The prototypical organophosphorus ylidion ·CH2PH3+

Article abstract of DOI:10.1021/ja9616342

The reactivity of the prototypical phosphorus-containing ylidion (α-distonic ion) ·CH₂PH₃⁺ has been investigated in the gas phase by using a dual cell Fourier-transform ion cyclotron resonance mass spectrometer. The ion ·CH₂PH₃⁺ and its more stable conventional isomer CH₃PH₂·⁺ show distinctly different reactivities toward neutral reagents. This observation contrasts the facile interconversion of the analogous sulfur- and oxygen-containing distonic ions ·CH₂SH₂⁺ and ·CH₂OH₂⁺ with their conventional isomers CH₃OH·⁺ and CH₃SH·⁺, respectively, within collision complexes in the gas phase. Bracketing experiments yield a proton affinity of 190.4 ± 3 kcal mol^-1 for the phosphorus atom in ·CH₂PH₂. Together with a calculated heat of formation for ·CH₂PH₂, this value yields a heat of formation of 217 ± 3 kcal mol^-1 (at 298 K) for the distonic ion ·CH₂PH₃⁺.

Full text of DOI:10.1021/ja9616342

J. Am. Chem. Soc. 1996, 118, 11893-11897
11893
•
+
The Prototypical Organophosphorus Ylidion CH2PH3
Andreas Schweighofer,† Phillip K. Chou,‡ Kami K. Thoen,‡
Vajira K. Nanayakkara,‡ Helmut Keck,† Wilhelm Kuchen,† and
Hilkka I. Kentta1maa*,‡
Contribution from the Department of Inorganic and Structural Chemistry I,
Heinrich-Heine-UniVersity Du¨sseldorf, 40225-Du¨sseldorf, Germany, and
Department of Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907-1393
ReceiVed May 15, 1996X
•
+
Abstract: The reactivity of the prototypical phosphorus-containing ylidion (R-distonic ion) CH2PH3 has been
investigated in the gas phase by using a dual cell Fourier-transform ion cyclotron resonance mass spectrometer. The
ion •CH2PH3+ and its more stable conventional isomer CH3PH2•+ show distinctly different reactivities toward neutral
reagents. This observation contrasts the facile interconversion of the analogous sulfur- and oxygen-containing distonic
ions •CH2SH2+ and •CH2OH2+ with their conventional isomers CH3OH•+ and CH3SH•+, respectively, within collision
complexes in the gas phase. Bracketing experiments yield a proton affinity of 190.4 ( 3 kcal mol-1 for the phosphorus
•
•
atom in CH2PH2. Together with a calculated heat of formation for CH2PH2, this value yields a heat of formation
•
+
of 217 ( 3 kcal mol-1 (at 298 K) for the distonic ion CH2PH3
.
•
Introduction
are greater than that of CH2OH (166 kcal mol-1),7 likely
•
+
because of interconversion of CH2OH2 and CH3OH•+ Via
intramolecular 1,2-hydrogen migration catalyzed by the neutral
Distonic radical cations1 (ionized ylides, zwitterions, and
biradicals) have been the focus of numerous recent experimental
gas-phase studies and theoretical investigations.2 How-
ever, examination of the reactivity of distonic ions containing
second-row elements is currently limited to some distonic
isomers of trimethyl and triethyl phosphate radical cations,3 i.e.,
the ions •CH2OP+(OH)(OCH3)2 and •CH2CH2OP+(OH)(OCH2-
•
+
reagent.7 In contrast, CH2SH2 reacts with many neutral
reagents by electron abstraction. This reaction is thought6 to
occur by a stepwise mechanism involving proton transfer from
the ion to the neutral reagent molecule, followed by hydrogen
atom transfer or proton transfer accompanied by electron transfer
within the collision complex. Further, compounds with ioniza-
tion energies greater than 9.9 eV and proton affinities greater
than 180 kcal mol-1 were found to induce base-catalyzed
CH3)2, and the ions •CH2CH2OP+(OCH3)3,4 •CH2S+(CH3)2,5 and
+ 6
•CH2SH2
.
