Catalytic effect of sulfoxides on the fluxional motion of 2,9-dimethyl-1,10-phenanthroline in a platinum(II) complex: Evaluation of steric and electronic contributions

Article abstract of DOI:10.1021/ic011206j

Addition of external weak nucleophiles to a chloroform solution of the cationic complex ion [Pt(Me)(dmphen)- (PPh₃)]⁺ (1) accelerates the fluxional motion of the symmetric chelating ligand 2,9-dimethyl-1,10 -phenanthroline (dmphen) between nonequivalent exchanging sites. The rates of the dynamic process can be measured by line-shape analysis of the ¹H NMR spectra. Concentration-dependent measurements were carried out with the ligands SOMe₂, SO(CH₂)₄, SO(n-Bu)₂, SO(sec-Bu)₂, SO(i-Pr)₂, SOEt(Ph), SOPr(Ph), SO(Bz)₂, SO(p-MeC₆H₄)₂, SOPh₂, SO-(p-ClC₆H₄)_2, SOMe(p-MeOC₆H₄), SOMe(p-MeC₆H₄), SOMe(Ph), SOMe(p-BrC₆H₄), and SOMe(p-ClC₆H₄). The rate constants k_obsd, when plotted against the concentration of the added ligands SOR(R'), give a family of straight lines with a common intercept, indicating that the two-term rate law k_obsd = k₁ + k₂[SOR(R′)] is obeyed. The same rate law applies to the displacement of SOMe₂ from [Pt(Me)(phen)(SOMe₂)]⁺ (2) (phen = 1,10-phenanthroline) by sulfoxides (SOMe₂, SO(i-Pr)₂, SOMe(p-MeOC₆H₄), SO(p-MeC₆H₄)₂, SOPh₂, SO(p-ClC₆H₄)₂, and SO(sec-Bu)₂). The fluxional rates in 1 are 6-7 orders of magnitude higher than the substitution rates in 2. The values of the rate constants for the two processes were resolved quantitatively into steric and electronic contributions by use of quantitative analysis of ligand effects (QALE). Inhibitory steric effects are linearly operative for the entire set of ligands, the rates of the reactions are enhanced with increasing electron donor capacity of the sulfoxides, and there is a small but significant E_ar effect that enhances the reactivity of the aryl sulfoxides. The strict similarity of the patterns of the two processes and of their dependence upon the stereoelectronic properties of the ligands, combined with the intrinsic lability of the platinum-nitrogen bonds, would suggest the operation of stereospecific consecutive ring-opening and ring-closure steps for the fluxional motion of dmphen in 1. However, the available evidence does not allow alternative mechanisms involving intramolecular rearrangements of the five-coordinate intermediate to be ruled out.

Full text of DOI:10.1021/ic011206j

Inorg. Chem. 2002, 41, 2839−2847
Catalytic Effect of Sulfoxides on the Fluxional Motion of
2,9-Dimethyl-1,10-phenanthroline in a Platinum(II) Complex: Evaluation
of Steric and Electronic Contributions
Raffaello Romeo,* Lucia Fenech, Salvatore Carnabuci, Maria Rosaria Plutino, and Andrea Romeo
Dipartimento di Chimica Inorganica, Chimica Analitica e Chimica Fisica and ISMN, CNR,
Sezione di Messina, UniVersita` di Messina, Salita Sperone 31, Villaggio S. Agata,
98166 Messina, Italy
Received November 27, 2001
Addition of external weak nucleophiles to a chloroform solution of the cationic complex ion [Pt(Me)(dmphen)-
(PPh3)]+ (1) accelerates the fluxional motion of the symmetric chelating ligand 2,9-dimethyl-1,10-phenanthroline
(dmphen) between nonequivalent exchanging sites. The rates of the dynamic process can be measured by line-
shape analysis of the 1H NMR spectra. Concentration-dependent measurements were carried out with the ligands
SOMe2, SO(CH ) , SO(n-Bu)2, SO(sec-Bu)2, SO(i-Pr)2, SOEt(Ph), SOPr(Ph), SO(Bz)2, SO(p-MeC H ) , SOPh2, SO-
2 4
6 4 2
(p-ClC H ) , SOMe(p-MeOC H ), SOMe(p-MeC H ), SOMe(Ph), SOMe(p-BrC H ), and SOMe(p-ClC H ). The rate
6
4 2
6
4
6
4
6
4
6 4
constants kobsd, when plotted against the concentration of the added ligands SOR(R′), give a family of straight lines
with a common intercept, indicating that the two-term rate law kobsd ) k1 + k2[SOR(R′)] is obeyed. The same rate
law applies to the displacement of SOMe2 from [Pt(Me)(phen)(SOMe2)]+ (2) (phen ) 1,10-phenanthroline) by
sulfoxides (SOMe2, SO(i-Pr)2, SOMe(p-MeOC H ), SO(p-MeC H ) , SOPh2, SO(p-ClC H ) , and SO(sec-Bu)2). The
6
4
6
4 2
6 4 2
fluxional rates in 1 are 6−7 orders of magnitude higher than the substitution rates in 2. The values of the rate
constants for the two processes were resolved quantitatively into steric and electronic contributions by use of
quantitative analysis of ligand effects (QALE). Inhibitory steric effects are linearly operative for the entire set of
ligands, the rates of the reactions are enhanced with increasing electron donor capacity of the sulfoxides, and
there is a small but significant E effect that enhances the reactivity of the aryl sulfoxides. The strict similarity of
ar
the patterns of the two processes and of their dependence upon the stereoelectronic properties of the ligands,
combined with the intrinsic lability of the platinum−nitrogen bonds, would suggest the operation of stereospecific
consecutive ring-opening and ring-closure steps for the fluxional motion of dmphen in 1. However, the available
evidence does not allow alternative mechanisms involving intramolecular rearrangements of the five-coordinate
intermediate to be ruled out.
Introduction
of the Pt-N1 (2.152(3) Å) and Pt-N2 (2.162(3) Å) bond
separations, caused by the trans influence of the methyl and
phosphine groups acting as strong σ donors. The interionic
solution structure was investigated in CDCl3 at low temper-
ature by 1H-NOESY and 19F{1H}-HOESY NMR spec-
troscopies, and the anions PF6- (1a) and BF4- (1b) showed
strong contacts with the cation [Pt(Me)(dmphen)(PPh3)]+,
being located preferentially on the side of the phenanthroline
ligand and thus forming tight ion pairs.
In a recent kinetic study1 we showed that the dinitrogen
ligand 2,9-dimethyl-1,10-phenanthroline (dmphen), in the
monoalkyl cationic complex ion [Pt(Me)(dmphen)(PPh3)]+
(1), is characterized by a fast fluxional motion between two
different exchanging sites. The X-ray structure of 1 showed
severe steric distortions in the coordination geometry and in
the phenanthroline ligand, as well as a significant elongation
* To whom correspondence should be addressed. Fax: ++39-090-
393756. Phone: ++39-090-391984. E-mail: Raf.Romeo@chem.unime.it.
(1) Romeo, R.; Fenech, L.; Monsu` Scolaro, L.; Albinati, A.; Macchioni,
A.; Zuccaccia, C. Inorg. Chem. 2001, 40, 3293-3302.
