Anation kinetics of pentacyanoaquachromate(III) by thiocyanate and azide

Article abstract of DOI:10.1016/j.ica.2006.06.002

The kinetics of the reaction of Cr(CN)₅(H₂O)^2- with NCS^- and N₃^- were studied at pH 5.0 and at pH 6.3-7.0, respectively, as a function of the temperature between 25.0 and 55.0 °C, and at various ionic strengths. Anation occurs in competition with aquation of CN^-, with rate constants that exhibit less-than-first-order dependence on the concentration of the entering anions. The results are interpreted in terms of ligand interchange in a context of association of the two reacting anions mediated by the Na⁺ or Ca²⁺ counterions. The degree of aggregation depends mainly on the total cationic charge rather than on the ionic strength, and is ca. 2-fold larger for N₃^- than for NCS^-. Within the associated species, N₃^- is a better entering ligand than NCS^- by a factor of 4.5. The Cr(CN)₅(NCS)^3- and Cr(CN)₅(N₃)^3- complexes were also synthesized, and the rates of aquation of NCS^- and N₃^-were measured at pH 5.0 and between 55.0 and 80.0 °C, over the same range of ionic strengths. The ionic strength enhances the anation rates but has little effect on the aquation rates. The average activation enthalpies of the interchange step are 80 ± 3 and 76 ± 3 kJ mol^-1 for entry of NCS^- and N₃^-, respectively. Those of the corresponding aquation reactions are 94 ± 4 and 107 ± 4 kJ mol^-1. Within error limits, all ΔH^? values are independent of the ionic strength. The results are consistent with an I_d mechanism for substitution in Cr(CN)₅X^z- complexes.

Full text of DOI:10.1016/j.ica.2006.06.002

Inorganica Chimica Acta 360 (2007) 897–907
www.elsevier.com/locate/ica
Anation kinetics of pentacyanoaquachromate(III) by
thiocyanate and azide
*
Pietro Riccieri, Edoardo Zinato
`
Dipartimento di Chimica, Universita di Perugia, Via Elce di Sotto 8, 06123 Perugia, Italy
Received 26 May 2006; accepted 4 June 2006
Available online 10 June 2006
Dedicated to Vincenzo Balzani.
Abstract
The kinetics of the reaction of Cr(CN)5(H2O)2ꢀ with NCSꢀ and N3 were studied at pH 5.0 and at pH 6.3–7.0, respectively, as a
function of the temperature between 25.0 and 55.0 ꢀC, and at various ionic strengths. Anation occurs in competition with aquation
of CNꢀ, with rate constants that exhibit less-than-first-order dependence on the concentration of the entering anions. The results are
interpreted in terms of ligand interchange in a context of association of the two reacting anions mediated by the Na+ or Ca2+ counte-
rioꢀns. The degree of aggregation depends mainly on the total cationic charge rather than on the ionic strength, and is ca. 2-fold larger for
ꢀ
N3 than for NCSꢀ. Within the associated species, N3 is a better entering ligand than NCSꢀ by a factor of 4.5. The Cr(CN)5(NCS)3ꢀ
ꢀ
and Cr(CN)5(N3)3ꢀ complexes were also synthesized, and the rates of aquation of NCSꢀ and N3ꢀwere measured at pH 5.0 and between
55.0 and 80.0 ꢀC, over the same range of ionic strengths. The ionic strength enhances the anation rates but has little effect on the aquation
rates. The average activation enthalpies of the interchange step are 80 3 and 76 3 kJ molꢀ1 for entry of NCSꢀ and N3ꢀ, respectively.
Those of the corresponding aquation reactions are 94 4 and 107 4 kJ molꢀ1. Within error limits, all DHꢀ values are independent of
the ionic strength. The results are consistent with an Id mechanism for substitution in Cr(CN)5Xzꢀ complexes.
ꢁ 2006 Elsevier B.V. All rights reserved.
Keywords: Chromium(III) cyano complexes; Kinetics and mechanism; Anation; Thiocyanate; Azide
1. Introduction
amounts and, sometimes, by their instability toward CNꢀ
loss. To date, relatively few such complexes have been
more or less characterized, and even fewer have been iso-
lated as solid salts [9].
The large majority of the investigations of chro-
mium(III) substitution reactions, both thermal [1–4] and
photochemical [3,5–8], have been performed on cationic
am(m)ine complexes. The model systems have, naturally,
We have been interested in pentacyanochromates(III)
with special regard to their photosolvolysis reactions [10–
12], and have reviewed the available information on the
ground- and excited-state chemistry of this class of com-
pounds [9]. As an extension of this work, we report here
on the kinetics of the interaction of Cr(CN)5(H2O)2ꢀ with
been the pentaammines of general formula Cr(NH3)5Xz+
.
By contrast, much less attention has been devoted to
mixed-ligand cyano complexes, and in particular to
Cr(CN)5Xzꢀ species that may be regarded as the anionic
counterpart of the pentaammines. Although these anions
look fairly simple, their quantitative study has presumably
been hampered by difficulties in the preparation of suitable
NCSꢀ and N3 in aqueous solution, as well as on the syn-
ꢀ
thesis, characterization and aquation of Cr(CN)5(NCS)3ꢀ
and Cr(CN)5(N3)3ꢀ. To our knowledge, this is the first ana-
tion study of a pentacyanochromate(III). Previous anation
studies of other anionic Cr(III) complexes have ꢀdealt with
Cr(NCS)5(Me2SO)2ꢀ [13] and CrðC2O4Þ ðH2OÞ [14].
*
Corresponding author. Tel.: +39 075 585 5517; fax: +39 075 585 5606.
E-mail address: zinato@unipg.it (E. Zinato).
2
2
0020-1693/$ - see front matter ꢁ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2006.06.002

898
P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
Besides looking into a little explored area, one purpose of
(H2O)2ꢀ, as well as all free NCSꢀ, as checked spectropho-
tometrically by sampling the effluent. Pure Cr(CN)5-
(NCS)3ꢀ was then recovered in a 10-mL fraction eluted
by 2.5 M NaClO4. Since precipitation of the complex anion
as a salt was not achieved, stock solutions (ca. 10ꢀ3 M)
were kept frozen at ꢀ10 ꢀC. These lasted unaltered for at
least 3 months. Analysis gave a Cr:CNꢀ:NCSꢀ proportion
of 1.00:(4.96 0.08):(1.01 0.03).
this research was to make some comparison with the more
extensively investigated cobalt(III) systems, the mechanistic
paradigms of which have been both the Co(NH3)5Xz+ cat-
ions and the Co(CN)5Xzꢀ anions. It is of interest in this
regard that while anation of Co(CN)5(H2O)2ꢀ was origi-
nally taken as a textbook example of limiting dissociative
substitution [15–18], later observations [19–21] have seri-
ously questioned such an occurrence.
K3[Cr(CN)5(N3)] was prepared in a similar fashiꢀon by
reaction of the pentacyanoaqua complex with the N3 ion.
A 300-mg portion of NaN3 (4.6 mmol) was dissolved in
2 mL of water. After addition of 100 mg of K3[Cr(CN)5-
(OH)] Æ 2H2O (0.3 mmol), the mixture was allowed to react
at room temperature for 12 h. The solution was diluted to
6 mL and adsorbed on a 1 · 5 cm column of Sephadex
QAE A-25 resin conditioned with 0.25 M NaClO4. The col-
umn was wꢀashed with 50 mL of 0.25 M NaClO4, until the
test for N3 in the effluent was negative. Treatment with
2.5 M NaClO4 led to recovery of the product in a 10-mL
volume. This was successively eluted with water through
a 1 · 10 cm column of Sephadex G 15 gel-filtration resin.