The prototypical R-distonic ions or ylidions containing
•
+
isomerization of the ylidion CH2SH2 to the more stable
conventional isomer CH3SH•+.6 These results are in keeping
with the theoretical prediction that second-row distonic ions with
a carbon radical center are generally less stable than their
conventional counterparts.8
The prototypical phosphorus-containing ylidion •CH2PH3+ has
been generated by electron ionization-induced McLafferty
rearrangement of n-hexylphosphine (Scheme 1), and structurally
second-row elements (e.g., CH2PH3 and CH2SH2+) are of
special interest since the knowledge concerning their properties
provides the basis for the understanding of more complex
second-row distonic ions. A recent study6 on the gas-phase
•
+
•
•
+
reactions of CH2SH2 in a Fourier-transform ion cyclotron
resonance (FT/ICR) mass spectrometer revealed interesting
reactivity that is quite different from that reported7 for the
oxygen-containing ylidion CH2OH2+. The latter ion reacts
•
predominantly by proton transfer and hydrogen atom abstrac-
tion.7 Moreover, it shows reactivity virtually identical to that
of CH3OH•+ toward neutral molecules whose proton affinities
Scheme 1
electron
CH3(CH2)4CH2PH2
8
† Heinrich-Heine-University Du¨sseldorf.
impact
‡ Purdue University.
•+
+
CH3(CH2)4CH2PH2
8 •CH2PH3
X Abstract published in AdVance ACS Abstracts, September 15, 1996.
(1) Yates, B. F.; Bouma, W. J.; Radom, L. Tetrahedron 1986, 42, 6225.
(2) (a) Hammerum, S. Mass Spectrom. ReV. 1988, 7, 123. (b) Smith, R.
L.; Chou, P. K.; H. I. Kentta¨maa In The Structure, Energetics, and Dynamics
of Ions: Experiments and Theory; Baer, T., Ed.; John Wiley and Sons:
New York, in press. (c) Drewello, Heinrich, N.; Maas, W. P. M.; Nibbering,
N. M. M.; Weiske, T.; Schwarz, H. J. Am. Chem. Soc. 1987, 109, 4810.
(d) Stirk, K. M.; Kiminkinen, L. K. M.; Kentta¨maa, H. Chem. ReV. 1992,
92, 1649.
-C5H10
characterized by collision-activated dissociation (CAD).9 The
+
CAD spectrum of •CH2PH3 features intense structurally
diagnostic fragment ion peaks at m/z 34 (PH3•+) and m/z 14
(CH2•+). In contrast, the dissociation product distribution
obtained upon collisional activation of CH3PH2•+ (generated by
electron ionization of methylphosphine) yields abundant signals
at m/z 15 (CH3+) and m/z 33 (PH2+).9 Further, neutralization-
reionization experiments involving •CH2PH3+ have demonstrated
that this ion has a lifetime of at least several microseconds and
that it is a suitable precursor for the generation of the hitherto
(3) Zeller, L.; Farrell, J., Jr.; Vainiotalo, P.; Kentta¨maa, H. I. J. Am. Chem.
Soc. 1992, 114, 1205.
(4) Kiminkinen, L. K. M.; Stirk, K. G.; Kentta¨maa, H. I. J. Am. Chem.
Soc. 1992, 114, 2027.
(5) (a) van Amsterdam, M. W.; Zappey, H. W.; Ingemann, S.; Nibbering,
N. M. M. Org. Mass Spectrom. 1993 28, 30. (b) Smith, R. L.; Chyall, L.
J.; Stirk, K. M.; Kentta¨maa, H. I. Org. Mass Spectrom. 1993, 28, 1623.
(6) Chou, P. K.; Smith, R. L.; Chyall, L. J.; Kentta¨maa, H. I. J. Am.
Chem. Soc. 1995, 117, 4374.
(7) (a) Mourgues, P.; Audier, H. E.; Leblanc, D.; Hammerum, S. Org.
Mass Spectrom. 1993, 28, 1098. (b) Audier, H. E.; Leblanc, D.; Mourgues,
P.; McMahon, T. B.; Hammerum, S. J. Chem. Soc., Chem. Commun. 1994,
2329. (c) Maquin, F.; Stahl, D.; Sawaryn, A.; von R. Schleyer, P.; Koch,
W.; Frenking, G.; Schwarz, H. J. Chem. Soc., Chem. Commun. 1984, 504.
(8) (a) Yates, B. F.; Bouma, W. J.; Radom, L. J. Am. Chem. Soc. 1984,
106, 5805. (b) Yates, B. F.; Bouma, W. J.; Radom, L. J. Am. Chem. Soc.