The dynamic exchange process was studied by variable-
temperature NMR spectroscopy, as a function of the nature
of the counteranion, of the solvent, or of added nucleophiles.
10.1021/ic011206j CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/04/2002
Inorganic Chemistry, Vol. 41, No. 11, 2002 2839

Romeo et al.
Scheme 1. Wire Frame Schematization of the Molecular Structures
of 1 and 2
The mechanism is switchable between dissociative and
associative pathways, depending on the interionic structure
of the complex and the action of potential nucleophiles. The
first process is prevalent within the ion pair formed by a
“noncoordinating” anion with the cationic complex and
involves (i) initial dissociation of the metal-nitrogen bond,
(ii) fast interconversion between two three-coordinate T-
shaped intermediates containing η1-coordinated dmphen, and
(iii) final ring closure. The associative mechanism is
promoted by addition of external ligands such as SO(n-Bu)2,
SO(sec-Bu)2, and S(sec-Bu)2, which have the effect of
accelerating the fluxionality rate without inducing intermo-
lecular exchange. We observed a significant nucleophilic
discrimination ability among these ligands, as a result of
differences in their electronic and steric properties. A variety
of associative mechanisms can be envisaged for the acceler-
ated flipping: (i) a sequence of nucleophilic substitution
reactions which take place with retention of configuration
or (ii) an intramolecular rearrangement from a five-coordinate
intermediate, such as a Berry pseudorotation or a “turnstile”
mechanism.2 We were inclined to think that the first process
was operative because of the correspondence observed
between the catalytic efficiency of the ligands in promoting
the fluxional motion and their nucleophilic power, even
though the comparison could have been made only on a fairly
low number of nucleophiles.
In this paper the dynamic exchange process of the
dinitrogen ligand on complex 1 in CDCl3 was investigated
by NMR spectroscopy, as a function of the nature and the
concentration of an extended series of sulfoxides of widely
different electronic and steric properties. The application of
QALE (quantitative analysis of ligand effects)3 to the rate
data provided a means of ascertaining the relative importance
of the electronic and steric properties of the sulfoxide ligands
in governing the fluxionality of the substrate. The way in
which the size and the electron-releasing ability of the
substituents on the sulfur atom influence the fluxionality of
1 was correlated with the effects observed in the bimolecular
substitution reaction of SOMe2 from [Pt(Me)(phen)(SOMe2)]+
(2) (phen ) 1,10-phenanthroline) with a number of different
sulfoxides. Although the mechanism for the fluxional motion
of dmphen in complex 1 could not be determined unambigu-
ously, the operation of a conventional consecutive displace-
ment mechanism would most readily account for the pattern
of behavior observed. This seems to be one of the first
attempts to use QALE as a diagnostic tool in assessing the
reaction mechanism.
Results
The structural features of compounds 1 and 2 have been
well established through X-ray diffraction studies.1,4 The
differences between the two complexes are significant (see
Scheme 1) as far as the distortion of the phenanthroline
ligand and the Pt-N separations are concerned. In the
4
complex [Pt(Me)(phen)(SOMe2)]PF6 the 1,10-phenanthro-
line ligand is planar and coordinates to the Pt atom with a
bite angle of 79.3(2)°; the Pt-N2 and Pt-N1 bond separa-
tions are significantly shorter than those of 1, being 2.135-
(4) vs 2.162(3) Å and 2.075(4) vs 2.152(3) Å, respectively.
The addition of external amounts of sulfoxides to 1 leads
simply to an acceleration in the intrinsic fluxional rate of
dmphen, while for 2 the only process observed is the
substitution of the coordinated SOMe2 by the entering
nucleophile.
Dynamic Behavior of Complex 1. The above-mentioned
dynamic process in the cationic [Pt(Me)(dmphen)(PPh3)]+
species was investigated in CDCl3 as a function of the nature
and of the concentration of the added ligands SOR(R′). At
low temperature (260 K), the NMR spectrum of 1, besides
the resonances of the triphenylphosphine (δ in the range of
7.54-7.41 ppm) and of a methyl group directly coordinated
(2) (a) Ugi, I.; Marquarding, D.; Klusacek, H.; Gillespie, P. Acc. Chem.
Res. 1971, 4, 288-296. (b) Casares, J. A.; Espinet, P. Inorg. Chem.
1997, 36, 5428-5431. (c) Stockland, R. A. Jr.; Anderson, G. K.
Organometallics 1998, 17, 4694-4699.
(3) (a) Wilson, M. R.; Liu, H.-Y.; Prock, A.; Giering, W. P. Organome-
tallics 1993, 12, 2044-2050. (b) Panek, J. S.; Prock, A.; Eriks, K.;
Giering, W. P. Organometallics 1990, 9, 2175-2176. (c) Rahman,
Md. M.; Liu, H.-Y.; Eriks, K.; Prock, A.; Giering, W. P. Organome-
tallics 1989, 8, 1-7. (d) Rahman, Md. M.; Liu, H.-Y.; Prock, A.;
Giering, W. P. Organometallics 1987, 6, 650-658. (e) Golovin, M.
N.; Rahman, M. M.; Belmonte, J. E.; Giering, W. P. Organometallics
1985, 4, 1981-1991. (f) Eriks, K.; Giering, W. P.; Liu, H.-Y.; Prock,
A. Inorg. Chem. 1989, 28, 1759-1763. (g) Tracey, A. A.; Eriks, K.;
Prock, A.; Giering, W. P. Organometallics 1990, 9, 1399-1405. (h)
Chen, L.; Poe¨, A. J. Inorg. Chem. 1989, 28, 3641-3647. (i) Brodie,
N. M.; Chen, L.; Poe¨, A. J. Int. J. Chem. Kinet. 1988, 20, 467-491.
(j) Poe¨, A. J. Pure Appl. Chem. 1988, 60, 1209-1216. (k) Dahlinger,
K.; Falcone, F.; Poe¨, A. J. Inorg. Chem. 1986, 25, 2654-2658. (l)
Farrar, D. H.; Hao, J.; Poe¨, A. J.; Stromnova, T. A. Organometallics
1997, 16, 2827-2832. (m) Hudson, R. H. E.; Poe¨, A. J. Organome-
tallics 1995, 14, 3238-3248. (n) Farrar, D. H.; Poe¨, A. J.; Zheng, Y.
J. Am. Chem. Soc. 1994, 116, 6252-6261. (o) Chen, L.; Poe¨, A. J.
Coord. Chem. ReV. 1995, 143, 265-295. (p) Neubrand, A.; Poe¨, A.
J.; van Eldik, R. Organometallics 1995, 14, 3249-3258.
2
to the metal (δ 0.73, JPtH ) 66.9 Hz), shows two singlets,
one for each methyl (at δ 3.08 and 1.95, respectively), four
doublets, for H3 (δ 7.90), H8 (δ 7.29), H4 (δ 8.64), and H7
(δ 8.46) protons, and an AB system, for H5 (δ 8.04) and H6
(δ 8.01) protons, as a result of the asymmetry of a firmly
bonded bidentate phenanthroline.
