Acidities of platinum(II) μ-hydroxo complexes bearing diphosphine ligands

Article abstract of DOI:10.1021/ic049463m

The pK_a values in DMSO of the monoprotic complexes [(L ₂Pt)₂(μ-OH)(μ-NMePh)]²⁺ (4) (L ₂ = Ph₂PCH₂CH₂PPh₂ (dppe), Ph₂PCMe₂PPh₂ (dppip)) are 11.9 ± 0.1 (L₂ = dppe) and 13.5 ± 0.2 (L₂ = dppip) as determined by ³¹P NMR equilibrium titration with bases of known pK_a. Complexes 4 were prepared by treatment of [L₂Pt(μ- OH)]₂²⁺ (1) with N-methylaniline. The oxo complexes [(L₂Pt)₂(μ-O)(μ-NMePh)]⁺, formed in the equilibrium titration reactions, were independently synthesized in THF by deprotonation of [(L₂Pt)₂(μ-OH)(μ-NMePh)] ²⁺ with NaN(SiMe₃)₂ and characterized as NaBF₄ adducts. Similar experiments with diprotic [L ₂Pt(μ-OH)]₂²⁺ (L₂ = dppe, Ph ₂PCH₂CH₂CH₂-PPh₂ (dppp)) were complicated by exchange processes and were less conclusive, giving pK _a1 < 18 and pK_a2 > 18 in DMSO.

Full text of DOI:10.1021/ic049463m

Inorg. Chem. 2004, 43, 6780−6785
Acidities of Platinum(II)
Ligands
µ-Hydroxo Complexes Bearing Diphosphine
U. Anandhi and Paul R. Sharp*
125 Chemistry, UniVersity of Missouri, Columbia, Missouri 65211
Received April 26, 2004
+
The pKa values in DMSO of the monoprotic complexes [(L2Pt)2(
µ
-OH)(
µ
±
-NMePh)]2 (4) (L2
0.2 (L2
)
Ph2PCH2CH2PPh2
(dppe), Ph2PCMe2PPh2 (dppip)) are 11.9
±
0.1 (L2
)
dppe) and 13.5
)
dppip) as determined by 31P
2
NMR equilibrium titration with bases of known pKa. Complexes 4 were prepared by treatment of [L2Pt(
with N-methylaniline. The oxo complexes [(L2Pt)2( -O)(
were independently synthesized in THF by deprotonation of [(L2Pt)2(
characterized as NaBF4 adducts. Similar experiments with diprotic [L2Pt(
µ
-OH)]2 + (1)
µ
µ
-NMePh)]+, formed in the equilibrium titration reactions,
+
µ
-OH)(
µ
µ
-NMePh)]2 with NaN(SiMe3)2 and
2+
-OH)]2 (L2
)
dppe, Ph2PCH2CH2CH2-
PPh2 (dppp)) were complicated by exchange processes and were less conclusive, giving pKa1 < 18 and pKa2 > 18
in DMSO.
Introduction
a quantitative knowledge of the acidity of late transition metal
hydroxo and amido complexes (i.e., pKa values) is needed.
In this paper, we report our efforts to determine the pKa
values of dinuclear platinum(II) bridging hydroxo complexes
supported by phosphine ligands. We have chosen the
nonprotic, polar solvent DMSO for these measurements for
the following reasons: (1) The hydroxo and oxo complexes
are readily soluble and stable in DMSO, (2) the strong
solvating properties of DMSO reduce concerns about ion
pairing effects, and (3) DMSO has been used extensively
for pKa determinations of very weak acids, giving a large
selection of comparison acids for our measurements.3,14,15
Successful pKa determinations are reported along with the
synthesis of new hydroxo and oxo complexes required to
optimize the system behavior. Acetonitrile, another common
nonaqueous, nonprotic solvent for pKa measurements (see
Discussion), could not be used because of poor stability of
the oxo complexes in solution.
Proton transfer is an important process in many areas of
chemistry.1-5 The tendency of a compound to undergo proton
transfer is usually expressed in terms of its Brønsted acidity
as indicated by the pKa value for the compound. The pKa
value of a compound is also an indication of the basicity of
its conjugate base. Late transition metal oxo and imido
complexes have been noted, in many cases, to display
reaction chemistry consistent with strong Brønsted basicity.6-9
Strong nonaqueous bases are frequently used to deprotonate
the conjugate hydroxo and amido complexes, a major
synthetic route to late transition metal oxo and imido
complexes, and reactions of oxo and imido complexes with
weak acids frequently lead to protonation of the oxo or imido
ligand.10-13 To understand the chemistry of these complexes,
* To whom correspondence should be addressed. E-mail: SharpP@
missouri.edu.
(1) Angelici, R. J. Acc. Chem. Res. 1995, 28, 51-60.
(2) Bartmess, J. E.; McIver, R. T. In Gas-Phase Basicities; Bowers, M.
T., Ed.; Academic Press: New York, 1979; p 87.
(3) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463.
(4) Pu, L.; Olmstead, M. M.; Power, P. P.; Schiemenz, B. Organometallics
1998, 17, 5602-5606.
