High-yield synthesis and crystal structure determination of sodium triphenylphosphane monosulfonate (TPPMSNa)

Article abstract of DOI:10.1002/ejic.200901002

A. simple, high-yield synthesis is described that leads to the sodium salt of monosulfonated triphenylphosphane (TPPMSNa) as a pure product in large quantities without complicated workup techniques. The single-crystal X-ray structure of TPPMS^Na-2.5H₂O is reported. The structure is built up by alternating layers of aquated sodium sulfonate units and hydrophobic triphenyl units. Thermogravimetric analyses show the loss of one molecule of water at 70 and. one at 105 °C. Thermal decomposition occurs at temperatures above 400 °C. The anhydrous ligand is hygroscopic. The preparation of the free acid TPPMS^H starting from the sodium, salt is also reported. TPPMSH is an extremely hygroscopic solid that could not be isolated in the crystalline form.

Full text of DOI:10.1002/ejic.200901002

FULL PAPER
DOI: 10.1002/ejic.200901002
High-Yield Synthesis and Crystal Structure Determination of Sodium
Triphenylphosphane Monosulfonate (TPPMSNa)
Arndt Karschin,[a] Wolfgang Kläui,*[a] Wilfried Peters,[a] and Bernhard Spingler[b]
Keywords: Phosphane ligands / Surfactants / Sulfur / Structure elucidation
A simple, high-yield synthesis is described that leads to
the sodium salt of monosulfonated triphenylphosphane
(TPPMSNa) as a pure product in large quantities without
complicated workup techniques. The single-crystal X-ray
structure of TPPMSNa·2.5H2O is reported. The structure is
built up by alternating layers of aquated sodium sulfonate
units and hydrophobic triphenyl units. Thermogravimetric
analyses show the loss of one molecule of water at 70 and
one at 105 °C. Thermal decomposition occurs at tempera-
tures above 400 °C. The anhydrous ligand is hygroscopic.
The preparation of the free acid TPPMSH starting from the
sodium salt is also reported. TPPMSH is an extremely hygro-
scopic solid that could not be isolated in the crystalline form.
Introduction
The synthesis of sodium triphenylphosphane monosul-
fonate (1), abbreviated in the following as TPPMSNa
(Scheme 1), was first described by Chatt et al. in 1958.[1]
They added triphenylphosphane slowly with cooling to
oleum (H2SO4/SO3), then heated the mixture, neutralized
carefully with sodium hydroxide after 1–2 h and isolated the
product after recrystallization from ethanol in a yield of
32%. Since this first appearance of TPPMSNa in the litera-
ture, many similar preparation methods have been pro-
posed.[2–5] Slow addition of triphenylphosphane and heat-
ing of the reaction mixture as described by Chatt were not
questioned over the years. Purification steps like extraction
of the sulfonated product with triisooctylamine in toluene
or the intermediate isolation of the ammonium salt of
TPPMS have been introduced to isolate TPPMSNa free of
impurities.
In this paper, we describe a simple procedure that leads
to TPPMSNa as a pure product in large quantities without
implementing complicated workup techniques. We deter-
mined the single-crystal X-ray structure of the sodium salt
and we further characterized this compound by thermogra-
vimetry and 31P NMR spectroscopy. In addition, we de-
scribe an easy way to obtain the free sulfonic acid TPPMSH
(2).
Scheme 1. Sulfonated triphenylphosphanes.
Results and Discussion
To synthesize TPPMSNa effectively, the reactions to
avoid are oxidation of the phosphane and the formation of
the di- and trisulfonated phosphane. We have found that
very slow addition of triphenylphosphane to the cooled
oleum solution, as it is described in most references, results
in higher amounts of disulfonated triphenylphosphane. Best
results were obtained when the triphenylphosphane was
added in one portion to the oleum cooled to 0 °C with very
effective stirring until all of the phosphane has dissolved.