1
Figure 1 shows the changes in the H NMR spectra of 1
upon the addition of increasing amounts of dmso-d6. The
(4) Bruno, G.; Nicolo`, F.; Scopelliti, R.; Arena, G. Acta Crystallogr., Sect.
C: Cryst. Struct. Commun. 1996, C52, 827-829.
2840 Inorganic Chemistry, Vol. 41, No. 11, 2002

Fluxional Motion of dmphen in a Pt(II) Complex
Figure 1. 1H NMR spectra of compound 1 at 298 K in CDCl3 as a function
of dmso-d6 concentration. Ligand concentrations ([dmso-d6]/m) and rate
constants (kobsd/s-1) for the fluxional motion of 2,9-dimethyl-1,10-phenan-
throline in 1 are as follows: (a) [dmso-d6] ) 0, kobsd ) 19; (b) [dmso-d6]
) 0.00295, kobsd ) 158; (c) [dmso-d6] ) 0.00589, kobsd ) 316; (d) [dmso-
d6] ) 0.00884, kobsd ) 484; (e) [dmso-d6] ) 0.0118, kobsd ) 633; (f) [dmso-
d6] ) 0.0177, kobsd ) 967.
initial spectrum, in the lower plot of the figure, was recorded
at 298 K in pure CDCl3 and maintains much of the
characteristics illustrated above for the spectrum taken at 260
K. On adding increasing amounts of dmso-d6, the two
methyls and the aromatic proton pairs H3-H8, H4-H7, and
H5-H6 become chemically equivalent, indicating a fast site
exchange of the two nitrogen atoms of the dmphen ligand.
The rate constant kobsd, measured on the methyl signals,
increased linearly with increasing concentration of dimethyl
sulfoxide. Concentration-dependent measurements, carried
out with the ligands SO(CH2)4, SO(n-Bu)2, SO(sec-Bu)2, SO-
(i-Pr)2, SOEt(Ph), SOPr(Ph), SO(Bz)2, SO(p-MeC6H4)2,
SOPh2, SO(p-ClC6H4)2, SOMe(p-MeOC6H4), SOMe(p-
MeC6H4), SOMe(Ph), SOMe(p-BrC6H4), and SOMe(p-
ClC6H4), exhibited a similar behavior. The rate constants kobsd
(see Supporting Information Table S1), when plotted against
the concentration of the added ligands SOR(R′), give straight
lines with a common intercept (Figure 2), indicating that the
two-term rate law
Figure 2. Dependence of the pseudo-first-order rate constants (kobsd/s-1
)
for the fluxional motion of dmphen in complex 1 upon sulfoxide concentra-
tion, at 298 K in CDCl3. The numbers refer to the sulfoxides as listed in
Table 1. Plot A is for the aliphatic SOR2 sulfoxides, plot B the mixed
sulfoxides, plot C SO(p-XC6H4)2, and plot D SOMe(p-XC6H4).
order rate constant k2 measures the efficiency of uncoordi-
nated sulfoxides in accelerating the fluxionality.
The rate constants for the flipping of complex 1 at different
temperatures (in the range 242-311 K) and at a constant
concentration of added dmso-d6 of 8.66 mm were determined
kobsd ) k1 + k2[SOR(R′)]
(1)
is obeyed. The values of k2 (m-1 s-1), from linear regression
analysis of the dependence of kobsd on [SOR(R′)], are listed
in Table 1 as log k2. The concentration-independent rate
constant k1 measures the rate of dmphen flipping within the
by line-shape analysis of the H NMR spectra (Supporting
1
Information Table S2 and Figure S1). The activation
parameters for the catalyzed fluxional process were derived
from the Eyring plot and are as follows: ∆Hq ) 24.2 ( 0.5
kJ mol-1 and ∆Sq ) -70 ( 2 J K-1 mol-1.
-
ion pair {[Pt(Me)(dmphen)(PPh3)]+PF6 },1 while the second-
Inorganic Chemistry, Vol. 41, No. 11, 2002 2841

Romeo et al.
Table 1. Selected Stereoelectronic Parameters of Sulfoxides 1-16,
Rate Constants for Their Catalytic Effect on the Fluxional Motion of the
Dinitrogen Ligand 2,9-Dimethyl-1,10-phenanthroline in
Discussion
The addition of external sulfoxide ligands to 1 in chloro-
form solution accelerates the fluxionality of dmphen without
promoting any detectable exchange between free and coor-
dinated ligands. The rates (log k2, in Table 1) are 6-7 orders
of magnitude greater than the rates of nucleophilic substitu-
tion of a ligand (SOMe2) from the parent 1,10-phenanthroline
complex 2 (log k2′, in Table 1). The catalytic efficiency of
the added ligands on the rearrangement decreases in the order
SOMe2 > SO(CH2)4 > SOMe(p-MeOC6H4) > SOMe(p-
MeC6H4) > SOMe(Ph) > SOMe(p-BrC6H4) > SOPr(Ph) >
SOMe(p-ClC6H4) > SOEt(Ph) > SO(n-Bu)2 > SO(p-
MeC6H4)2 > SOPh2 > SO(Bz)2 > SO(p-ClC6H4)2 > SO-
(sec-Bu)2 > SO(i-Pr)2 and is strongly affected by the nature
of the substituents bonded to the sulfur atom.
[Pt(Me)(PPh3)(dmphen)]PF6, and Rate Constants of SOMe2 for SOR(R′)
Substitution from [Pt(Me)(Me2SO)(phen)]TFTBa
f
g
h
sulfoxide
SOMe2 (1)
SO(CH2)4 (2)
SO(n-Bu)2 (3)
SO(sec-Bu)2 (4)
SO(i-Pr)2 (5)
SOEt(Ph) (6)
SOPr(Ph) (7)
SO(Bz)2 (8)
∑σ* b
øc
θd θe Ear log k2 log k2′
0.00 5.70 79 111
0
4.731 -2.220
4.623
-0.18
110
-0.23 3.95 91 128
-0.19 3.44 107 141
-0.38 2.30 107 141
0.50 6.52 92 130
0.48 6.22 92 130
0.44
0
0
0
1
1
4.130i
3.389i -2.947
3.339 -2.533
4.167
4.293
3.711
SO(p-MeC6H4)2 (9)
SOPh2 (10)
SO(p-ClC6H4)2 (11)
SOMe(p-MeOC6H4) (12) 0.33 6.35 88 120
SOMe(p-MeC6H4) (13)
SOMe(Ph) (14)
0.86 7.66 97 124
1.20 8.84 97 124
1.60 11.20 97 124
2
2
2
1
1
1
4.040 -2.695
3.758 -2.991
3.414 -3.910
4.597 -2.428
4.522
0.43 6.68 88 120
0.60 7.30 88 120
4.392
QALE Analysis. Giering and Prock3a-g have suggested
the equations and the protocol to perform a quantitative
analysis (QALE) of the steric and electronic effects of
ligands. A large body of kinetic, thermodynamic, NMR, and
IR data and other physicochemical properties that depend
on the stereoelectronic characteristics of the ligands have
been analyzed successfully with QALE. Very recently these
authors have collected all the available material (i.e., sets of
data, protocol for the analysis, program package, parameters
of the ligands, leading references, etc.) in a flexible tool such
as a Web site, subjecting each set of data to a QALE analysis
with commentary on how each analysis is done and how
successful it is.5 Essentially, the protocol implies a combined
method, based on the use of data obtained from graphical
analysis to control the results of the regression analysis.