Results
Dihydroxo Complexes. We began this investigation with
the diprotic dihydroxo complexes [L2Pt(µ-OH)]22+ (1) (L )
a phosphine), which are known to be doubly deprotonated
in THF by two or more equivalents of MN(SiMe3)2 (M )
(5) Berners-Price, S. J.; Appleton, T. G. Platinum-Based Drugs Cancer
Ther. 2000, 3-35.
(6) Sharp, P. R. J. Chem. Soc., Dalton Trans. 2000, 2647-2657.
(7) Sharp, P. R. Comments Inorg. Chem. 1999, 21, 85-114.
(8) Dobbs, D. A.; Bergman, R. G. Organometallics 1994, 13, 4594-4605.
(9) Glueck, D. S.; Wu, J.; Hollander, F. J.; Bergman, R. G. J. Am. Chem.
Soc. 1991, 113, 2041-2054.
(12) Link, H.; Fenske, D. Z. Anorg. Allg. Chem. 1999, 625, 1878-1884.
(13) Mindiola, D.; Hillhouse, G. J. Am. Chem. Soc. 2001, 123, 4623-
4624.
(14) Bordwell, F. G.; Zhang, X. M. Acc. Chem. Res. 1993, 26, 510-517.
(15) Taft, R. W.; Bordwell, F. G. Acc. Chem. Res. 1988, 21, 463-469.
(10) Fulton, J. R.; Holland, A. W.; Fox, D. J.; Bergman, R. G. Acc. Chem.
Res. 2002, 35, 44-56.
(11) Bergman, R. G. Polyhedron 1995, 14, 3227-3237.
6780 Inorganic Chemistry, Vol. 43, No. 21, 2004
10.1021/ic049463m CCC: $27.50
© 2004 American Chemical Society
Published on Web 09/11/2004

Acidities of Platinum(II) µ-Hydroxo Complexes
Li, Na, K) to give the dioxo complexes [L2Pt(µ-O)]2 (2).16-18
Presumably, this occurs in two separate steps, giving the
hydroxo-oxo complexes [(L2Pt)2(µ-OH)(µ-O)]+ (3) in the
first deprotonation (eq 1) and the dioxo complexes 2 in the
second deprotonation (eq 2). Each of these deprotonations
will have an associated pKa for the acid: pKa1 for the
dihydroxo complex 1 and pKa2 for the oxo hydroxo complex
3. Most likely, pKa2 is significantly greater than pKa1.
Chart 1. Acid Form of the Bases Used in This Study (Ionizable
Hydrogen Atoms Shown)
Scheme 1
As in THF, double deprotonation of dihydroxo complex
1 (L2 ) dppp ) Ph2PCH2CH2CH2PPh2) by 2 equiv of MN-
(SiMe3)2 is also observed in DMSO, indicating that pKa1 and
pKa2 in DMSO are less than pKa of HN(SiMe3)2, which is
26.19 The fluorenyl anion (pKa ) 22 for fluorene),3 also
completely deprotonates 1 (L2 ) dppp) but the weaker base
9-phenyl fluorenyl anion, derived by deprotonation of
9-phenyl fluorene (pKa ) 17.9),3 gives different results. (See
Chart 1 for the structures of the acid form of the bases used
in this study.) With excess 9-phenyl fluorenyl, two equal
area doublets with satellites appear in the 31P NMR spectrum.
This same set of doublets, although considerably broadened,
is observed in the reverse reaction of the dioxo complex
[(dppp)Pt(µ-O)]2 (2) with excess 9-phenyl fluorene and in a
1:1 mixture of dioxo complex [(dppp)Pt(µ-O)]2 (2) and
dihydroxo complex 1 (Scheme 1). We therefore assign these
signals to the hydroxo-oxo complex [(L2Pt)2(µ-OH)(µ-O)]+
(3) (L2 ) dppp). 195Pt-31P coupling constants are consistent
with this formulation where one doublet, corresponding to
the P trans to the strong donor oxo ligand, shows a small
195Pt-31P coupling constant of 2488 Hz and the other, trans
to the weaker donor hydroxo ligand, shows a larger 195Pt-
31P coupling constant of 3948 Hz. Unfortunately, attempts
to isolate this complex have given only dihydroxo 1.
However, these observations indicate that 9-phenyl fluorenyl
to proton exchange between the various solution species),
we sought the preparation of more sterically hindered
monoprotic analogues of the dihydroxo complexes. Steric
hindrance would also protect the Pt centers from nucleophilic
attack, which we thought might be causing the instability of
the complexes in solution. The dihydroxo complexes are
known to undergo aminolysis with primary amines giving
2+
amido complexes [L2Pt(µ-NHR)]2 and [(L2Pt)2(µ-OH)(µ-
NHR)]2+ where the amido groups can also be deprotonated.20
Secondary amines should give amido complexes without
ionizable hydrogen atoms. Treatment of [L2Pt(µ-OH)]22+ (L2
) dppip (Ph2PCMe2PPh2), dppe) with N-methyl aniline
readily yields the amido-hydroxo complexes [(L2Pt)2(µ-
NMePh)(µ-OH)]2+ (4) (eq 3, L2 ) dppip, dppe).