1987, 109, 2250. (c) Gauld, J. W.; Radom, L. J. Phys. Chem. 1994, 98,
777.
(9) Weger, E.; Levsen, K.; Ruppert, I.; Burgers, P. C.; Terlouw, J. K.
Org. Mass. Spectrom. 1983, 18, 327.
S0002-7863(96)01634-4 CCC: $12.00 © 1996 American Chemical Society

11894 J. Am. Chem. Soc., Vol. 118, No. 47, 1996
Schweighofer et al.
unknown neutral ylide CH2dPH3 (methylenephosphorane).10
These experimental findings support a theoretical prediction that
the distonic and conventional CH5P•+ radical cations correspond
diselenide which was vacuum distilled. Their purity was verified by
mass spectrometry. n-Hexylphosphine and methylphosphine were
synthesized by established procedures.14 The solid adduct CH3PH2‚HI
was generated15 by adding dry HI gas to frozen CH3PH2, which was
maintained at -196 °C. The gaseous HI was generated by adding a
water solution of HI dropwise into P4O10.
•+
to minima on the potential energy surface. The ion CH3PH2
has been calculated to be 8 kcal mol-1 more stable than
•CH2PH3+ (at the QCISD/6-311+G(3df,2p)//MP2/6-311+G(2df,p)
level of theory)8c and to be separated from the latter ion by a
barrier of 52.6 kcal mol-1 (at the UMP3/6-31G(d,p)//6-31G(d,p)
level of theory).8b
The reaction rate constants (kexp) were obtained from the slope of
the plot of the natural logarithm of the relative abundance of the reactant
ion as a function of reaction time. Collision rate constants (kcoll) were
calculated using the parametrized trajectory theory.16 The reaction
efficiency is given by kexp/kcoll. For rate measurements, the ion gauges
were calibrated for their sensitivity17 toward the neutral reagent.
Correction for the pressure gradient between the ion gauge and the
dual cell was obtained by measuring rates of reactions with known
rate constants involving each neutral reagent. The branching ratios of
competitive reaction channels were derived from constant relative
abundances of the product ions measured at short reaction times.
The ab initio molecular orbital calculations were carried out utilizing
the Gaussian 92 suite of programs18 on a Convex C220 supercomputer
at the University of Du¨sseldorf. Geometry optimizations of stationary
points were executed at the UMP2(FU)/6-311+G(2df,p) level of theory
by using the analytical gradient technique (Berny optimization).19-21
Vibrational frequencies were calculated at the UMP2(FU)/6-
311+G(2df,p)//UMP2(FU)/6-311+G(2df,p) level of theory in order to
obtain zero point energies (scaled by 0.9) and to ensure that the
optimized structures were true minima on the potential energy surfaces.
The spin contamination, as expressed by S2 , was found to be within
an acceptable range (0.763-0.772) for all the structures examined ( S2
) 0.75 for the pure spin state).
We report here the first systematic study of the intrinsic
chemical properties of the prototypical phosphorus-containing
ylidion •CH2PH3+. The ion •CH2PH3+ was found to have very
different chemical properties from those of CH3PH2•+ and hence
to be stable toward isomerization. The proton affinity of the
•
phosphorus atom in CH2PH2 was experimentally determined
to be 190.4 ( 3 kcal mol-1. This value, combined with a
•
calculated heat of formation for CH2PH2, yields a heat of
•
+
formation of 217 ( 3 kcal mol-1 for CH2PH3
.
Experimental Section
All experiments were conducted utilizing a Model 2001 Extrel
Fourier transform ion cyclotron resonance mass spectrometer that has
been described elsewhere.3,11 The instrument consists of a dual cell
(two 4.7 cm cubic cells) that is aligned collinearly with the magnetic
field produced by a 3 T superconducting magnet operated at 2.2-2.8
T. The dual cell is differentially pumped to a nominal base pressure
of less than 10-9 Torr by using two Balzers turbomolecular pumps
(330 L s-1), each backed with an Alcatel 2012 mechanical pump. A
typical experiment consisted of five steps: ion formation in one cell,
transfer of the ions into the other cell, collisional cooling, isolation of
the desired ion, and detection after a variable reaction time.
