carry out the reaction in one or two steps, whereas
the one-pot reaction is significantly advantageous regarding
effort and reaction time. Yields are between 62% and 69% for
the pure white crystalline 6a. Yet, the crude salt can be obtained
as a brownish crystalline solid in a yield of 96% and a purity of
>99% (determined by 31P NMR spectroscopy), which is
sufficient for most of the following chemistry. From the crude
6a the free acid 6b can be released with hydrochloric acid and
isolated (crude 95%, 79% after recrystallization) or used for
further reactions without isolation. The reaction was conducted
on different scales, the largest of them in a 10-L reactor, starting
from 3 mol of 5. In this case, 5 was obtained by direct
conversion of 4 with sulfur in a solvent-free reaction (see
Experimental Section). As the yield of 5 by this pathway was
100%, both steps can also be easily combined to a one-pot
reaction. To this point, the scale-up was accomplished without
C
dx.doi.org/10.1021/op300147f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
methane and sodium ethanolate to form 7.23 A feasible lab-scale
synthesis has been developed in which 6c is converted to 7 in
high yield (92%) by treatment with iodine/potassium iodide.22
While this synthesis works well on a laboratory scale, an
industrial application is not feasible because of the high price of
iodine. A feasible and scalable synthesis of diphenylphosphi-
nothioyl disulfide (7) has now been developed (Scheme 9).
Scheme 8. Synthesis of 6b via treatment of 5 with sodium
hydrogen sulfide
Scheme 9. Preparation of 7 starting from the sodium salt 6c
or from the free acid 6b
Scale-up was successfully carried out in a 10-L reactor, the
largest batch starting from 6.5 mol of 5. Since no decreases in
yield or purity were observed, the reaction seems to be well-
suited for industrial employment. The evolved H2S is easily
absorbed by sodium hydroxide to form sodium hydrogen
sulfide which can be reused in this process. Thus, the necessity
for disposal of H2S can be avoided which further enhances the
efficiency of the process.
Synthesis of Bis(diphenylphosphinothioyl) Disulfide.
Good availability of dithiophosphinic acids 3 and 6 as starting
materials enables a broad variety of subsequent reactions.
Recently, the Michael-like addition of 3a to different acrylates as
well as benzoquinone was reported.14 Also the preparation of
bis(diphenylphosphinothioyl) disulfide has been investigated,
and different synthetic approaches have been reported.20−23
However, none of the known procedures is suitable for scale-
up or industrial application as noted in Table 1. In an earlier
approach, the expensive trichloroacetonitrile was applied as
oxidation agent, leading to an overall yield of only 49%.21 Even
lower yields (40%) were achieved in a more cumbersome
synthesis in which 6b was allowed to react with tetrachloro-
7 is accessible from free acid 6b via simple oxidation with
H2O2 in methyl-tert-butyl ether (MTBE) in nearly quantitative
yield (99.7%, purity 99% determined by 31P NMR). It is
noteworthy that no elaborate workup is required. The product
7 precipitates from the reaction mixture and is obtained in
excellent purity after collection by filtration and washing with
MTBE and water. The acid 6b can also be released in situ from
the sodium salt 6c by treatment with hydrochloric acid
solution. In this case, the purity and yield of the product are
slightly lower (yield 98.8%, purity >98% determined by 31P
NMR). Both routes are easy to scale up. Therefore, no
obstacles are left for an industrial application of this process.