2+
is capable of only the first deprotonation of [L2Pt(µ-OH)]2
1 (L2 ) dppp) and that therefore pKa1 < 18 and pKa2 > 18.
Similar results were obtained with 1 (L2 ) dppe ) Ph2PCH2-
CH2PPh2) and the fluorenyl anion but the DMSO solutions
are unstable and produce an unidentified precipitate.
Amido-Hydroxo Complexes. To simplify this system and
to eliminate the broadness in the 31P NMR spectra (attributed
Spectroscopic data are consistent with the expected
structure and are similar to the amido-hydroxo complexes
derived from primary amines.20 The 31P NMR spectrum of
isolated 4 (L2 ) dppe) in CH2Cl2 shows two doublets flanked
by 195Pt satellites. The small P-P coupling of 5 Hz is
(16) Li, W.; Barnes, C. L.; Sharp, P. R. J. Chem. Soc., Chem. Commun.
1990, 1634-1636.
(17) Li, J. J.; Li, W.; Sharp, P. R. Inorg. Chem. 1996, 35, 604-613.
(18) Anandhi, U.; Holbert, T.; Lueng, D.; Sharp, P. R. Inorg. Chem. 2003,
42, 1282-1295.
(19) Bordwell, F. G. as cited in: Grimm, D. T.; Bartmess, J. E. J. Am.
Chem. Soc. 1992, 114, 1227-1231.
(20) Li, J. J.; Li, W.; James, A. J.; Holbert, T.; Sharp, T. P.; Sharp, P. R.
Inorg. Chem. 1999, 38, 1563-1572.
Inorganic Chemistry, Vol. 43, No. 21, 2004 6781

Anandhi and Sharp
Table 1. Crystallographic and Data Collection Parameters for 4 (L2 )
Dppe)
formula
fw
C59H57NOP4Pt2B2F8‚2CH2Cl2‚C7H8 a, Å
12.9775(19)
13.933(2)
19.257(3)
1745.72
b, Å
c, Å
space group P21/m
V, Å3
3478.6(9)
â, deg 92.492(2)
λ, Å 0.71070
T, °C -100
d
calc, g/cm3 1.67
µ, mm-1
4.33
R1a, wR2b
0.0306, 0.1028
Z
2
2
2
2
a R1 ) (Σ||Fo| - |Fc||)/Σ|Fo|. b wR2 ) [(Σw(Fo - Fc )2)/Σw(Fc )2]1/2
.
consistent with the cis-phosphine geometry at the Pt centers.
The P-Pt coupling constants for 4 are similar in magnitude
to those of related dihydroxo and diamido complexes.17,18,20
The upfield doublet shows larger P-Pt coupling (3960 Hz)
than the downfield doublet (2980 Hz) and is assigned to the
P atom trans to the more weakly donating hydroxo group.
The downfield doublet is then assigned to the P atom trans
to the more strongly donating amido group. Changing the
NMR solvent to DMSO markedly alters the appearance of
the 31P NMR spectrum of 4 (L2 ) dppe). Instead of two
doublets only a singlet is observed. However, two sets of
195Pt satellites are observed, yielding similar JPt-P values to
those in CH2Cl2 and indicating that the 31P NMR resonances
for the phosphorus trans to the hydroxo group and the
phosphorus trans to the amido group are accidentally
coincident in DMSO.
The 31P NMR spectrum of 4 (L2 ) dppip) in both DMSO
and CH2Cl2 shows two doublets for the phosphorus atoms
trans to the hydroxo group and the amido group. The small
P-P coupling of 61 Hz is consistent with a cis-phosphine
geometry at the Pt centers. Each doublet displays 195Pt
satellites with P-Pt coupling constants, indicative of a trans
hydroxo and a trans amido group (3432 and 2518 Hz).
1H NMR spectra of both derivatives of 4 in DMSO-d6
show a broad triplet for the N-Me protons at 2.93 (L2 )
dppe) or 3.30 ppm (L2 ) dppip) and peaks corresponding
to the chelated bidentate phosphine ligand. Decoupling
experiments indicate that the triplet pattern is due to coupling
of the methyl group to two of the phosphorus nuclei
(presumably those trans to the amido group). The peak for
the hydroxo group appears at -0.61 (L2 ) dppe) or 2.63
ppm (L2 ) dppip). The structure of 4 (L2 ) dppe) was
confirmed by a single-crystal X-ray diffraction study (Table
1). A drawing of the cationic portion is shown in Figure 1.
Metrical parameters are presented in Table 2 and closely
resemble those of related complexes.20
Figure 1. Thermal ellipsoid (50%) plot of the cationic portion of 4 (L )
dppe). Labeled and unlabeled atoms are related by a mirror in the O1-
N1-C3 plane. Hydrogen atoms (except H1O) have been omitted for clarity.