We took advantage of an essential tenet of QALE which
allows the stereoelectronic parameters to be transferred from
AZ3 (A ) P, phosphines) to AZ2 (A ) SO, sulfoxides), as
done before by Giering et al. for thioethers.6 However, we
were unable with this method to confer suitable steric or
electronic parameters to SO(Bz)2, SO(CH2)4, and SOMe(p-
BrC6H4). Therefore, the parameter ø in Table 1 describes
the σ electron donor capacity of the ligands (a small ø value
is associated with a good σ donation),7 θ is Tolman’s cone
angle,8 which measures the size of the ligand, and Ear is a
secondary electronic effect of unknown origin which is
generally associated with the number of pendent aryl groups,9
even if it is not limited to it. The response of the kinetic
data to ø and Ear is assumed to be linear over the entire set
of ligands, while the response to the steric parameter θ is
not linear for the possibility of the onset of a steric threshold.
A steric threshold marks the point below or above which no
steric effects are evident.10
SOMe(p-BrC6H4) (15)
SOMe(p-ClC6H4) (16)
0.83
120
4.336
4.252
0.83 8.45 88 120
1
a At 298 K in CDCl3. b Sum of the Taft σ* constants for the radicals R
and R′ bonded to sulfur in SOR(R′). c Fractional values calculated from
the phosphorus(III) values (see ref 6); ø values (cm-1) were taken from ref
5. d Fractional values calculated from phosphorus(III) values. θ values (deg)
were taken from ref 5. e Values of circular cone apertures taken from ref
13. f Number of aryl rings bonded to the sulfur atom. g Logarithm form of
second-order rate constants (m-1 s-1) for the fluxional motion of dmphen
in 1 under the accelerating effect of SOR(R′). h Logarithm form of second-
order rate constants (m-1 s-1) for the displacement of SOMe2 from 2 by
SOR(R′). i From ref 1.
Substitution Reactions. The reactions
[Pt(Me)(phen)(SOMe2)]+ + SOR(R′) f
[Pt(Me)(phen)(SOR(R′))]+ + SOMe2 (2)
(SOR(R′) ) SOMe2, SO(sec-Bu)2, SO(i-Pr)2, SO(p-MeC6H4)2,
SOPh2, SO(p-ClC6H4)2, and SOMe(p-MeOC6H4)) were car-
ried out in CDCl3 as solvent at 298 K. The analysis of the
1
changes of the H NMR spectra during the course of the
reaction showed that the process under study was indeed the
simple substitution of the bound SOMe2 by SOR(R′). The
systematic kinetics of this reaction were studied, under
pseudo-first-order conditions, at different ligand concentra-
1
tions and were followed by H NMR spectroscopy. The
changes in the intensity of the signals of bound and free
SOMe2 with time were used for the calculation of the rate
constants. In some cases the reactions went to completion
(with dmso-d6, SO(i-Pr)2, and SOMe(p-MeOC6H4) as enter-
ing groups), and the pseudo-first-order rate constants kobsd
(see Supporting Information Table S3), plotted against the
concentration of the entering sulfoxide, gave straight lines
with a zero intercept, indicating that in the two-term rate
equation (1) the k1 term makes little if any contribution to
the reactivity (k1 arises from the solvolytic path, and k2′ is
the second-order rate constant for the bimolecular attack of
SOR(R′) on the substrate). The values of k2′ were obtained
from linear regression analysis of the rate law, and log k2′
values are collected in Table 1, together with those of the
other reagents calculated as discussed in the Experimental
Section.
We start the interpretation of the kinetic data by analyzing
the plot of log k2 versus ø in Figure 3. The line determined
(5) http://www.bu.edu/qale/.
(6) Fernandez, A. L.; Prock, A.; Giering, W. P. Organometallics 1996,
15, 2784-2789 and references therein.
(7) Bartik, T.; Himmler, T.; Schulte, H.; Seevogel, K. J. J. Organomet.
Chem. 1984, 272, 29-41.
(8) Tolman, C. A. Chem. ReV. 1977, 77, 313-348.
(9) Wilson, M. R.; Woska, D. C.; Prock, A.; Giering, W. P. Organome-
tallics 1993, 12, 1742-1752.
(10) Liu, H.-Y.; Eriks, K.; Prock, A.; Giering, W. P. Organometallics 1990,
9, 1758-1766.
2842 Inorganic Chemistry, Vol. 41, No. 11, 2002

Fluxional Motion of dmphen in a Pt(II) Complex
obtained using the SCIENTIST program are listed in Table
2. As a measure of the goodness of fit, we used the value of
r2, which measures the fraction of the total variance
accounted for by the model. When dealing with a multiplicity
of possible models, a more appropriate index of the goodness
of fit is the MSC (model selection criterion) value.12 The
most appropriate model will be that with the largest MSC.
A supplementary statistical criterion is given by the value
of r2 ) 0.995 for the linear plot of log k2(calcd) vs log k2(obsd)
,
where log k2(calcd) is obtained by introducing the calculated
coefficients in eq 3. Another appropriate index of the
goodness of fit is given by the correspondence between the
value of R (-0.17 ( 0.03) obtained from full regression
analysis and that (-0.17 ( 0.02) obtained from the plots of
log k2 vs ø for the isosteric ligands SO(p-XC6H4)2 and SOMe-
(p-XC6H4), characterized by the values of the θ and Ear
constants. Attempts to apply the sum of Taft’s σ* constants
as electronic parameters, or the Calligaris circular cone
apertures13 as a measure of the steric requirement of the
ligands, led to a poorer correlation than with ø and θ.
The results of the QALE analysis can now be displayed
as electronic, steric, and Ear profiles (Figure 4). The electronic
profile (Figure 4A) indicates that the rate of fluxional motion
of dmphen in 1 is enhanced as the sulfoxides become better
electron donors (smaller ø), and can be constructed by
subtracting the contributions of all the terms of the regression
equation, except that of the variable of interest, i.e., R(ø),
from the experimental data, according to the equation
Figure 3. Graphical analysis of log k2(obsd) for the fluxional motion
promoted by sulfoxides in complex 1 (see the text). The datum for SO-
(sec-Bu)2 is considered an outlier. The intersections of the lines for SOMe-
(p-XC6H4) and SO(p-XC6H4)2 with that for SOR2 occur at ø ) 5.7 and ø
) 4.9, rather than at ø ) 4.5 and ø ) 3.4, respectively.
by the data for SO(p-XC6H4)2 has a slope of -0.17 ( 0.02.