CONCLUSION
■
A novel route to 3a and 6a (and 3b and 6b, respectively) has
been established. With the employment of very inexpensive,
readily available elemental sulfur and triethylamine as starting
materials, this represents a high-yielding, one-pot synthesis
which is easy to scale up and is therefore suitable for industrial
application. This method is superior to the routes that have
been previously reported. A second development is even more
remarkable. The conversion of 5 to 6c by treatment with
inexpensive technical grade sodium hydrogen sulfide in water as
a solvent and subsequent acidification with hydrochloric acid
proceeds smoothly and rapidly to produce diphenylphosphi-
nothioic acid (6b) in excellent yield and purity. Scale-up was
successfully carried out (3 mol scale). On an industrial scale,
the evolved H2S is easily recyclable in the form of sodium
hydrogen sulfide, which leads to sodium chloride as the sole
byproduct of this reaction and results in a very high atom
efficiency and low overall cost of the process. In a subsequent
reaction the disulfide 7 was synthesized by simple oxidation of
6b with hydrogen peroxide, using MTBE as a solvent. Since 7
precipitates from the reaction mixture in excellent yield and
purity, workup is quite simple. Again this reaction is superior to
known procedures.
Table 1. Comparison of different synthetic pathways to 6b
advantages
drawbacks
Grignard Reaction
low yield (24%)
expensive, high effort
Electrophilic Substitution (Friedel−Crafts Reaction)
−
large-scale published
moderate yield (54%)
8-fold molar excess of AlCl3
poor atom economy, large
amount of waste, high effort
Sulfidation of Ph2H
high price of Ph2PH
one-pot reaction
crude yield: 91% (compared to 80% for earlier handling of Ph2PH
approaches)
Treatment of 4 with S8 and NEt3
inexpensive reactants, small effort
additional step necessary for
the release of 6b
high crude yield of 6a (96%), 99% purity
acidification affords 79% pure 6b (95% crude)
Treatment of 5 with H2S
very high yield (95%)
additional step necessary for
the release of 6b
EXPERIMENTAL SECTION
■
no further purification necessary
H2S handling and availability
General Experimental. Unless stated otherwise, all
reactants and solvents were purchased from commercial
sources and used without further purification or drying. 1 was
synthesized according to previously reported procedures.24
Standard glass apparatus was used; reactions were carried out
with magnetic stirring and, if moisture sensitive, in flame-dried
glassware under argon. For reactions that were conducted in a
Treatment of 5 with NaSH
pure 6b is easily accessible in a one-pot
reaction in very high yield (>98%) and
purity
H2S evolution
H2O as solvent, inexpensive reactants
possible recycling of H2S, NaCl as sole
byproduct
D
dx.doi.org/10.1021/op300147f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
reactor, the following reactor components were used: (i) 10-L
reactor, equipped with a three-step cross-beam agitator and
baffle, condenser, and dropping funnel; (ii) thermostat for the
thermoregulation of the reactor jacket; (iii) pump and
controller for the dosage of the feed; (iv) flexible-tube pump
for the dosage of MTBE and HCl solution; (v) alkaline
scrubbing tower (hold-up 4 L, 15% NaOH solution) for flue
gas scrubbing; (vi) H2S-monitoring by an H2S sensor in the
hood.
60 °C, and triethylamine (934 g, 9.25 mol) was added within
20 min. The reaction mixture was heated to reflux and stirred at
reflux temperature for 6 h. The reaction mixture was then
cooled down to rt, and water (5 L) was added while stirring.
The resulting suspension was stirred for 1 h at ambient
temperature. The precipitate was isolated via suction filtration
and subsequently washed with cold water (1 L) and cold
toluene (1 L). The crystals were collected and dried in a
vacuum oven (T = 60 °C) overnight to afford the crude
product as a brownish crystalline solid (1012 g, 96%). The
crude product was divided into two batches of 506 g each and
batchwise purified via recrystallization from hot water (14 mL
water/g) and charcoal (150 mg charcoal/g). The product
crystallized as white crystals. The two batches were combined
to afford 725 g (69%) of triethyl ammonium dithiodiphenyl-
phosphinate.
NMR spectra were obtained at room temperature. Chemical
shifts are reported as δ values relative to the solvent absorption.
All 31P NMR spectra were obtained with proton decoupling. All
13C NMR spectra were obtained with proton decoupling and
1
phosphorus coupling. H NMR spectra were obtained with
phosphorus coupling. Melting points are uncorrected. High-
resolution mass spectrometry (HR-MS) analyses were
performed using electron ionization (EI, 70 eV) or electrospray
ionization (ESI) and time-of-flight (TOF) analyzers.