Table 2. Selected Distances (Å) and Angles (deg) for 4 (L2 ) Dppe)
Pt1-P1
Pt1-P2
Pt1-O1
2.2178(11)
2.2353(11)
2.100(3)
Pt1-N1
N1-C3
N1-C51
2.138(3)
1.479(7)
1.442(8)
P1-Pt1-P2
O1-Pt1-P1
N1-Pt1-P1
O1-Pt1-P2
N1-Pt1-P2
85.24(4)
176.82(13)
100.58(11)
95.01(9)
O1-Pt1-N1
Pt1-O1-Pt1
Pt1-N1-Pt1
C3-N1-Pt1
C51-N1-C3
79.57(15)
101.17(17)
98.7(2)
113.6(2)
113.6(5)
170.89(13)
The 31P NMR spectrum of 5 (L2 ) dppe) in DMSO shows
two singlets flanked by 195Pt satellites. The cis coupling
between the two phosphorus atoms is evidently too small to
be observed. The 31P NMR spectrum of 5 (L2 ) dppip) in
DMSO shows a pair of doublets flanked by 195Pt satellites.
The cis coupling between the two phosphorus atoms (41 Hz)
is 20 Hz less than the starting amido-hydroxo complex 4
(L2 ) dppe). For both derivatives, the phosphorus atom trans
to the oxo group is assigned to the downfield resonances
with the smaller P-Pt coupling constants (dppip: 2344 Hz,
dppe: 2754 Hz) and the phosphorus atom trans to the amido
group to the resonances with the larger coupling constants
(dppip: 3185 Hz, dppe: 3436 Hz). The 1H NMR spectra of
5 show peaks corresponding to the bidentate phosphine
ligands and broad triplet N-Me peaks at 2.77 (L2 ) dppe)
and 2.83 ppm (L2 ) dppip).
pKa Determinations. The pKa values for 4 are obtained
by determining Keq for the reaction shown in eq 5 when the
pKa for the acid BH is known (see Chart 1 for the structures
of BH). The relevant mathematical relationships appear in
eqs 6 and 7. The equilibrium ratios of 4 and 5 are obtained
by 31P NMR spectroscopy and from these the equilibrium
concentrations of 4, 5, BH, and B- are calculated given the
initial concentrations. Practically, the pKa for the acid BH
must be similar ((2) to the pKa of 4 to give modest Keq
values and reasonable concentrations of each species. As-
suming that pKa1 of the dihydroxo complexes 1 is similar to
the pKa of 4, our initial experiments suggest a pKa for 4 of
less than 18. This is born out by the complete deprotonation
of 4 (L2 ) dppe, dppip) (i.e., Keq is too large to measure) by
fluorenyl anion (B-), giving 5 and fluorene (pKa ) 22.6).3
Similarly, 9-phenyl fluorenyl anion (pKa ) 17.9 for fluo-
rene)3 and 2,6-di-tert-butylphenoxy anion (pKa ) 16.85 for
Amido-Oxo Complexes. As anticipated, 4 are readily
deprotonated by NaN(SiMe3)2 in THF or DMSO to give
solutions of the amido-oxo complexes [(L2Pt)2(µ-NMePh)-
(µ-O)]+ (5) (eq 4). Both complexes are isolated from THF
and contain 1 equiv of NaBF4 probably through oxo group
adduct formation similar to LiBF4 adducts isolated for related
oxo complexes.17,18,20
6782 Inorganic Chemistry, Vol. 43, No. 21, 2004

Acidities of Platinum(II) µ-Hydroxo Complexes
Table 3. pKa of 4 (DMSO, 25 °C)
[(L2Pt)2(µ-NMePh)(µ-OH)]2+ 4
a
pKa
L2 ) dppe
L2 ) dppip
11.9 (0.1)b
13.5 (0.2)c
a Average of 4 measurements, standard deviation in parentheses. b BH
) 9-carbomethoxy fluorene (pKa ) 10.35). c BH ) pentaphenylcylcopen-
tadiene (pKa ) 12.5).
2,6-di-tert-butylphenol) give 5 and the organic acid with no
detectable equilibrium, indicating that the pKa for 4 is less
than ca. 15. A detectable equilibrium mixture is finally
obtained with the base 9-carbomethoxy fluorenyl anion (pKa
) 10.35)3 for 4 (L2 ) dppe). Successive additions of this
base allow multiple determinations of Keq. The averaged
value is given in Table 3. With time and with increasing
concentrations of the base small additional peaks grow into
the 31P NMR spectra, indicating reactions other than depro-
tonation (probably attack of base at the platinum centers)
are occurring. However, the extent of these reactions is small
and does not significantly affect the determined pKa values
as shown by the good agreement between successive values
(see Supporting Information). Poor stability of the solutions
is more serious with acetate ion (pKa ) 12.3).3 Sodium
acetate addition to a DMSO solution of 4 (L2 ) dppe) gives
5 but the solution is unstable, yielding (dppe)Pt(OAc)2.
probably not more than a few pKa units below that for HN-
(SiMe3)2 (pKa ) 2619). However, the poor solubility of 1 in
the reaction solvent (THF) prevented an assessment of the
pKa values. This problem is eliminated in DMSO, which
readily dissolves all of the complexes. Although we were
unable to determine pKa1 values for 1, they are probably very
similar to the pKa values for 4.
Despite the unexpectedly low pKa values for 4, the values
still indicate a basic µ-oxo complex relative to other
µ-hydroxo complexes for which the pKa has been determined.