Since the SO(p-XC6H4)2 ligands are isosteric and possess
the same number of aryl groups, the slope of this plot affords
an estimate of the coefficient of ø for the full set of data in
the QALE equation. In practice, from the pattern of behavior
of a subset of ligands having θ and Ear constants, we get
fast information on the sensitivity of the system to the
electron release of substituents on sulfur. The line determined
by the data for SOMe(p-XC6H4) appears to be parallel to
the line for SO(p-XC6H4)2. The data for three of the SOR2
ligands fit a reasonably good straight line, while the datum
for SO(sec-Bu)2 appears to be an outlier.11 Since the latter
line, in contrast to those for the SOMe(p-XC6H4) and SO-
(p-XC6H4)2 ligands, has a positive slope, we can conclude
that there is a dominant and inhibitory steric effect associated
with the SOR2 ligands. In the absence of an aryl effect, the
two parallel lines would have intersected the SOR2 line at
points 4.5 and 3.4. The magnitude of the deviation observed
is a measure of the Ear effect. The latter enhances the
reactivity of SO(p-XC6H4)2 as compared to those of SOMe-
(p-XC6H4) and the SOR2 ligands. Plots of log k2 versus the
number of aryl groups (i) in the series SOPhiMe2-i are
reasonably linear (see Supporting Information Figure S2),
suggesting the absence of a steric threshold. Application of
the steric threshold program supports this conclusion.
Since there is no evidence for a steric threshold, and only
steric (θ) and electronic (σ + Ear) effects were found to be
operative, the appropriate equation to analyze the data is
log k2(ø) ) log k2(obsd) - (â(θ) + γ(Ear) + ω)
(4)
Likewise, the steric profile (Figure 4B) and the Ear profile
(Figure 4C) can be constructed using similar equations.
Figure 4B indicates that the rate of dmphen flipping
diminishes as the size of the sulfoxide increases, and Figure
4C indicates that the presence of an aryl group enhances the
reactivity of the sulfoxide and, thereby, the rate of reaction.
Parts A-C of Figure 4 are presented with the same absolute
scale to give a visual comparison of the relative importance
of the three parameters. However, another useful way to
display the extent to which a parameter affects a property is
to calculate the percentage contribution of that parameter,
using the regression coefficients derived from eq 3 and
defining the maximum range of variation of the parameters
(∆ø ) 26 cm-1; ∆θ ) 60°; ∆Ear ) 2),5 according to the
equation
log k2 ) ω + R(ø) + â(θ) + γ(Ear)
(3)
%ø ) 100(|R|∆ø)/[(|R|∆ø) + (|â|∆θ) + (|γ|∆Ear)] (5)
where R, â, and γ are regression coefficients that measure
the relative importance of electronic, steric, and Ear factors
in the process. All the kinetic data in Table 1 for which we
have parameters, except that for SO(sec-Bu)2 considered an
outlier, were analyzed with the above equation. The results
Therefore, log k2 exhibits a 45% electronic contribution, a
45% steric contribution, and only a 10% Ear contribution.
The kinetic data (log k2′) for the reaction of sulfoxides
with 2 were analyzed in the same manner. The results are
(12) When comparing two models with different numbers of parameters,
this criterion not only places a burden on the model with more
parameters to have a better coefficient of determination, but also
quantifies how much better it must be for the model to be deemed
more appropriate.
(13) Calligaris, M.; Faleschini, P.; Todone, F.; Alessio, E.; Geremia, S. J.
Chem. Soc., Dalton Trans. 1995, 1653-1661.
(11) The steric and electronic properties of SO(sec-Bu)2 should resemble
those of SO(i-Pr)2 rather than those of SO(n-Bu)2. As a matter of fact
Bartik’s value of ø for P(sec-Bu)3 (on which the ø value of SO(sec-
Bu)2 is based) is closer to that of P(n-Bu3) than it is to that of P(i-
Pr)3. The datum for SO(sec-Bu)2 fits the graphical analysis when it is
assigned the same ø value as that for SO(i-Pr)2.
Inorganic Chemistry, Vol. 41, No. 11, 2002 2843

Romeo et al.
Table 2. Selected Coefficients Derived from QALE Analysis on the Second-Order Rate Constants (k2) of Some Reactions on Platinum(II) Complexes
type of
reaction
complex
reagent
na
Rb
âc
γ
ω
r2
MSCd
2.37
[Pt(Me)(dmphen)(PPh3)]+ e
% contribution of params
SOR1R2
fluxionality
12
-0.17 ( 0.03
-0.076 ( 0.006
0.49 ( 0.1
11.9 ( 0.7
0.952
45
45
10
[Pt(Me)(phen)(SOMe2)]+ e
% contribution of params
[PtPh2(CO)(5-Aq)]f
% contribution of params
[Pt(NNC)Cl]g
SOR1R2
PR1R2R3
PR1R2R3
substitution
substitution
substitution
7
13
14
-0.34 ( 0.03
-0.052 ( 0.004
0.61 ( 0.08
3.9 ( 0.4
14.6 ( 0.8
13.8 ( 0.8
0.99
3.55
3.62
2.70
67
24
9
-0.07 ( 0.03
22
-0.099 ( 0.005
66
0.3 ( 0.1
12
0.983
0.962
-0.134 ( 0.02
38
-0.075 ( 0.005
48
0.4 ( 0.1
14
% contribution of params
a Number of data points. b Centimeters. c Inverse degrees. d See the text and ref 12. e This work. f From ref 14, leaving group ) 5-aminoquinoline (5-Aq).
g From ref 15, leaving group ) chloride (N-N-CH ) 6-(1-methylbenzyl)-2,2′-bipyridine).
Figure 4. Electronic profile (A), steric profile (B), and Ear plot (C) for
sulfoxides acting as catalysts on the fluxional motion of dmphen in complex
1. Filled squares refer to dialkyl sulfoxides, open circles to SOEt(Ph) and
SOPr(Ph), open squares to SO(p-XC6H4)2, and open triangles to SOMe(p-
XC6H4).
Figure 5. Electronic profile (A), steric profile (B) and Ear plot (C) for
sulfoxides acting as entering groups in the nucleophilic substitution reaction
of SOMe2 from complex 2. Filled squares refer to dialkyl sulfoxides, open
squares to SO(p-XC6H4)2, and open triangles to SOMe(p-XC6H4).