B. One-Pot Reaction. In a three-necked, round-bottomed
flask equipped with a condenser, thermometer, and argon inlet,
was dissolved 4 (20 g, 90.65 mmol) in 120 mL of dry toluene.
Elemental sulfur (8.72 g, 272 mmol) and triethylamine (27.52
g, 272 mmol) were added, and the reaction mixture was then
heated at solvent reflux temperature for 5.5 h. During this time
a precipitate formed, and the color changed from light yellow to
dark brown. Afterwards, the suspension was cooled to 0 °C in
an ice bath, and triethylammonium chloride was removed by
filtration. From the filtrate, the product crystallized overnight.
The crystals were collected by filtration at reduced pressure and
washed with cold toluene and cold water. Recrystallization (see
above) yielded 19.3 g (62%) of white crystals. Note: instead of
letting the product crystallize from the mother liquor, the same
workup as in A may be chosen, to yield 96% of crude or 69% of
Synthesis of Triethylammonium 6H-Dibenzo[c,e][1,2]-
oxaphosphinine-6-thiolate 6-Sulfide (3a). 1 (65.73 g, 280
mmol) was charged to a three-necked, round-bottomed flask
equipped with condenser, thermometer, magnetic stirrer bar,
and argon inlet and dissolved in anhydrous toluene (300 mL)
at room temperature. Elemental sulfur (26.95 g, 840 mmol)
and triethylamine (85.05 g, 840 mmol) were added, and the
reaction mixture was heated to solvent reflux temperature and
kept there for 8 h. During this time, a dark-brown suspension
developed. Samples for 31P NMR analysis were removed after
2, 4, 6, and 8 h. After 6 h 1 was fully consumed, and the
intermediate 2 was exclusively formed (δ (CDCl3) = 74.5).
After 8 h complete conversion to the target product was
achieved (δ (CDCl3) = 99.2). The reaction mixture was then
cooled to 0 °C in an ice bath and filtered. The precipitate was
washed with cold toluene and cold water and dried at reduced
pressure to yield the target compound as a brownish crystalline
solid (89.37 g, 87%). Mp 128−130 °C; 31P NMR: δ (CDCl3) =
99.2; 1H NMR: δ (CDCl3) = 1.36 (t, J = 7.3 Hz, 9H), 3.22 (m,
6H), 7.11−7.19 (m, 2H), 7.27−7.32 (m, 1H), 7.38−7.47 (m,
2H), 7.72−7.77 (m, 1H), 7.83−7.86 (m, 1H), 8.07 (dd, J =
15.5 Hz, J = 7.0 Hz, 1H), 9.69 (s, 1H); 13C NMR: δ (CDCl3) =
8.5 (s, 3C), 46.0 (s, 3C), 120.7 (d, JPC = 5.3 Hz, 1C), 123.2 (s,
1C), 123.4 (s, 1C), 124.6 (d, JPC = 12.8 Hz, 1C), 124.9 (d, JPC
1
the pure 6a. Mp 109 °C; 31P NMR: δ (CDCl3) = 62.6; H
3
NMR: δ (CDCl3) = 1.29 (t, J = 7.3 Hz, 9H), 3.22 (m, 6H),
7.26−7.33 (m, 6H), 8.12 (dd, J = 13.8 Hz, J = 6.0 Hz, 4H),
10.17 (s, 1H); 13C NMR: δ (CDCl3) = 8.3 (s, 3C), 45.7 (s,
3C), 127.4 (d, JPC = 12.6 Hz, 4C), 129.0 (d, JPC = 3 Hz, 2C),
130.4 (d, JPC = 11.4 Hz, 4C), 143.4 (d, JPC = 80.3 Hz, 2C);
Anal. Calcd for C18H26NPS2: C 61.5, H 7.4, N 4.0, S 18.2, P
8.8; found C 60.8, H 7.5, N 3.9, S 17.9, P 8.5. HR-EI: calcd
(C12H10PS2): m/z = 248.9957, found: m/z = 248.9962.