Results and Discussion
•+
Intrinsic Chemical Properties of •CH2PH3+ and CH3PH2
.
Examination of the gas-phase reactivity of the distonic ion
+
The ion •CH2PH3 was generated by electron ionization of n-
+
•+
•CH2PH3 and its conventional isomer CH3PH2 reveals that
these ions are distinct species: the two ions possess very
different chemical properties. Radical-type abstraction reactions
hexylphosphine added into the instrument with a heated batch inlet
•+
system equipped with a leak valve. CH3PH2 was generated by
electron ionization of CH3PH2‚HI that was introduced into the system
with a heated solids probe. The nominal pressure was typically 1 ×
10-7 Torr in the cell wherein the ions were generated. The ion signal
was optimized by varying the ionization energy (20-70 eV), the
ionization time (20-50 ms), and the emission current (4-8 µA). At
all times except during unselective ejection of unwanted ions prior to
ion transfer and during ion transfer, the ions were trapped in the center
of one side of the dual cell by applying +2.0 V potential onto the
plates perpendicular to the magnetic field (trapping plates). Ions were
transferred between the two sides of the dual cell through a 2 mm hole
in the middle plate (conductance limit) by grounding this plate for 100-
200 µs. Ions were kinetically and internally cooled by multiple
collisions with argon atoms pulsed12 into the cell at a peak nominal
pressure of 1 × 10-5 Torr. The ion of interest was isolated by using
the stored waveform inverse Fourier transform method13 (with an Extrel
SWIFT module) to generate rf pulses that ejected all undesired ions
from the cell. The isolated ion was allowed to undergo reactions for
variable time intervals with neutral reagents introduced into the cell at
a nominal pressure of 1 × 10-7 Torr. Detection of ions was achieved
by exciting all the ions present in the cell into larger orbits through the
use of a radio frequency sweep (“chirp”) of 124 Vp-p amplitude and a
3.2 kHz µs-1 sweep rate. The spectra were recorded as 32k data points
subjected to one zero fill prior to Fourier transformation.
•
+
and proton transfer reactions dominate for CH2PH3 (Table
1). In sharp contrast, CH3PH2•+ reacts by predominant electron
abstraction with most of the reagents studied (Table 1). All
the reactions examined were found to follow pseudo-first-order
kinetics (for example, see Figures 1 and 2). This finding further
supports the conclusion that CH2PH3 and CH3PH2 do not
isomerize within the collision complexes involving the neutral
reagent molecules studied. The reactivity of the isomeric ions
is discussed in detail below.
•
+
•+
The reagents CH3SSCH3 and CH3SeSeCH3 commonly react
with distonic ions by transfer of CH3S• and CH3Se•,
respectively.2bd,22,24,25 Indeed, CH3S• abstraction was observed
•
upon reaction of CH2PH3 with CH3SSCH3 (the elemental
composition of the product ion was verified by an accurate mass
(14) (a) Pass, F; Schindelbauer, H. Monatsh. Chem. 1959, 90, 148. (b)
Bissey, J. E.; Goldwhite, H.; Rowsell, D. G. J. Org. Chem. 1967, 32, 1542.
(15) Tommes, P. Ph.D. Dissertation, University Du¨sseldorf, Germany,
1995.
(16) Su, T.; Chesnavich, W. J. J. Chem. Phys. 1982, 76, 5183.
(17) Bartmess, J. E.; Georgiadis, R. M. Vacuum 1983, 33, 149.
(18) Frisch, M. J.; Trucks, G. W.; Head-Gordon, M.; Gill, P. M. W.;
Wong, M. W.; Foresman, J. B.; Johnson, B. G.; Schlegel, H. B.; Robb,
Meploge, E.; Gomberts, R.; Andres, J. L.; Raghavachari, K.; Binkley, J.
S.; Gonzales, C.; Martin, R. L.; Fox, D. J.; Defrees, D. J.; Baker, J.; Stewart,
J. J. P.; Pople, J. A.; Gaussian, Inc., Pittsburgh, PA, 1992.
(19) Schlegel, H. B. J. Comput. Chem. 1982, 3, 214.
All reagents, with the exception of methyl- and n-hexylphosphines,
were commercially obtained and used as received except dimethyl
(10) Keck, H.; Kuchen, W.; Tommes, P.; Terlouw, J. K.; Wong, T.