and electronic contributions are of about equal importance
for the fluxional process in 1, while the electronic effect is
prevalent (67%) with respect to the steric effect (24%) in
the substitution reactions in complex 2. The sensitivity to
electronic effects is a measure of the extent of metal-ligand
bond making in the transition state. Thus, there is a greater
platinum-sulfur interaction in the transition state of 2 than
in the transition state of 1. The QALE analysis shows that
the five-coordinate transition state of 2 is more associative
in nature than the transition state of 1. These conclusions
are in line with the structural features of the two complexes
under investigation. Other examples of associative substitu-
tion reactions on platinum complexes for which a complete
listed in Table 2 and are illustrated in Figure 5. We can see
that the signs of the coefficients of ø and θ of both processes
(fluxionality on 1 and substitution on 2) are characteristic
of entering-ligand-dependent substitution reactions. Thus, the
negative sign of the R coefficient indicates that the transition
state is electron demanding relative to the ground state. The
negative sign of the â coefficient indicates that the increasing
steric requirement of the ligand impedes the reaction. The
relative magnitudes of the coefficients give insight into the
nature of the transition states of these reactions. By compar-
ing the data in Table 2, we can see that the contribution of
Ear is small if not negligible for both processes. The steric
2844 Inorganic Chemistry, Vol. 41, No. 11, 2002

Fluxional Motion of dmphen in a Pt(II) Complex
Scheme 2. Sequence of Berry Pseudorotations (Left Side) and
Turnstile Intramolecular Rearrangements (Right Side) of a
Five-Coordinate Intermediate, Leading to the Exchange of Coordination
Sites of the Pt-N Bonds in 1
QALE analysis has been performed refer to the use of
phosphines as reagents in the substitution of 5-aminoquino-
line (5-Aq) from [PtPh2(CO)(5-Aq)]14 and of chloride from
the cyclometalated complex [Pt(N-N-C)Cl] (N-N-CH )
6-(1-methylbenzyl)-2,2′-bipyridine).15 The results are sum-
marized in Table 2. A survey of the literature data suggests
that a number of other entering-ligand-dependent substitution
reactions have been analyzed with eq 3, including the heat
of reaction of trans-[Pt(Me)L2(THF)]+ with phosphines,16
the displacement of acetone from [η5-(C5R5)(CO)Fe(Ac)-
(COMe)]+ with thioethers,6 etc.
The Mechanism. The fluxional motion of dmphen in [Pt-
(Me)(dmphen)(PPh3)]+, catalyzed by sulfoxides, exhibits a
pattern strictly similar to that of bimolecular nucleophilic
substitution reactions.17 Akin to the displacement of a ligand
in a square-planar complex, the rate-determining step is the
attack of a nucleophile to the metal, which takes place with
a low enthalpy of activation and a largely negative entropy
of activation (∆Hq ) 24.2 ( 0.5 kJ mol-1 and ∆Sq ) -70
( 2 J K-1 mol-1, in the presence of dmso-d6 8.66 mm). The
free energy for the catalyzed process (∆Gq ) 45 kJ mol-1)
is at least 20 kJ mol-1 lower than that for the dissociative
uncatalyzed flipping of dmphen in 1 (66.5 kJ mol-1)1 (see
Supporting Information Figure S1). The kinetics clearly
indicate that a key step in the fluxional mechanism of 1 is
the formation of a five-coordinate intermediate upon sul-
foxide coordination. Therefore, in addressing the question
of how the fluxional motion of the dinitrogen ligand occurs,
besides the obvious traditional consecutive displacement
mechanism, we must take into account the possibility of
intramolecular rearrangements from a five-coordinate inter-
mediate, such as some form of Berry pseudorotation or a
turnstile mechanism (Scheme 2). No distinction can be made
among the three possible reaction pathways on the basis of
the rate law or the structural effects observed. The difficulty
of differentiating between these mechanisms, using kinetic
data alone, has been known for many years. Anderson and
Cross,18 in an excellent review paper on the isomerization
mechanism of square-planar complexes, have pointed out
that different authors have used the same evidence to support
radically different opinions. We avoid adding a new case in
contest, being aware that the one-step transfer of the sulfoxide
to the substrate leads the five-coordinate intermediate to
occupy a point of high energy along the energy profile, and
any discussion on the nature of the subsequent minima along
this profile risks becoming semantic in nature. However, it
is useful to comment briefly on the various alternatives.
Pseudorotation of five-coordinated species is difficult to
reconcile with the highly stereospecific nature of substitutions
on Pt(II). Its operation here could be favored by the fact
that all the potential rearrangements of the trigonal bipyra-
mids (left side of Scheme 2) could involve the participation
of the N-N ligand in an axial-equatorial coordination. This
could appear to be an easy situation for a rigid ligand with
a small bite angle such as dmphen. The same reasoning
applies for the sequence of turnstile rearrangements necessary
to originate a site exchange for the two nitrogen atoms of
the dmphen ligand (right side of Scheme 2). Interestingly,
the best documented example of operation of such a
mechanism2b requires a square-pyramidal coordination, since
the ligands used make the trigonal-bipyramidal coordination
very strained.
The most likely consecutive displacement mechanism that
can be envisaged consists of a sequence of nucleophilic
substitution reactions, which take place with complete
retention of configuration (Scheme 3). The process features
a catalyzed geometrical isomerization, many examples of
which can be found in the literature.18-23 The mechanism
involves the opening of one arm of the bidentate N-N ligand
upon nucleophilic attack by the ligand L, and the intercon-
version of two η1-coordinated complexes via formation of a
common five-coordinate chelated intermediate. The latter
requires the participation of the phenanthroline as an equato-
rial ligand, in the trigonal Ph3P-Pt{η2-(N1-N2)} moiety,
(14) Romeo, R.; Arena, G.; Monsu` Scolaro, L. Inorg. Chem. 1992, 31,
4879-4884.
(15) Romeo, R.; Plutino, M. R.; Monsu` Scolaro, L.; Stoccoro, S. Inorg.
Chim. Acta 1997, 265, 225-233.
(16) Manzer, L. E.; Tolman, C. A. J. Am. Chem. Soc. 1975, 97, 1955-
1956.
(19) Romeo, R. Comments Inorg. Chem. 1990, 11, 21-57.
(20) Cattalini, L.; Martelli, M. J. Am. Chem. Soc. 1969, 91, 312-316.
(21) Roulet, R.; Barbey, C. HelV. Chim. Acta 1973, 56, 2179-2186.
(22) Annibale, G.; Bonivento, M.; Canovese, L.; Cattalini, L.; Michelon,
G.; Tobe, M. L. Inorg. Chem. 1985, 24, 797-800.
(23) Favez, R.; Roulet, R.; Pinkerton, A. A.; Schwatzenbach, D. Inorg.
Chem. 1980, 19, 1356-1365.
(17) (a) Tobe, M. L.; Burgess, J. Inorganic Reaction Mechanisms; Addison-
Wesley Longman: Essex, England, 1999. (b) Wilkins, R. G. Kinetics
and Mechanisms of Reactions of Transition Metal Complexes; VCH:
Weinheim, Germany, 1991. (c) Basolo, F. Coord. Chem. ReV. 1996,
154, 151-161.
(18) Anderson, G. K.; Cross, R. J. Chem. Soc. ReV. 1980, 9, 185-215.
Inorganic Chemistry, Vol. 41, No. 11, 2002 2845

Romeo et al.
Scheme 3. Sequence of Stereospecific Ring-Opening Reaction Steps,
Leading to the Exchange of Coordination Sites of the Pt-N Bonds in 1,
through a Chelated Five-Coordinated Intermediate
tably takes place with traces of the most nucleophilic ligands,
such as phosphines or halide ions. Compounds similar to 1,
where PPh3 has been substituted by less efficient trans-
activating groups (amine, pyridine, and thiourea) behave as
stereochemically rigid N,N-chelated complexes.1 Likewise,
a static structure characterizes complexes containing 1,10-
phenanthroline such as 2, and the removal of the chelated
ligand hardly occurs, even by adding an excess of strong
nucleophiles. A further impetus to ring opening comes from
the release of steric congestion in forming a complex in
which dmphen is η1-coordinated. In conclusion, although the
kinetics do not prove or disprove the operation of alternative
mechanisms, such as those depicted in Scheme 2, where
N-N remains η2-coordinated, a consecutive stepwise dis-
placement mechanism is to be preferred. This can be inferred
by the strict correlation between the facility of Pt-N bond
breaking and the fluxional motion of the dinitrogen ligand
and the well-known preference of platinum(II) complexes
to undergo stereospecific substitutions.