(Terminal hydroxo complex pKa values are not comparable
as these are generally expected to be higher than µ-hydroxo
complexes.21) Two Mn complexes (6 and 7, Chart 2) that
are structurally related to 1 and 4 in that they contain a M(µ-
OH)2M core have pKa1 values in MeCN of 11.5 and 6.5.22,23
Comparisons of pKa values of various organic acids, includ-
ing cationic acids, in DMSO and MeCN show that pKa values
in MeCN are 6-10 units higher than those in DMSO.3,24
Assuming that these results apply to the Mn complexes 6
and 7, their pKa values in DMSO should be far below those
for 4. Even pKa2 (13.4 in MeCN) for 7 should be below the
pKa values for 4.23 However, reducing 7 from a Mn(IV),-
Mn(IV) state to a Mn(III),Mn(IV) state increases pKa2
dramatically to 24.5 in MeCN, likely above the pKa values
for 4. Reduction produces a neutral Mn(µ-OH)(µ-O)Mn
complex while all the other Mn complexes are cationic and
this likely contributes to the higher pKa2 value. Although
not as structurally similar to 4, complexes containing the
adamantine-like {Mn4O6}4+ core have pKa values in MeCN
that can range from 1.54 to 12.525 and a singly bridged
µ-hydroxo Fe-Cu complex (8, Chart 2) has a pKa value in
MeCN between 16.7 and 17.6, which in DMSO would be
below those of 4.26
In contrast to 4 (L2 ) dppe), 9-carbomethoxy fluorenyl
does not deprotonate 4 (L2 ) dppip). This suggests that the
dppip derivative has a higher pKa than the dppe derivative
and that bases with higher pKa values are needed to observe
equilibrium. Pentaphenylcyclopentadienyl (pKa ) 12.5)3 is
such a base and equilibrium is observed again, allowing
multiple determinations of the pKa (Supporting Information).
As with the dppe derivative, there is good agreement between
the successive values and the average is given in Table 3.
As a final check on the measurements, equilibrium
mixtures of hydroxo 4 (L2 ) dppip) and oxo 5 (L2 ) dppe)
were examined (eq 8). This gives a determination of the
relative pKa values for the two derivatives where Keq ) Ka-
(dppe)/Ka(dppip). Again, good agreement from successive
measurements was observed with an average ∆pKa (pKa-
(dppe) - pKa(dppip)) of -1.6 ( 0.1, in excellent agreement
with the difference in the measured pKa values (11.6 - 13.9
) -1.6).
pKa values for several other µ-hydroxo complexes have
been measured in water.25,27-34 Unfortunately, comparison
(21) For an example of a terminal hydroxo complex determination in DMSO
see: Gupta, R.; Borovik, A. S. J. Am. Chem. Soc. 2003, 125, 13234-
13242.
(22) Wang, K.; Mayer, J. M. J. Am. Chem. Soc. 1997, 119, 1470-1471.
(23) Baldwin, M. J.; Pecoraro, V. L. J. Am. Chem. Soc. 1996, 118, 11325-
11326.
(24) Kolthoff, I. M.; Chantooni, M. K.; Bhowmik, S. J. Am. Chem. Soc.
1968, 90, 23-28.
Discussion
(25) Dube, C. E.; Wright, D. W.; Bonitatebus, P. J., Jr.; Pal, S.; Armstrong,
W. H. J. Am. Chem. Soc. 1998, 120, 3704-3716.
(26) Kopf, M.-A.; Neuhold, Y.-M.; Zuberbuehler, A. D.; Karlin, K. D.
Inorg. Chem. 1999, 38, 3093-3102.
(27) Lebeau, E. L.; Adeyemi, S. A.; Meyer, T. J. Inorg. Chem. 1998, 37,
6476-6484.
The synthesis of [L2Pt(µ-O)]2 (2) by deprotonation of the
dihydroxo complex 1 with LiN(SiMe3)2 and the reaction
chemistry of 2 suggested that 2 is highly basic and therefore
1 is a very weak acid.17 We suspected that pKa1 for 1 was
Inorganic Chemistry, Vol. 43, No. 21, 2004 6783

Anandhi and Sharp
platinum hydroxo and amido complexes.36 Synthetic work
by our group indicates a greater acidity for the diamido
complexes [L2Pt(µ-NHAr)]2 with the small bite angle
Chart 2
2+
diphosphine, L2 ) dppip, over analogous complexes with
larger bite angle diphosphines (L2 ) dppp, dppb).18,20 The
dppip derivatives are readily deprotonated by LiN(SiMe3)2
while the dppp and dppb derivatives are unreactive. However,
this difference could result from either kinetic or thermo-
dynamic factors. A kinetic acidity increase would be expected
from decreased steric hindrance at the acidic centers. The
single carbon bridge of dppip results in a retraction of the
phosphorus phenyl groups away from the amido groups,
allowing greater access to the acidic protons. The pKa
measurements on 4 do show a thermodynamic acidity
difference between the dppip and the dppe complexes.