Synthesis of Diphenylphosphinodithioic Acid (6b).
A. Conversion of 6a with Hydrochloric Acid. In a round-
bottomed flask equipped with a condenser was dissolved the
crude 6a (8.4 g, 23.9 mmol) in ethanol (40 mL) at 80 °C. At
this temperature concentrated aqueous hydrochloric acid
solution (50 mL) was added. A brownish solid precipitated,
and the resulting suspension was stirred for 5 min. Brine was
added (50 mL), and the reaction mixture was cooled to 5 °C in
an ice bath. It was filtered, and the residue was washed with
cold water and dried at reduced pressure to yield 5.7 g of the
crude acid 6b (95%) as a brownish solid. The purity of the
crude product was already at 99% as determined by 31P NMR
spectroscopy. Recrystallization from isopropanol yielded 4.7 g
(79%) of the pure compound as white crystals.
B. Via Conversion of 5 with Hydrogen Sulfide. A 750-mL
glass reactor equipped with stirrer, thermometer, gas-inlet tube,
gas-outlet, and bubble counter was coupled to a weighted H2S-
bottle and an argon purging line via a security bottle. The gas-
outlet was coupled to a wash tower filled with 10% aqueous
NaOH solution. In the glass reactor, 5 (75 g, 0.3 mol) was
dissolved in pyridine (300 mL) at room temperature, forming a
clear yellowish solution in a slightly exothermic reaction. The
reaction mixture was heated to 30 °C, and H2S (20.7 g, 0.6
= 1.4 Hz, 1C), 127.9 (d, JPC = 14.6 Hz, 1C), 128.3 (d, JPC
13.8 Hz, 1C), 129.3 (s, 1C), 130.1 (d, JPC = 2.7 Hz, 1C), 132.1
(d, JPC = 5.5 Hz, 138.7 (d, JPC = 103.4 Hz, 1C), 150.9 (d, JPC
=
=
10.3 Hz, 1C), HR-EI calcd (C12H9OS2P) m/z = 263.9832,
found m/z = 263.9836.
Synthesis of Triethyl Ammonium Diphenylphosphi-
nodithioate (6a). A. Two-Step Reaction via 5. Under an inert
atmosphere, elemental sulfur (90.1 g, 0.35 mol) was added to 4
(620.8 g, 2.81 mol) and heated to 130 °C. At this temperature,
the reaction mixture was stirred for 6.5 h and then allowed to
cool to room temperature. 5 was obtained as clear yellowish oil
(713 g, 100%). The product was used for the next step without
1
further purification. 31P NMR: δ (CDCl3) = 81.6; H NMR: δ
(CDCl3) = 7.39−7.52 (m, 6 H), 7.93−8.03 (dd, J = 15.6 Hz, J
= 7.9 Hz, 4H); 13C NMR: δ (CDCl3) = 128.6 (d, JPC = 14.6
Hz, 4C), 130.9 (d, JPC = 12.6 Hz, 4C), 132.6 (d, JPC = 3.4 Hz,
2C), 135.3 (d, JPC = 96.4 Hz, 2C); HR-EI: calcd (C12H10ClPS)
m/z = 251.9929, found m/z = 251.9941.