Angew. Chem., Int. Ed. Engl. 1992, 31, 86.
(11) (a) Leeck, D. T.; Stirk, K. M.; Zeller, L. C.; Kiminkinen, L. K. M.;
Castro, L. M.; Vainiotalo, P.; Kentta¨maa, H. I. J. Am. Chem. Soc. 1994,
116, 3028. (b) Stirk, K. M.; Smith, R. L.; Orlowski, J. C.; Kentta¨maa, H.
I. Rapid Commun. Mass Spectrom. 1993, 7, 392. (c) Stirk, K. G.; Kentta¨maa,
H. I. J. Phys. Chem. 1992, 96, 5272. (d) Lin, P.; Kentta¨maa, H. I. Org.
Mass Spectrom. 1992, 27, 1155. (e) Farrell, Jr., J. T.; Lin, P.; Kentta¨maa,
H. I. Anal. Chim. Acta 1991, 246, 227.
(12) Carlin, T. J.; Freiser, B. S. Anal. Chem. 1983, 55, 571.
(13) Chen, L.; Wang, T.-C. L.; Ricca, T. L.; Marshall, A. G. Anal. Chem.
1987, 59, 449.
(20) (a) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin,
R. D.; Mallard, W. G. J. Phys. Chem. Ref. Data, Suppl. 1 1988, 17. (b)
Berkowitz, J.; Ellison, G. B.; Gutman, D. J. Phys. Chem. 1994, 98, 2744.
(21) (a) Leeck, D. T.; Kentta¨maa, H. I. Org. Mass Spectrom. 1994, 29,
106. (b) Li, W.-K.; Chiu, S.-W.; Ma, Z.-X.; Liao, C.-L.; Ng, C. Y. J. Am.
Chem. Soc. 1993, 99, 8440.
(22) Thoen, K. K.; Beasley, B. J.; Smith, R. L.; Kentta¨maa, H. I. J. Am.
Soc. Mass Spectrom., in press.
(23) Leeck, D. T.; Li, R.; Chyall, L. J.; Kentta¨maa, H. I. J. Phys. Chem.
1996, 100, 6608.

•
+
Prototypical Organophosphorus Ylidion CH2PH3
J. Am. Chem. Soc., Vol. 118, No. 47, 1996 11895
•
+
•+
Table 1. Reactions of CH2PH3 and CH3PH2 with Selected Neutral Reagents
+
•+
reactions of •CH2PH3
reactions of CH3PH2
reagent
IEa (eV)
8.1b
products (%)
efficiencye
0.4
products (%)
efficiencye
•+
dimethyl disulfide
CH3SSCH3•+ (50)
CH3SSCH3
=1.2
CH3SCH2PH3+ (50)
+
•+
dimethyl diselenide
allyl iodide
benzeneselenol
acetaldehyde
7.9c
9.3
CH3SeCH2PH3
0.7
0.3
g
CH3SeSeCH3
CH5IP+
0.7
1.0
g
0.3
1.3
CH5IP+
+
8.3d
10.2
7.5
CH3PH3
C6H5SeH•+ (100)
+
no reaction
CH3PH3
triethylamine
(CH3CH2)3NH+ f
(CH3CH2)3N•+ (48)
(CH3CH2)2NdCH2+ (52)
1.3
(CH3CH2)3N•+ (73)
(CH3CH2)2NdCH2+ (27)
e
a Values taken from ref 20a unless otherwise noted. b Value taken from ref 21. c Value taken from ref 22. d Value taken from ref 23. kexp/kcoll
.
f This is a major secondary product; some of these ions may also be formed as a primary product. g Not measured.
Many distonic ions have been reported5b to abstract
CH2dCHCH2• and/or I• from CH2dCHCH2I. In keeping with
these findings, I• abstraction was observed when •CH2PH3+ was
allowed to react with CH2dCHCH2I. In this case, the same
•+
reaction was observed for CH3PH2
. This result was not
unexpected since some conventional radical cations are known
to abstract I• from allyl iodide.5b
Proton Transfer Reactions. The proton affinity of the
•
phosphorus atom in CH2PH2 was experimentally determined
by reacting •CH2PH3+ with a series of neutral compounds with
known proton affinities and monitoring the occurrence of proton
transfer (Table 2). Given that proton transfer occurs from
+
•CH2PH3 to propionic acid and 2-propanol (proton affinities
191.8 and 191.2 kcal mol-1, respectively) but not to propional-
dehyde (proton affinity 189.6 kcal mol-1), the proton affinity
of •CH2PH3+ at the phosphorus atom is concluded to be 190.4
( 3 kcal mol-1
.