However, it is in the opposite direction to that observed in
the diamido complexes, strongly suggesting that the diamido
differences are indeed kinetic with a steric origin.
The lower acidity of 4 with L2 ) dppip as compared to
the dppe complex is consistent with a bite angle effect.
Reducing the P-Pt-P angle of the L2Pt fragment raises the
energy of the occupied frontier orbital that are involved in
bonding to the OH group.37,38 This transfers more electron
density to the OH group, decreasing its acidity as is observed.
Another possible contributing factor in the lower acidity of
the dppip derivative is the presence of the methyl groups on
the bridging carbon atom of the dppip ligand. This should
increase the electron donation from the phosphorus centers
as compared to the dppe derivative. Greater electron donation
results in a more electron-rich platinum center and more
electron density on the OH group. Unfortunately, separation
of these two factors by elimination of the methyl groups from
dppip is not possible. Previous work on the dppm hydroxo
complexes 1 (L2 ) dppm) has shown that the dppm
methylene group of the complex is more readily deprotonated
than the hydroxo groups.17,39
Finally, it should be noted that platinum oxo complexes,
including 5, are often, but not always,20 isolated in association
with Group 1 salts.17,39 Interaction of the oxo complexes with
these salts could influence the measured pKa values. X-ray
crystallography has revealed that in the solid-state Li ion
coordinates to the oxo group of platinum oxo complexes.
Similar interactions are likely for other Group 1 ions but
are expected to weaken as the Group 1 ion size increases.
Ion coordination to the oxo group is most likely also present
in THF solution. Pt-P coupling constants in THF have been
observed to decrease (∼100 Hz) in the dioxo complexes [L2-
Pt(µ-O)]2 (L ) a phosphine) as the associated salt ion is
changed from Li to Na.17 A weaker Na ion interaction with
the oxo group results in a more strongly donating oxo ligand,
which in turn weakens the Pt-P bond trans to the oxo ligand.
However, in more strongly donating solvents coordination
of the group 1 ion by solvent molecules is expected to
of these pKa values with DMSO results are problematic since
variations in hydrogen bonding can dramatically shift relative
values.3 For example, pKa for PhNH3+ in water and DMSO
differ by only one unit whereas acetic acid is a markedly
weaker acid in DMSO with a pKa of 12.3 versus 4.75 in
water.3
Such dramatic changes in pKa on transfer to water suggest
the possibility that 4 and 1 could be much more acidic in
water. Poor aqueous solubility prevents us from examining
our platinum hydroxo complexes in water but the aqueous
2+
chemistry of the bipyridyl analogue of 1, [L2Pt(µ-OH)]2
(L2 ) bpy), has been studied.35 This dihydroxo complex is
stable in neutral solution but on increasing the pH to 8 the
yellow complex converts to a red complex. This red complex
was assigned the tri-hydroxo bridged formulation [(L2Pt)2-
(µ-OH)3]+ (L2 ) bpy) primarily on the basis of elemental
analysis results. We propose that the complex should be
reformulated as the hydroxo-oxo complex [(L2Pt)2(µ-OH)-
(µ-O)]+ 3 (L2 ) bpy) formed by deprotonation of the
dihydroxo complex. The elemental analysis data fit this
formulation if a water of solvation is included. Future work
in our group will attempt to confirm this proposed formula-
tion.
One of the objectives of this work was to establish the
effect of the diphosphine bite angle on the acidity of dinuclear
(28) Thorp, H. H.; Sarneski, J. E.; Brudvig, G. W.; Crabtree, R. H. J. Am.
Chem. Soc. 1989, 111, 9249-9250.
(29) Manchanda, R.; Thorp, H. H.; Brudvig, G. W.; Crabtree, R. H. Inorg.
Chem. 1992, 31, 4040-4041.
(30) Springborg, J.; Zehnder, M. Acta Chem. Scand., Ser. A: Phys. Inorg.
Chem. 1987, 41, 484-95.
(31) Dalley, N. K.; Kou, X.; O’Connor, C. J.; Holwerda, R. A. Inorg. Chem.
1996, 35, 2196-201.
(32) Springborg, J.; Toftlund, H. Acta Chem. Scand., Ser. A: Phys. Inorg.
Chem. 1979, 33, 31-9.
(36) For a review of bite angle effects see: Dierkes, P.; van Leeuwen, P.
W. N. M. J. Chem. Soc., Dalton Trans. 1999, 1519-1529.
(37) Hofmann, P., In Organometallics in Organic Synthesis; de Meijiere,
A., tom Dieck, H., Eds.; Springer: Berlin, 1987; p 1.
(38) Yoshida, T.; Tatsumi, K.; Otsuka, S. Pure Appl. Chem. 1980, 52, 713-
727.
(33) Turowski, P. N.; Armstrong, W. H.; Liu, S.; Brown, S. N.; Lippard,
S. J. Inorg. Chem. 1994, 33, 636-645.
(34) Hage, R.; Krijnen, B.; Warnaar, J. B.; Hartl, F.; Stufkens, D. J.; Snoeck,
T. L. Inorg. Chem. 1995, 34, 4973-4978.
(35) Wimmer, S.; Castan, P.; Wimmer, F. L.; Johnson, N. P. J. Chem.