In the next step, toluene (2.4 L), elemental sulfur (192.3 g,
6.1 mol), and 5 (757.5 g, 3.0 mol) were placed in a 10-L reactor
at room temperature. The reaction mixture was warmed up to
E
dx.doi.org/10.1021/op300147f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
mol) was bubbled into the solution (beneath the surface)
within 45 min. During this process, the temperature rose to 40
°C until 1 equiv of H2S had been added, and then the
temperature fell. The apparatus was purged with argon, and the
reaction mixture was heated to 40 °C and stirred for 1 h at this
temperature. Then two-thirds of the pyridine (200 mL) was
removed by evaporation at reduced pressure, and MTBE (270
mL) was added to the viscous residue. An aqueous solution
(37%) of hydrochloric acid (150 g, 1.52 mol) was poured into
the suspension to dissolve the solid ingredients, thereby
forming an emulsion. This emulsion was stirred for 1 h, after
which the phases were separated. Evaporation of the organic
phase yielded 6b as a brittle yellow solid (70.6 g, 95%).
ASSOCIATED CONTENT
■
S
* Supporting Information
Copies of 1H, 13C, and 31P NMR spectra of all the compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Manfred.doering@lbf.fraunhofer.de; christoph.a.fleckenstein@
basf.com
■
Notes
The authors declare no competing financial interest.
REFERENCES
■
C. Via Conversion of 5 with Sodium Hydrogen Sulfide. In a
10-L reactor, a suspension of sodium hydrogen sulfide (962 g,
13.0 mol) in 2 L of water was heated to 95 °C. When this
temperature was reached, the salt was completely dissolved and
chlorodiphenylphosphine sulfide (1642 g, 6.5 mol) was added
within 1 h via a ProMinent-Pump. After a retardation period,
H2S-evolution started and continued until the addition was
completed. Afterwards, the reaction mixture was stirred for 1
additional hour at 95 °C. With the addition of ice (500 g) the
solution was then cooled to 50 °C. At this temperature, MTBE
(2.5 L) was poured into the reactor via a flexible-tube pump
over a period of 15 min. The reaction mixture was cooled to 25
°C, and 15% hydrochloric acid solution (1581.7 g, 6.5 mol) was
added via flexible-tube pump. After another 30 min of stirring at
room temperature, the phases were separated, and the organic
phase (MTBE) was cooled to 2 °C in an ice bath while stirring.
In the process, the diphenylphosphinedithioic acid crystallized.
After 1 additional hour of stirring, the crystals were collected via
filtration at reduced pressure and afterwards dried in an oven at
reduced pressure to afford 1073 g (93%) of a white crystalline
(1) Klaehn, J. R.; Peterman, D. R.; Harrup, M. K.; Tillotson, R. D.;
Luther, T. A.; Law, J. D.; Daniels, L. M. Inorg. Chim. Acta 2008, 361,
2522−2532.
(2) Xu, Q.; Wu, J.; Chang, Y.; Zhang, L.; Yang, Y. Radiochim. Acta
2008, 96 (12), 771−779.
(3) Sherif, A. Schr. Forschungszent. Juelich, Reihe Energietech./Energy
Technol. 2004, 29 (i-viii), 1−198.
(4) Kabanova, L. L.; Usova, S. V. Zh. Anal. Khim. 1974, 29 (11),
2248−2249.
(5) Assavathorn, N.; Davies, R. P.; White, A. J. P. Polyhedron 2008,
27, 992−998.
(6) Fernandez Gonzalez, M. ; Hickl, M.; Goethlich, A.;
Vandermeulen, G.; Biehler, M.; Bergmann, H.; Roschmann, K.;
Menzel, K.; PCT Int. Appl., WO/2008/017647, A1/2008/0214, 2008.
(7) Chromy, L.; Marek, K. DEFAZET − Dtsch. Farben−Z. 1978, 32
(11), 434−437.
(8) Minabe, Y.; Kawakatsu, Y.; Takahashi, H.; Yamaguchi, T. Jpn.
Kokai Tokkyo Koho, JP 51031798, A 19760318, 1976.
(9) Lenke, G. U.S. Patent 3,580,894, A19710525, 1971.
(10) BASF SE. PCT Int. Appl. WO/2011/095540, A2 20110811,
2011.
(11) Kopacki, A. F.; Horodecky, J. U.S. Patent 3,424,715,A19690128,
1969.