Removal of NaClO4 preceded the displacement of the yel-
low–orange band of Cr(CN)5(N3)3ꢀ. The fraction of ca.
20 mL was rotary evaporated at 30 ꢀC under reduced pres-
sure, and the residue was vacuum dried over P2O5. Elemen-
tal analysis gave a Cr:C:N proportion of 1.00:(5.03 0.06):
(7.82 0.08). A 20–25% amount of inert material remained
in the solid. Spectra and HPLC warranted the absence of
extraneous complexes, so that the product was employed
without further crystallization to avoid a yield reduction
of as much as 50%.
2. Experimental
2.1. Materials
Caution! Manipulations involving cyanide require a
well-ventilated hood.
K3[Cr(CN)5(OH)] Æ 2H2O was obtained, mixed with
some amounts of K3[Cr(CN)6], starting from chro-
mium(III) acetate and KCN, and was separated from the
hexacyano complex by gel filtration chromatography
through a Sephadex G-15 resin, as described elsewhere
[11]. The maxima of the absorption spectrum in aqueous
solution at pH > 12 are given in Table 1. Protonation
reversibly turns the hydroxo complex to Cr(CN)5(H2O)2ꢀ
,
causing a blue shift of the long-wavelength bands. Table 1
includes the spectral details of the protonated form at
pH < 5. All data are in agreement with literature values
[9]. As this material is subject to slow decomposition even
in the solid state, chromatographic purification is necessary
after long storage.
Cr(CN)5(NCS)3ꢀ was synthesized by NCSꢀ anation of
Cr(CN)5(H2O)2ꢀ in water and was isolated in solution. All
operations were performed in dim light. A 100-mg portion
of K3[Cr(CN)5(OH)] Æ 2H2O (0.3 mmol) was dissolved in
10 mL of aqueous 3 M NaNCS at pH 5 (0.2 M acetate buf-
fer), and the solution was kept at 40 ꢀC for 60 min. Follow-
ing 8-fold dilution with water to lower the concentration of
NCSꢀ, the mixture was loaded onto a 1 · 5 cm column of
Sephadex QAE A-25 anion exchanger previously treated
with 0.25 M NaClO4. The bulk of the yellow–orange com-
plex mixture remained adsorbed at the top of the column,
while minor quantities of Cr(CN)3(H2O)3 moved through
the resin. Elution with 50 mL of 1.0 M NaClO4 removed
a yellow band mainly consisting of unreacted Cr(CN)5-
2.1.1. Reactants
Reagent grade NaNCS was employed as purchased.
Stock solutions of Ca(NCS)2 were prepared by mixing stoi-
chiometric amounts of aqueous KNCS and Ca(ClO4)2,
removing precipitated KClO4 and concentrating by rotary
evaporation. NaN3 was recrystallized from water before use.
2.2. Analyses and instrumentation
Depending on the moiety to be determined, preliminary
decomposition of the complexes followed different
procedures.
Table 1
Absorption maximaa of pentacyanochromate(III) complexes in aqueous solution
Complex
LF bandsb
CT(Xꢀ) bandc
CT(CNꢀ) bandsc
Cr(CN)5(H2O)2ꢀd
Cr(CN)5(OH)3ꢀe
Cr(CN)5(NCS)3ꢀ
Cr(CN)5(N3)3ꢀ
a
428 (110)
436 (118)
416 (171)
440 (220)
334 (50)
368 (73)f
264 (2500)f
252 (2900)
244 (2800)f
246 (3600)
240 (8550)
218 (2800)f
319 (6200)
327 (2150)
260 (5400)f
260 (5300)
226 (9200)f
226 (7300)
Wavelengths in nm; molar extinction coefficients, Mꢀ1 cmꢀ1, in parentheses.
b
c
d
e
f
Ligand-field.
Charge-transfer.
pH < 5.
pH 12.
Absorption shoulder.

P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
899
Cr was analyzed as chromate ion by spectrophotometry
at 374 nm. Calibration with CrO4 standards at pH 13
and all aquation reactions, the pH was set at 5.0 by means
of 0.04 M acetate buffer. For N3 anation, a 0.04 M phos-
2ꢀ
ꢀ
gave e = (4.75 0.02) · 103 Mꢀ1 cmꢀ1. As in basic solu-
tion cyanide remains firmly coordinated, samples were first
heated in 1 M HClO4 at 80 ꢀC for 10 min to promote CNꢀ
release. After cooling, the pH was brought to ca. 13 with
NaOH pellets, and boiling for another 10 min in the pres-
ence of excess H2O2 completed decomposition and oxidized
chromium(III).
phate buffer having pH 6.0 was employed. Its mixing with
the moderately basic azide ion ðKHN ¼ 3:6 ꢂ 10ꢀ5Þ gave
3
rise to the equilibrium
H2POꢀ þ Nꢀ ¢ HPO2ꢀ þ HN3
ð1Þ
4
3
4
the constant of which is KH PO =KHN ¼ 1:7 ꢂ 10ꢀ3. This
ꢀ
2
3
caused a decrease of N3 by 4amounts varying between
1.1% for ½N3ꢀꢃ ¼ 0:2 and 0.3% for ½N3ꢀꢃ ¼ 5:0, and
brought the actual pH between 6.3 and 7.0, respectively.
The solutions containing all tꢀhe ingredients (ionic
strength, buffer, and NCSꢀ or N3 salt when required)
but the complexes were degassed in an ultrasound bath
to prevent bubble formation, and were brought to the
desired temperature. The complexes were then added as
follows. For studying anation, ꢁ0.2-mg aliquots of solid
K3[Cr(CN)5(OH)] Æ 2H2O were introduced in the cell, yield-
ing concentrations of 0.2–0.3 mM. In the case of NCSꢀ
aquation, mixing of a 100-lL volume of a ꢁ10ꢀ3 M
Cr(CN)5(NCS)3ꢀ stock solution ꢀ gave final samples
ꢁ0.05 mM in complex. For N3 aquation, ꢁ0.2-mg
amounts of K3[Cr(CN)5(N3)] were used, to obtain 0.2–
0.3 mM solutions. Under the conditions of fastest change,
reliable absorbance values, At, could be recorded within 5 s
from mixing. Data points (120–160) were collected during
each run, at intervals that varied from 0.5 to 400 s, depend-
ing on the reaction rates. The reactions were monitored
under pseudo-first-order conditions, generally for 4–5
half-lives. All chromium(III) species involved present
intense spectral bands between 200 and 280 nm (Fig. 1),
and light absorption in this region was found to cause effi-
cient photolysis. In order to preclude such an interference,
the radiation below 300 nm from the diode-array spectro-
photometer source was eliminated by a cutoff filter. For
the experiments performed in duplicate, the reproducibility
was generally within 2.5%. Rate constants kobsd were
obtained by nonlinear least-squares fits of the At values
to the equation At = A0 + (A1 ꢀ A0)exp(ꢀkobsdt). When
ꢀ
CNꢀ was measured potentiometrically at pH ca. 12 by
an Orion 94-06 selective electrode and a Radiometer
PHM-84 potentiometer. To avoid cyanide loss, the reac-
tion with 1 M HClO4 was carried out in stoppered flasks
at room temperature for 24 h. Each analysis was standard-
ized by a series of fresh KCN solutions.
