Organometallics in acidic media: Catalytic dimerization of ethylene by (perfluoroalkyl)phosphine complexes of platinum and palladium in trifluoroacetic acid

Article abstract of DOI:10.1021/om990179q

Catalytic ethylene dimerization by (dfepe)Pt(Me)X complexes in aprotic and trifluoroacetic acid solvents is described. In CH₂Cl₂, catalyst deactivation due to acid loss and generation of (dfepe)Pt(η²-C₂H₄) is observed. In contrast, long-term ethylene dimerization activity takes place in trifluoroacetic acid at 80°C and produces 2-(trifluoroacetato)butane as the sole organic product. Under these reaction conditions, the catalyst resting state is (dfepe)Pt(Et)-(O₂CCF₃). At 20°C, reversible acid elimination from (dfepe)Pt(Et)(O₂CCF₃) and ethylene ligand exchange results in catalytic vinylic H⁺/D⁺ exchange with CF₃CO₂D (t_1/2 ≈ 40 min). Analogous palladium systems exhibit enhanced dimerization activity at 25°C (340 turnovers/ h, 100 psi C₂H₄) and form (dfepe)₂Pd as a catalyst resting state. Inhibition of catalytic activity in the presence of added dfepe was noted. Comparisons between catalytic runs in CF₃CO₂H and CF₃CO₂D gave an apparent solvent kinetic effect of 3.5(2), based on an initial second-order rate dependence on ethylene. In CF₃CO₂D, regioselective acid addition to give only CH₃CH(O₂CCF₃)CH(D)CH₃ indicates that isomerization to 2-butene has occurred prior to 1,2-acid addition. A general olefin dimerization mechanism incorporating protonation preequilibria is presented.

Full text of DOI:10.1021/om990179q

2536
Organometallics 1999, 18, 2536-2542
Or ga n om eta llics in Acid ic Med ia : Ca ta lytic Dim er iza tion
of Eth ylen e by (P er flu or oa lk yl)p h osp h in e Com p lexes of
P la tin u m a n d P a lla d iu m in Tr iflu or oa cetic Acid †
Shannon White, Byron L. Bennett, and Dean M. Roddick*
Department of Chemistry, Box 3838, University of Wyoming, Laramie, Wyoming 82071
Received March 12, 1999
Catalytic ethylene dimerization by (dfepe)Pt(Me)X complexes in aprotic and trifluoroacetic
acid solvents is described. In CH2Cl2, catalyst deactivation due to acid loss and generation
of (dfepe)Pt(η2-C2H4) is observed. In contrast, long-term ethylene dimerization activity takes
place in trifluoroacetic acid at 80 °C and produces 2-(trifluoroacetato)butane as the sole
organic product. Under these reaction conditions, the catalyst resting state is (dfepe)Pt(Et)-
(O2CCF3). At 20 °C, reversible acid elimination from (dfepe)Pt(Et)(O2CCF3) and ethylene
ligand exchange results in catalytic vinylic H+/D+ exchange with CF3CO2D (t1/2 ≈ 40 min).
Analogous palladium systems exhibit enhanced dimerization activity at 25 °C (340 turnovers/
h, 100 psi C2H4) and form (dfepe)2Pd as a catalyst resting state. Inhibition of catalytic activity
in the presence of added dfepe was noted. Comparisons between catalytic runs in CF3CO2H
and CF3CO2D gave an apparent solvent kinetic effect of 3.5(2), based on an initial second-
order rate dependence on ethylene. In CF3CO2D, regioselective acid addition to give only
CH3CH(O2CCF3)CH(D)CH3 indicates that isomerization to 2-butene has occurred prior to
1,2-acid addition. A general olefin dimerization mechanism incorporating protonation
preequilibria is presented.
In tr od u ction
chelate (C2F5)2PCH2CH2P(C2F5)2 (dfepe), as well as
monodentate phosphines (C2F5)2(R)P,7 generally exhibit
greatly enhanced oxidative and thermal stabilities rela-
tive to donor phosphine analogues. Compared to corre-
sponding carbonyl systems, (fluoroalkyl)phosphine com-
plexes also closely approximate (CO)nM electronics
while providing increased steric control and thermal
stability. In recent reports, we have demonstrated (1)
the compatibility of (dfepe)M systems with strongly
acidic media and (2) the exceptional resistance of
(dfepe)M-alkyl bonds to protonolysis in neat acidic and
superacidic media.4,6 While these preliminary studies
have demonstrated the protic compatibility of (dfepe)-
Pt-alkyl bonds, the question remained whether these
compounds would still undergo fundamental organo-
metallic transformations in acidic media (Scheme 1). In
this paper, we report that (dfepe)M (M ) Pd, Pt)
systems function as efficient ethylene dimerization
catalysts in trifluoroacetic acid under mild conditions,
affording the ester product 2-(trifluoroacetato)butane.
Studies concerning the nature of the catalyst resting
state for these palladium and platinum systems are also
presented, as well as labeling experiments which indi-
There is extensive interest in extending organome-
tallic chemistry and metal-mediated homogeneous cata-
lytic processes to aqueous and other “nontraditional”
media.1-3 Impetus for this work has been the develop-
ment of desirable environmentally benign chemical
conversions, as well as the potential discovery of new
reactivity modes which exploit a wider range of solvent
properties. One approach taken in aqueous/protic sol-
vent chemistry has been the shift to electrophilic late
transition metal systems which form covalent metal-
carbon bonds that are more resistant to protonolysis and
more tolerant of substrate functionality.
Our research program has focused on the synthesis
and properties of (fluoroalkyl)phosphine metal com-
plexes of the middle and late transition metal triads.4-6
Complexes incorporating the fluorinated phosphine
† Dedicated to Professor Warren Roper on the occasion of his 60th
birthday.