Soc., Dalton Trans. 1989, 403-412.
(39) Li, J. J.; Sharp, P. R. Inorg. Chem. 1994, 33, 183-184.
6784 Inorganic Chemistry, Vol. 43, No. 21, 2004

Acidities of Platinum(II) µ-Hydroxo Complexes
6H, CMe2) Anal. Calcd (Found) for C61H61B2F8NOP4Pt2‚2CH2-
Cl2: C, 44.99 (44.99); H, 4.12 (3.89); N, 0.83 (0.68).
become more important and ultimately dominate. This is
evident in the decrease (200 Hz) in the Pt-P coupling
constant observed when the dioxo complex [(dppp)Pt(µ-O)]2‚
2LiOTf is dissolved in strongly solvating DMSO as com-
pared to THF.40 Data on 5 indicate a weak to nonexistent
interaction with all group 1 ions in DMSO. Pt-P coupling
constants trans to the oxo group are nearly invariant for 5 in
the presence of Li (2757 Hz), Na (2754 Hz), or K (2745
Hz) ions. Furthermore, the Pt-P coupling constant of 5 in
DMSO solution in the presence of K ions remains unchanged
on addition of 18-crown-6, a strong binder of K ions.41 Ion
interactions with 5 in DMSO do not appear to be important
in the pKa values for 4 determined here.
[(dppe)2Pt2(µ-NMePh)(µ-O)](BF4)‚NaBF4 (5) (L2 ) dppe).
NaN(SiMe3)2 (0.009 g, 0.0491 mmol) in 1 mL of THF is added
dropwise to a stirred suspension of [(dppe)2Pt2(µ-NMePh)(µ-OH)]-
(BF4)2 (4) (0.050 g, 0.033 mmol) in 3 mL of THF. During the base
addition the colorless suspension slowly becomes yellow and after
complete addition a clear yellow solution forms. With continued
stirring, the dark yellow crystalline product precipitates. This is
filtered off, washed with petroleum ether, and dried in vacuo.
Yield: 0.035 g (70%). 31P {1H} NMR (THF): 35.8 (s with satellites,
1
1JPt-P ) 2768 Hz), 21.2 (s with satellites, JPt-P ) 3430 Hz). 31P
1
{1H} NMR (DMSO): 35.8 (s with satellites, JPt-P ) 2754 Hz),
21.2 (s with satellites, 1JPt-P ) 3436 Hz). 1H NMR (CD2Cl2): 8.18-
6.56 (m, 45H, Ph), 2.77 (br t, 4JP-H ) 5 Hz, 3H, N-Me), 2.18 (br,
4H, CH2), 1.80 (br, 4H, CH2). 19F {1H} NMR (DMSO): -154.1
(s, BF4). Anal. Calcd (Found) for C59H56B2F8NNaOP4Pt2‚C4H8O:
C, 47.96 (47.93); H, 4.08 (3.89); N, 0.89 (1.16).
Experimental Section
Experiments were performed under a dinitrogen atmosphere in
a Vacuum Atmospheres Corporation drybox. DMSO was dried
according to the literature procedure.42 Other solvents were dried
by standard techniques and stored under dinitrogen over 4 Å
molecular sieves or sodium metal. The precursor complexes 1 [L2-
Pt(u-OH)]2+ (L2 ) dppe, dppip) were synthesized by known
methods.17,18 N-methyl aniline, MN(SiMe3)2 (M ) Na, K), fluorene,
9-phenyl fluorene, 2,6-di-tert-butylphenol, and pentaphenylcyclo-
pentadiene were obtained from commercial sources. 9-Car-
bomethoxy fluorene was prepared according to the reported
procedure.42 Bases were prepared by treating the neutral organic
acids with 1 equiv of KN(SiMe3)2 in ether. The precipitated product
was washed thoroughly with hexane or petroleum ether and dried
in vacuo. NMR spectra were recorded on a Bruker AMX-250
spectrometer at 25 °C. Shifts are given in ppm with positive values
downfield of TMS (1H), external H3PO4 (31P), or CFCl3 (19F). Desert
Analytics performed the microanalyses (inert atmosphere). The
presence of CH2Cl2 or THF of crystallization in the analyzed
[(dppip)2Pt2(µ-NMeC6H4)(µ-O)](BF4)‚NaBF4 (5) (L2 ) dp-
pip). The procedure is the same as that used for 5 (L2 ) dppe).
1
Yield: 0.040 g (78%). 31P {1H} NMR (DMSO): -14.7 (d, JPt-P
2
1
2
) 2344 Hz, JP-P ) 40.5 Hz), -21.9 (d, JPt-P ) 3185 Hz, JP-P
) 40.5 Hz). 1H NMR (DMSO-d6): 8.64-6.97 (m, 45H, Ph), 2.83
4
3
(br t, JP-H ) 5 Hz, 3H, N-Me), 1.36 (t, JP-H ) 15.0 Hz, 12H,
CMe2), 1.03 (t, JP-H ) 15.0 Hz, 12H, CMe2). 19F {1H} NMR
3
(DMSO): -148.1 (s, BF4). Anal. Calcd (Found) for C61H60B2F8-
NNaOP4Pt2‚0.5C4H8O: C, 48.20 (48.18); H, 4.11 (3.93); N, 0.89
(1.24).
General Procedure for the pKa Determinations. Solid hydroxo
complex 4 (40-50 mg) was placed in a 5 mL vial and ca. 1.5 mL
of DMSO was added to dissolve the solid. A portion of solid base
was then added with stirring. After the base had dissolved, an NMR
aliquot was removed and the 31P NMR spectrum recorded. The
aliquot was returned to the vial and another sample of the base
added. An NMR aliquot was removed and the 31P NMR spectrum
recorded. This procedure was repeated until the 31P NMR spectrum
indicated almost complete conversion of 4 to 5. The mmol of
hydroxo complex 4 and oxo complex 5 present in the solution are
calculated from the integrated intensity ratios. The amount of base
consumed is determined by subtracting the amount of oxo complex
formed from the total amount of base added. (See Tables S1 and
S2 in the Supporting Information.)
Procedure for 4/5 (L2 ) dppe and dppip) Equilibrium
Measurements. The procedure was similar to that described for
the pKa determinations except that oxo complex 5 (L ) dppip)
was added in place of the base to a solution of hydroxo complex
4 (dppe). The ratio of all species is given by NMR integration and
is used to directly calculate the equilibrium constant. (See Table
S3.)
1
samples was confirmed by H NMR spectroscopy in DMSO-d6.
[(dppe)2Pt2(µ-NMeC6H4)(µ-OH)](BF4)2 (4) (L2 ) dppe). To
a stirred solution of [Pt(µ-OH)(dppe)]2(BF4)2 (0.100 g, 0.0717
mmol) in 5 mL of CH2Cl2 is added N-methyl aniline (0.012 g, 0.11
mmol). The yellow solution is stirred overnight, concentrated under
reduced pressure, filtered, and layered with an equal amount of
toluene. Pale yellow crystals of the product, suitable for X-ray
analysis, form overnight. Yield: 0.080 g (75%). 31P {1H} NMR
(DMSO): -34.1 (s, 1JPt-P ) 3740 Hz, 1JPt-P ) 2980 Hz). 31P {1H}
NMR (CH2Cl2): -37.1 (d, 1JPt-P ) 3960 Hz, 2JP-P ) 5 Hz), -32.4
2
1
(d,1JPt-P ) 2888 Hz, JP-P ) 5 Hz). H NMR (CD2Cl2): 7.79-
4
6.58 (m, 45H, Ph), 2.93 (br t, JP-H ) 5 Hz, N-Me, 3H), 1.77-
2.83 (m, 8H, CH2), -0.61 (s, 1H, OH). Anal. Calcd (Found) for
C59H57B2F8NOP4Pt2‚2CH2Cl2: C, 44.29 (44.33); H, 3.72 (3.68);
N, 0.848 (1.30).
Acknowledgment. Support from the Chemical Sciences,
Geosciences and Biosciences Division, Office of Basic
Energy Sciences, Office of Science, U.S. Department of
Energy (DE-FG02-88ER13880) is gratefully acknowledged.
A grant from the National Science Foundation (9221835)
provided a portion of the funds for the purchase of the NMR
equipment. We thank Dr. C. Barnes for assistance with the
X-ray experiment.
[(dppip)Pt2(µ-NMeC6H4)(µ-OH)](BF4)2 (4) (L2 ) dppip). The
procedure is the same as that used for 2 (L2 ) dppip) except the
white crystalline product is obtained by layering the reaction mixture
with diethyl ether. Yield: 0.075 g (71%). 31P {1H} NMR (DMSO):
2
1
-27.4 (d,1JPt-P ) 3432 Hz, JP-P ) 61 Hz), -24.3 (d, JPt-P
)
2
1
2518 Hz, JP-P ) 61 Hz). H NMR (DMSO-d6): 8.48-6.49 (m,
45H, Ph), 3.30 (t, JP-H ) 5 Hz, 3H, N-Me), 2.63 (s, 1H, OH),
4
3
3
1.37 (t, JP-H ) 17.5 Hz, 6H, CMe2), 0.76 (t, JP-H ) 17.5 Hz,
(40) Flint, K.; Sharp, P. R. Unpublished results.
(41) Greenwood, N. N.; Earnshaw, A., Chemistry of the Elements, 2nd ed.;
Butterworth-Heinemann: Woburn, MA, 1997.
(42) Matthews, W. S.; Bares, J. E.; Bartmess, J. E.; Bordwell, F. G.;
Cornforth, F. J.; Drucker, G. E.; Margolin, Z.; McCallum, R. J.;
McCollum, G. J.; Vanier, N. R. J. Am. Chem. Soc. 1975, 97, 7006-
7014.
Supporting Information Available: X-ray crystallographic data
for 4 (L2 ) dppe) in the form of CIF files and example 31P NMR
spectra and data for the equilibrium measurements. This material
is available free of charge via the Internet at http://pubs.acs.org.
IC049463M
Inorganic Chemistry, Vol. 43, No. 21, 2004 6785