NCSꢀ was determined as the iron(III) complex with
maximum absorbance at 450 nm (e 4300 Mꢀ1 cmꢀ1). In
this case, the samples were treated with 10 M HClO4 at
40 ꢀC for 2 h, since higher temperatures were found to
deplete the thiocyanate ion, presumably by oxidation. Fur-
ther reaction in NaOH for 2 h at 40 ꢀC caused complete
ligand aquation and formation of chromium(III) hydrox-
ide. The precipitate was dissolved with 10 M HClO4 and
the solution was then diluted at a ratio of 5:1 with iron
reagent (0.1 M Fe3+ in 0.5 M HClO4).
ꢀ
N3 was determined spectrophotometrically at 460 nm
(e ꢁ 1900 Mꢀ1 cmꢀ1) using 0.2 M Fe3+ + 0.1 M HClO4 as
reagent. Calibration with fresh NaN3 samples was done
each time.
C and N were determined by standard organic
microanalysis.
Absorption spectra were measured by a Cary 2300 spec-
trophotometer. The time dependence of optical densities
either at specific wavelengths or within given spectral
regions was recorded by a Hewlett–Packard 8452A
diode-array instrument. HPLC experiments were per-
formed by use of a Dionex GP40 chromatograph and an
AD20 detector. A 25-cm Ionpac AS-11 anion-exchange
column was eluted by gradients of aqueous NaClO4/NaOH
from 80/20 up to 320 mM/20 mM.
A
1
was experimentally measured, the difference between
the calculated and the experimental values was less than
3% of the total change in optical density. Activation
parameters were determined by least-squares analyses of
log(k/T) versus 1/T.
2.3. Kinetic procedures
The rates of anation of Cr(CN)5(H2O)2ꢀ by NCSꢀ and
those of aquation of Cr(CN)5(NCS)3ꢀ were measured by
spectrophotometry at 318 nm, where the extinction coeffi-
cients are 40 Mꢀ1 cmꢀ1 for the former complex and
6200 Mꢀ1 cm ꢀ1 for the latteꢀr. In a similar manner, anation
of Cr(CN)5(H2O)2ꢀ by N3 and aquation of Cr(CN)5-
(N3)3ꢀ were monitored at 328 nm, where the e values are
48 and 2150 Mꢀ1 cmꢀ1, respectively. In general, samples
of 2.0-mL volume were made up in stoppered cells of 10-
mm path length, thermostated to within 0.1 ꢀC and stirred
by a magnetic bar. The anion sources were NaNCS,
Ca(NCS)2, and NaN3, and the total cation concentration
was maintained at the desired values by addition of either
NaClO4 or Ca(ClO4)2 as appropriate. For NCSꢀ anation
Some measurements of the CNꢀ aquation rate of
Cr(CN)5(H2O)2ꢀ were made by essentially the same proce-
dure. The wavelength of analysis was 428 nm, in corre-
spondence of a ligand-field (LF) band maximum of the
reactant (e 110 Mꢀ1 cmꢀ1) and close to a minimum of the
first observable product, Cr(CN)3(H2O)3 (e 53 Mꢀ1 cmꢀ1).
Because of the relatively low extinction coefficients
involved, complex concentrations were 2–3 mM, or 10-fold
higher than in the experiments described above. In addi-
tion, since the tricyano product is subject to successive,
slower aquation, only the initial absorption changes, up
to no more than 15% conversion, were considered for rate
evaluation.

900
P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
Fig. 1. Ligand-field (right scale) and charge-transfer (left scale) absorption spectra of Cr(CN)5Xzꢀ complexes in aqueous solution. ꢅ ꢅ ꢅX = H2O; -----
X = NCSꢀ; —— X ¼ N3
.
ꢀ
3. Results
CNꢀ removal in 2 M HClO4. The spectral maxima of the
terminal aquation product occurred at 570, 412 and
294 nm, like those reported for Cr(H2O)5(NCS)2+ [27,28].
Cr(CN)5(N3)3ꢀ was likewise synthesized from Cr(CN)5-
(H2O)2ꢀ and N3ꢀ. As in the thiocyanato system, the inten-
sity of the LF absorption increases during reaction, with
the maximum now undergoing a red shift. Another analogy
is the development of the intense band in the 300–350-nm
region. After chromatographic isolation, Cr(CN)5(N3)3ꢀ
was obtained as the potassium salt. Characterization was
supplemented by the spectrum of the final product of
CNꢀ aquation in 0.2 M acid with maxima at 582, 432
and 271 nm, that matched the literature data for
Cr(H2O)5(N3)2+ [29].
3.1. The anation products
In order to authenticate the products observed under
kinetic conditions, both Cr(CN)5(NCS)3ꢀ and Cr(CN)5-
(N3)3ꢀ were independently synthesized and characterized.
Attempts to prepare Cr(CN)5(NCS)3ꢀ were first made in
methanol solution by the method that proved successful with
Cr(CN)5(NH3)2ꢀ and Cr(CN)5(py)2ꢀ [10,12]. Cr(CN)5-
(MeOH)2ꢀ was obtained by H+-promoted CNꢀ labilization
of CrðCNÞ63ꢀ. Its anation by NCSꢀ was not selective, how-
ever, as substantial amounts of other species were formed,
with more than one NCSꢀ group. Furthermore, separation
was made unduly difficult by the identical 3ꢀ charge of all
products. Analogous complexities were reported for the
reaction of K3[Cr(NCS)6] with KCN in acetonitrile, result-
ing in mixtures of all CrðCNÞ6ꢀnðNCSÞn3ꢀ anions, analytical
quantities of which were isolated by either ionophoresis
[22,23] or chromatography on alumina [24].
On the contrary, the reaction of aqueous
Cr(CN)5(H2O)2ꢀ with NCSꢀ is fairly specific and com-
plete. Buffering at pH 5 prevents deprotonation of the
H2O ligand (pKa ꢁ 9 [11,25,26]) that would otherwise
induce fast decomposition of the reactant. The formation
of Cr(CN)5(NCS)3ꢀ is accompanied by intensification of
the long-wavelength LF band with a slight blue shift and,
more significantly, by the growth of a strong absorption
between 300 and 350 nm, typical of coordinated NCSꢀ.
Chromatographic purification was facilitated by the charge
difference between reactant and product. Although all
attempts to isolate a solid were unsuccessful, the consider-
able inertness of the complex (vide infra) enabled storage of
frozen stock solutions. The composition and charge of
Cr(CN)5(NCS)3ꢀ were verified by elemental analysis and
by the elution behavior through the anion exchanger. A
single HPLC peak confirmed the absence of any extraneous
chromium(III) species. Further validation was by means of
Fig.