∆Hf(•CH2PH3+) can be derived from the proton affinity of
•
Figure 1. Temporal variation of the abundances of the reactant ion
the phosphorus atom in the radical CH2PH2 and the heats of
+
•CH2PH3 (m/z 48, b) and the product ions (CH3CH2)3N•+ (m/z 101,
formation for •CH2PH2 and H+ (Scheme 2). Since ∆Hf(•CH2PH2)
+
2), (CH3CH2)2NdCH2 (m/z 86, O), and (CH3CH2)3NH+ (m/z 102,
3) upon reaction with (CH3CH2)3N. The lines are arbitrary smooth
curves drawn through the data points.
Scheme 2
•CH2PH3+ f •CH2PH2 + H+ ∆Hreactn ) PA
measurement. However, the reaction is accompanied by
electron abstraction. CH3SeSeCH3 is more reactive toward
organic radical cations than CH3SSCH3 and often reacts by
exclusive CH3Se• transfer. 22,25 Also •CH2PH3+ reacts with CH3-
SeSeCH3 by exclusive CH3Se• abstraction (Table 1). In sharp
contrast, the only reaction channel observed for the conventional
+
∆Hf(•CH2PH3 ) ) ∆Hf(•CH2PH2) + ∆Hf(H+) - PA
is not known, an estimate for this value was obtained by using
the isodesmic reaction shown in Scheme 3. At the UMP2(FU)/
Scheme 3
•+
radical cation CH3PH2 with both of these neutral reagents is
∆Hreactn
electron transfer (Table 1). This result is in keeping with the
earlier findings that most known conventional radical cations
react with CH3SSCH3 and CH3SeSeCH3 by exclusive electron
abstraction if the reaction is exothermic.2b
•CH2PH2 + CH4
8 CH3PH2 + •CH3
∆Hf(•CH2PH2) ) ∆Hf(CH3PH2) + ∆Hf(•CH3) -
∆Hf(CH4) - ∆Hreactn
The isomeric ions show different reactivity toward most of
the neutral reagents studied. The efficient hydrogen atom donor
6-311+G(2df,p) level of theory (including ZPVE correction),
the enthalpy change of the isodesmic process was calculated to
be 5.0 kcal mol-1 (at 0 K). Using the known20a,b experimental
C6H5SeH readily transfers a hydrogen atom to CH2PH3+, in
•
analogy with the observations made earlier for other distonic
•+
ions.5b In contrast, the conventional radical cation CH3PH2
heats of formation (∆Hf298) of CH3PH2, CH4, and CH3 ,
•
∆Hf298(•CH2PH2) was concluded to be 42.2 kcal mol-1 (Table
3). Together with the experimentally determined proton affinity
of the phosphorus atom in •CH2PH2, this value yields
reacts by electron abstraction (Table 1). The distonic ion is
unreactive toward CH3CHO while the conventional ion CH3PH2•+
abstracts a hydrogen atom from this reagent.
∆Hf298(•CH2PH3+) ) 217 ( 3 kcal mol-1
. This value is
somewhat greater than an earlier computational prediction of
(24) (a) Smith, R. L.; Kentta¨maa, H. I. J. Am. Chem. Soc. 1995, 117,
1393. (b) Chyall, L. J.; Kentta¨maa, H. I. J. Am. Chem. Soc. 1994, 116,
3135. (c) Smith, R. L.; Chyall, L. J.; Chou, P. K.; Kentta¨maa, H. I. J. Am.
Chem. Soc. 1994, 116, 781. (d) Stirk, K. M.; Orlowski, J. C.; Leeck, D. T.;
Kentta¨maa, H. I. J. Am. Chem. Soc. 1992, 114, 8604.
(25) Beasley, B. J.; Smith, R. L.; Kentta¨maa, H. I. J. Mass Spectrom.
1995, 30, 384.
212 kcal mol-1 (G2′ method).8c
•
+
Electron Transfer Reactions. The reactions of CH2PH3
with (C2H5)3N and CH3SSCH3, yield abundant electron transfer
products. These reactions are not likely to occur via simple
electron transfer to yield CH2dPH3 (calculated ionization energy

11896 J. Am. Chem. Soc., Vol. 118, No. 47, 1996
Schweighofer et al.
•
+
Table 2. Reactions between CH2PH3 and Neutral Reagents with
Scheme 4
Known Proton Affinities (PA) and Ionization Energies (IE)
PAa
(kcal
IEa
(eV)
reagent
mol-1
)
reaction
benzaldehyde
2-butanol
2-methyl-2-propanol 193.7
200.2
195
9.4
predominant proton transfer
9.88 proton transfer
9.90 proton transfer
propionic acid
2-propanol
propionaldehyde
acetaldehyde
methanol
191.8 10.53 proton transfer
191.2 10.12 proton transfer
189.6 10.23 no reaction
186.6 10.2
no reaction
181.9 10.82 no reaction
a Values taken from ref 20a. b Value taken from ref 21.
Table 3. Calculated Total Energies (hartrees) and Experimental
Heats of Formation (∆Hf298) of the Species Used To Calculate
∆Hf298(•CH2PH2)
∆Hf298
∆Hf298
energya
(hartrees)
(kcal
energya
(hartrees)
(kcal
species
mol-1
)
species
mol-1
)
•CH3
CH4
-39.718 06
35.0b CH3PH2 -381.965 41 -4.0c
-40.377 47 -16.2b •CH2PH2 -381.313 99
42.2d
the different reaction rates of CH2PH3 and CH3PH2+•, for
•
+
a UMP2-Full/6-31+G(2df,p) + ZPVE (scaled by 0.9). b Value taken
from ref 20b. c Value taken from ref 20a. d Obtained in this work using
the equation in Scheme 3 and the other values shown in this table.
example, toward CH3SSCH3 and CH2dCHCH2I (Table 1).
•
+
The electron transfer reactions observed for CH2PH3 can
be readily explained on the basis of the recently proposed
“shuttle” mechanism.26 The same mechanism was used to
rationalize the electron transfer reactions involving the ion
+ 6
•CH2SH2
. This mechanism involves an initial proton transfer
between the ion and the neutral reagent. If the reaction is highly
exothermic, the resulting complex II (Scheme 4) may decom-
pose. Alternatively, hydrogen atom transfer may occur to the
carbon radical site in •CH2PH2, generating the complex V that
consists of neutral CH3PH2 and the radical cation of the reagent.
Decomposition of this complex yields the observed electron
transfer product. Another possible reaction channel involves
proton transfer within complex II, yielding complex III. Since
•+
the recombination energy of CH3PH2 (9.12 eV)20a is greater
than the ionization energies of the reagents of interest
((CH3CH2)3N, 7.5 eV;20a CH3SSCH3, 8.1 eV21), electron transfer
within complex III is likely to occur readily to produce complex
V that then decomposes to yield the electron transfer product
(III to IV transition is also posible but less likely).
Figure 2. Temporal variation of the abundances of the reactant ion
•CH2PH3+ (m/z 48, b) and the product ion CH3PH3+ (m/z 49, 0) upon
reaction with C6H5SeH. The lines are arbitrary smooth curves drawn
through the data points.
Most of the reactions of the sulfur-containing distonic ion
•CH2SH2+ are dominated by electron transfer and indistinguish-
able from those of its conventional isomer CH3SH•+.6 In
6.8 eV) since this process would be highly endothermic.8b An
intramolecular 1,2-hydrogen shift in •CH2PH3+ to yield the more
stable conventional isomer CH3PH2•+ prior to reaction with the
above neutral reagents would explain the observed reactivity,
•
+
contrast, CH2PH3 shows reactivity distinctly different from
that of CH3PH2+•. This is likely explained by several facts.
•
+
First, CH2SH2 is significantly more acidic (PA(•CH2SH) e
176 kcal mol-1 at S) than •CH2PH3+ (PA(•CH2PH2) ) 190 kcal
mol-1 at P) and hence more easily deprotonated within a
collision complex with a base. Second, the driving force for
•+
since CH3PH2 was found to react by predominant electron
abstraction with these neutral reagents. However, the results
discussed above rule out this isomerization. Furthermore, ab
initio molecular orbital calculations carried out8b at the UMP3/
+
isomerization of •CH2SH2 to the conventional isomer is
•
+
calculated to be -19 kcal mol-1, while that for CH3PH2 is
6-31G(d,p) level of theory predict that CH2PH3 is separated
•+
•+
from CH3PH2 by a very high barrier (43 kcal mol-1).8b
A
only -11 kcal mol-1 (Table 4).8b Finally, CH2SH has a
•
barrier of this size cannot be overcome even when assisted by
the ion-dipole energy of an ion-molecule collision complex
(typically in the range of 10-20 kcal mol-1). Partial isomer-
ization is experimentally ruled out on the basis of the finding
significantly greater dipole moment (1.3 D) than •CH2PH2 (0.9
D; both values were calculated at the UHF/3-21G(d) level of
theory; Table 5). Hence, solvation of a protonated base BH+
by •CH2SH within the collision complex is more efficient than
by •CH2PH2. This results in a deeper well and a longer lifetime
•
that the decay of the natural logarithm of the signal of CH2-
+
PH3 as a function of time follows the expected pseudo-first-
for the collision complex corresponding to BH+ and CH2SH,
•
order kinetics for all the reactions studied, as exemplified by
making further reactions within this collision complex more
likely.
•
+
the reaction of CH2PH3 with triethylamine in Figure 1 and
with benzeneselenol in Figure 2. These observations strongly
suggest that the ion population does not consist of mixtures of
isomeric ions. Further support for this conclusion comes from
(26) Reddy, A. C.; Danovich, D.; Ioffe, A.; Shaik, S. J. Chem. Soc.,
Perkin Trans. 2 1995, 1525-1539.

•
+
Prototypical Organophosphorus Ylidion CH2PH3
J. Am. Chem. Soc., Vol. 118, No. 47, 1996 11897
•+
Table 4. Calculated (G2′) and Experimental Heats of Formation
(∆Hf298) for CH5P•+ and CH4S•+
The conventional ion CH3PH2 reacts by predominant
electron abstraction with most of the neutral molecules studied.
•
+
∆Hf298 (kcal mol-1
)
∆Hf298 (kcal mol-1
)
In contrast, CH2PH3 shows a variety of reactions, including
proton transfer, atom abstraction, and group abstraction reac-
tions. The proton affinity of the phosphorus atom in •CH2PH2
was determined to be 190 ( 3 kcal mol-1. This value, together
ion
calcda
exptl
ion
calcda
exptl
+
+
•CH2PH3
CH3PH2
212
204
217b
206c
•CH2SH2
232
213
g227d
•+
CH3SH•+
211c
•
with a calculated heat of formation for CH2PH2, yields an
a Value taken from ref 8c. b This work. c Value taken from ref 20a.
d Value taken from ref 6.
estimate of 217 ( 3 kcal mol-1 for the heat of formation of the
•
+
distonic ion CH2PH3
.
•
•
•
+
Table 5. Calculated Dipole Moments of CH2PH2 and CH2SH
The ion CH2PH3 was found to react with (C2H5)3N and
CH3SSCH3 to yield electron transfer products. These reactions
are proposed to take place via a shuttle mechanism26 that
involves consecutive proton and either hydrogen or proton and
electron transfers within the collision complex. This mechanism
(UHF/3-21G* Level of Theory)
species
D
•CH2PH2
•CH2SH
0.8728
1.3173
is in keeping with that recently reported6 for the reactions of
+
the sulfur-containing distonic ion •CH2SH2
. The shuttle
Conclusion
mechanism seems to constitute a general alternative for the
traditional direct electron transfer mechanism for small organic
ions.
The work described here represents the first study on the
reactivity of the prototypical phosphorus-containing distonic ion
•CH2PH3+. The distinct reactivity of •CH2PH3+ and that of its
conventional isomer CH3PH2•+ demonstrate that these isomeric
ions are stable, noninterconverting species. This finding
contrasts the recent suggestion7b that the interconVersion of
R-distonic ions and their conVentional isomers within collision
complexes of neutral reagents is a widespread phenomenon.
Acknowledgment. The National Science Foundation (Grant
CHE9409644) is acknowledged for financial support of this
work. Professor Sason Shaik is thanked for thoughtful insights.
This work is part of the Ph.D. Thesis of A.S.
JA9616342