(1) Aqueous systems: (a) Haggin, J . Chem. Eng. News 1994, 72 (41),
28. (b) Ferrara, M. L.; Orabona, I.; Ruffo, F.; Funicello, M.; Panunzi,
A. Organometallics 1998, 17, 3832. (c) Chen, J .; Alper, H. J . Am. Chem.
Soc. 1997, 119, 893. (d) Baxley, G. T.; Miller, W. K.; Lyon, D. K.; Miller,
B. E.; Nieckarz, G. F.; Weakley, T. J . R.; Tyler, D. R. Inorg. Chem.
1996, 35, 6688. (e) Wang, L.; Lu, R. S.; Bau, R.; Flood, T. C. J . Am.
Chem. Soc. 1993, 115, 6999. (f) Nguyen, S. T.; J ohnson, L. K.; Grubbs,
R. H. J . Am. Chem. Soc. 1992, 114, 3974.
(2) Supercritical solvent systems: (a) J essop, P. G.; Hsiao, Y.;
Ikariya, T.; Noyori, R. J . Am. Chem. Soc. 1996, 118, 344. (b) J essop,
P. G.; Ikariya, T.; Noyori, R. Organometallics 1995, 14, 1510. (c) Burk,
M. J .; Feng, S.; Gross, M. F.; Tumas, W. J . Am. Chem. Soc. 1995, 117,
8277.
(3) (a) Horva´th, I. T.; Ra´bath, J . Science 1994, 266, 72. (b) Gladysz,
J . A. Science 1994, 266, 55. (c) Hughes, R. P.; Trujillo, H. A.
Organometallics 1996, 15, 286. (d) Brookhart, M.; Chandler, W.;
Kessler, R. J .; Liu, Y.; Pienta, N. J .; Santini, C. C.; Hall, C.; Perutz, R.
N.; Timney, J . A. J . Am. Chem. Soc. 1992, 114, 3802. (e) R. C. Schnabel,
R. C.; Roddick, D. M. ACS Symp. Ser. 1994, 555, 421.
(4) Houlis, J . F.; Roddick, D. M. J . Am. Chem. Soc. 1998, 120, 11020.
(5) (a) Ontko, A. C.; Houlis, J . F.; Schnabel, R. C.; Roddick, D. M.;
Fong, T. P.; Lough, A. J .; Morris, R. H. Organometallics 1998, 17, 5467.
(b) Peters, R. G.; White, S.; Roddick, D. M. Organometallics 1998, 17,
4493. (c) Schnabel, R. C.; Roddick, D. M. Organometallics 1996, 15,
3550. (d) Bennett, B. L.; Roddick, D. M. Inorg. Chem. 1996, 35, 4703.
(e) Schnabel, R. C.; Carroll, P. S.; Roddick, D. M. Organometallics 1996,
15, 655. (f) Schnabel, R. C.; Roddick, D. M. ACS Symp. Ser. 1994, 555,
421.
(6) Bennett, B. L.; Birnbaum, J .; Roddick, D. M. Polyhedron 1995,
14, 187.
(7) Peters, R. G.; Bennett, B. L.; Schnabel, R. C.; Roddick, D. M.
Inorg. Chem. 1997, 36, 5962.
10.1021/om990179q CCC: $18.00 © 1999 American Chemical Society
Publication on Web 06/06/1999

Organometallics in Acidic Media
Organometallics, Vol. 18, No. 13, 1999 2537
Sch em e 1
equiv of CF3CO2H, while the addition of excess
CF3CO2H (∼10 equiv) to benzene solutions of (dfepe)-
Pt(η2-C2H4) at 20 °C reformed the Pt(II) ethyl complex
5 (eq 1).
cate that vinylic-H+/D+ exchange with ethylene can
occur prior to dimerization.
Resu lts
The reversibility of ethylene/ethyl group interchange
between 2 and 5 was also apparent in H/D exchange
studies between (dfepe)Pt(Et)(O2CCF3) and neat
CF3CO2D: At room temperature, facile conversion of
5 to (dfepe)Pt(Et-d5)(O2CCF3) (5-d5) was indicated by
1H NMR (t1/2 ≈ 40 min). Upon exposure of 5-d5 to 1 atm
C2H4, both the methylene and methyl proton signals
increased in intensity, then diminished over the course
of several hours. A concomitant decrease in the free
ethylene resonance and increase in the CF3CO2H protic
signal were also observed during this time period. Thus,
(dfepe)Pt(Et)(O2CCF3) serves as a catalyst for H/D
exchange between ethylene and trifluoroacetic acid
under mild conditions (Scheme 4).
E t h ylen e Dim er iza t ion in Ap r ot ic a n d P r ot ic
Media. Exposure of dichloromethane solutions of (dfepe)-
Pt(Me)(OTf) (1) to 2 atm of ethylene at 65 °C resulted
in ethylene uptake and the generation of 1 equiv of
propene and a mixture of cis- and trans-2-butenes (∼1
turnover/h). After several hours, 31P NMR indicated the
complete conversion of 1 to an approximately 1:1
mixture of previously characterized (dfepe)Pt(η2-C2H4)
(2)5d and a new species, 3 (Scheme 2). Although 3 was
not isolated, 31P NMR spectral data (δ 77.8, 1J PtP ) 1178
1
Hz; δ 54.0, J PtP ) 5847 Hz) are consistent with the
monoalkyl triflate product, (dfepe)Pt(Et)(OSO2CF3). A
decrease in catalytic activity correlated with the ap-
pearance of 2 as a product.