1 compares the absorption spectra of the
Cr(CN)5Xzꢀ ions with X = H2O, NCSꢀ and N ꢀ, and Table
3
1 reports the respective band maxima. As noted above, the
NCSꢀ and N3 species show analogous features. A single
ꢀ
4
4
LF band is apparent, denoting the A2g ! T2g transition
4
in Oh approximation. The higher-energy 4A2g ! T1g band
usually seen in Cr(III) is hidden by the intense absorption
ꢄ
at 320–330 nm, associated with the pðXÞ ! eg ðCrÞ
charge-transfer (CT) transition peculiar to all NCSꢀ- and
N3ꢀ-containing complexes. N-coordination of ambidentate
NCSꢀ is indicated by the displacement to higher energy of
4
the T2g maximum upon replacement of X = H2O by
X = NCSꢀ, as predicted by LF theory [30]. S-coordination
would have caused an appreciable spectral shift in the oppo-
site direction, let alone the general scarce affinity of Cr(III)
toward the soft S end of thiocyanate. The even stronger
absorption deeper in the UV common to all three species
is readily ascribed to CT transitions involving cyanide, on
the grounds of its close resemblance in position, shape and
intensity with that of CrðCNÞ63ꢀ. At variance with the long
accepted Cr ! CN assignment [31], recent theoretical
results infer a CN ! Cr charge shift [32,33].
Some related absorption data are found in the literature.
While our results confirm the spectrum of aqueous

P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
901
Cr(CN)5(NCS)3ꢀ reported earlier in a graphic form [24],
some discrepancy arises with Cr(CN)5(N3)3ꢀ. The kmax
and e values we have determined for this complex were
attributed to the Cr(CN)4(N3)(OH)3ꢀ ion [34], the nature
of which was apparently deduced from a Cr:C:N propor-
tion bearing a rather large uncertainty. Besides agreeing
with the present experimental CNꢀ:Cr ratio, our attribu-
tion is supported by several circumstances. First, on the
Cr(CN)5(H2O)2ꢀ mediated by protonation of CNꢀ had
to be totally inhibited. (b) The H3O+ concentration was
sufficient to maintain protonated the H2O ligand of the
complex (pKa ꢁ 9 [11,25,26]). At pH > 10 the deprotonated
form tends to decompose, so that even traces of Cr(OH)3
may thwart spectrophotometry. (c) In the Xꢀ ¼ N3 sys-
ꢀ
tem, the formation of HN3 (pKa ꢁ 4.4 [38]) had to be min-
ꢀ
imized. Note that the large excess of N3 was modified by
4
4
basis of the average LF strength, the A2g ! T2g maxi-
mum of a tetracyano species is expected to occur at a some-
what lower energy than found. In addition, the very same
spectrum is approached according to plain first-order
behavior during anation of Cr(CN)5(H2O)2ꢀ under kinetic
conditions. Finally, compelling evidence for the absence of
any OHꢀ ligand is the invariance of the whole spectrum
upon changing the pH from 3 to12.
the reaction with the buffer (Eq. (1)) to an extent (0.3–
1.1%) that was well within experimental uncertainty.
The reactions were occasionally followed through the
evolution of the whole spectrum. In the range of [Xꢀ] cov-
ered, the approach to the absorption of the Cr(CN)5X3ꢀ
species showed anation to be essentially complete. For
Xꢀ ¼ N3ꢀ the spectra of the reactant and the product never
cross, while for Xꢀ = NCSꢀ an isosbestic point occurs at
450 nm. Its persistence is suggestive of uncomplicated
kinetics. Possible secondary anation of minor amounts of
Cr(CN)3(H2O)3 generated by reactions (2) and (3) was
kinetically unimportant, as first-order behavior was always
obeyed up to at least 95% conversion.
ꢄ
The distinct pðXÞ ! eg ðCrÞ band of both Cr(CN)5-
(NCS)3ꢀ and Cr(CN)5(N3)3ꢀ was the key to the kinetic
experiments. High sensitivity was, in fact, ensured by the
large differences in molar absorbancy at 320–330 nm
between these complexes and Cr(CN)5(H2O)2ꢀ
.
The pseudo-first-order rate constants for the appearance
of Cr(CN)5X3ꢀ, kobsd, are listed in Tables 2–4 (Supplemen-
tary material) and typical series of kobsd at various total
[Na+] and [Ca2+] are displayed in Figs. 2 and 3 for
Xꢀ = NCSꢀ, and in Fig. 4 for Xꢀ ¼ N3ꢀ. The data of each
series are adequately fitted by nonlinear regressions to the
three-parameter equation:
3.2. Anation kinetics
The aqueous chemistry of Cr(CN)5(H2O)2ꢀ is one of
stepwise loss of CNꢀ to ultimately yield CrðH2OÞ
[35].
3þ
6
The first aquation step (Eq. (2)) is known to be followed
by very fast release of a second CNꢀ (Eq. (3)), leading to
Cr(CN)3(H2O)3 as the first observable product:
b½Xꢀꢃ
kobsd ¼ a þ
¼ kaq þ kX
ð5Þ
c þ ½Xꢀꢃ
kaq
2ꢀ
CrðCNÞ ðH2OÞ þ H2O ! CrðCNÞ ðH2OÞꢀ þ CNꢀ
ð2Þ
ð3Þ
5
4
2
where the nonzero intercept a represents the first-order rate
constant for CNꢀ aquation (Eq. (2)), designated as kaq, and
the composite [Xꢀ]-dependent expression indicated as kX,
describes anation (Eq. (4)). The sum of two terms in Eq.
(5) reflects depletion of the same substrate by two concur-
rent processes. While interpretation of the b and c param-
eters is discussed below, some features are immediately
apparent. (i) At a given [Xꢀ] and in the presence of the
same counterion, kobsd increases with the ionic strength.
(ii) On changing cation, equal or close results are obtained
at an equal total charge ([Ca2+] = 1/2[Na+]) rather than at
an equal ionic strength. (iii) The less-than-first-order
dependence of kx on [Xꢀ], that is, the downward curvature
of the plots is enhanced by the ionic strength. (iv) The kaq
parameters bear large uncertainties (in some cases up to
50%), arising from extrapolation of kobsd data that, of
necessity, lie far off the origin of the plots.
Less inaccurate kaq values were independently deter-
mined by direct monitoring of the aquation of
Cr(CN)5(H2O)2ꢀ in the absence of Xꢀ. Even so, the preci-
sion was not better than 17%, because the spectral sensi-
tivity in the LF region (the only suited to monitor this
process) was much lower than that attainable in the CT
region. Moreover, only the initial absorption changes could
be utilized as the Cr(CN)3(H2O)3 product undergoes suc-
cessive aquation. The two sets of results were consistent,
fast
CrðCNÞ ðH2OÞꢀ þ H2O ! CrðCNÞ ðH2OÞ þ CNꢀ
4
2
3
3
Cr(CN)3(H2O)3 can accumulate in the time scale of reac-
tion (2), since the successive aquation stages are by 50-fold
slower [36,37]. All steps involve acid-independent and acid-
dependent pathways, typical of complexes with strongly
basic ligands. Anation by Xꢀ (Eq. (4)) must obviously
compete with aquation of CNꢀ (Eq. (2)):
kX
2ꢀ
CrðCNÞ5ðH2OÞ þ Xꢀ ! CrðCNÞ5X3ꢀ þ H2O
ð4Þ
The rates of reaction (4) were measured under pseudo-
first-order conditions as a function of the Xꢀ concentration
up to 5.00 M, the ionic strength, and the temperature
between 25.0 and 55.0 ꢀC. The need for this process to be
reasonably competitive with process (2) imposed a lower
limit of 0.2 M or higher to [Xꢀ]. In most of the kinetic runs,
the only cation present was Na+. So as to assess possible
counterion effects, some sets of rate data were also
obtained by the use of Ca2+ salts.