The often mentioned heating of the reaction mixture after
dissolution of triphenylphosphane is counterproductive, as
it increases the relative amounts of the phosphane oxide. In
our hands, the reaction time shortens to a few hours but
the oxidation ratio reaches up to 30%. Running the reaction
at or below room temperature suppresses the oxidation of
the phosphane ligand almost completely.[6] The reaction
[a] Institut für Anorganische Chemie und Strukturchemie,
Lehrstuhl I: Bioanorganische Chemie und Katalyse, Heinrich
Heine Universität Düsseldorf,
Universitätsstraße 1, 40225 Düsseldorf, Germany
Fax: +49-211-8112287
E-mail: klaeui@uni-duesseldorf.de
[b] Institute of Inorganic Chemistry, University of Zürich,
Winterthurerstr. 190, 8057 Zürich, Switzerland
942
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 942–946

Sodium Triphenylphosphane Monosulfonate (TPPMSNa
)
mixture obtained after rapid addition of triphenylphos- TPPMSH
phane and no heating contains after 18 h mono- and di-
sulfonated triphenylphosphane and small amounts (Ͻ5%)
When an aqueous solution of TPPMSNa is treated with
concentrated hydrochloric acid, a dense white precipitate
forms. After about 30 min separation into two clear, colour-
less liquid phases is observed. The viscous lower phase con-
tains the free acid TPPMSH that can be isolated as a
noncrystalline and extremely hygroscopic off-white solid by
evaporation of the solvent. Shaking the two-phase system
leads again to a white emulsion. Addition of more concen-
trated hydrochloric acid to the milky emulsion yields a
homogeneous solution. The reason for this behaviour is not
yet clear. One can assume that the first protonation occurs
at the phosphorus atom to form a phosphonium sulfonate,
that is, a betaine-type structure. The formation of the white
precipitate is not restricted to hydrochloric acid. It is ob-
served also upon addition of sulfuric acid or nitric acid to
of the oxides (Table 1).
Table 1. Composition of the reaction mixture after 18 h quantified
by integration of the 31P{1H} NMR signals.
TPPMS (1)
TPPDS (2)
OTPPMS[a]
2%
OTPPDS[a]
1%
75%
22%
[a] Phosphane oxides of 1 and 2.
Further important changes were made in the workup
procedure. It was proposed not to neutralize the reaction
mixture completely, but to raise the pH value only to 2–3
to suppress oxidation. It is assumed that the acidic medium
protects the phosphane by protonation. Our experiments
show that complete neutralization does not increase the
amount of oxidized product. However, very effective cool-
ing conditions during neutralization are important so that
local overheating of the reaction mixture is minimized.
When we tried to separate TPPMSNa from the neutralized
reaction mixture we found out that TPPMSNa is practically
insoluble in a saturated solution of sodium sulfate. An im-
portant step was therefore to estimate the amount of ad-
ditional water during neutralization so that the final mix-
ture becomes a saturated solution of sodium sulfate. So-
dium sulfate is therefore more than a byproduct; it is the
adjuvant for separation, when used appropriately. After fil-
tration, the product was recrystallized from water to get rid
of small amounts of sodium sulfate. The solubility of both
sodium sulfate and TPPMSNa are strongly temperature de-
pendent. From the data in Table 2 it is clear that the
recrystallization should be done at 0 °C. After drying, the
ligand is stirred in n-pentane to wash out unreacted tri-
phenylphosphane. This procedure has to be done thor-
oughly, because TPPMSNa is known to be a surface-active
agent[3] and probably helps to keep triphenylphosphane in
solution. To support this separation an ultrasonic bath can
be used.
an aqueous solution of TPPMSNa
.
Crystal Structure of TPPMSNa
The sodium salt of TPPMS crystallized in the triclinic
Na
¯
space group P1 with 4 TPPMS and 10 water molecules
in the asymmetric unit. As a representative example, one
TPPMSNa molecule is shown in Figure 1. The structure is
built up by alternating layers of aquated sodium sulfonate
units and the hydrophobic triphenyl units (Figure 2). These
layers are oriented perpendicular to the c axis of the cell.
The sodium cations are all octahedrally coordinated by six
oxygen atoms with Na–O distances varying between 2.25
and 2.61 Å. Na1 and Na21 are coordinated by three sulfo-
nate and three water oxygen atoms, whereas Na41 and
Na61 are coordinated by one sulfonate and five water oxy-
gen atoms (Figure 3).
Table 2. Solubility of Na2SO4 and TPPMSNa in water.
Temperature Solubility TPPMSNa·2H2O
Solubility Na2SO4
[mmolg–1][a]
[°C]
[mmolg–1]
0
1.5
10
–
32
–
154
229
22
25
40
52
[a] Data taken from ref.[7]
TPPMSNa crystallizes with 2.5 molecules of water (see
crystal structure determination). Thermogravimetric analy-
ses show a weight loss of 4.1 and 4.9% corresponding to
the loss of ca. one molecule of water each at 70 and one
at 105 °C. Thermal decomposition occurs at temperatures
above 400 °C. The ligand dried in vacuo is hygroscopic and
reabsorbs water over a period of several hours.
Figure 1. ORTEP presentation of one out of four crystallographic
independent TPPMSNa molecules including all five other oxygen
atoms that are coordinated to Na1. Ellipsoids are drawn at 50%
probability level. Used symmetry operator: $1: –x – 1, –y – 2, –z +
1. O43 and O62_$1 are oxygen atoms of other sulfonate groups
whereas O71, O72 and O80_$1 are oxygen atoms of water mole-
cules. Hydrogen atoms of water molecules and one disordered
water molecule were omitted for clarity.
Eur. J. Inorg. Chem. 2010, 942–946
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
943

A. Karschin, W. Kläui, W. Peters, B. Spingler
FULL PAPER
Characteristics in NMR Spectroscopy
TPPMSNa, TPPDSNa and TPPTSNa give rise to signals
at –5.9, –5.5 and –5.1 ppm, respectively, in the 31P{1H}
NMR spectrum in H2O at pH 5 or above. The introduction
of each sulfonate group shifts the signal downfield by
0.4 ppm. A similar shift of 0.6 ppm per sulfonate group was
reported by Atwood et al. for the spectra of triphenylphos-
phane, TPPMSNa, TPPDSNa and TPPTSNa in [D6]-
DMSO.[8] Below pH 5, the chemical shifts of the 31P NMR
signals of the sulfonated phosphanes become pH dependent
(Figure 4). The downfield shift of the signal of the mono-
sulfonated triphenylphosphane TPPMS is most pronounced.
In 5  hydrochloric acid the TPPMS chemical shift is
5 ppm, which is 11 ppm lower than that at pH 7. The signal
of the trisulfonated phosphane moves only by 6.3 ppm. The
31P{1H} NMR signals of the oxidized forms of the sulfo-
nated phosphanes appear at +37.3 (OTPPMS), +36.3
(OTPPDS) and +35.2 ppm (OTPPTS) at neutral pH. In 5 
HCl they are shifted to 38.6, 37.1 and 35.7 ppm, respec-
tively. Again, the signal of the monosulfonated species
shows the strongest downfield shift. It is reasonable to as-
sume that this reflects the influence of the electron-with-
drawing sulfonate groups on the degree of protonation of
the phosphorus in the phosphanes and the oxygen in the
phosphane oxides. The proton coupled 31P NMR spectrum
of TPPMS in 5  and even in concentrated HCl does not
1
show JP,H coupling. The lifetime of the protonated phos-
phane is obviously too short. However, in ca. 10  sulfuric
acid and in nonaqueous acids like trifluoracetic acid the
coupling constant JP,H = 510 Hz is observed, which con-
firms protonation at phosphorus.
Figure 2. Packing diagram of sodium triphenylphosphane mono-
sulfonate. Only the major components of the disordered water
molecules are shown.
1
Figure 4. Downfield shift and peak intensities of sulfonated phos-
phanes in the 31P{1H} NMR spectra at different pH values.
The second effect when lowering the pH value is that
with increasing downfield shift the signals appear with in-
creasing intensity. The reason for this artefact is the relax-
ation time of the phosphorus nuclei in the phosphanes and
its dependence on the degree of protonation. We measured
the relaxation time T1 in neutral aqueous solution by using
inversion recovery with power gated decoupling (Table 3).
It was surprising for us to see that the phosphorus relax-
Figure 3. Diagram of one of the four TPPMSNa molecules and the
four crystallographically independent oxygen coordinated sodium
atoms. The hydrogen atoms and the minor components of the dis-
ordered water molecules are omitted for clarity. The NaO6 octahe-
dra are connected through shared vertices and edges. Oxygen atoms
that belong to sulfonate groups of TPPMS are marked with front
ellipses.
944
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 942–946

Sodium Triphenylphosphane Monosulfonate (TPPMSNa
)
distance restraints. The structure was checked for higher symmetry
with help of the program Platon.[17] CCDC-748773 contains the
supplementary crystallographic data for this paper. These data can
be obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
ation times of the three sulfonated phosphanes are not only
rather long but also quite different. In addition, proton-
ation shortens the relaxation times significantly. These two
effects lead to inaccurate and pH-dependent signal inten-
sities under standard measuring conditions (2 s delay time,
30° pulses).
Sodium (Diphenylphosphanyl)benzenesulfonate (1): Oleum (25 mL,
25% free SO3) was placed in a 500-mL, three-necked flask charged
with a 200-mL dropping funnel, and cooled in an ice bath to 0 °C.
In order to get very good cooling conditions during the whole syn-
thesis the ice bath was stirred magnetically and the reaction mixture
was stirred mechanically with a KPG-stirring unit. Triphenylphos-
phane (10 g) was added in one step. The reaction mixture was co-
oled until all of the triphenylphosphane was dissolved (≈2 h). Then,
the mixture was stirred at room temperature for 18 h. Then, the
reaction mixture was cooled again to 0 °C, and water (150 mL)
was added dropwise with very effective stirring to prevent local
overheating. Using a pH-glass electrode and 7.5  NaOH solution
(ca. 130 mL) the mixture was neutralized. During neutralization,
precipitation of the product as a white, fine solid was observed.
This was filtered off at room temperature through a large glass
filter funnel with no or only gentle suction. The white solid was
transferred into an Erlenmeyer flask and recrystallized overnight
from water (80 mL) at 4 °C to remove sodium sulfate impurities.
The solid was filtered off, dried and suspended in n-pentane
(50 mL) for several hours to extract unreacted starting material un-
til the product was completely soluble in water. Yield 9.1 g (59%)
of monosulfonated triphenylphosphane. 1H NMR (200 MHz, D2O,
25 °C): δ = 4.8 (s, water protons), 6,9–7.1 (m, 12 H), 7,6–7.8 (m, 2
H, 2-C and 4-C protons) ppm. 31P{1H} NMR (81 MHz, D2O,
25 °C): δ = –5.7 (s) ppm. 13C{1H} NMR (126 MHz, D2O, 25 °C):
Table 3. Relaxation times T1 [s] of the phosphorus nuclei in the
sulfonated phosphanes and the corresponding oxides in aqueous
solution.
TPPMSNa
TPPDSNa
TPPTSNa
Water (pH ≈ 7)
1  HCl (pH ≈ 0)
5.3
2.1
10.3
5.7
8.9
9.2
OTPPMSNa
OTPPDSNa
OTPPTSNa
Water (pH ≈ 7)
8.9
7.3
5.9
Conclusions
We have presented here a simple, fast and high-yield pre-
parative method for the monosulfonated triphenylphos-
phane TPPMSNa. Our procedure gives analytically pure
TPPMSNa with a yield (59%) twice as high as that reported
in Inorganic Syntheses (29%).[5] There are discrepancies in
the literature about the water content of TPPMSNa. We
characterized this sodium salt by thermogravimetric
measurements and determined its crystal structure with the
composition TPPMSNa·2.5H2O. The easy access of
1
2
δ = 140.6 (d, JC,P = 12.2 Hz, 1 C, 1-C), 132.8 (d, JC,P = 24.0 Hz,
TPPMSNa can be the basis for further studies of this mo- 1 C, 2-C), 145.8 (d, JC,P = 7.4 Hz, 1 C, 3-C), 128.6 (s, 1 C, 4-C),
3
3
2
131.7 (d, JC,P = 6.8 Hz, 1 C, 5-C), 138.1 (d, JC,P = 15.6 Hz, 1 C,
lecule that can act as an anionic surfactant and as a phos-
phane ligand at the same time. This property might be espe-
cially helpful in the development of metal complexes in bi-
phasic catalysis.[9–11]
6-C), 138.2 (d, 1JC,P = 8.8 Hz, 2 C, 7-C), 136.2 (d, JC,P = 19.6 Hz,
4 C, 8-C), 131.3 (d, JC,P = 7.2 Hz, 4 C, 9-C), 131.8 (s, 2 C, 10-C)
2
3
ppm.
(Diphenylphosphanyl)benzenesulfuric Acid (2): TPPMSNa (1; 2.0 g)
was dissolved in water (50 mL). The solution was transferred into
a separating funnel and treated dropwise with concentrated hydro-
chloric acid solution (≈5.2 mL) until no additional milky emulsion
was formed. The emulsion was chilled overnight, and the highly
viscous lower phase was separated from the upper phase. Drying
of the collected viscous oil resulted in a glassy, amorphous white
solid, which was quickly pestled and vacuum dried again. The
product was analyzed as TPPMSH monohydrate. Yield: 70%. 1H
NMR (200 MHz, CD3CN, 25 °C): δ = 5.6 (br. s, water/acid pro-
tons), 7,4–7.8 (m, 12 H), 7,9–8.1 (m, 2 H, C-SO3-ortho protons)
ppm. TPPMSH is very hygroscopic, soluble in water, organic
alcohols, acetone, acetonitrile and dichloromethane and insoluble
in nonpolar organic solvents like hexane or toluene. Dimethyl sulf-
oxide causes oxidation of TPPMSH.
Experimental Section
General: TPPDS[12] and TPPTS,[13] used as reference materials,
were synthesized according to literature methods. NMR spectra
were recorded with a Bruker Avance DRX 200 and a Bruker Av-
ance DRX 500 spectrometer. 31P{1H} NMR chemical shifts were
referenced to external phosphoric acid (85%) with downfield values
taken as positive. 1H NMR and 13C{1H} NMR chemical shifts
were referenced to 3-(trimethylsilyl)propanesulfonic acid sodium
salt. DEPT-135 and gs-HMQC spectroscopy were used for unam-
biguous correlation of 13C chemical shifts and structure of TPPMS.
Spin-lattice relaxation times T1 were determined by the inversion
recovery experiment. Crystallographic data were collected at
183(2) K with an Oxford Diffraction Xcalibur system with a ruby
detector by using Mo-Kα radiation (λ = 0.7107 Å) that was graph-
ite-monochromated. A suitable crystal was covered with oil (Infi-
neum V8512, formerly known as Paratone N), mounted on top of
a glass fibre and immediately transferred to the diffractometer. The
program suite CrysAlisPro was used for data collection, semiempir-
ical absorption correction and data reduction.[14] The structure was
solved with direct methods by using SIR97[15] and refined by full-
matrix least-squares methods on F2 with SHELXL-97.[16] Two
water molecules were disordered and refined with 50:50 and 55:45
occupancy, respectively. The hydrogen atoms of three water mo-
lecules could be localized and were refined with appropriate O–H
[1] S. Ahrland, J. Chatt, N. R. Davies, A. A. Williams, J. Chem.
Soc. 1958, 276–288.
[2] G. Schmid, N. Klein, L. Korste, U. Kreibig, D. Schonauer,
Polyhedron 1988, 7, 605–608.
[3] F. Joo, J. Kovacs, A. Katho, A. C. Benyei, T. Decuir, D. J. Dar-
ensbourg, Inorg. Synth. 1998, 32, 2–4.
[4] H. Gulyas, Z. Bacsik, A. Szollosy, J. Bakos, Adv. Synth. Catal.
2006, 348, 1306–1310.
[5] T. Suarez, B. Fontal, M. Reyes, F. Bellandi, R. R. Contreras,
A. Bahsas, G. Leon, P. Cancines, B. Castillo, React. Kinet. Ca-
tal. Lett. 2004, 82, 317–324.
Eur. J. Inorg. Chem. 2010, 942–946
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
945

A. Karschin, W. Kläui, W. Peters, B. Spingler
FULL PAPER
[6] B. M. Bhanage, S. S. Divekar, R. M. Deshpande, R. V. Chaud-
hari, Org. Process Res. Dev. 2000, 4, 342–345.
[7] H. Furkert, H. Schütt in Ullmann’s Encyclopedia of Industrial
Chemistry, 3rd ed. (Ed.: Wilhelm Foerst), Wiley, Berlin, 1960,
vol. 12, p. 682.
[8] M. R. Barton, Y. G. Zhang, J. D. Atwood, J. Coord. Chem.
2002, 55, 969–983.
[9] M. Saoud, A. Romerosa, S. M. Carpio, L. Gonsalvi, M. Peruz-
zini, Eur. J. Inorg. Chem. 2003, 1614–1619, and references cited
therein.
[10] J. Diez, M. P. Gamasa, E. Lastra, A. Garcia-Fernandez, M. P.
Tarazona, Eur. J. Inorg. Chem. 2006, 2855–2864, and references
cited therein.
[12]
[13]
[14]
T. Thorpe, S. M. Brown, J. Crosby, S. Fitzjohn, J. P. Muxwor-
thy, J. M. J. Williams, Tetrahedron Lett. 2000, 41, 4503–4505.
T. Bartik, B. Bartik, B. E. Hanson, T. Glass, W. Bebout, Inorg.
Chem. 1992, 31, 2667–2670.
Oxford Diffraction Ltd., 171.32 ed., Oxford, UK, 2007, Xcali-
bur CCD system.
[15] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C.
Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R.
Spagna, J. Appl. Crystallogr. 1999, 32, 115–119.
[16] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[17] A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7–13.
Received: October 13, 2009
Published Online: January 15, 2010
[11] G. Kovacs, G. Ujaque, A. Lledos, F. Joo, Eur. J. Inorg. Chem.
2007, 2879–2889, and references cited therein.
946
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 942–946