P a lla d iu m Ch em istr y. The generally low kinetic
activity of platinum catalysts prompted us to examine
the chemistry of analogous palladium systems. At-
tempts to prepare (dfepe)Pd(Me)2, either from dfepe
addition to (cod)Pd(Me)2 or alkylation of (dfepe)PdCl2
with MeMgBr, were not successful, most likely due to
facile reductive elimination of ethane. The presumed
instability of (dfepe)Pd(Me)2 led us to investigate ap-
proaches to (dfepe)Pd(Me)X systems beginning with
(cod)Pd(Me)Cl. Treatment of (cod)Pd(Me)Cl with AgX
(X ) O2CCF3, OSO2CF3) in dichloromethane afforded
the corresponding (cod)Pd(Me)X complexes 6 and 7,
respectively, in moderate yield. Addition of dfepe to
(cod)Pd(Me)Cl or 6 at 20 °C gave the corresponding
dfepe monomethyl complexes (dfepe)Pd(Me)X (X ) Cl
(8); O2CCF3 (9) (eq 2). Attempts to prepare (dfepe)Pd-
(Me)(OTf) from 7 and dfepe gave (dfepe)2Pd (see later)
as the only detectable product. Both 8 and 9 are stable
in air and in the solid state and, like their platinum
analogues, do not undergo protonolysis in neat
CF3CO2H at ambient temperature. After 2 days at 55
°C, decomposition of 9 occurred to form a mixture of
(dfepe)2Pd and several unidentified products.
The generation of 2 in dichloromethane suggested
that loss of HX from a catalytic intermediate (dfepe)-
Pt(H)(X) is competitive with ethylene dimerization, and
therefore the catalytic activity of (dfepe)Pt(R)(X) sys-
tems should be sensitive to acid concentration. Indeed,
treatment of either 1 or (dfepe)Pt(Me)(O2CCF3) (4) with
2 atm ethylene in neat CF3CO2H at 80 °C resulted in a
single metal monoalkyl product that was tentatively
identified as (dfepe)Pt(Et)(O2CCF3) (5). Under these
conditions, however, 1 equiv of 2-(trifluoroacetato)pro-
pane and the catalytic production of 2-(trifluoroacetato)-
butane (∼6 turnovers/h) were observed as products
(Scheme 3). No 1-butene or 2-butene products were
detected throughout the course of reaction. The propyl
and butyl ester products were identified by comparison
of 1H and 13C NMR data to authentic samples prepared
by the addition of propene and 1- or 2-butene, respec-
tively, to trifluoroacetic acid at 20 °C (see Experimental
Section). No ethane or butane protonolysis products
were detected by NMR under these reaction conditions.
An independent synthesis of (dfepe)Pt(Et)(O2CCF3)
was carried out in order to confirm the identify of 5:
Addition of dfepe to (cod)Pt(Et)2 afforded (dfepe)Pt(Et)2,
which when treated with excess CF3CO2H converted
cleanly to (dfepe)Pt(Et)(O2CCF3). Essentially identical
spectral data for (dfepe)Pt(Et)(O2CCF3) and solutions
of 5, as well as nearly identical catalytic activities for
CF3CO2H solutions of 1 and (dfepe)Pt(Et)(O2CCF3) at
80 °C, support the assignment of (dfepe)Pt(Et)(O2CCF3)
as the catalyst resting state for ethylene dimerization.
Catalytic Vin yl-H+/D+Exch an ge.Complexes (dfepe)-
Pt(η2-C2H4) and (dfepe)Pt(Et)(O2CCF3) are readily in-
terconverted: warming (dfepe)Pt(Et)(O2CCF3) in benzene-
d6 to 80 °C cleanly afforded (dfepe)Pt(η2-C2H4) and 1
Both (dfepe)Pd(Me)X complexes were fully character-
ized by NMR. The 1H methyl proton resonances of 8 and
9 appear as doublets due to coupling with the trans
phosphorus (3J PH ≈ 7.5 Hz). For each compound two 31
P
singlet resonances are observed, a high-field resonance

2538 Organometallics, Vol. 18, No. 13, 1999
White et al.
Sch em e 2
Sch em e 3
Sch em e 4
(δ ) 56.3, 8; and 55.6, 9) for the phosphorus atoms trans
to X and a low-field resonance (δ ) 72.9, 8; and 74.3, 9)
for the phosphorus atoms trans to the methyl group.
These assignments were made on the basis of the
corresponding platinum chemical shifts.
Like the platinum analogue, (dfepe)Pd(Me)(O2CCF3)
is catalytically active: treatment of 9 with ∼2 atm
ethylene in trifluoroacetic acid at 25 °C resulted in rapid
gas uptake and the clean production of 2-(trifluoro-
acetato)butane. A qualitatively similar activity was also
observed for (dfepe)Pt(Me)Cl. In contrast to platinum,
which forms (dfepe)Pt(Et)(O2CCF3) as a catalyst resting
state, the growth of a new metal species having a single
phosphorus resonance at δ ) 49.1 ppm was observed
during the course of the reaction. In an effort to estab-
lish the nature of this symmetrical (dfepe)Pd species,
three likely candidates, (dfepe)Pd(O2CCF3)2, (dfepe)2Pd,
and (dfepe)Pd(η2-C2H4), were considered. Addition of
excess dfepe to either (Cp)Pd(allyl) or Pd(dba)2 afforded
the bis-chelate (dfepe)2Pd (10) in good yield (eq 3).
F igu r e 1. Ethylene consumption in trifluoroacetic acid
versus time. Double exponential decay line-fits are shown.
Initial parameters: [(dfepe)2Pd] ) 1.61 mM, [(dfepe)Pd-
(Me)(O2CCF3)] ) 3.23 mM, P0 ) 90 psi, T ) 25 °C.
(∼2 atm) to (dfepe)2Pd in aprotic solvents gave no
spectroscopic evidence for (dfepe)Pd(η2-C2H4) formation.
In addition to exhibiting a single phosphorus resonance
for 10 coinciding with that observed in ethylene dimer-
izations using 9 as the catalyst, (dfepe)2Pd also exhibited
essentially identical catalytic activity in trifluoroacetic
acid and is therefore the likely dimerization catalyst
resting state. Significantly, (dfepe)2Pd was catalytically
inactive when acetic acid was used as a solvent.
A series of kinetic experiments were performed to
probe the nature of palladium catalysis (Figures 1-3).
For each kinetic run, a 100 psi ethylene charge was
admitted to a Fisher-Porter bottle containing well-
stirred solutions of (dfepe)Pd(Me)(O2CCF3) (3.23 mM)
or (dfepe)2Pd (1.61 mM) in 5 mL of trifluoroacetic acid
at 25 °C. Taking into account an initial 10 psi pressure
drop due to gas dissolution, the extent of ethylene
dimerization was monitored by following the corre-
sponding decrease in reactor pressure (Figure 1). Con-
siderable deviations from second-order behavior oc-
curred below ∼20 psi ethylene. However, inverse pressure
plots of ethylene uptake above this threshold for
(dfepe)2Pd in CF3CO2H and CF3CO2D were reasonably
Attempts to prepare (dfepe)Pd(O2CCF3)2 by treatment
of (cod)Pd(O2CCF3)2 (11) with 1 equiv of dfepe resulted
only in low yields of (dfepe)2Pd. Similarly, efforts to
prepare (dfepe)Pd(η2-C2H4) by the addition of ethylene

Organometallics in Acidic Media
Organometallics, Vol. 18, No. 13, 1999 2539
addition or competitive propylene insertions would allow
for dimerization in this solvent. Prolonged thermolysis
at 100 °C of (dfepe)Pt(Me)(O2CCF3) under ∼2 atm pro-
pylene gave no evidence for organic products other than
iPr(O2CCF3). In contrast, treatment of (dfepe)Pd(Me)-
(O2CCF3) in CF3CO2H with 2 atm propylene at 70 °C
resulted in the clean formation of (dfepe)2Pd and 1 equiv
t
of Bu(O2CCF3), as judged by 1H and 13C NMR spec-
troscopy. In a separate experiment, no dimerization ac-
tivity was observed for (dfepe)2Pd under similar condi-
tions at temperatures up to 80 °C.
F igu r e 2. Plots of inverse pressure (psi) versus time at
25 °C. Linear regression fits are for data below a threshold
of 0.05 psi-1. Derived second-order rate constants for
(dfepe)2Pd in CF3CO2xH (1.61 mM), 0.201(4) mol-1‚min-1
in CF3CO2D (1.61 mM), 0.0570(3) mol-1‚min-1; in CF3CO2H
+ 1 equiv dfepe (1.61 mM), 0.122(1) mol-1‚min-1; (dfepe)-
;
Pd(Me)(O2CCF3) in CF3CO2H (3.23 mM), 0.356(8) mol-1
‚
min-1
.
Discu ssion
A variety of palladium systems have been well
established as olefin dimerization catalysts.8 In contrast,
very few reports of platinum-based olefin dimerization
systems have appeared.9 Lewis or Bro¨nsted acid cocata-
lysts and/or initiators have been employed in a number
of cases to induce activity,10,11 but to our knowledge the
use of an acid as both a solvent medium and a co-
reactant is not precedented.12
The compatibility of (dfepe)M-alkyl bonds with acidic
media has been noted in our previous work.4,6,13 De-
pending on the specific mechanism of protonolysis
(stepwise protonation of the metal center followed by
reductive elimination or direct electrophilic attack of the
M-C bond), the ease of LnM-alkyl bond protonolysis
should be related to the ease of protonation of available
metal lone-pairs and/or the electron density and polarity
of the M-alkyl bond.13,14 For late metal and electron-
poor metal systems such as (dfepe)M(R)(X), a “conven-
tional” M(δ+)-C(δ-) bond polarity is not an accurate
description, and therefore proton transfer to a carban-
ionic alkyl group is not anticipated. Nevertheless, the
results of our current study as well as earlier work by
Flood1e,15 clearly show that olefin insertion into these
types of M-C bonds is not precluded by such shifts in
M-C bond electron density.
F igu r e 3. Plot of inverse pressure (psi) versus time at 25
°C, expanded to show data for the initial 350 min.
linear and gave effective second-order rate constants of
0.201(4) mol-1‚min-1 and 0.0570(3) mol-1‚min-1, re-
spectively, corresponding to a solvent kinetic isotope
effect of 3.5(2) (Figures 2 and 3). The catalytic activity
of (dfepe)2Pd in CF3CO2H at 100 psi is calculated to be
340 turnovers/h. In the presence of 1 equiv of dfepe, a
significantly slower (60%) rate constant of 0.122(1)
mol-1‚min-1 was observed in CF3CO2H. The initial
catalytic activity of (dfepe)Pd(Me)(O2CCF3) was calcu-
(8) (a) Al-J arallah, A. M.; Anabtawi, J . A.; Siddiqui, M. A. B.; Aitani,
A. M.; AL-Sa’Doun, A. W. Catal. Today 1992, 14, 1. (b) Pillai, S. M.;
Ravindrathan, M.; Sivaram, S. Chem. Rev. 1986, 86, 353. (c) Keim,
W.; Behr, A.; Ro¨per, M. In Comprehensive Organometallic Chemistry
I; Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.; Pergamon: New
York, 1982; Vol. 8, p 371.
(9) de Renzi, A.; Panunzi, A.; Vitagliano, A.; Paiaro, G. J . Chem.
Soc., Chem. Commun. 1976, 47.
1
lated to be approximately 300 turnovers/h. H and 13C
(10) Cramer, R. J . Am. Chem. Soc. 1965, 87, 4717.
(11) Guibert, I.; Neibecker, D.; Tkatchenko, I. J . Chem. Soc., Chem.
Commun. 1989, 1850.
(12) Acetic acid has been used as a solvent for olefin dimerization
using Pd(II) salts: see Table 2, in ref 8b.
(13) The mechanism of (dfepe)Pt-R protonolyses has been addressed
in more detail: Bennett, B. L.; J . M. Hoerter, J . M.; Roddick, D. M.
Manuscript in preparation.
(14) J enkins, H. A.; Yap, G. P. A.; Puddephatt, R. J . Organometallics
1997, 16, 1946, and references therein.
NMR analysis of the 2-trifluoroacetatobutane product
formed in the CF3CO2D kinetic run revealed that the
sole observable deuterated product produced in this
reaction was CH3CH(O2CCF3)CH(D)CH3.
The dimerization of higher olefins by dfepe-substi-
tuted platinum and palladium systems was briefly
i
examined. Although propylene readily converts to Pr-
(O2CCF3) in trifluoroacetic acid, reversible 1,2-acid
(15) Wang, L.; Flood, T. C. J . Am. Chem. Soc. 1992, 114, 3169.

2540 Organometallics, Vol. 18, No. 13, 1999
White et al.
Sch em e 5
A catalytic cycle consistent with all of our observa-
tions for both the platinum and palladium systems may
be accommodated by the general mechanistic framework
shown in Scheme 5. Prior dimerization studies with late
transition metal systems have either proposed or di-
rectly observed alkyl complexes as reaction intermedi-
ates, in accord with a catalytic cycle involving olefin
insertion into M-H and M-alkyl bonds. Although an
alternative cycloaddition dimerization mechanism may
be considered, several observations argue against this
potential pathway and support a Cossee-Arlman alkyl
insertion mechanism for the systems presented in this
study: (1) the decrease in catalytic activity upon forma-
tion of (dfepe)Pt(η2-C2H4) during the course of ethylene
dimerization in aprotic solvents (Scheme 2), (2) the
observation of (dfepe)Pt(Et)(O2CCF3) as a catalyst rest-
ing state, and (3) the catalytic inactivity of (dfepe)2Pd
in aprotic solvents or in the weaker acid CH3CO2H.
In addition to the conventional ethylene dimerization
catalytic cycle depicted in Scheme 5, a number of
interesting additional features merit consideration.
First, despite the presence of both M-ethyl and M-bu-
tyl intermediates, no ethane or butane protonolysis
products are observed in trifluoroacetic acid after pro-
longed reaction. From this it can be concluded that the
rates of ethylene insertion and â-elimination are much
greater than protonolysis side reactions. A second
unusual feature in these systems is the role of the acid
solvent itself in substrate and product discrimination.
For simple alkenes H2CdCH(R), 1,2-addition of trifluo-
roacetic acid (pKa ) 0.23) is favored for all cases where
R * H. As a consequence, only ethylene is stable under
the reaction conditions. The absence of any significant
quantities of free 1- or 2-butenes throughout the course
of dimerization indicates that 1,2-HX addition is fast
relative to butene generation. The observation that
deuterium is incorporated into only the 3-position of the
2-butyl ester product in (dfepe)2Pd-mediated catalysis
in CF3CO2D is also quite significant, since it indicates
that exclusively 2-butene is intercepted by acid, rather
than 1-butene or an equilibrated thermodynamic mix-
ture of 1- and 2-butenes. Thus, not only is olefin
isomerization occurring in these systems (as in Scheme
2) but the presence of acid serves to selectively discrimi-
nate in favor of 2-butene. An additional conclusion
drawn from the selective formation of CH3CH(O2CCF3)-
CH(D)CH3 is that prior scrambling of deuterons into
ethylene is not competitive with dimerization by
(dfepe)2Pd in CF3CO2D.
The resistance of (dfepe)Pt-alkyl bonds to protonoly-
sis also leads to catalytic deuteration of ethylene in
trifluoroacetic acid under mild conditions (Scheme 4).
While this type of exchange should generally be pro-
moted by sufficiently electrophilic late metal systems
with reversible olefin insertion/elimination and proto-
nation equilibria, only scattered reports of this reaction
mode have appeared in the literature. Cramer, in
particular, has reported ethylene deuteration by (acac)-
Rh(C2H4)2 in DCl/CH3OD at -24 °C.10 Although intrigu-
ing from a fundamental standpoint, competing ethylene
dimerization (especially for palladium) and the limited
stability of higher olefins in trifluoroacetic acid place
constraints on the utility of this scrambling mechanism.
The observed catalyst resting states for palladium and
platinum catalytic systems differ. This is most likely due
to a combination of a lower ethylene binding affinity for
(dfepe)Pd(0) relative to (dfepe)Pt(0) and a more unfavor-
able equilibrium between (dfepe)Pd(η2-C2H4) and (dfepe)-
Pd(Et)X for palladium (Scheme 6). While both dfepe
displacement by C2H4 and protonation are required to

Organometallics in Acidic Media
Organometallics, Vol. 18, No. 13, 1999 2541
Sch em e 6
enter the catalytic cycle from (dfepe)2Pd, the activity of
(dfepe)2Pd is still substantially higher than (dfepe)Pt-
(R)X. The similar activities of (dfepe)2Pd and (dfepe)-
Pd(Me)(O2CCF3) indicate that these preequilibria are
not severely restrictive; however, the catalytic inactivity
of (dfepe)2Pd in the weaker acid MeCO2H, rate inhibi-
tion by added dfepe, and the rate differences observed
between CF3CO2D and CF3CO2H runs indicate that
both dfepe dissociation and proton transfer are kineti-
cally important in these systems.
that (dfepe)Pt-alkyl bonds are stable in even supera-
cidic solvents such as CF3SO3H or FSO3H for hours at
room temperature, the range of potential organometallic
transformations based on electrophilic metal systems
may in principle be extended even further to extremely
acidic media.
Exp er im en ta l Section
Gen er a l P r oced u r es. All manipulations were conducted
under an atmosphere of purified nitrogen using high-vacuum
and/or glovebox techniques. Dry oxygen-free solvents were
prepared using standard procedures. Aprotic deuterated sol-
vents used in NMR experiments were dried over activated 3
Å molecular sieves. CF3CO2D was obtained from Cambridge
Isotope Laboratories (Massachusetts) and used as received.
AgO2CCF3 and MeMgBr were obtained from Aldrich and
used as obtained. Elemental analyses were performed by
Desert Analytics. High-resolution mass spectrum data were
obtained from the Nebraska Center for Mass Spectrometry.
IR spectra were recorded on a Perkin-Elmer 1600 FTIR in-
strument as Nujol mulls using NaCl plates. NMR spectra were
obtained with a J EOL GSX-270 or a Bru¨ker DRX-400 instru-
ment. 31P spectra were referenced to an 85% H3PO4 external
standard. (dfepe)Pt(Me)(X) (X ) OTf (1), O2CCF3 (4)),6 (cod)-
Pt(Et)2,16 (cod)Pd(Me)Cl (cod ) 1,5-cyclooctadiene),17 dfepe,18
The stoichiometric reaction of propylene with (dfepe)-
Pd(Me)(O2CCF3) and the inactivity of (dfepe)2Pd in
trifluoroacetic acid reflect a lower activity toward higher
t
olefins. While the stoichiometric production of Bu-
(O2CCF3) confirms that propylene insertion into a
Pd-Me bond has occurred, the lack of any C6 products
suggests that either (1) the insertion of propylene into
a Pd-propyl bond is less favorable or (2) (dfepe)Pd-
(propyl)+ is not accessible from the (dfepe)2Pd catalyst
resting state. The lower binding affinity generally
observed for higher olefins relative to ethylene is likely
to be the limiting factor in both of these processes.
The number of preequilibria involved in ethylene
dimerization by (dfepe)2Pd complicates the kinetic
analysis for this system. If dfepe displacement by
ethylene (K1 ) k1/k-1) is rapid relative to CF3CO2H
addition (K2 ) k2/k-2) and the addition of ethylene to a
(dfepe)Pd(Et)O2CCF3 intermediate is rate limiting (k3),
then the following rate law can be derived:
CpPd(allyl),19 and Pd(dba)2 were prepared using literature
20
methods.
(d fep e)P t(Et)2. Approximately 0.5 g of (cod)Pt(Et)2 was
dissolved in 20 mL of Et2O, and 0.95 g (1.68 mmol) of dfepe
was added via syringe. After stirring the reaction mixture
overnight at ambient temperature, cooling to -78 °C and
filtration afforded 1.1 g (∼95%) of (dfepe)Pt(Et)2 as an off-white
solid. Anal. Calcd for C14H14F20P2Pt: C 20.53, H 1.72. Found:
C 20.53, H 1.58. 1H NMR (CD2Cl2, 270 MHz, 27 °C): δ 2.42
k1k2k3[(dfepe)2Pd][CF3CO2H][C2H4]2
rate )
k-1[dfepe](k-2 + k3[C2H4])
2
(m, 4H; PCH2), 2.12 (m, J PtH ) 86 Hz, 2H; Pt(CH2CH3), 1.22
3
(m, J PtH ) 82 Hz, 3H; Pt(CH2CH3)). 13C{1H} NMR (acetone-
Under conditions where k-2 . k3[C2H4] this expression
simplifies to a second-order dependence in ethylene and
an inverse dependence on added dfepe, in qualitative
agreement with the observed kinetic behavior. Never-
theless, the assumption that the preequilibrium K1 is
established is unsupported, and the deviation from
second-order behavior observed in Figure 2 is not
explained. A more complete understanding of the un-
derlying mechanism relating to Scheme 5 clearly re-
quires more detailed study.
1
1
d6, 100.6 MHz, 27 °C): δ 121.2 (td, J CF ) 292 Hz, J CP ) 39
Hz; CF2CF3), 116.9 (qd, 1J CF ) 284 Hz, 2J CP ) 18 Hz; CF2CF3),
1
1
2
23.4 (t, J CP ) 40 Hz; PCH2), 15.9 (d, J CPt ) 713 Hz, J CP
)
92 Hz; CH2CH3), 15.2 (s, J CPt ) 28 Hz; CH2CH3). 19F NMR
2
(CD2Cl2, 254.05 MHz, 27 °C): δ -79.87 (s, PCF2CF3) (NOTE:
3
in all dfepe complexes, there are no observable J FF vicinal
couplings), -111 (overlapping ABX multiplets, PCF2CF3). 31P
1
NMR (CD2Cl2, 109.25 MHz, 27 °C): δ -8.70 (m, J PPt ) 1095
Hz). IR (cm-1): 1296(s), 1226(sh), 1200(s), 1146(sh), 1127(s),
1090(sh), 956(s), 871(w), 809(w), 749(m).
(d fep e)P t(Et)(O2CCF 3) (5). A 100 mL flask was charged
with 1.100 g (1.343 mmol) of (dfepe)Pt(Et)2 and 1 mL of
CF3CO2H under a nitrogen atmosphere. Rapid evolution of gas
indicated the reaction was essentially complete upon dissolu-
In summary, we have shown that (perfluoroalkyl)-
phosphine group 10 systems can serve as useful orga-
nometallic catalyst precursors which are compatible
with trifluoroacetic acid as a reaction medium. In the
case of palladium, a sufficiently high level of acidity is
actually required for hydride formation and the initia-
tion of insertion activity. This is in marked contrast with
catalysts where adventitious protons are required to
initiate catalytic activity. Since we have recently shown
(16) Clark, H. C.; Manzer, L. E. J . Organomet. Chem. 1973, 59, 411.
(17) Rulke, R. E.; Ernsting, J . M.; Spek, A. L.; Elsevier, C. J .; van
Leeuwen, P. W. N. M.; Vrieze, K. Inorg. Chem. 1993, 32, 5769.
(18) Ernst, M. F.; Roddick, D. M. Inorg. Chem. 1989, 28, 1625.
(19) Tatsuno, Y.; Toshida, T.; Otsuka, S. Inorg. Synth. 1990, 28, 343.
(20) Rettig, M. F.; Maitlis, P. M. Inorg. Synth. 1990, 28, 110.

2542 Organometallics, Vol. 18, No. 13, 1999
White et al.
tion of the solid (1-2 min). All volatiles were removed, and
15 mL of diethyl ether was added. Filtration of the resulting
slurry and drying under vacuum afforded 1.017 g of white solid
(84% yield). Anal. Calcd for C14H9F23O2P2Pt: C, 18.62; H, 1.00.
multiplets, PCF2CF3). IR (cm-1): 1693(s), 1420(w), 1298(m),
1234(s), 1196(m), 1137(s), 962(s).
(d fep e)2P d (10), Meth od A. To a flask charged with 0.05
g (0.2 mmol) of (Cp)Pd(allyl) was added 20 mL of toluene and
0.2 mL (0.7 mmol) of dfepe. The reaction mixture was stirred
at room temperature for 1 h, and the solution was then cooled
to -78 °C, upon which a white precipitate formed. The solid
was collected by cold filtration and dried under vacuum to yield
0.20 g (80%) of 10. Anal. Calcd for C20H8F40Pd: C, 19.40; H,
1
Found: C, 18.61; H, 0.99. H NMR (CD2Cl2, 400 MHz, 27 °C):
2
δ 2.70 (m, 2H; PCH2), 2.44 (m, 2H; PCH2), 2.32 (q, J HPt ) 53
3
3
3
Hz, J HH) 6 Hz, 2H; Pt(CH2CH3)), 1.16 (t, J HH ) 6 Hz, J HPt
) 29 Hz Pt(CH2CH3)). 19F NMR: (CD2Cl2, 376.5 MHz, 27 °C):
δ -78.6 (s, 6F, PCF2CF3), -77.9 (s, 6F; PCF2CF3), -108.4 (m,
overlapping ABX multiplets, 8F; PCF2CF3), -73.2 (s, 3F;
CF3CO2). 31P NMR (CF3CO2H, 109.25 MHz, 27 °C): δ 76.7 (m,
1
0.651. Found: C, 19.14; H, 0.40. H NMR (C6D6, 400 MHz, 23
°C): δ 1.82 (m). 31P NMR (C6D6, 400 MHz, 23 °C): δ 48.7 (m).
(d fep e)2P d , Meth od B. To a flask charged with 0.25 g of
Pd(dba)2 (0.04 mmol) was added 30 mL of CH2Cl2 and 0.35
mL (0.11 mmol) of dfepe. The reaction mixture was stirred at
room temperature for 3 h and evaporated to dryness. CH3OH
(10 mL) was added to precipitate a white solid, which was
collected by filtration and dried under vacuum to yield 0.35 g
(66%) of (dfepe)2Pd.
1J PPt ) 1171 Hz), 57.4 (m, J PPt ) 5302 Hz). In acetone-d6: δ
1
79.8 (s, 1J PPt ) 1135 Hz), 60.73 (s, 1J PPt ) 4861 Hz). IR (cm-1):
1704(m), 1408(w), 1295(s), 1191(s), 1136(s), 963(m), 751(w).
(cod )P d (Me)(O2CCF 3) (6). A flask charged with 0.50 g (1.9
mmol) of (cod)Pd(Me)(Cl), 0.50 g (2.3 mmol) of AgO2CCF3, and
20 mL of CH2Cl2 was stirred at 20 °C in the absence of light
for 5 h. The resulting brown-gray slurry was filtered, and the
CH2Cl2 was removed from the filtrate. Addition of 20 mL of
Et2O yielded a yellow solution, which upon cooling to -78 °C
formed a white precipitate. Filtration and drying under
vacuum yielded 0.48 g (74%) of (cod)Pd(Me)(O2CCF3) (6). 6 is
thermally sensitive and was stored at -40 °C to prevent
(cod )P d (O2CCF 3)2 (11). To a flask charged with 0.26 g (0.89
mmol) of (cod)PdCl2 and 0.394 g (1.78 mmol) of AgO2CCF3 was
added 25 mL of CH2Cl2. After stirring at room temperature
for 24 h in the absence of light a red-brown solid was filtered.
The yellow filtrate was concentrated and precipitated with
petroleum ether to give a light yellow powder (0.393 g (73%)).
7 was isolated by filtration, dried under vacuum, and stored
at -40 °C under nitrogen. 7 was not sufficiently stable for
analysis. 1H NMR (CD2Cl2, 400 MHz, 20 °C): δ 6.36 (s, 4H,
olefinic cod), 3.05 (s, 4H, aliphatic cod), 2.54 (s, 4H, aliphatic
cod). 19F NMR (CD2Cl2, 400 MHz, 23 °C): δ -73.64 (s). IR
(cm-1): 1785(s), 1670(s), 1203(w), 1150(w).
1
plating of Pd0 metal; no elemental analysis was obtained. H
NMR (C6D6, 270 MHz, 20 °C): δ 5.54 (s, 2H, olefinic cod),
4.07 (s, 2H, olefinic cod), 1.61 (m, 4H, aliphatic cod), 1.44 (m,
4H, aliphatic cod), 1.02 (s, 3H, PdCH3). IR (cm-1): 1685(s),
1581(w), 1415(m), 1194(s), 1179(s), 1160(s), 1137(s), 1087(w),
1020(w), 1003(w), 986(w), 906(w), 840(m), 790(w), 771(w),
748(w), 726(s), 668(w).