For Xꢀ = NCSꢀ, all samples were uniformly buffered at
pH 5.0. For Xꢀ ¼ N3ꢀ, interaction of the phosphate buffer
with the different concentrations of azide resulted in a pH
range of 6.3–7.0. These pH values represented the best
compromise for complying with the following contrasting
requirements. (a) The fast acid-assisted aquation path of

902
P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
Fig. 2. Pseudo-first-order rate constants for the anation of Cr(CN)5(H2O)2ꢀ by NCSꢀ as a function of the thiocyanate ion concentration at 40 ꢀC, pH 5.0,
and at various total [Na+] (NaNCS + NaClO4).
Fig. 3. Pseudo-first-order rate constants for the anation of Cr(CN)5(H2O)2ꢀ by NCSꢀ as a function of the thiocyanate ion concentration at 40 ꢀC, pH 5.0,
and at various total [Ca2+] (Ca(NCS)2 + Ca(ClO4)2). For comparison, dashed lines represent the data of Fig. 2 at [Na+] = 2 · [Ca2+].
and the kaq values thus obtained were employed to evaluate
kX from kobsd by Eq. (5). Inasmuch as kX exceeds kaq by 1
order of magnitude for Xꢀ = NCSꢀ, and by 2 orders of
magnitude for Xꢀ ¼ N3ꢀ, the error introduced by kaq is
smaller than experimental imprecision in the former sys-
tem, and is quite negligible in the latter.
ues at the different ionic strengths, oscillate with no appar-
ent trend.
The parameters of Eq. (5) for the two anation reactions
at various cation concentrations and temperatures are col-
lected in Table 5. Figs.
2 and 3 show that at
[Na+] = 2.00 M or [Ca2+] = 1.00 M, the curvature of the
kobsd plot is hardly perceptible for Xꢀ = NCSꢀ, denoting
c ꢆ [NCSꢀ] in the whole workable range; therefore, only
the b/c ratios, but not the individual parameters, were
determinable. Under identical conditions, the less-than-
first-order dependence of kX on [Xꢀ] is, instead, clearly
observed for Xꢀ ¼ N3ꢀ. It is also noted that anation by
For NCSꢀ anation, additional sets of kobsd were
obtained at either pH 4.2 or pH 6.0. There was no signifi-
cant pH dependence of both kaq and kNCS. The lack of
acidity effects on kNCS is not unexpected. The random dis-
tribution of kaq inside the above error band indicates that
acid-assisted CNꢀ release is not the dominant aquation
path in this pH interval. Within the same band the kaq val-
N3 is ca. 6-fold faster than anation by NCSꢀ.
ꢀ

P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
903
Fig. 4. Pseudo-first-order rate constants for the anation of Cr(CN)5(H2O)2ꢀ by N3 as a function of the azide ion concentration at 40 ꢀC, pH 6.3–7.0, and
ꢀ
at various total [Na+] (NaN3 + NaClO4).
Table 5
Parameters of rate equation (5)a and derived quantities for the reaction of Cr(CN)5(H2O)2ꢀ with Xꢀ
d
Xꢀ
t (ꢀC)
[Na+]b (M)
[Ca2+]c (M)
103b (=103kI)d,e (sꢀ1
)
cd (M)
104b/c (Mꢀ1sꢀ1
)
Qe (Mꢀ1
)
NCSꢀ f
25.0
2.00
5.00
2.00
3.50
4.25
5.00
0.289 0.024
1.12 0.05
7.56 0.40
0.132 0.007
40.0
1.62 0.04
3.50 0.17
4.48 0.11
5.24 0.11
8.42 0.51
7.08 0.19
6.16 0.20
0.119 0.008
0.141 0.004
0.162 0.005
1.00
1.27
1.75
2.50
1.63 0.04
2.04 0.02
3.47 0.51
4.96 0.31
6.74 1.48
5.29 0.92
0.148 0.035
0.189 0.035
55.0
2.00
3.50
5.00
7.45 0.32
14.9 0.9
23.1 1.4
7.88 0.49
5.57 0.42
0.127 0.008
0.180 0.013
ꢀ g
N3
25.0
40.0
2.00
5.00
2.00
3.50
5.00
2.00
5.00
2.17 0.09
5.38 0.40
10.4 0.7
15.4 1.0
23.8 1.2
41.7 2.6
96.1 5.2
5.06 0.13
3.21 0.24
4.84 0.30
3.72 0.33
3.05 0.18
4.43 0.30
2.88 0.22
0.198 0.005
0.312 0.024
0.206 0.013
0.269 0.024
0.328 0.020
0.225 0.017
0.348 0.027
55.0
a
The parameter a, denoting the pseudo-first-order rate constant kaq for CNꢀ aquation of Cr(CN)5(H2O)2ꢀ, is (2.3 0.4) · 10ꢀ6 sꢀ1 at 25.0 ꢀC,
(4.5 0.5) · 10ꢀ5 sꢀ1 at 40.0 ꢀC and (4.3 0.7) · 10ꢀ4 sꢀ1 at 55.0 ꢀC. No dependence of kaq on the ionic strength is observable within experimental
uncertainty (see text).
b
Anion source: NaX; the ionic strength, adjusted with NaClO4, is equal to [Na+].
c
Anion source: Ca(NCS)2; the ionic strength, adjusted with Ca(ClO4)2, is 3 · [Ca2+].
d
Each entry is calculated from 7 to 16 rate constant determinations at various [Xꢀ]. Uncertainties are standard deviations.
Quantity defined by Eq. (12).
pH 5.0 (acetate buffer).
pH 6.3–7.0 (phosphate buffer, see text).
e
f
g
Table 6 reports the activation parameters for the coeffi-
cient b (and for the b/c ratio when the two quantities are
not separable). Concerning the intercept a of Eq. (5), it is
worthwhile to note that the activation enthalpy for CNꢀ
aquation (DHꢀaq = 136 5 kJ molꢀ1) is about 70% larger
than the DHꢀ values of the anation reactions. This bears
on the competition between processes (2) and (4). The
lower the temperature, the more preferred anation over
cyanide loss from Cr(CN)5(H2O)2ꢀ. For example, at
l = 2.00 and [NCSꢀ] = 1.00, kNCS/kaq ꢁ 12 at 25 ꢀC and
ꢁ0.7 at 70 ꢀC. The difference was especially relevant to
the preparative conditions of Cr(CN)5(NCS)3ꢀ
.

904
P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
Table 6
Activation parameters for anation of Cr(CN)5(H2O)2ꢀ by Xꢀ and aquation of Cr(CN)5X3ꢀ
Xꢀ
[Na+]a (M)
Anation (b/c = kIQ)
Anation (b = kI)
Aquation ðkH O
Þ
2
DHꢀ (kJ molꢀ1
)
DSꢀ (J molꢀ1 Kꢀ1
)
DHꢀ (kJ molꢀ1
)
DSꢀ (J molꢀ1 Kꢀ1
ꢀ37 10b
)
DHꢀ (kJ molꢀ1
)
DSꢀ (J molꢀ1 Kꢀ1
)
NCSꢀ
2.00
3.50
5.00
85.4 3.4
83.5 3.4b
87.7 3.4
ꢀ45 11
ꢀ43 11b
ꢀ24 11
ꢀ37 10
95.9 3.8
93.0 4.6
93.6 3.2
ꢀ32 11
ꢀ39 11
79.9 3.2b
79.3 2.9
ꢀ35
ꢀ36
9
9
ꢀ37
8
ꢀ
N3
2.00
3.50
5.00
81.0 2.9
77.5 2.8
75.6 3.5
107
106
107
4
5
3
14 12
9
14
14
8
78.3 3.9
ꢀ35 13
ꢀ35 11
a
The ionic strength, adjusted with NaClO4, is equal to [Na+].