Within this mechanistic framework, it would be of interest
to extend the study to the fluxionality of 1 promoted by other
families of ligands such as thioethers or amines. However,
a fascinating aspect of this system appears to be the
uncatalyzed intrinsic fluxionality of [Pt(Me)(dmphen)(L)]+
cationic complexes, which involves a dissociative mechanism
and is connected to the capacity of the ligand L in promoting
Pt-N bond dissociation. A detailed kinetic study is underway
on complexes in which L represents an extended series of
phosphines, with the aim of ascertaining the relative impor-
tance of steric and electronic factors of “spectator” ligands
in governing the fluxionality of the substrates.
with the two nitrogen donors being equivalent with respect
to the coordinated sulfoxide. This coordinating geometry
apparently looks uneasy for a rigid ligand with a small bite
angle. In fact, a number of X-ray structural studies of five-
coordinate trigonal-bipyramidal platinum(II) complexes have
shown that dmphen lies in the trigonal plane, forming a flat
unconstrained five-membered ring,24 even if a rare case for
axial-equatorial N-N coordination has been reported.25
Panunzi and Natile26 have discussed at length the reasons
why a chelate ring, either flat or puckered, produced by a
bidentate dinitrogen ligand, occupies two equatorial sites in
a five-coordinate trigonal-bipyramidal platinum(II) complex
and have shown evidence of how an increase of the bite angle
destabilizes the trigonal bipyramid, which shifts to a square-
planar geometry by chelate ring opening.
If one does not object to the equatorial chelate, which we
do not, the displacement mechanism in Scheme 3 becomes
the most reasonable interpretation for the following rea-
sons: (i) it involves only stereospecific steps, and (ii) it
accounts for the lability of the nitrogen-platinum bond. The
weakening of the Pt-N1 and Pt-N2 bonds, under the trans
influence of the strong σ donors methyl and phosphine
groups, appears to be a factor of overwhelming importance
in dictating the dynamic motion of dmphen. Easy Pt-N bond
dissociation facilitates the ring opening either in the original
square-planar complex (spontaneous fluxional motion of the
phenanthroline in 1) or in a bipyramidal trigonal coordinative
environment (catalyzed flipping of dmphen). Accordingly,
the addition of external ligands to 1 must be carried out with
care to avoid dmphen displacement, which, however, inevi-
Experimental Section
Synthesis of the Complexes. The complex [Pt(Me)(dmphen)-
(PPh3)]PF6 (1) was prepared following the published procedure.1
[Pt(Me)(phen)(SOMe2)]TFPB (2) (TFPB ) B[3,5-(CF3)2C6H3]4)
was obtained by ion exchange, by adding a stoichiometric amount
of NaTFPB27 to a dichloromethane solution of [Pt(Me)(phen)-
(SOMe2)]PF6.28 The salt NaPF6 that separated out as a solid from
the solution was filtered off, and the complex was obtained by
evaporation of the residual solvent. 1H NMR (CDCl3, T ) 298 K):
3
3
4
δ 10.09 (dd, JPtH ) 18.7 Hz, JHH ) 5.5 Hz, JHH ) 1.1 Hz, 1H,
H9), 9.04 (dd, 3JPtH ) 46.2 Hz, 3JHH ) 5.5 Hz, 4JHH ) 1.1 Hz, 1H,
H2), 8.62 (d, 3JHH ) 8.2 Hz, 1H, H7), 8.58 (d, 3JHH ) 8.2 Hz, 1H,
H4), 8.01 (dd, 3JHH ) 8.2, 5.5 Hz, 1H, H8), 7.99 (AB system, 3JHH
) 8.8 Hz, 1H, H5), 7.93 (AB system, 3JHH ) 8.8 Hz, 1H, H6), 7.89
(dd, 3JHH ) 8.2, 5.5 Hz, 1H, H3), 7.71 (s, br, 8H, Ho,o′), 7.48 (s, br,
3
2
(24) (a) Fanizzi, F. P.; Intini, F. P.; Maresca, L.; Natile, G.; Lanfranchi,
M.; Tiripicchio, A. J. Chem. Soc., Dalton Trans. 1991, 1007-1015.
(b) Albano, V. G.; Ferrara, M. L.; Monari, M.; Panunzi, A.; Ruffo, F.
Inorg. Chim. Acta 1999, 285, 70-75. (c) De Felice, V.; Ferrara, M.
L.; Giordano, F.; Ruffo, F. Gazz. Chim. Ital. 1994, 124, 117-119.
(d) Giordano, F.; Ruffo, F.; Saporito, A.; Panunzi, A. Inorg. Chim.
Acta 1997, 264, 231-237. (e) Fanizzi, F. P.; Natile, G.; Lanfranchi,
M.; Tiripicchio, A.; Pacchioni, G. Inorg. Chim. Acta 1998, 275-276,
500-509. (f) Albano, V. G.; Castellari, C.; Monari, M.; De Felice,
V.; Panunzi, A.; Ruffo, F. Organometallics 1996, 15, 4012-4019.
(g) Albano, V. G.; Castellari, C.; Monari, M.; De Felice, V.; Panunzi,
A.; Ruffo, F. Organometallics 1992, 11, 3665-3669.
(25) Albano, V. G.; Monari, M.; Orabona, I.; Ruffo, F.; Vitagliano, A.
Inorg. Chim. Acta 1997, 265, 35-46.
(26) (a) Albano, V. G.; Natile, G.; Panunzi, A. Coord. Chem. ReV. 1994,
133, 67-114. (b) Maresca L.; Natile, G. Comments Inorg. Chem. 1993,
14, 349-366.
4H, Hp), 3.50 (s, JPtH ) 36.8 Hz, 6H, S-CH3), 0.90 (s, JPtH
)
72.0 Hz, 3H, Pt-CH3). Anal. Calcd for PtBC47H29F24N2OS: C, 42.4;
H, 2.2; N, 2.1. Found: C, 42.4; H, 2.4; N, 2.1.
General Procedures and Chemicals. Sulfoxides and sulfides
were purchased from Lancaster or Aldrich Chemical Co. The
crystalline sulfoxide SOPh2 was dried by being allowed to stand
for many hours under vacuum over P4O10. Pure SO(sec-Bu)2, SOEt-
(Ph), SOMe(p-ClC6H4), and SOMe(p-MeOC6H4) were obtained by
oxidation of the corresponding sulfides with 3-chloroperoxybenzoic
(27) Brookhart, M.; Grant, B.; Volpe, A. F. Organometallics 1992, 11,
3920-3922.
(28) Romeo, R.; Monsu` Scolaro, L.; Nastasi, N.; Arena, G. Inorg. Chem.
1996, 35, 5087-5096.