(cod )P d (Me)(OTf) (7). To a flask charged with 0.500 g
(1.89 mmol) of (cod)Pd(Me)(Cl) and 0.484 g (1.89 mmol) of
AgOTf (OTf ) O3SCF3) was added 25 mL of CH2Cl2. After
stirring at room temperature for 24 h in the absence of light,
a gray solid was filtered off to give a yellow-brown filtrate,
which was concentrated to ca. 5 mL and precipitated with Et2O
to yield white crystalline 7 (0.716 g, 74%). The solid was
isolated by cold filtration, dried under vacuum, and stored at
-40 °C under nitrogen. 7 was not sufficiently stable for
elemental analysis. 1H NMR (CD2Cl2, 400 MHz, 20 °C): δ 6.14
(s, 2H, olefinic cod), 5.50 (s, 2H, olefinic cod), 2.71 (m, 8H,
aliphatic cod), 1.40 (s, 3H, PdCH3). 19F NMR (CD2Cl2, 400 MHz,
23 °C): δ -77.12 (s, O3SCF3). IR (cm-1): 1311(m), 1231(s),
1208(m), 1156(m), 1017(s), 906(s), 853(s), 815(s).
2-(Tr iflu or oa cet a t o)b u t a n e Syn t h esis. To 10 mL of
CF3CO2H in a 3 oz Fisher-Porter bottle was admitted 45 psi
1-butene. After complete reaction, as judged by 1H and 13C
NMR, the solution was quenched with H2O (100 mL) and the
ester product was extracted with 100 mL of CH2Cl2. Removal
of the CH2Cl2 under vacuum gave essentially pure 2-(trifluo-
roacetyl)butane as a light brown liquid (crude yield 1.78 g).
High-resolution mass spectrum (CI, isobutane) for MH+ ion:
176.0637 (176.0633 calcd, -2.3 ppm). 1H NMR (CF3CO2H, 400
MHz, 20 °C): δ 5.07 (m, 1H; CH(O2CCF3)), 1.66 (m, 2H; CH2),
1.29 (d, 3H; CH3), 0.87 (t, 3H; CH3). 13C NMR (CF3CO2H, 400
2
MHz, 20 °C): δ 82.7 (d, J CH ) 150 Hz; CH3CH(O2CCF3)-
2
CH2CH3), 31.1 (t, J CH ) 123 Hz; CH3CH(O2CCF3)CH2CH3),
20.5 (q, 2J CH ) 128 Hz; CH3CH(O2CCF3)CH2CH3), 10.9 (q, 2J CH
) 115 Hz; CH3CH(O2CCF3)CH2CH3).
(d fep e)P d (Me)(Cl) (8). To a flask charged with 0.345 g
(1.30 mmol) of (cod)Pd(Me)(Cl) and 25 mL of CH2Cl2 was added
0.550 mL (1.95 mmol) of dfepe via syringe at 20 °C. After 24
h the solution was concentrated to 10 mL and cooled to -78
°C. The resulting white solid, 8 (0.64 g (68%)), was collected
by filtration and dried under vacuum. Anal. Calcd for C11H7-
ClF20P2Pd: C, 18.27; H, 0.98. Found: C, 17.91; H, 0.75. IR
(cm-1): 1459(m), 1377(w), 1301(m), 1228(s), 1135(m), 963(m),
866(w), 808(w), 752(m). 1H NMR (acetone-d6, 270 MHz, 20
Eth ylen e Dim er iza tion Stu d ies. A typical experiment
was as follows: 5 mL of CF3CO2H and (dfepe)2Pd (20 mg, 0.016
mmol) were placed in a 3 oz Fisher-Porter bottle fitted with a
magnetic stirbar, gas inlet valve, and 160 psi pressure gauge
(1 psi increments); 100 psi was admitted at 25 °C, and after
the initial pressure drop due to the equilibration of ethylene
into solution, the uptake of gas was monitored by following
the pressure drop as a function of time.
3
°C): δ 3.33 (m, 2H, PCH2), 3.03 (m, 2H, PCH2), 1.63 (d, J PH
Rea ction of (d fep e)P d (Me)(O2CCF 3) w ith P r op ylen e.
Trifluoroacetic acid (1 mL) and (dfepe)Pd(Me)(O2CCF3) (20 mg,
0.025 mmol) were placed in an NMR tube fitted with a Teflon
valve, pressurized with ∼2 atm propylene, and heated to 70
°C. After 24 h, 1H and 13C NMR indicated the formation of
) 7 Hz, 3H, PdCH3). 31P NMR (acetone-d6, 270 MHz, 20 °C):
δ 72.9 (m), 56.3 (m).
(d fep e)P d (Me)(O2CCF 3) (9). To a flask charged with 0.206
g (0.601 mmol) of (cod)Pd(Me)(O2CCF3) (prepared as described
above) and 20 mL of toluene was added 0.25 mL (0.90 mmol)
of dfepe via syringe at 20 °C. An immediate reaction took place
to give a tan solid. Removal of toluene and addition of a
petroleum ether/Et2O (ca. 2:1, 30 mL) gave 9 (0.345 g (72%),
which was isolated by filtration and dried under vacuum. 9
was stored at -40 °C under nitrogen. Anal. Calcd for
t
iPr(O2CCF3), 1 equiv of Bu(O2CCF3), and (dfepe)2Pd.
Ack n ow led gm en t. This work has been supported
by the National Science Foundation (Grant CHE-
9615985), the Wyoming DOE-EPSCoR Program, and
the donors of the Petroleum Research Fund, adminis-
tered by the American Chemical Society. The Nebraska
Center for Mass Spectrometry is acknowledged for
MSCI data.
C
13H7F23O2P2Pd: C, 19.50; H, 0.88. Found: C, 19.13; H, 0.81.
1H NMR (CF3CO2D, 270 MHz, 20 °C): δ 2.99 (m, 2H, PCH2),
2.67 (m, 2H, PCH2), 1.84 (d, J PH ) 8.1 Hz, 3H, PdCH3). 31P
3
NMR (C6D6, 270 MHz, 20 °C): δ 74.3 (m), 55.6 (m). 19F NMR
(C6D6, 400 MHz, 20 °C): δ -72.39 (s, O2CCF3), -78.27 (s,
PCF2CF3), -78.91 (s, PCF2CF3), -107.82 (overlapping ABX
OM990179Q