(12) Higgins, Wm.A.; Vogel, P. W.; Craig, W. G. J. Am. Chem. Soc.
1955, 77, 1864.
(13) Krems Chemie Chemical Services AG. PCT Int. Appl. WO/
2011/000019, A1 20110106, 2011.
1
solid. Mp 58 °C; 31P NMR: δ (CDCl3) = 56.6; H NMR: δ
(CDCl3) = 2.88 (s, 1H), 7.43−7.55 (m, 6H), 7.91−8.01 (dd, J
= 15.0 Hz, J = 8.1 Hz, 4H); 13C NMR: δ (CDCl3) = 128.5 (d,
JPC = 13.7 Hz, 4C), 130.9 (d, JPC = 11.9 Hz, 4C), 131.9 (d, JPC
= 3.3 Hz, 2C), 135.4 (d, JPC = 85.8 Hz, 2C); Anal. Calcd for
C12H11PS2: C 57.6, H 4.4, S 25.6, P 12.4; found C 57.1, H 4.4, S
25.4, P 12.1; HR-EI: calcd (C12H11PS2) m/z = 250.0040, found
m/z = 250.0036.
(14) Rakotomalala, M.; Ciesielski, M.; Walter, O.; Doering, M.
ARKIVOC 2012, In print.
(15) Wallach, O. Liebigs Ann. 1890, 259, 300−309.
(16) Levesque, C. L. U.S. Patent 2,560,296, 1951.
(17) Moore, C. G.; Saville, R. W. J. Chem. Soc. 1954, 2082−2089.
(18) Malatesta, L. Gazz. Chim. Ital. 1947, 76, 167.
(19) Peters, G. J. Org. Chem. 1962, 27, 2198.
(20) Kazan State University, SU 433787, 1978.
(21) Zimin, M. G.; Zabirov, N.; Smirnov, V. N.; Cherkasov, R. A.;
Pudovik, A. N. J. Gen. Chem. USSR (Engl. Transl.) 1980, 50 (1), 18−
30.
(22) Kuchen, W. Chem. Ber. 1968, 101 (10), 3454.
(23) Bergemann, K.; Hesselbarth, F.; Wenschuh, E.; Baumeister, U.;
Hartung, H. Phosphorous, Sulfur Silicon Relat. Elem. 1993, 79 (1,4),
131.
Synthesis of Bis(diphenylphosphinothioyl) Disulfide
(7). In a three-necked, round-bottomed flask equipped with a
condenser, thermometer, dropping funnel and argon inlet, 6b
(325.4 g, 1.3 mol) was dissolved in MTBE (2.5 L) and cooled
to 2−5 °C in an ice bath. At this temperature aqueous
hydrogen peroxide (73.6 g, 0.64 mol) was dropped into the
reaction mixture over a period of 15 min. During this time a
white solid precipitated. The newly formed suspension was
stirred for one hour at 5 °C and then filtered. The residue was
washed with a small amount of MTBE and subsequently dried
at reduced pressure at 50 °C, to yield 323.1 g (99.7%) of a
white crystalline solid. Mp 145 °C 31P NMR: δ (CDCl3) =
(24) Pastor, S. D.; Spivack, J. D.; Steinhuebel, L. P. Phosphorous,
Sulfur Silicon Relat. Elem. 1987, 31, 71−76.
1
69.5; H NMR: δ (CDCl3) = 7.32−7.48 (m, 12H), 7.76−7.85
(dd, J = 14.3 Hz, J = 7.0 Hz, 8H); 13C NMR: δ (CDCl3) =
128.4 (d, JPC = 13.8 Hz, 4C), 131.9 (d, JPC = 11.6 Hz, 4C),
132.2 (d, JPC = 3.1 Hz, 2C), 132.8 (d, JPC = 70.0 Hz, 2C). HR-
EI: calcd (C24H20P2S4) m/z = 497.9923, found m/z =
497.9968.
F
dx.doi.org/10.1021/op300147f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX