A versatile protocol for the quantitative and smooth conversion of phosphane oxides into synthetically useful pyrazolylphosphonium salts

Article abstract of DOI:10.1002/cssc.201100503

A convenient protocol for the smooth conversion of the resistant P-O bond in phosphane oxides into a reactive P-N bond of synthetically useful pyrazolylphosphonium salts is described. A highly charged, oxophilic, phosphorus-centered trication is employed and the reactions are conducted at room temperature with quantitative yields. The resulting pyrazolylphosphonium cations are valuable synthetic intermediates and are used for the synthesis of a variety of organophosphorus compounds. This represents a new approach towards the transformation of the rather inert phosphoryl group under very mild reaction and workup conditions and aims towards alternatives to existing reduction methods for phosphane oxide functionalization.

Full text of DOI:10.1002/cssc.201100503

DOI: 10.1002/cssc.201100503
A Versatile Protocol for the Quantitative and Smooth
Conversion of Phosphane Oxides into Synthetically Useful
Pyrazolylphosphonium Salts
Kai-Oliver Feldmann, Stephen Schulz, Felix Klotter, and Jan J. Weigand*[a]
Dedicated to Professor Joseph Grobe on the occasion of his 80th birthday.
A convenient protocol for the smooth conversion of the resist-
ant PꢀO bond in phosphane oxides into a reactive PꢀN bond
of synthetically useful pyrazolylphosphonium salts is described.
A highly charged, oxophilic, phosphorus-centered trication is
employed and the reactions are conducted at room tempera-
ture with quantitative yields. The resulting pyrazolylphosphoni-
um cations are valuable synthetic intermediates and are used
for the synthesis of a variety of organophosphorus com-
pounds. This represents a new approach towards the transfor-
mation of the rather inert phosphoryl group under very mild
reaction and workup conditions and aims towards alternatives
to existing reduction methods for phosphane oxide functional-
ization.
Introduction
The Wittig reaction is the key step of the industrial-scale pro-
duction of carotenoids, such as b-carotene (provitamin A;
Scheme 1).[1] However, the formation of stoichiometric
yields, and more viable reaction conditions. Notably, the reduc-
tion of phosphane oxides still represents the only major excep-
tion to the paradigm that the phosphoryl group generally re-
sists chemical transformations.[6a] Alternative, nonreductive
transformations of the resistant PꢀO bond of phosphane
oxides are sought and are addressed in this contribution.[5m]
The selective cleavage of the thermodynamically favoured
PꢀO bond [E(D)PꢀO =128–139 kcalmolꢀ1][6a] in close vicinity to a
much weaker PꢀC bond [E(D)PꢀC ꢁ65 kcalmolꢀ1 for tetracoordi-
nate phosphorus atoms][6b] is a major challenge. More than
half a century after the Wittig reaction gained industrial impor-
tance,[7] examples of such transformations are still extremely
rare (Scheme 2). Phosphane oxides can be transformed into
the corresponding dihalophosphoranes (Scheme 2a) by react-
ing them with Group IV–VI halides and oxyhalides (e.g., thionyl
chloride,[8] phosphoryl chloride,[9] phosphorus pentachloride,[10]
oxalyl chloride,[11] diphosgene,[11] and phosgene[12]) to imple-
ment reactive PꢀCl bonds. Phosphane sulfides (strong PꢀO vs.
weak PꢀS bonds) have been obtained from the corresponding
oxides upon reaction with P4S10,[13,5l] Lawesson’s reagent,[14] or
related derivatives[15] (Scheme 2b)[5l]. Iminophosphoranes are
intermediates in the industrial manufacturing of carbodiimides
from isocyanates with phosphane oxides as catalysts;[16] these
highly reactive intermediates are only isolable in rare cases.
Scheme 1. The Wittig reaction as the key step in the industrial production of
b-carotine.[1d]
amounts of triphenylphosphane oxide (TPPO) is a major prob-
lem because there are very few applications for this by-prod-
uct. The recycling of TPPO has been identified as a major
factor for the economic efficiency of the Wittig reaction.[2] Al-
though facilities for the recycling of TPPO have been instal-
led[1b] and catalytic Wittig reactions are being developed,[3]
most TPPO is still disposed of as waste.[1b] There is a persistent
need for more efficient recycling methods and alternative ap-
plications of TPPO and other phosphane oxides, which moti-
vates research in this field.[4]
[a] K.-O. Feldmann, S. Schulz, F. Klotter, Dr. J. J. Weigand
Institut fꢀr Anorganische und Analytische Chemie
and Graduate School of Chemistry
WWU Mꢀnster
Supplemented by academic research, which has been largely
triggered by the development of new synthetic routes to terti-
ary phosphanes as ligands in transition-metal-mediated reac-
tions, a plethora of reduction methods for functionalized phos-
phane oxides were developed.[5] Such methods are being con-
tinuously optimized towards better cost efficiency, higher
Corrensstr. 30, 48149 Mꢀnster (Germany)
Fax: (+49)251-83-33126
E-mail: jweigand@uni-muenster.de
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201100503.
ChemSusChem 2011, 4, 1805 – 1812
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1805

J. J. Weigand et al.
(Scheme 3).[19,20] Compound 4[OTf]2 itself represents a very
viable reagent, however, its applicability will be addressed else-
where. By-product 4[OTf]2 can be conveniently hydrolyzed
with water to quantitatively yield 3,5-dimethylpyrazolium tri-
flate and P4O6, which can be easily separated. Oxidation of
P4O6 with O2 to P4O10 and subsequent reaction with water
yields pure H3PO4. After removal of the solvent from the fil-
trate, the colourless residue obtained was investigated by
means of multinuclear NMR spectroscopy in CD2Cl2. The
31P{1H} NMR spectrum exclusively displayed a singlet (d=
41.3 ppm) at lower field than that of 2a (d=26.0 ppm)[21] The
1H NMR spectrum showed only one set of resonances, reveal-
ing the presence of phenyl groups and an asymmetrically sub-
stituted 3,5-dimethylpyrazolyl fragment (d=7.97 (m, 3H), 7.79
Scheme 2. Selected examples for the transformation of phosphane oxides.
4
(m, 6H), 7.71 (m, 6H), 6.37 (d, J(H,P)=3.2 Hz, 1H), 2.29 (s, 3H),
4
1.75 ppm (d, J(H,P)=0.7 Hz, 3H). In accordance with the re-
The example shown in Scheme 2c, represents one of the very
rare cases for which a PꢀO bond is substituted by a much
weaker PꢀN bond.[17] A reversed Wittig reaction to form an
ylide is observed when dicyanoalkyne is reacted with TPPO at
elevated temperatures in benzene to exchange a PꢀO bond
for a PꢀC bond (Scheme 2d).[18]
sults from the 13C and 19F NMR spectroscopy investigations, the
formation of triphenyl-3,5-dimethyl-1-pyrazolyl-phosphonium
triflate (3a[OTf]; Scheme 3) was deduced. Compound 3a[OTf]
was obtained in quantitative yield and was analytically pure
without the need for further purification. Suitable crystals for
X-ray single-crystal structure determination were obtained by
slow diffusion of n-hexane into a CH2Cl2 solution at low tem-
peratures (ꢀ328C).[20] In cation 3a+ (Figure 1) the coordination
sphere at the phosphorus atom possesses almost ideal tetrahe-
dral geometry (angles span 108.2(9)–111.2(1)8). The PꢀN bond
length (1.683(2) ꢁ) is in the typical range of P(tetravalent)ꢀ
N(sp2) bonds if the P atom is tetravalent (e.g., [(PMe3)2NH]2+
1.655 and 1.661 ꢁ;[22] [tBu2P(H)pyrrole]+ 1.669 ꢁ;[23] Ph2P-
(BH3)pyrrole 1.711 ꢁ;[24] and [PMe3DMAP]2+, DMAP=4-dimethy-
laminopyridine, 1.720 ꢁ[25]) and does not significantly depend
on the substituents on phosphorus (compare 3c+ and 3e+;
Figure 1; see the Supporting Information for further details
and natural bond orbital analysis of 3a+). We extended our ap-
proach to a series of phosphane oxides (Scheme 3). Under
identical conditions, compounds 2b–g were transformed into
the corresponding pyrazolylphosphonium salts b–g[OTf] in
only moderate purity (90%). Further purification reduced the
yields drastically. However, by applying a slight excess of
1[OTf]3 (1.1 equiv),[20] phosphonium salts 3b–g[OTf] were ob-
tained in quantitative yields in satisfying purities (>97%) after
filtration and removal of the solvent. To the best of our knowl-
edge, only derivative 3b[I] has been described in the literature
and was prepared in very low yield (28%) from (3,5-dimethyl-
pyrazolyl)diphenylphosphane and a 25-fold excess of MeI.[26]
Having prepared pyrazolylphosphonium salts 3a–g[OTf] on a
multigram scale, we proceeded to investigate the synthetic po-
tential of this class of compounds, which had not been as-
Herein, we report a versatile and attractive concept to easily
obtain value-added products from unreactive phosphane
oxides. This novel approach uses the recently reported Janus
head type diphosphorus trication 13+ (Scheme 3), which is
easily accessible in one step and on a large scale from compa-
rably cheap and commercially available starting materials. In
previous work we showed that trication 13+ exhibited very un-
usual reactivity towards water.[19] We have now found that tri-
cation 13+ is highly oxophilic and can be used for the facile
and clean transformation of phosphane oxides 2a–g into pyra-
zolylphosphonium cations 3a–g+ under very smooth reaction
conditions at room temperature. This approach allows the
transformation of the PꢀO bond in phosphane oxides into a
much more reactive PꢀN bond and offers a convenient proto-
col for the transformation of functional phosphane oxides. The
resulting pyrazolylphosphonium triflate salts 3a–g[OTf] can be
easily isolated in quantitative yields and a series of selected
transformation reactions prove their utility as precursors for
further syntheses.
Results and Discussion
The reaction of TPPO (2a) with 1[OTf]3 in CH2Cl2 in a ratio of
2:1 at ambient temperature resulted in the formation of a col-
ourless precipitate, which was isolated by filtration in quantita-
tive yield and identified as analytically pure 4[OTf]2
Scheme 3. Preparation of pyrazolylphosphonium salts 3a–g[OTf]. a: CH2Cl2, ambient temperature, 15–20 h.[31]
1806
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 1805 – 1812

Conversion of Phosphane Oxides into Pyrazolylphosphonium Salts
[B3LYP/6-311G(2d)][27] show that protonation of the free lone
pair of the pyrazole moiety in 3a+ (gas-phase proton affinity
of 3a+: 160.3 kcalmolꢀ1 [28]
)
leads to weakening of the PꢀN
bond. The optimized molecular structure of 3aH2+ features a
significantly longer PꢀN bond (1.7699 ꢁ) than that in 3a+
(calcd: 1.6998 ꢁ, exptl: 1.683(2) ꢁ) and a smaller Wiberg bond
index[29] (3a+: WBI=0.7785; 3aH2+: WBI=0.6708). Hence, the
addition of a Brønsted acid renders the pyrazol moiety an ex-
cellent leaving group. Attempts to isolate 3aH[OTf]2 from reac-
tion mixtures containing 3a[OTf] and HOTf failed. Instead, we
were able to verify the formation of phosphonium anhydride
6[OTf]2 (isolated yield 86%) and trifluoromethanesulfonic anhy-
dride (Scheme 4, bottom) in excellent yields. The latter origi-
nates from the dehydration of HOTf, the oxygen atom of
which is transferred to form phosphonium anhydride 6[OTf]2.
Furthermore, compound 3a[OTf] could be reduced to TPP elec-
trochemically (Scheme 5). A cyclic voltammogram of 3a[OTf]
Figure 1. ORTEP plot of the molecular structures of cations a) 3a+, b) 3c+
and c) 3e+ in 3a[OTf], 3c[OTf], and 3e[OTf], respectively. Thermal ellipsoids
are drawn at the 50% probability level (hydrogen atoms and counterions
are omitted for clarity). Selected bond lengths [ꢁ] and angles [8]: 3a+: P1ꢀ
N1 1.683(2); N1-P1-C6 108.16(9), N1-P1-C12 109.86(8), N1-P1-C18 108.26(9),
C6-P1-C12 109.52(9), C6-P1-C18 110.74(9), C12-P1-C18 110.26(9); 3c+: P1ꢀN1
1.688(2); N1-P1-C6 110.7(1), N1-P1-C12 108.6(1), N1-P1-C18 106.5(1), C6-P1-
C12 110.7(1), C6-P1-C18 111.0(1), C12-P1-C18 109.2(1); 3e+: P1ꢀN1 1.683(2);
N1-P1-C6 106.4(1), N1-P1-C8 108.1(1), N1-P1-C10 109.0(1), C6-P1-C8 109.6(1),
C6-P1-C10 110.6(1), C8-P1-C10 113.0(1).
sessed previously. We found that phosphonium cation 3a+
was prone to nucleophilic substitution of the pyrazole moiety,
that is, it served well as the synthetic equivalent of a [R3P2+
]
synthon. Treating 3a[OTf] with diphenylphosphane or 2a in
the presence of trifluoromethanesulfonic acid (HOTf) gave
phosphanylphosphonium salt 5[OTf] (79%) and phosphonium
anhydride 6[OTf]2 (83%), respectively, in very good yields
(Scheme 5). The addition of HOTf is necessary to sequester the
3,5-dimethylpyrazole moiety into 3,5-dimethylpyrazolium tri-
flate. 3,5-Dimethylpyrazole may otherwise act as a nucleophile
and initiate side reactions.[32] Furthermore, protonation of
monocation 3a+ to dication 3aH2+ may also occur under
these conditions (Scheme 4). Quantum chemical calculations
Scheme 5. Preparation of phosphanylphosphonium salt 5[OTf], bis(triphenyl-
phosphonium) anhydride 6[OTf]2, Ph3P, ylide 7, and alkene 8 from 3a[OTf].
a) HOTf, Ph2PH, ꢀ 3,5-dimethylpyrazolium triflate, CH2Cl2, RT, 12 h, 79%;
b) Ph3PO, 2HOTf, ꢀ 3,5-dimethylpyrazolium triflate, CH2Cl2, RT, 12 h 83%;
c) galvanostatic electrolysis: 25 mA, 2.86 mAcmꢀ2, Pt mesh, divided cell;
d) LiAlH4, Et2O, RT, 48 h, 90%; e) 2nBuLi, ꢀ lithium 3,5-dimethylpyrazolate,
Et2O, ꢀ958C!RT, 4 h; f) 2-nitrobenzaldehyde, ꢀ 2a, RT, 12 h, 74%.
showed a single irreversible peak at ꢀ2.11 V (CH3CN, support-
ing electrolyte [nBu4N][OTf], Pt electrode, 0.1 Vsꢀ1, range: ꢀ2.3
to ꢀ0.5 V, referenced to ferrocene). Galvanostatic electrolysis
(25 mA, 2.86 mAcmꢀ2, Pt mesh, divided cell) afforded conver-
sion to TPP, as confirmed by NMR spectroscopy.[31] As anticipat-
ed, compound 3a[OTf] was also cleanly reduced to TPP with
LiAlH4 in Et2O (90% yield).[30,31] In addition, we could also dem-
onstrate that the reaction of 3a[OTf] with two equivalents of
nBuLi (Scheme 5) gave ylide 7 quantitatively in solution. Addi-
tion of 2-nitrobenzaldehyde resulted in the formation of
alkene 8 (74% yield, 7:3 ratio of Z to E isomer) through a
Wittig reaction.[31]
Scheme 4. a) Hypothetical protonation of monocation 3a+ and b) observed
reaction of 3a[OTf] with HOTf; a: 1,2-difluorobenzene, RT, ꢀ2 3,5-dimethyl-
pyrazolium triflate, 86% isolated yield of 6[OTf]2.
ChemSusChem 2011, 4, 1805 – 1812
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
1807

J. J. Weigand et al.
Synthesis of 3a[OTf]: Compound
3a[OTf] was prepared according
to the general procedure; howev-
er, excess 1[OTf]3 was not re-
quired. A 2:1 stoichiometry of 2a/
1[OTf]3 was applied. Typically, this
reaction was performed on a scale
of 10 mmol of 1[OTf]3 or more.
M.p. 148–1518C; 1H NMR (CD2Cl2,
300 K): d=7.97 (m, 3H; C9-H),
Conclusion
We have developed a versatile method for the conversion of
phosphane oxides into pyrazolylphosphonium cations, which
were shown to be valuable synthetic intermediates amenable
to a number of subsequent transformations to a variety of or-
ganophosphorus compounds. Our protocol enables the clean
transformation of the PꢀO bond in phosphane oxides into a
much more reactive PꢀN bond under very mild reaction condi-
tions at room temperature. The resulting pyrazolylphosphoni-
um triflate salts are easily isolated in quantitative yields. This
represents a new approach towards the transformation of the
rather inert phosphoryl group and aims to provide alternatives
to existing reduction methods for phosphane oxide functional-
ization. Furthermore, our protocol might also be applicable to
the transformation of functional phosphane oxides by provid-
ing a novel synthetic route to tertiary phosphanes, which are
used as ligands in transition metal-mediated reactions. A de-
tailed investigation to address this aspect is currently under-
way.
4
7.79 (m, 6H; C8-H), 7.71 (m, 6H; C7-H), 6.37 (d, J(H,P)=3.2 Hz, 1H;
4
C2-H), 2.29 (s, 3H; C4-H), 1.75 ppm (d, J(H,P)=0.7 Hz, 3H; C5-H);
13C NMR (CD2Cl2, 300 K): d=159.0 (d, 3J(C,P)=12.9 Hz, 1C; C1),
2
4
150.3 (d, J(C,P)=7.6 Hz, 1C; C3), 137.2 (d, J(C,P)=3.1 Hz, 3C; C9),
134.9 (d, 2J(C,P)=11.6 Hz, 6C; C7), 131.0 (d, 3J(C,P)=13.9 Hz, 6C;
C8), 121.4 (q, 1J(C,F)=321.7 Hz, 1C; CF3), 118.0 (d, 1J(C,P)=
3
102.4 Hz, 3C; C6), 114.9 (d, J(C,P)=5.5 Hz, 1C; C2), 14.0 (s, 1C; C4),
13.7 ppm (s, 1C; C5); 31P NMR (CD2Cl2, 300 K): d=41.3 ppm;
19F{1H} NMR (CD2Cl2, 300 K): d=ꢀ78.8 ppm (s, CF3); Raman
(300 mW, 300 K): 3074 (56), 2994 (6), 2929 (23), 1587 (77), 1480 (6),
1437 (13), 1226 (5), 1189 (7), 1101 (21), 1029 (55), 1000 (100), 755
(16), 696 (14), 616 (19), 585 (9), 573 (14), 349 (14), 314 (16), 282 (9),
266 (6), 250 (14), 207 (9), 179 cmꢀ1 (10); IR (300 K, ATR): n˜ =2961
(w), 1611(w), 1581 (m), 1482 (w), 1440 (s), 1410 (w), 1376 (w), 1259
(vs), 1223 (w), 1146 (w), 1110 (m), 1028 (m), 996 (w), 960 (w), 821
(w), 755 (w), 726 (s), 687 (m), 635 (m), 571 (m), 534 (w), 515 (m),
475 (vw), 437 (vw), 425 cmꢀ1 (w); MS (ESI-EM): m/z calcd for
C23H22N2P [M+]: 357.15151; found: 357.1512; m/z calcd for C18H15P
[M+ꢀ3,5-dimethylpyrazole radical]: 262.0906; found: 262.0903; m/z
calcd for C12H10P [M+ꢀ3,5-dimethylpyrazole radicalꢀphenyl radi-
cal]: 185.0515; found: 185.0511; elemental analysis calcd (%) for
C24H22F3N2O3PS: C 56.98, H 4.38, N 5.53; found: C 56.78, H 4.30, N
5.94.
Experimental Section
All reactions were performed in a glove box or by using standard
Schlenk techniques under an inert Ar atmosphere. Dry, oxygen-free
solvents were employed. Compound 1[OTf]3 was prepared accord-
ing to a previously described method.[19] All phosphane oxides
were sublimed prior to use. All new compounds were fully charac-
terized by 31P, 1H, 19F, and 13C NMR spectroscopy; Raman and IR
spectroscopy; mass spectrometry; melting points; and elemental
analyses. NMR spectra were measured on a Bruker AVANCE 400
spectrometer (1H: 400.03 MHz; 13C: 100.59 MHz; 31P: 161.94 MHz)
at 300 K. Chemical shifts were referenced to tetramethylsilane
(TMS) or H3PO4 (85%). Assignments of individual resonances were
possible by using 2D techniques (HMBC, HSQC) and phosphorus-
decoupled proton NMR spectroscopy investigations (1H{31P}) if nec-
essary. IR and Raman spectra were recorded by using a Bruker
Vertex 70 instrument equipped a RAM II module. An attenuated
total reflectance (ATR) unit (diamond) was used to record IR spec-
tra. The intensities are reported relative to the most intense peak.
Raman spectra were obtained by employing an Nd:YAG laser
(1064 nm, 10–200 mW). The intensities are reported in percent rela-
tive to the most intense peak and are given in parentheses. Mass
spectra were recorded on a Thermo Scientific Orbitrap LTQ XL in-
strument at the Organisch-Chemisches Institut, WWU Mꢂnster.
Parent cations were separated and fragmented by using appropri-
ate potentials to obtain high-resolution mass data of the fragment
cations. Melting points were recorded on a thermoelectric melting
point apparatus in sealed capillaries under an argon atmosphere.
Elemental analyses were performed on a Vario EL III CHNS elemen-
tal analyzer at the IAAC, University of Mꢂnster
1
Synthesis of 3b[OTf]: M.p. 133–1368C; H NMR (CD2Cl2, 300 K): d=
7.94 (m, 2H; C9-H), 7.77 (m, 4H; C8-H), 7.67 (m, 4H; C7-H), 6.29 (d,
4J(H,P)=2.8 Hz, 1H; C2-H), 3.08 (d, 3J(H,H)=13.4 Hz, 2H; C10-H),
2.30 (s, 3H; C4-H), 1.82 ppm (3H;
C5-H); 13C NMR (CD2Cl2, 300 K): d=
158.2 (d, 3J(C,P)=13.0 Hz, 1C; C1),
149.1 (d, 2J(C,P)=7.6 Hz, 1C; C3),
136.6 (d, 4J(C,P)=3.0 Hz, 2C; C9),
132.8 (d, 2J(C,P)=12.0 Hz, 4C; C7),
130.7 (d, 3J(C,P)=13.9 Hz, 4C; C8),
120.9 (q, 1J(C,F)=321.2 Hz, 1C; CF3),
118.1 (d, 1J(C,P)=98.4 Hz, 2C; C6),
114.2 (d, 3J(C,P)=5.1 Hz, 1C; C2),
13.5 (s, 1C; C4), 12.9 (d, 1J(C,P)=
68.6 Hz, 1C; C10), 12.8 ppm (s, 1C;
C5); 31P NMR (CD2Cl2, 300 K): d=49.3 ppm; 19F{1H} NMR (CD2Cl2,
300 K): d=ꢀ78.9 ppm (s, CF3); Raman (60 mW, 300 K): 3069 (52),
3031 (9), 2937 (26), 2918 (8), 1590 (45), 1576 (10), 1440 (11), 1227
(8), 1191 (8), 1164 (8), 1116 (22), 1032 (66), 1020 (7), 999 (100), 757
(22), 694 (21), 615 (25), 585 (16), 574 (6), 553 (8), 384 (5), 350 (21),
316 (18); IR (300 K, ATR): n˜ =3073 (vw), 1581 (m), 1482 (w), 1440
(s), 1408 (w), 1376 (w), 1259 (vs), 1224 (w), 1193 (w), 1145 (w), 1110
(m), 1066 (w), 1029 (m), 996 (w), 960 (w), 821 (m), 756 (w), 727 (vs),
687 (s), 635 cmꢀ1 (s); MS (ESI-EM): m/z calcd for C18H20N2P [M+]:
295.13586; found: 295.13559; m/z calcd. for C14H14P [M+ꢀ3,5-dime-
thylpyrazole]: 199.06711; found: 199.06675.
General procedure for preparation of compounds 3a–g[OTf]: A so-
lution of phosphane oxide (3.00 mmol, 2 equiv) in CH2Cl2 (10 mL)
was added to
1.1 equiv) in CH2Cl2 (20 mL). The resulting pale yellow solution was
stirred for 15 h at ambient temperature. Precipitate 4[OTf]2 was fil-
tered off. The reaction of tri-n-butylphosphane oxide required an
additional reaction time of 5 h. After removal of all volatile com-
pounds from the filtrate in vacuo, the targeted pyrazolylphospho-
nium salts 3a–g[OTf] were obtained in quantitative yield.
a suspension of 1[OTf]3 (1.311 g, 1.65 mmol,
Synthesis of 3c[OTf]: M.p. 82–858C; 1H NMR (CD2Cl2, 300 K): d=
7.93 (m, 2H; C9-H), 7.78 (m, 4H; C8-H), 7.70 (m, 4H; C7-H), 6.30
4
4
3
(dq, J(H,P)=2.4 Hz, J(H,H)=0.7 Hz, 1H; C2-H), 3.43 (dq, J(H,H)=
1808
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 1805 – 1812

Conversion of Phosphane Oxides into Pyrazolylphosphonium Salts
1
7.4 Hz, 2J(H,P)=13.2 Hz, 2H; C10-H),
2.32 (s, 3H; C4-H), 1.79 (d, 4J(H,P)=
0.7 Hz, 3H; C5-H), 1.43 ppm (dt,
3J(H,H)=7.4 Hz, 3J(H,P)=21.9 Hz, 3H;
C11-H); 13C NMR (CD2Cl2, 300 K): d=
158.5 (d, 3J(C,P)=12.3 Hz, 1C; C1),
149.3 (d, 2J(C,P)=6.3 Hz, 1C; C3),
136.9 (d, 4J(C,P)=3.1 Hz, 2C; C9),
133.4 (d, 2J(C,P)=11.5 Hz, 4C; C7),
131.1 (d, 3J(C,P)=13.2 Hz, 4C; C8),
1C; C2), 16.0 (d, J(C,P)=54.1 Hz, 3C; C6), 13.7 (s, 1C; C4), 13.0 (s,
1C; C5), 5.6 ppm (d, 2J(C,P)=5.5 Hz, 3C; C7); 31P NMR (CD2Cl2,
300 K): d=79.0 ppm (1P); 19F{1H} NMR (CD2Cl2, 300 K): d=
ꢀ78.9 ppm (s, CF3); Raman (80 mW, 300 K): 2993 (17), 2970 (7),
2.952 (10), 2932 (100), 2.894 (8), 2761 (8), 1576 (5), 1467 (26), 1442
(6), 1392 (7), 1225 (13), 1064 (7), 1032 (77), 978 (9), 758 (28), 650
(22), 584 (32), 522 (18), 441 (7), 348 (37), 315 cmꢀ1 (20); IR (300 K,
ATR): n˜ =2969 (w), 2931 (m), 1612 (vw), 1575 (m), 1467 (w), 1408
(m), 1315 (vw), 1303 (w), 1258 (vs), 1223 (w), 1150 (m), 1085 (w),
1026 (s), 961 (m), 872 (w), 811 (w), 793 (s), 766 (m), 731 (m),
635 cmꢀ1 (m); MS (ESI-EM): m/z calcd for C11H22N2P [M+]:
213.15151; found: 213.15117; m/z calcd for C6H14P [M+ꢀ3,5-dime-
thylpyrazole]: 117.08276; found: 117.08239. Single crystals suitable
for single-crystal X-ray analysis were obtained by slow diffusion of
Et2O into a solution of 3e[OTf] in CH2Cl2.
1
1
121.2 (q, J(C,F)=320.8 Hz, 1C; CF3), 117.1 (d, J(C,P)=94.5 Hz, 2C;
3
1
C6), 114.4 (d, J(C,P)=5.4 Hz, 1C; C2), 21.2 (d, J(C,P)=61.8 Hz, 1C;
2
C10), 13.8 (s, 1C; C4), 13.3 (s, 1C; C5), 6.7 ppm (d, J(C,P)=5.3 Hz,
1C; C11); 31P NMR (CD2Cl2, 300 K): d=51.8 ppm; 19F{1H} NMR
(CD2Cl2, 300 K): d=ꢀ78.9 ppm (s, CF3); Raman (81 mW, 300 K):
3063 (67), 2937 (59), 2899 (8), 1589 (62), 1579 (5), 1443 (13), 1225
(11), 1194 (11), 1163 (9), 1115 (40), 1033 (80), 998 (100), 755 (32),
686 (29), 616 (34), 585 (22), 574 (8), 545 (14), 348 (22), 313 (28),
264 cmꢀ1 (13); IR (300 K, ATR): n˜ =3065 (vw), 2929 (w), 1580 (m),
1440 (m), 1408 (w), 1258 (vs), 1222 (w), 1145 (m), 1118 (w), 1075
(w), 1028 (s), 996 (w), 961 (m), 844 (w), 747 (m), 689 (m), 634 cmꢀ1
(vs); MS (ESI-EM): m/z calcd for C19H22N2P [M+]: 309.1515; found:
309.1512; m/z calcd. for C14H14P [M+ꢀ3,5-dimethylpyrazole]:
213.0828; found: 213.0823, m/z calcd for C8H8P [M+ꢀ3,5-dimethyl-
pyrazoleꢀbenzene]: 135.0358; found: 135.0355.
1
Synthesis of 3f[OTf]: M.p. oil at RT; H NMR (CD2Cl2, 300 K): d=6.20
4
(d, J(H,P)=2.4 Hz, 1H; C2-H), 2.74 (m, 6H; C6-H), 1.54 (m, 6H; C7-
H), 1.50 (m, 6H; C8ꢀH), 0.95 ppm (t, 3J(H,H)=7.1 Hz, 9H; C9-H),
2.48 (3H, s, C5–H), 2.24 (3H, s, C4–H); 13C NMR (CD2Cl2, 300 K): d=
157.4 (d, 3J(C,P)=11.3 Hz, 1C; C1),
147.7 (d, 2J(C,P)=6.5 Hz, 1C; C3),
121.3 (q, 1J(C,F)=320.7, 1C; CF3),
3
113.8 (d, J(C,P)=5.0 Hz, 1C; C2), 23.9
(d, 3J(C,P)=16.7 Hz, 3C; C8), 23.5 (d,
2J(C,P)=4.7 Hz, 3C; C7), 22.6 (d,
1
Synthesis of 3d[OTf]: M.p. liquid at RT; H NMR (CD2Cl2, 300 K): d=
1J(C,P)=52.4 Hz, 3C; C6), 13.4 ppm (s,
7.94 (m, 2H; C12-H), 7.77 (m, 4H; C11-H), 7.68 (m, 4H; C10-H), 6.31
(m, J(H,P)=2.8 Hz, coupling to C4-H and C5-H not resolved, 1H;
3C; C9); 31P NMR (CD2Cl2, 300 K): d=
73.5 ppm (s); 19F{1H} NMR (CD2Cl2,
4
C2-H), 5.84 (m, 1H; C7-H), 5.40–
5.49 (m, 2H; C8-H), 4.29 (m, 1H;
C6), 2.34 (s, 3H; C4-H), 1.74 ppm
(d, 4J(H,H)=1.0 Hz, 3H; C5-H);
13C NMR (CD2Cl2, 300 K): d=158.7
(d, 3J(C,P)=12.6 Hz, 1C; C1),
300 K): d=ꢀ78.9 ppm (s, CF3); Raman
(80 mW, 300 K): 3112 (5), 2937 (100), 2876 (25), 1449 (30), 1302 (7),
1224 (12), 1101 (7), 1051 (8), 1031 (91), 868 (8), 755 (32), 587 (24),
574 (8), 348 (21), 312 cmꢀ1 (16); IR (300 K, ATR): n˜ =2963 (w), 2935
(vw), 2875 (w), 2361 (w), 1577 (w), 1465 (w), 1408 (w), 1256 (vs),
1222 (w), 1150 (s), 1099 (vw), 1079 (w), 1028 (s), 967 (w), 903 (w),
832 (w), 778 (vw), 755 (w), 718 (w), 636 (vs), 573 (w), 516 (m),
449 cmꢀ1 (w); MS (ESI-EM): m/z calcd for C17H34N2P [M+]: 297.2454;
found: 297.2445; m/z calcd. for C12H26P [M+ꢀ3,5-dimethylpyrazole]:
201.1767; found: 201.1760; m/z calcd for C8H18P [M+ꢀ3,5-dimethyl-
pyrazoleꢀbutene]: 145.1141; found: 145.1136.
2
149.6 (d, J(C,P)=6.1 Hz, 1C; C3),
137.1 (d, 4J(C,P)=3.0 Hz, 2C;
C12), 133.7 (d, 2J(C,P)=10.7 Hz,
4C; C10), 131.1 (d, 3J(C,P)=
13.7 Hz, 4C; C11), 127.1 (d, 3J(C,P)=14.6 Hz, 1C; C8), 122.8 (d,
1
2J(C,P)=10.0 Hz, 1C; C7), 121.1 (q, J(C,F)=322.1 Hz, 1C; CF3), 116.9
1
3
Synthesis of 3g[OTf]: M.p. 49–528C; 1H NMR (CD2Cl2, 300 K): d=
(d, J(C,P)=94.2 Hz, 2C; C9), 114.6 (d, J(C,P)=5.3 Hz,1C; C2), 32.7
1
(s, J(C,P)=60.8 Hz, 1C; C6), 13.9 (s, 1C; C4), 13.3 ppm (s, 1C; C5);
7.83 (m, 2H; C9-H), 7.67 (m, 4H; C8-H), 7.62 (m, 4H; C7-H), 6.58
31P NMR (CD2Cl2, 300 K): d=45.2 ppm (s); 19F{1H} NMR (CD2Cl2,
300 K): d=ꢀ78.9 ppm (s, CF3); Raman (100 mW, 300 K): 3067 (34),
2934 (25), 1637 (18), 1589 (36), 1440 (9), 1307 (5), 1224 (7), 1165
(8), 1111 (15), 1031 (61), 999 (100), 756 (19), 701 (11), 616 (15), 586
(16), 348 (17), 313 (14), 252 cmꢀ1 (7); IR (300 K, ATR): n˜ =2930 (w),
2361 (w), 1580 (w), 1440 (m), 1408 (w), 1258 (vs), 1222 (w), 1146
(m), 1118 (w), 1076 (vw), 1027 (vs), 996 (w), 962 (m), 834 (w), 742
(m), 727 (w), 688 (m), 645 (vs), 571 (vw), 551 (w), 516 (m), 500 (vw),
450 cmꢀ1 (vw); MS (ESI-EM): m/z calcd for C20H22N2P [M+]:
321.1515; found: 321.1497; m/z calcd for C15H14P [M+ꢀ3,5-dime-
thylpyrazole]: 225.0828; found: 225.0812.
2
4
4
(ddt, pseudo sext, J(H,P)=30.6 Hz, J(H,H)=1.5 Hz, J(H,H)=1.5 Hz,
1H; C10-H), 6.23 (dt, 4J(P,H)=
2.4 Hz, 4J(H,H)=0.7 Hz, 1H; C2-H),
3.71
(dddd,
2J(H,H)=16.3 Hz,
3J(H,H)=8.3 Hz,
2J(H,P)=16.3 Hz,
3J(H,H)=2.7 Hz, 1H; C13-H), 3.34
3
(dddd, broad, n1/2 =6.2 Hz, J(H,P)=
17.8 Hz, 2J(H,H)=17.8 Hz 3J(H,H)=
9.2 Hz, second 3J(H,H) is not re-
solved, 1H; C12-H), 3.22 (m, 1H;
C12-H), 3.13 (dddd, 2J(H,H)=
16.3 Hz, 2J(H,P)=6.7 Hz, 3J(H,H)=9.2 Hz, 3J(H,H)=4.6 Hz, 1H; C13-
H), 2.37 (s, 3H, C14-H), 2.29 (s, 3H, C4-H), 2.18 ppm (d 4J(H,H)=
0.7 Hz, 3H; C5-H); the assignment of the coupling constants was
Synthesis of 3e[OTf]: M.p. 55–588C; 1H NMR (CD2Cl2, 300 K): d=
4
6.20 (d, J(H,P)=2.4 Hz, 1H; C2-H), 2.80 (dq,
3J(H,H)=7.6 Hz, 2J(H,P)=11.8 Hz, 6H; C6-H),
2.49 (s, 3H; C5-H), 2.24 (s, 3H; C4-H),
1.28 ppm (dt, 3J(H,H)=7.6 Hz, 3J(H,P)=
20.3 Hz, 9H; C7-H); 13C NMR (CD2Cl2, 300 K):
d=157.7 (d, 3J(C,P)=11.2 Hz, 1C; C1), 148.0
confirmed by a H{31P} spectrum; 13C NMR (CD2Cl2, 300 K): d=183.2
1
(d, 2J(C,P)=27.5 Hz, 1C; C14), 158.1 (d, 3J(C,P)=12.2 Hz, 1C; C1),
148.0 (d, 2J(C,P)=7.3 Hz, 1C; C3), 136.6 (d, 4J(C,P)=3.4 Hz, 1C;
2
3
C19), 132.4 (d, J(C,P)=12.9 Hz, 2C; C7), 130.8 (d, J(C,P)=14.5 Hz,
2C; C8), 121.3 (q, 1J(C,F)=321.1 Hz, 1C; CF3), 120.6 (d, 1J(C,P)=
97.8 Hz, 1C; C6), 113.5 (d, 3J(C,P)=5.3 Hz, 1C; C2), 107.2 (d,
1J(C,P)=95.9 Hz, 1C; C10), 36.7 (d, 2J(C,P)=8.4 Hz, 1C; C12), 24.1
2
1
(d, J(C,P)=6.2 Hz, 1C; C3), 121.2 (q, J(C,F)=
321.1 Hz, 1C; CF3), 113.8 (d, 4J(C,P)=5.2 Hz,
ChemSusChem 2011, 4, 1805 – 1812
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
1809

J. J. Weigand et al.
(d, 1J(C,P)=65.1 Hz, 1C; C13), 22.3 (d, 3J(H,P)=18.5 Hz, 1C; C14),
13.8 (s, 1C; C4), 12.8 ppm (s, 1C; C5); 31P NMR (CD2Cl2, 300 K): d=
77.3 ppm; 19F{1H} NMR (CD2Cl2, 300 K): d=ꢀ78.9 ppm (s, CF3);
Raman (300 mW, 300 K): 3107 (10), 3063 (53), 3002 (12), 2929 (69),
1591 (60), 1576 (8), 1480 (7), 1442 (22), 1396 (6), 1335 (6), 1303 (7),
1223 (10), 1126 (21), 1031 (86), 998 (100), 778 (7), 756 (31), 714
(18), 615 (22), 598 (13), 574 (16), 510 (6), 456 (6), 391 (6), 348 (20),
328 (7), 314 (23), 285 (8), 263 (14), 228 cmꢀ1 (7); IR (300 K, ATR): n˜ =
3082 (w), 2930 (vw), 1593 (vw), 1577 (m), 1541 (vw), 1440 (m), 1410
(w), 1375 (w), 1333 (vw), 1260 (vs), 1222 (w), 1199 (vw), 1152 (m),
1124 (w), 1082 (w), 1028 (s), 997 (vw), 963 (m), 922 (w), 892 (m),
855 (w), 811 (w), 776 (w), 756 (s), 727 (w), 696 (w), 635 cmꢀ1 (s);
MS (ESI-EM): m/z calcd for C16H20N2P [M+]: 271.1359; found:
271.1349; m/z calcd. for C11H12P [M+ꢀ3,5-dimethylpyrazole]:
175.0671; found: 175.0664.
with 1,2-difluorobenzene, the colourless residue (0.418 g) was iden-
tified as analytically pure 6[OTf]2 by means of multinuclear NMR
spectroscopy experiments and elemental analysis. The filtrate was
reduced to approximately 30% of its original volume. Addition of
Et2O resulted in the formation of a colourless precipitate. This ma-
terial was isolated dissolved in CH2Cl2 and recrystallized by slow
addition of Et2O to yield a second batch of 6[OTf]2 (0.304 g). The
formation of trifluoromethanesulfonic anhydride was confirmed by
a
19F NMR spectroscopy investigation of the supernatant of the re-
action mixture (1,2-difluorobenzene/C6D6 capillary, 300 K). A singlet
observed at d=ꢀ73.0 ppm increased in intensity upon addition of
a
commercial sample of trifluoromethanesulfonic anhydride
(0.02 mL). Yield of 6[OTf]2 86% (0.722 mg); m.p.>2958C
(decomp.); 1H NMR (CD2Cl2, 300 K): d=7.96 (m, 6H; C4-H),
7.72 ppm (m, 24H; C2-H, C3-H); 13C NMR (CD2Cl2, 300 K): d=138.7
2
2
(s, 6C; C4), 134.9 (d, J(C,P)=11.9 Hz, 12C; C2), 131.6 (d, J(C,P)=
1
Preparation of TPP from 3a[OTf]: LiAlH4 (0.19 g, 5.00 mmol) was
added to a suspension of 3a[OTf] (2.53 g, 5.00 mmol) in Et2O
(50 mL). The resulting suspension was stirred for 48 h and filtered.
H2O (15 mL) was added to the filtrate and the aqueous phase was
extracted with Et2O (2ꢃ5 mL). The combined organic layers were
extracted with HCl (1m, 10 mL), separated, and dried over MgSO4.
All volatile compounds were removed in vacuo. The colourless
solid obtained (90%, 1180 mg) showed exclusively the resonances
14.1 Hz, 12C; C3), 121.2 (q, J(C,F)=320.5 Hz, 6C; CF3), 116.1 ppm
(d, 1J(C,P)=104.0 Hz, 6C1); 31P{1H} NMR (CD2Cl2, 300 K): d=
75.0 ppm (s, 2P); 19F{1H} NMR (CD2Cl2, 300 K): d=ꢀ78.7 ppm (s, 3F;
CF3); Raman (80 mW, 300 K): 3068 (52), 1587 (65), 1225 (6), 1.170
(9), 1.121 (6), 1108 (29), 1.031 (40), 1001 (100), 755 (14), 638 (12),
614 (17), 573 (6), 349 (12), 315 (13), 283 (7), 258 (16), 232 (38),
195 cmꢀ1 (6); IR (300 K, ATR): n˜ =3067 (vw), 1586 (w), 1440 (w),
1258 (vs), 1223 (w), 1189 (vw), 1152 (m), 1119 (s), 1104 (vw), 1044
(w), 1028 (vs), 991 (m), 735 (vs), 682 (m), 635 (vs), 572 (w), 525 (m),
516 cmꢀ1 (w).
1
expected for Ph3P in 31P and H NMR spectra.
Preparation of 5[OTf] from 3a[OTf]: HOTf (29.5 mL, 0.050 g,
0.33 mmol, 1 equiv) and Ph2PH (58.0 mL, 0.062 g, 0.33 mmol,
1 equiv) were added successively to a solution of 3a[OTf] (0.169 g,
0.33 mmol, 1 equiv) in CH2Cl2 (5 mL). The mixture was stirred over-
night and all volatile compounds were removed in vacuo. The pale
yellow solid obtained was dissolved in CH2Cl2 and precipitated by
slow addition of Et2O. The resulting suspension was filtered and
the precipitate washed with Et2O (3ꢃ2 mL) to give 5[OTf] (79%,
157 mg). 1H, 31P, and 31P{1H} NMR spectroscopy investigations
(CD2Cl2, 300 K) showed exclusively the resonances that were previ-
ously reported for 5[OTf].[32]
Preparation of 7 from 3a[OTf] and subsequent Wittig reaction to
8: n-Butyllithium (1.6m in n-hexane; 5 mL, 8.00 mmol, 2 equiv) was
slowly added to a suspension of 3a[OTf] (2.027 g, 4.00 mmol,
1 equiv) in Et2O (100 mL) at ꢀ958C. The mixture was allowed to
warm to ambient temperature and was stirred for 5 h. The 31P NMR
spectrum of the resulting bright orange solution was recorded and
showed only one singlet at d=13.4 ppm (Et2O/C6D6 capillary,
300 K; see Figure S2.1 in the Supporting Information) and was con-
sistent with a previously reported chemical shift for 7.[34] A solution
of 2-nitrobenzaldehyde (0.605 g, 4.00 mmol, 1 equiv) in Et2O
(10 mL) was added to the ylide solution and the mixture was
stirred overnight. The resulting brown suspension was treated with
aqueous HCl (1m, 3ꢃ30 mL). After separation of the organic layer
and drying with MgSO4, Et2O was removed in vacuo and the resi-
due subjected to column chromatography on silica gel (EtOAc/n-
pentane=5:95), which gave cis/trans-7 (see Figure S2.2 in the Sup-
porting Information)[35] as a 70:30 isomeric mixture (74%, 567 mg).
Preparation of 6[OTf]2 from 3a[OTf], Method A: A mixture of
Ph3PO (0.417 g, 1.50 mmol, 1 equiv) and 3a[OTf] (0.760 g,
1.50 mmol, 1 equiv) were dissolved in CH2Cl2 (15 mL) and HOTf
(0.266 mL, 0.450 g, 3.00 mmol,
2 equiv) was added dropwise. The
mixture was stirred overnight at
ambient temperature and re-
duced to two thirds of its original
volume in vacuo. Slow addition of
Et2O gave a colourless precipitate,
which was filtered off and washed
with Et2O (2ꢃ3 mL) and dried in
vacuo. As characterization data for
X-ray data collection and reduction: Suitable single crystals for
3a[OTf], 3c[OTf], 3e[OTf] were obtained by diffusion of n-hexane
into solutions of the compound in a minimum amount of CH2Cl2 at
ꢀ308C. The crystals were coated with Paratone-N oil, mounted by
using a glass fibre pin, and frozen in the cold nitrogen stream of
the goniometer. X-ray diffraction data for all compounds were col-
lected on a Bruker AXS APEX CCD diffractometer equipped with a
rotation anode at 153(2) K using graphite-monochromated CuKa
radiation (l=1.54178 ꢁ). Diffraction data were collected over the
full sphere with a scan width of 0.58 and exposure times of 20/40 s
for 3a[OTf], 30/60 s for 3c[OTf], and 5/10 s for 3e[OTf]. The genera-
tor settings were 45 kV and 110 mA. Diffraction data were collected
over the full sphere and the frames were integrated by using the
Bruker SMART[36] software package with the narrow frame algo-
rithm. Data were corrected for absorption effects by using the
SADABS routine (empirical multiscan method). For further crystal
and data collection details, see Table 1.
6[OTf]2 has never been comprehensively reported, it is given
below. X-ray structural analysis of suitable crystals obtained by
slow diffusion of n-hexane into a solution of 6[OTf]2 in CH2Cl2 at
ꢀ358C gave comparable results (a=8.6199(4), b=11.1712(5), c=
19.5323(8) ꢁ; a=94.6070(10), b=93.0320(10), g=98.9920(10)8; V=
1847.64(14) ꢁ3) to those previously reported (a=8.620(2), b=
11.193(3), c=19.584(6) ꢁ; a=94.669(5), b=92.894(5), g=
99.068(7)8; V=1856.0(9) ꢁ3).[33]
Preparation of 6[OTf]2 from 3a[OTf], Method B: HOTf (0.354 mL,
0.600 g, 4.00 mmol, 2 equiv) was slowly added to a solution of
3a[OTf] (1.013 g, 2.00 mmol, 1 equiv) in 1,2-difluorobenzene
(10 mL). The reaction mixture was stirred for 24 h, resulting in the
formation of a colourless precipitate. After filtration and washing
Structure elucidation and refinement: Atomic scattering factors for
non-hydrogen elements were taken from the literature. The struc-
1810
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 1805 – 1812

Conversion of Phosphane Oxides into Pyrazolylphosphonium Salts
Table 1. Crystallographic data and structure refinements of compounds
3a[OTf], 3c[OTf], and 3f[OTf].
Keywords: phosphorus · phosphanes · synthetic methods ·
structure elucidation · green chemistry
3a[OTf]
3c[OTf]
3e[OTf]
[1] a) W. Reif, H. Grassner, Chemie-Ing.-Tech. 1973, 45, 646–652; b) C. Christ,
Production Integrated Environmental Protection and Waste Management
in the Chemical Industry, Wiley-VCH, Weinheim, 1999, pp. 90–93; c) J. D.
Surmantis, US patent 3600473, 1971; d) H. Pommer, Angew. Chem.
1977, 89, 437–443; Angew. Chem. Int. Ed.1977, 16, 423–429.
[2] D. B. Malpass, G. S. Yeargin, German Patent 2714721, 1977.
[3] a) C. J. O’Brien, J. L. Tellez, Z. S. Nixon, L. J. Kang, A. L. Carter, S. R.
Kunkel, K. C. Przeworski, G. A. Chass, Angew. Chem. 2009, 121, 6968–
6971; Angew. Chem. Int. Ed. 2009, 48, 6836–6839; b) S. P. Marsden, Nat.
Chem. 2009, 1, 685–687.
Formula
Molecular weight
Colour, habit
C24H22F3N2O3PS C20H22F3N2O3PS C12H22F3N2O3PS
506.47
458.43
362.35
colourless,
block
triclinic
colourless,
irregular
monoclinic
P21/c
colourless, plate
Crystal system
Space group
a [ꢁ]
b [ꢁ]
c [ꢁ]
a [8]
b [8]
triclinic
¯
P1
¯
P1
8.8402(1)
10.2945(2)
26.1495(4)
99.395(1)
97.473(1)
92.098(1)
2323.81(6)
4
8.8201(1)
11.4535(1)
21.3641(2)
90
93.729(1)
90
8.4915(1)
8.4969(1)
12.2513(2)
81.026(1)
85.299(1)
81.743(1)
862.47(2)
2
[4] H. Mayer, Pure Appl. Chem. 1994, 66, 931–938.
[5] For a general introduction, see: a) J. W. Hartley, The Chemistry of Orga-
nophosphorus Compounds, Vol. 2, Phosphine Oxides, Sulphides, Selenides
and Tellurides (Ed.: F. R. Hartley), 1992, Wiley, Chichester; for selected
laboratory methods for reducing phosphane oxides by using the indi-
cated reagents, see: b) H.-C. Wu, J.-Q. Yu, J. B. Spencer, Org. Lett. 2004,
6, 4675–4678 (Ph3P/HSiCl3 and (EtO)3P/HSiCl3); c) T. Imamoto, S. Kikuchi,
T. Miura, Y. Wada, Org. Lett. 2001, 3, 87–90 (MeOTf/LiAlH4); d) S. Griffin,
L. Heath, P. Wyatt, Tetrahedron Lett. 1998, 39, 4405–4406 (AlH3·THF);
e) C. A. Busacca, R. Raju, N. Grinberg, N. Haddad, P. James-Jones, H. Lee,
J. C. Lorenz, A. Saha, C. H. Senanayake, J. Org. Chem. 2008, 73, 1524–
1531 [diisobutylaluminium hydride (DIBAL-H)]; f) T. Imamoto, T. Takeya-
ma, T. Kusumoto, Chem. Lett. 1985, 1491–1492 (CeCl3/LiAlH4); g) E.
Cernia, G. M. Marcati, W. Marconi, N. Palladino, Inorg. Chim. Acta 1974,
11, 195–200 (aminoalanes); h) M. Zablocka, B. Delest, A. Igau, A. Skow-
ronska, J.-P. Majoral, Tetrahedron Lett. 1997, 38, 5997–6000 (Schwartz’
reagent); i) Y. Handa, J. Inanaga, M. Yamaguchi, J. Chem. Soc. Chem.
Commun. 1989, 298–299 (SmI2/HMPA); j) F. Mathey, R. Maillet, Tetrahe-
dron Lett. 1980, 21, 2525–2526 (Cp2TiCl2/Mg); k) M. Berthod, A. Favre-
Rꢄguillon, J. Mohamed, G. Mignani, G. Docherty, M. Lemaire, Synlett
2007, 1545–1548 (Ti(OiPr)/polymethylhydrosiloxanes); l) J. L. Lecat, M.
Devaud, Tetrahedron Lett. 1987, 28, 5821–5822 (Me2SO4/electrochemi-
cally); for selected patented methods for reducing phosphane oxides
directly, i.e. without transformation into dichlorophosphoranes, see:
m) T. Wettling, European Patent 0548682, 1992 (H2, catalyst: Bi/TiO2);
n) T. Dockner, German Patent 3523320, 1985 (paraffin oil/activated char-
coal); o) F. Frey, J. Y. Lee, US patent 4507503, 1983 (trialkyalane/trialkyl-
borane); p) Ref. [2] (DIBAL-H).
g [8]
V [ꢁ3]
2153.65(4)
4
Z
T [K]
153(1)
153(1)
153(1)
Crystal size [mm]
0.08ꢃ0.08ꢃ0.07 0.08ꢃ0.07ꢃ0.04 0.22ꢃ0.12ꢃ0.04
1calcd [gcmꢀ3
F(000)
]
1.448
1048
1.414
952
1.395
380
l [ꢁ]
qmin/max[8]
Index range
1.54178 (CuKa) 1.54178 (CuKa) 1.54178 (CuKa)
1.73/68.24
ꢀ10ꢂhꢂ10
ꢀ12ꢂkꢂ12
ꢀ30ꢂlꢂ31
2.371
4.15/68.24
ꢀ10ꢂhꢂ9
ꢀ12ꢂkꢂ10
ꢀ25ꢂlꢂ22
2.491
3.66/68.18
ꢀ9ꢂhꢂ10
ꢀ9ꢂkꢂ10
ꢀ14ꢂlꢂ14
2.944
m [mmꢀ1
]
Absorption
correction
Reflections
collected
Reflections unique
Rint
SADABS
SADABS
SADABS
13584
9856
5056
7582
3477
2810
0.0332
6481
0.0415
2851
0.0320
2494
Reflections
observed, F>2s(F)
Residual density
0.361/ꢀ0.398
0.400/ꢀ0.353
0.378/ꢀ0.380
[eꢁꢀ3
]
Parameters
GooF
R1, I>2s(I)
wR2, all data
CCDC
617
274
204
1.063
0.0407
0.1156
828014
1.053
0.0428
0.1229
828015
1.079
0.0457
0.1315
828016
[6] L. D. Quin, A Guide to Organophosphorus Chemistry, 2000, Wiley, New
York, p. 95 and p. 11.
[7] G. Wittig, H. Pommer, German Patent 954247, 1956.
[8] a) K. Akiba, K. Okada, K. Ohkata, Tetrahedron Lett. 1986, 27, 5221–5224;
b) R. Appel, W. Heinzelmann, German patent 1192205, 1965.
[9] M. V. Kazantseva, B. V. Timokhin, A. V. Rokhin, D. G. Blazhev, A. I. Golubin,
Y. V. Rybakova, Russ. J. Gen. Chem. 2001, 71, 1233–1235.
[10] a) G. Mꢅrkl, D. E. Fischer, H. Olbrich, Tetrahedron Lett. 1970, 11, 645–
648; b) J. G. Natoli, German Patent 1259883, 1965.
[11] a) M. Masaki, K. Fukui, Chem. Lett. 1977, 151–152; b) T. Yano, M. Kuro-
boshi, H. Tanaka, Tetrahedron Lett. 2010, 51, 698–701; c) T. Yano, M.
Hoshino, M. Kuroboshi, H. Tanaka, Synlett 2010, 801–803.
[12] a) G. Wunsch, K. Wintersberger, H. Geierhaas, Z. Anorg. Allg. Chem. 1969,
369, 33–37; this transformation is also relevant for industrial processes;
the reduction of phosphane oxides to phosphanes is commonly ach-
ieved by transformation to the significantly more reactive dichlorophos-
phoranes and subsequent reduction, e.g.: ref. [5m]; b) G. Wunsch, K.
Wintersberger, H. Geierhaas, German patent 1247310, 1967; c) H.
Tanaka, T. Yano, K. Kobayashi, S. Kamenoue, M. Kuroboshi, H. Kawakubo,
Synlett 2011, 582–584.
ture was elucidated by using direct methods as implemented in
the SHELXS-97 package[37] and were refined with SHELXL-97[38]
against F2 by using first isotropic and later anisotropic thermal pa-
rameters for all non-hydrogen atoms. Hydrogen atom positions
were calculated and allowed to ride on the carbon atom to which
they were bonded by assuming a CꢀH bond length of 0.95 ꢁ. Hy-
drogen atom temperature factors were fixed at 1.20 times the iso-
tropic temperature factor of the carbon atom to which they were
bonded. The locations of the largest peaks in the calculated final
difference Fourier map, as well as the magnitude of the residual
electron densities in each case, were of no chemical significance.
CCDC 828014 (3a[OTf]), 828015 (3c[OTf]), and 828016 (3e[OTf])
contain the supplementary crystallographic data for this paper.[39]
[13] E. Deschamps, F. Mathey, J. Chem. Soc. Chem. Commun. 1984, 1214–
1215.
Acknowledgements
[14] C. C. Santini, J. Fischer, F. Mathey, A. Mitschler, J. Am. Chem. Soc. 1980,
102, 5809–5815.
[15] A. M. Z. Slawin, D. J. Williams, P. T. Wood, J. D. Woollins, J. Chem. Soc.
Chem. Commun. 1987, 1741.
[16] a) T. W. Campbell, J. J. Verbanc, USA Patent 2853473, 1958; b) T. W.
Campbell, J. J. Monagle, V. S. Foldi, J. Am. Chem. Soc. 1962, 84, 3673–
We gratefully acknowledge funding from the FCI (fellowship for
K.-O.F.), the DFG (WE 4621/2-1), and the European PhosSciNet
(CM0802). J.J.W. thanks Prof. F. E. Hahn for support and Prof. Dr.
R. Wolf and T. W. Kemnitzer for helpful discussions.
ChemSusChem 2011, 4, 1805 – 1812
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
1811

J. J. Weigand et al.
3677; c) J. J. Monagle, T. W. Campbell, H. F. McShane, Jr., J. Am. Chem.
Soc. 1962, 84, 4288–4295.
[17] H. Hoffmann, H. Fçrster, G. Tor-Poghossian, Monatsh. Chem. 1969, 100,
311–315.
[29] All calculations were performed at the B3LYP/6-311G(2d) level of theory
by using Gaussian 09. See the Supporting Information for further details
and full citation.
[30] The structural changes observed on protonation in the gas phase are
very likely to be translated into solution because the quantitative gas-
phase proton affinity will have little relevance to the behaviour in
CH2Cl2. However, the absolute value is strongly dependent on a solvate
model and we refrain here from a detailed calculation.
[18] E. Ciganek, J. Org. Chem. 1970, 35, 1725–1729.
[19] J. J. Weigand, K.-O. Feldmann, A. K. C. Echterhoff, A. W. Ehlers, K. Lam-
mertsma, Angew. Chem. 2010, 122, 6314–6317; Angew. Chem. Int. Ed.
2010, 49, 6178–6181.
[31] a) See the Supporting Information for details on natural bond orbital
(NBO) analyses of 3a+ and 3aH2+, calculated by using NBO Version 3.1,
developed by E. D. Glendening, A. E. Reed, J. E. Carpenter, and F. Wein-
hold; b) A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899.
[32] Direct reduction of 2a with LiAlH4 typically gives yields of 60–80%; The
Chemistry of Organophosphorus Compounds, Vol. 2, Phosphine Oxides,
Sulphides, Selenides and Tellurides (Ed.: F. R. Hartley), Wiley, Chichester,
1992, p. 292.
[20] A slight excess of 1[OTf]3 remains in the precipitate and can also be hy-
drolyzed quantitatively with water to yield 3,5-dimethylpyrazolium tri-
flate and P4O6, which can be easily separated. Accordingly, oxidation of
P4O6 with O2 yields P4O10, which can be hydrolyzed to H3PO4.
[21] 2a 31P{1H} NMR shift, d=26.0 ppm, was obtained in dry CD2Cl2 at 300 K;
varying chemical shifts have been reported for 2a in the literature, for
example, a) d=23.7 ppm (in CDCl3): H. Rao, Y. Jin, H. Fu, Y. Jiang, Y.
Zhao, Chem. Eur. J. 2006, 12, 3636–3646; b) d=29.8 ppm (in CDCl3): F.-
H. Hu, L.-S. Wang, S.-F. Cai, J. Chem. Eng. Data 2009, 54, 1382–1384.
[22] J. Grebe, G. K. Harms, F. Weller, K. Dehnicke, Z. Anorg. Allg. Chem. 1995,
621, 1489–1495.
[33] S.-L. You, H. Razavi, J. W. Kelly, Angew. Chem. 2003, 115, 87–89; Angew.
Chem. Int. Ed. 2003, 42, 83–85.
[34] S. Grim, J. H. Ambrus, J. Org. Chem. 1968, 33, 2993–2999.
[35] R. J. Sundberg, J. Org. Chem. 1965, 30, 3604–3609.
[36] a) SAINT 7.23A, Bruker AXS, Inc, Madison, Wisconsin (USA), 2006; b)
G. M. Sheldrick, SADABS, Bruker AXS, Inc., Madison, Wisconsin (USA),
2004.
[23] E. E. Nifantyev, M. K. Gratchev, S. Y. Burmistrov, L. K. Vasyanina, M. Y. Anti-
pin, Y. T. Struchkov, Tetrahedron 1991, 47, 9839–9860.
[24] I. M. Dixon, E. Lebon, G. Loustau, P. Sutra, L. Vendier, A. Igau, A. Juris,
Dalton Trans. 2008, 5627–5635.
[37] G. M. Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467.
[38] G. M. Sheldrick, SHELXL-97; Universitꢅt Gçttingen, Gçttingen (Germany),
1997.
[39] These data can be obtained free of charge from the Cambridge Crystal-
lographic Data Centre at www.ccdc.cam.ac.uk/data_request/cif.
[25] J. J. Weigand, N. Burford, A. Decken, A. Schulz, Eur. J. Inorg. Chem. 2007,
4868–4872.
[26] S. Fischer, J. Hoyano, L. K. Peterson, Can. J. Chem. 1976, 54, 2710.
[27] See the Supporting Information for further details.
[28] N. Burford, D. E. Herbert, P. J. Ragogna, R. McDonald, M. J. Ferguson, J.
Am. Chem. Soc. 2004, 126, 17067; N. Burford, P. J. Ragogna, R. McDo-
nald, M. J. Ferguson, J. Am. Chem. Soc. 2003, 125, 14404–14407.
Received: August 28, 2011
1812
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 1805 – 1812