Evaluated from rate constants at two temperatures. The deviations are assumed to be the average of those pertaining to the same group of data.
b
3.3. Aquation kinetics
Eq. (6)). A practical advantage of the relatively high inert-
ness of the two Cr(CN)5X3ꢀ complexes was that their stock
solutions could be kept unaltered for fairly long periods.
The rates of Xꢀ release from the Cr(CN)5X3ꢀ complexes
(Eq. (6)) were measured at pH 5.0 and in the same NaClO4
media used in the anation studies.
4. Discussion
kH2
O
2ꢀ
CrðCNÞ5X3ꢀ þ H2O ! CrðCNÞ5ðH2OÞ þ Xꢀ
ð6Þ
A complexity of the Cr(CN)5(H2O)2ꢀ system, compared
with Co(CN)5(H2O)2ꢀ is that under the conditions, where
anation is reasonably efficient, the former is unstable
toward CNꢀ loss, while the latter is not. Therefore, rate
measurements are precluded at low [Xꢀ], where anation
does not sufficiently compete with decomposition of
Cr(CN)5(H2O)2ꢀ. This shortcoming, in turn, brings about
uncertainty on the indirect determination of kaq (intercepts
of Figs. 2–4), although its consequence on the precision of
As Xꢀ aquation is appreciably slower than Xꢀ entry, a
higher temperature interval was chosen (55.0–80.0 ꢀC).
Even at the low acidity employed, subsequent CNꢀ loss
from the Cr(CN)5(H2O)2ꢀ primary product (Eqs. (2) and
(3)) is much faster than reaction (6). This occurrence did
not affect the kinetic measurements, since at the monitoring
wavelengths the extinction coefficients of all species
encountered in the cascade that ultimately leads to
CrðH2OÞ6 are less than 1% of those of the Cr(CN)5X3ꢀ
3þ
the anation parameters is minimal for Xꢀ = NCSꢀ and
anions [35–37]. A pH of 5.0 also excludes any aquation
pathway of Cr(CN)5(N3)3ꢀ via protonation of the azido li-
quite negligible for Xꢀ ¼ N3
.
ꢀ
The general reaction pattern can thus be summarized. (1)
Anation of Cr(CN)5(H2O)2ꢀ by Xꢀ is faster than Xꢀ
aquation of Cr(CN)5X3ꢀ. (2) Theꢀrates of ꢀboth reactions
are larger for Xꢀ ¼ N3ꢀ than for X = NCS . (3) The ionic
strength enhances anation, but has little effect on aquation.
(4) The activation enthalpies for anation are lower than those
for aquation. (5) The DHꢀ values of all processes are not
significantly affected by the ionic strength.
gand [29]. The release of both NCSꢀ and N3 obeyed first-
ꢀ
order kinetics virtually to completion. The mean reproduc-
ibility of the rate constants kH O was 5%. The kH O values
2
2
at various temperatures are given in Table 7, and the asso-
ciated activation parameters are included in Table 6.
Compared with the related anation rates, the aquation
rates are smaller by 1 order of magnitude for Xꢀ = NCSꢀ,
and by 2 orders of magnitude for Xꢀ ¼ N3ꢀ. This made
negligible any possible contribution to the rate expression
Regarding a plausible reaction mechanism, an issue that
has been discussed in detail [4] is that a metal ion may not
be subject to a single type of mechanism in different ligand
and medium environments. There now appears to be an
agreement that Cr(H2O)5Xz+ [39,40] and Cr(NH3)5Xz+
[38,41] complexes are generally subject to Ia and Id substi-
tutions, respectively. For the pentaammine family the
assignment has been controversial [1,3], but more recent
results [42,43] have strengthened the Id attribution. In light
of these considerations, our findings are compared with
two possible schemes. The first (Eqs. (7) and (8)) postulates
kobsd for anation (Eq. (5)) by the reverse process (kH O
,
2
Table 7
a
Aquation rate constants of Cr(CN)5X3ꢀ complexes at pH 5.0
c
Xꢀ
[Na+]b (M]
104kH O ðsꢀ1
Þ
2
55.0 ꢀC
70.0 ꢀC
80.0 ꢀC
NCSꢀ
2.00
3.50
5.00
0.78 0.04
0.99 0.05
1.01 0.04
3.68 0.09
4.12 0.16
4.44 0.16
10.2 0.5
12.2 0.6
12.5 0.4
ꢀ
2ꢀ
N3
2.00
3.50
5.00
2.93 0.14
3.16 0.15
2.98 0.13
15.3 0.9
15.8 1.2
16.3 0.6
52.3 3.2
54.2 3.6
52.9 1.5
formation (k1) of a CrðCNÞ5 intermediate capable to dis-
criminate between H2O (kꢀ1) and Xꢀ (k2):
k1
a
2ꢀ
CrðCNÞ ðH2OÞ ¢ CrðCNÞ25ꢀ þ H2O
ð7Þ
ð8Þ
Acetate buffer.
5
b
c
The ionic strength, established by NaClO4, is equal to [Na+].
k
ꢀ1
k2
Average of two or more independent determinations. Uncertainties are
2ꢀ
CrðCNÞ5 þ Xꢀ ! CrðCNÞ5X3ꢀ
standard deviations.

P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
905
In principle, this hypothesis is not unreasonable, keep-
ing in mind that on passing from pentaaqua to pentaam-
mine species, an important factor in inducing the change
from Ia to Id must be the increased electron density on
the central metal. In pentacyano complexes, the even stron-
ger r donation of CNꢀ, the related trans effect, and the
The alternative possibility is ligand interchange within
ion triplets:
2ꢀ
CrðCNÞ5ðH2OÞ þ Mzþ þ Xꢀ
Q
2ꢀ
¢ fCrðCNÞ ðH2OÞ ꢅ Mzþ ꢅ Xꢀg
ð10Þ
5
kI
2ꢀ
fCrðCNÞ ðH2OÞ ꢅ Mzþ ꢅ Xꢀg !fCrðCNÞ5X3ꢀ ꢅ Mzþ ꢅ H2Og
negative charge may well push the situation toward the D
5
3þ
¢ CrðCNÞ5X3ꢀ þ Mzþ
ð11Þ
limit. The acid dissociation constants of CrðH2OÞ6
-
ðpKa ꢁ 4Þ [44], Cr(NH ) (H2O)3+ (pKa ꢁ 5) [45] and
Cr(CN)5(H2O)2ꢀ (pKa ꢁ39)5 [11,25,26] are in line with an
increasing tendency to Cr–OH2 bond breakage. Further-
where Mz+ = Na+ or Ca2+. Even though ion pairing be-
tween negative reactants is obviously excluded, formation
of ion triplets or higher associates cannot be disregarded
at the high [Mz+] at which anation is efficient and the plots
of Figs. 2–4 are curved enough to enable reliable separation
of the b and c parameters. According to this scheme, kX is
expressed by Eq. (12),
2ꢀ
more, a possible CrðCNÞ5 tbp intermediate would be sta-
bilized by the p system of the cyanides and by the reduction
of the mutual repulsion of the charged ligands. A limiting
D pathway of Cr(III) has previously been proposed for
NCSꢀ anation of Cr(NCS)5(Me2SO)2ꢀ in sulfolane [13].