2846 Inorganic Chemistry, Vol. 41, No. 11, 2002

Fluxional Motion of dmphen in a Pt(II) Complex
acid (Aldrich).29 The compounds were obtained as pure products
after silica gel chromatography with petroleum ether and ethyl
acetate (9:1, v/v), and their purity was checked by IR and 1H NMR
measurements. All the other sulfoxides were used as received from
Lancaster or Aldrich. Solvents (from Aldrich) used in the synthetic
procedures were distilled under nitrogen from the appropriate drying
agents (diethyl ether from sodium benzophenone ketyl, dichlo-
romethane from barium oxide, and dimethyl sulfoxide, at low
pressure, from CaH2, after preliminary filtration through an alumina
column) and then stored in N2-filled flasks over activated 4 Å
molecular sieves. Chloroform-d (+99.96% D, Cambridge Isotope
Laboratories) was dried by being allowed to stand for many days
over CaH2, distilled under nitrogen over activated magnesium
sulfate and sodium carbonate, and then stored over activated 4 Å
molecular sieves. K2PtCl4 (Strem) was purified by dissolving it in
water and filtering. Commercial reagent grade chemicals dmphen
and phen were used without further purification. Triphenylphosphine
(Strem) was crystallized from ethanol by dissolving it in the hot
solvent, filtering, and cooling the filtrate to 0 °C.
complex in CDCl3. At least a 5-fold excess of dmso-d6 over the
complex was ensured in any run. The isotopic exchange was
followed by monitoring the increase in intensity of the proton NMR
signal of free nondeuterated SOMe2 at δ 2.60 ppm and the matching
decrease of the signal of the sulfoxide coordinated to the metal.
The spectra were recorded at appropriate intervals of time. The
mole fraction F ) [SOMe2]f/([SOMe2]f + [SOMe2]b) of the free
dimethyl sulfoxide was obtained by integration of the signals, and
the first-order rate constants kexch (s-1) for the exchange of the label
were obtained from a nonlinear least-squares fit of the experimental
data to F ) c1 + c2 exp(-kexcht), with c1, c2, and kexch as the
parameters to be optimized. A similar analysis can be performed
by using the mole fraction of the bound dimethyl sulfoxide. From
the McKay equation, Rexch ) kexchab(a + b)-1 (where Rexch is the
rate of the exchange process, a the concentration of the complex,
and b the concentration of free sulfoxide), the pseudo-first-order
rate constants, kobsd ) Rexch/a, were calculated.
Sulfoxide Substitutions. The kinetic study of ligand substitution
on 2 was carried out in CDCl3 at 298 K by following the same
procedure described for the isotopic exchange. After addition of
the appropriate amount of the proper sulfoxide, the 1H NMR spectra
were recorded at various time intervals. The signals of released
and bound SOMe2 were monitored, and the first-order rate constants
Instrumentation. Infrared spectra were recorded as Nujol mulls
or neat using CsI disks on a Perkin-Elmer FT-IR model 1730
1
spectrophotometer. H and 31P{1H} NMR spectra were obtained
in CDCl3 solution on a Bruker AMX R-300 spectrometer equipped
with a broad-band probe operating at 300.13 and 121.49 MHz,
respectively. 1H chemical shifts were measured relative to the
residual solvent peak and are reported in δ units downfield from
Me4Si. 31P chemical shifts, in parts per million, are given relative
to that of external 85% phosphoric acid. The temperature within
the probe was checked using the methanol or ethylene glycol
method.30
k
obsd (s-1) were calculated from a nonlinear least-squares fit of the
experimental data to F ) c1 + c2 exp(-kobsdt), where F is the mole
fraction of one of the reagents, calculated as described above. In
some reactions (SO(sec-Bu)2, SO(p-MeC6H4)2, SOPh2, and SO(p-
ClC6H4)2), it was necessary to account for the possible reentry of
free SOMe2 into the coordination sphere of the metal, and therefore
the rate data were calculated by using the SCIENTIST software
package32 and the differential equation -d[A]/dt ) kf[A][Y] - kr-
[B][SOMe2], where [A] and [B] are the concentrations of complex
2 and of the final product, respectively, while [Y] represents the
concentration of the added sulfoxide. kf and kr (m-1 s-1) are the
second-order rate constants for substitution of SOMe2 from 2 and
for the reverse reaction, respectively.
Kinetics. The rates of the fluxional motion of dmphen in complex
1 and of sulfoxide exchange and substitution in complex 2 were
1
obtained by various dynamic H NMR techniques.
Line-Shape Analysis. In a typical set of ligand-concentration-
dependent experiments, carried out at constant temperature (298
K), approximately 10 mg of complex 1 was dissolved in 0.5 mL
of dried CDCl3, and the initial NMR spectrum was recorded. The
subsequent NMR spectra were taken after the addition of the
appropriate amount of sulfoxide. All concentrations of the ligand
are expressed in moles per kilogram of solvent (m ) mol kg-1).
The methyl protons of the substituted phenanthroline were used
for the line-shape analysis. The spectral simulation was performed
using the computer program gNMR 4.0.31 A satisfactory fit between
the simulated and observed spectra was judged both by visual
comparison and by a least-squares fit. The variable-temperature
experiments were carried out by recording the spectrum of a
solution, prepared as described above, over a proper temperature
range. The activation parameters ∆Gq, ∆Hq, and ∆Sq were derived
from linear regression analysis of the Eyring plot.
Acknowledgment. We thank the Ministero dell’Istruzione
dell’Universita` e della Ricerca (MIUR), Programmi di
Ricerca Scientifica di Rilevante Interesse Nazionale, Cofi-
nanziamento 2000-2002, the Universita` degli Studi di
Messina, and CNR for funding this work. We also gratefully
thank Prof. W. P. Giering for helpful discussions.
Supporting Information Available: Tables S1 and S3 giving
primary kinetic data (pseudo-first-order rate constants (kobsd/s-1
)
for the fluxional motion in 1 and for nucleophilic substitution on
compound 2, as a function of [SOR(R′)]), Table S2 and Figure S1
for the temperature dependence of the uncatalyzed and catalyzed
flipping in 1, and Figure S2 illustrating the dependence of the
fluxional rate of compound 1 upon the number of pendent aryl
groups in SOR(R′). This material is available free of charge via
the Internet at http://pubs.acs.org.
Isotopic Exchange. The kinetic study of SOMe2 exchange on 2
was performed by adding with a microsyringe a known volume of
dimethyl sulfoxide-d6 on a prethermostated solution of weighted
(29) Aversa, M. C.; Barattucci, A.; Bonaccorsi, P.; Giannetto, P.; Jones,
D. N. J. Org. Chem. 1997, 62, 4376-4384 and references therein.
(30) (a) Van Geet, A. L. Anal. Chem. 1968, 40, 2227-2229. (b) Van Geet,
A. L. Anal. Chem. 1970, 42, 679-680.
IC011206J
(31) gNMR 4.0, Cherwell Science, Oxford.
(32) SCIENTIST, Micro Math Scientific Software, Salt Lake City, UT.
Inorganic Chemistry, Vol. 41, No. 11, 2002 2847