Eqs. (7) and (8) lead to rate equation (9).
kIQ½Xꢀꢃ
kobsd ꢀ kaq ¼ kX ¼
ð12Þ
1 þ Q½Xꢀꢃ
k1k2½Xꢀꢃ
kobsd ꢀ kaq ¼ kX
¼
ð9Þ
k
½H2Oꢃ þ k2½Xꢀꢃ
ꢀ1
where the parameter kI for ligand interchange (Eq. (11))
coincides with the b factor of Eq. (5), and the equilibrium
quotient Q for association of Cr(CN)5(H2O)2ꢀ with Xꢀ
at each given [Mz+] is equal to 1/c.
In this context, the b parameter of Eq. (5) represents the
rate constant k1 for generation of the intermediate, and c
(= kꢀ1[H2O]/k2) describes its relative preference for the
two nucleophiles. In spite of its attractiveness, this interpre-
tation is not congruent with the data given in Table 5. A
requirement of the D mechanism is that the b parameter
be the same for both NCSꢀ and N3ꢀ, contrary to observa-
tion. The linearized form of Eq. (5), illustrated in Fig. 5,
clearly shows that the intercepts 1/b of the plots of
1/(kobsd ꢀ kaq) versus 1/[Xꢀ] are not the same for the two
entering ions. By the way, even for anation of the Co(III)
analogue Co(CN)5(H2O)2ꢀ, long regarded as a classic case
of D-type substitution [15–18], serious doubt has been cast
as to the evidence for a genuine pentacoordinate species
[19–21].
The increase of kI with the ionic strength (Mz+ being the
same) is qualitatively consistent with the expectation for a
reaction between like-charged ions. The magnitude of Q
suggests a possibly more appropriate picture in terms of
random association between reacting anions. Association
constants smaller than ca. 1 Mꢀ1 have been reported even
for anation of cationic complexes [38,41]. Yet, it has been
argued [46] that attribution of such small values to the for-
mation of definite 1:1 associated species may be of doubtful
significance. From this viewpoint, Eqs. (10)–(12) continue
to formally account for the observations, with Q taken as
an operational quantity and the species in brackets as
Fig. 5. Plots of 1/(kobsd ꢀ kaq) vs. the reciprocal of the anion concentration at 40.0 ꢀC, total [Na+] = 5.00, pH 5.0 for Xꢀ = NCSꢀ, and pH 6.3–7.0 for
Xꢀ ¼ N3
.
ꢀ

906
P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
encounter complexes between Xꢀ and mostly Cr(CN)5-
(H2O)2ꢀ Æ Mz+ ion pairs. The cations would no longer be
part of specific triplets, but would mainly act as attenuators
of the repulsion between reactants. Of course, intermediate
situations are possible. In addition to the general increase
of Q with [Mz+], in agreement with this picture is the effect
of changing counterion. The condition [Ca2+] = 1/2[Na+]
at constant [Xꢀ] results in close kinetic behaviors that tend
to coincide at the lower [Mz+] employed, despite the large
differences in ionic strength (Fig. 3). The total positive
charge thus appears to be a more important overall factor
than the ionic strength. The tendency of Q to increase with
the temperature (Table 5) is consistent with the random
aggregation model. The variations are rather small and,
specially for Xꢀ ¼ N3ꢀ, are within standard deviation,
allowing the temperature dependence to be semiquantita-
tively assessed as DHQ = 5 3 kJ molꢀ1 for both systems.
Comparison of Figs. 2 and 4 shows that, all conditions
Table 8 reports the kH O values extrapolated at 25.0 ꢀC,
2
along with the second-order anation rate constants kIQ, the
equilibrium quotients for the interchange process (Eq. (11))
QI ¼ kI=kH O and the overall reaction quotients (Eq. (4))
2
Q0 ¼ kIQ=kH O. Given the small ionic strength dependence
2
of kH O, evaluation of QI would not be too inaccurate even
2
in case aquation did not proceed mainly from
{Cr(CN)5X3ꢀ Æ Mz+} ion-paired species. Both QI and Q0
increase substantially with the ionic strength as a conse-
quence of its larger influence on kI. Note that the
Xꢀ ¼ N3 system, besides being more reactive in both
ꢀ
directions, features a larger anation equilibrium quotient.
The average temperature dependencies of kIQ and kH O
2
(Table 6) yield the enthalpy of reaction (4). Formation of
the Cr(CN)5X3ꢀ ions is exothermic, with DH0 values of
ꢀ9 3 kJ molꢀ1 for Xꢀ = NCSꢀ, and ꢀ27 8 kJ molꢀ1
for Xꢀ ¼ N3ꢀ. While the large deviations obscure any pos-
sible ionic strength dependence of DH0, the data confirm
the larger stability of the azido complex.
being equal, N3 is a better entering group than NCSꢀ.
ꢀ
The data of Table 5 attribute such a propensity to larger
values of both Q and kI. The ca. 2-fold difference between
the association parameters can be explained by the smaller
As to the type of interchange, the above-discussed trend
related to the variation of the ancillary ligands would itself
imply an Id step in Eq. (11). The temperature dependencies
of Table 6 argue in favor of such an assignment, even
though comparison is necessarily restricted to the two
Cr(CN)5(H2O)2ꢀ/Xꢀ systems that could be studied with
suitable accuracy.
size of N3 with respect to isostructural NCSꢀ. The inter-
ꢀ
change rate constants differ by a factor of ca. 4.5; this, how-
ever, is halved if a statistical coefficient of 2 is entered in
ꢀ
favor of N3 capable to biꢀnd with both N ends. The higher
intrinsic reactivity of N3 parallels earlier findings with
The activation enthalpies for anation are quite similar
and overlap within the margins of error. Their minor
dependence on the nature of Xꢀ suggests that Cr–OH2
bond breaking is more important than Cr–X bond mak-
ing in the transition state. Conversely, the DHꢀ parame-
ters for aquation show differences that are definitely
larger than the uncertainties involved and are consistent
with Cr–X bond labilization being energetically dominant
over Cr–OH2 bond formation. Also the similarity of the
DSꢀ parameters for anation and the diversity of those
for aquation may be of some significance in this respect.
That the DHꢀ values for the two aquation reactions differ
by a limited extent is not surprising, since both leaving
groups are N-coordinated. Finally, the present interpreta-
tion is consonant with literature data on the aquation of
3þ
Cr(III) cationic systems. Examples are CrðMe2SOÞ6 [47]
3þ
and CrðDMFÞ6 [48] in the respective aprotic solvents
(where the reactants are highly ion-paired), as well as aque-
ous Cr(NH3)5(H2O)3+ [38,41]. In the pentaammine case, the
overall kIQ parameters are similar, but the difference of the
association constants of N3ꢀ (estimated 60.1 Mꢀ1 [38]) and
NCSꢀ (0.7 Mꢀ1 [41]) implies a higher kI value for azide. The
same order of nucleophilicity was also observed with Co(N-
H3)5(H2O)3+ [49–51] and Co(CN)5(H2O)2ꢀ [15,16].
Differently from anation, the aquation rate constants
kH O are little sensitive to the cationic charge and the ionic
2
strength (Table 7). Under the chosen conditions, aquation
is likely to take place largely from {Cr(CN)5X3ꢀ Æ Mz+} ion
pairs. Any cation interaction with the neutral H2O mole-
cules of the second coordination sphere is certainly less
important than the interactions thought to promote a com-
bination of Cr(CN)5(H2O)2ꢀ and Xꢀ.
3ꢀ
the N-coordinate CrðNCSÞ6 anion, inferred to proceed
by an Id mechanism on the basis of a moderately positive
volume of activation [52].
Table 8
Evaluation of the equilibrium quotients for anation of Cr(CN)5(H2O)2ꢀ by Xꢀ at 25.0 ꢀC
b
Xꢀ
[Na+]a
104kIQ (Mꢀ1 sꢀ1
)
106kH O (sꢀ1
)
QI = kI/kH O
Q0 = kIQ/kH O (Mꢀ1
)
2
2
2
NCSꢀ
2.00
3.50
5.00
0.289 0.024
0.786 0.052b
1.48 0.08
2.03 0.10
2.80 0.14
2.85 0.11
14
28
52
2
2
3
254 17
393 18
ꢀ
N3
2.00
5.00
4.29 0.17
16.8 1.3
4.89 0.30
5.04 0.20
444 27
1070 80
88
330 26
6
a
The ionic strength, adjusted with NaClO4, is equal to [Na+].
Extrapolated from data at higher temperatures.
b

P. Riccieri, E. Zinato / Inorganica Chimica Acta 360 (2007) 897–907
907
[22] E. Blasius, H. Augustin, U. Wenzel, J. Chromatogr. 49 (1970) 520.
Appendix A. Supplementary data
[23] E. Blasius, H. Augustin, U. Wenzel, J. Chromatogr. 50 (1970) 319.
[24] E. Blasius, H. Augustin, Z. Anorg. Allg. Chem. 417 (1975) 47, 55.
[25] L. Jeftic, S.W. Feldberg, J. Am. Chem. Soc. 92 (1970) 5272.
[26] Y. Sakabe, Y. Matsumoto, Bull. Chem. Soc. Jpn. 54 (1981) 1253.
[27] E.L. King, E.B. Dismukes, J. Am. Chem. Soc. 74 (1952) 1674.
[28] C. Postmus, E.L. King, J. Phys. Chem. 59 (1955) 1208.
[29] T.W. Swaddle, E.L. King, Inorg. Chem. 3 (1964) 234.
[30] C.K. Jørgensen, Modern Aspects of Ligand Field Theory, North
Holland, Amsterdam, 1970 (Chapter 26).
Tables 2–4 listing kinetic data for the anation reactions.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.ica.2006.
06.002.
References
[31] J.J. Alexander, H.B. Gray, J. Am. Chem. Soc. 90 (1968) 4260.
[32] E. Wasielewska, Inorg. Chim. Acta 122 (1986) L1.
[33] M.F.A. Hendrickx, V.S. Mironov, L.F. Chibotaru, A. Ceulemans, J.
Am. Chem. Soc. 125 (2003) 3694.
[34] H.B. Quang, Z. Stasicka, B. Mainusch, F. Wasgestian, J. Photochem.
Photobiol. A: Chem. 98 (1996) 149.
[35] W.B. Schaap, R. Krishnamurthy, D.K. Wakefield, W.F. Coleman,
in: S. Kirschner (Ed.), Coordination Chemistry, Plenum Press, New
York, 1969, p. 177.
[36] D.K. Wakefield, W.B. Schaap, Inorg. Chem. 8 (1969) 512.
[37] D.K. Wakefield, W.B. Schaap, Inorg. Chem. 10 (1971) 306.
[38] S. Castillo-Blum, A.G. Sykes, Inorg. Chem. 23 (1984) 1049.
[39] T.W. Swaddle, G. Guastalla, Inorg. Chem. 7 (1968) 1915.
[40] D. Thusius, Inorg. Chem. 10 (1971) 1106.
[1] T.W. Swaddle, Coordin. Chem. Rev. 14 (1974) 217.
[2] D.A. House, Coordin. Chem. Rev. 23 (1977) 223.
[3] L. Mønsted, O. Mønsted, Coordin. Chem. Rev. 94 (1989) 109.
[4] P.A. Lay, Coordin. Chem. Rev. 110 (1991) 213.
[5] E. Zinato, in: A.W. Adamson, P.D. Fleischauer (Eds.), Concepts of
Inorganic Photochemistry, Wiley, New York, 1975 (Chapter 4).
[6] A.D. Kirk, Coordin. Chem. Rev. 39 (1981) 225.
[7] E. Zinato, Coordin. Chem. Rev. 129 (1994) 195.
[8] A.D. Kirk, Chem. Rev. 99 (1999) 1607.
[9] E. Zinato, P. Riccieri, Coordin. Chem. Rev. 211 (2001) 5.
[10] P. Riccieri, E. Zinato, Inorg. Chem. 29 (1990) 5035.
[11] J. Sotomayor, A.J. Parola, F. Pina, E. Zinato, P. Riccieri, M.F.
Manfrin, L. Moggi, Inorg. Chem. 34 (1995) 6532.
[12] P. Riccieri, E. Zinato, A. Aliboni, Inorg. Chem. 36 (1997) 2279.
[13] S.T.D. Lo, D.W. Watts, Aust. J. Chem. 28 (1975) 1907.
[14] D. Banerjea, B. Chakravarti, Inorg. Chim. Acta 240 (1995) 117.
[15] A. Haim, W.K. Wilmarth, Inorg. Chem. 1 (1962) 573, 583.
[16] A. Haim, R.J. Grassi, W.K. Wilmarth, Adv. Chem. Ser. 49 (1965) 31.
[17] C.H. Langford, H.B. Gray, Ligand Substitution Processes, Benjamin,
New York, 1965.
[41] T. Ramasami, A.G. Sykes, Inorg. Chem. 11 (1976) 2885.
[42] P.A. Lay, Inorg. Chem. 26 (1987) 2144.
[43] D.C. Gaswick, S.M. Malinak, Inorg. Chem. 32 (1993) 175.
[44] L.G. Sillen, A.E. Martell, Chem. Soc., Spec. Publ. No. 25, 1971.
[45] C.S. Garner, D.A. House, Transition Met. Chem. 6 (1970) 59.
[46] K.A. Connors, Binding Constants, J. Wiley, New York, 1987, p. 89
and references therein.
[18] F. Basolo, R.G. Pearson, Mechanisms of Inorganic Reactions, 2nd
ed., J. Wiley, New York, 1967.
[19] M.G. Burnett, W.M. Gilfillan, J. Chem. Soc., Dalton Trans. (1981)
1578.
[47] S.T.D. Lo, D.W. Watts, Aust. J. Chem. 28 (1975) 491.
[48] S.T.D. Lo, T.W. Swaddle, Inorg. Chem. 15 (1976) 1881.
[49] C.H. Langford, W.R. Muir, J. Am. Chem. Soc. 89 (1967) 3141.
[50] T.W. Swaddle, G. Guastalla, Inorg. Chem. 8 (1969) 1604.
[51] A. Haim, Inorg. Chem. 9 (1970) 426.
[20] A. Haim, Inorg. Chem. 21 (1982) 2887.
[21] M.H.M. Abou-El-Wafa, M.G. Burnett, J.F. McCullagh, J. Chem.
Soc., Dalton Trans. (1987) 2311.
[52] K.E. Hyde, D.A. Palmer, H. Kelm, Inorg. Chim. Acta 29 (1978) L278.