A mechanistic study of the [La2(OCH3) 2]4+-and [(1,5,9-triazacyclodo-decane):Zn:(OCH 3)]+-catalyzed methanolysis of carbonates: Possible application for the recycling of bisphenol A polycar

Article abstract of DOI:10.1139/cjc-2013-0270

The kinetics of the methanolysis of seven methyl aryl carbonates (3) and two methyl alkyl carbonates (4) promoted by [12[ane]N3:Zn:(OCH ₃)]⁺ and [La₂(OCH₃) ₂]⁴⁺ catalysts (1 and 2, respecti

Full text of DOI:10.1139/cjc-2013-0270

1139
ARTICLE
A mechanistic study of the [La2(OCH3)2]4+- and [(1,5,9-triazacyclodo-
decane):Zn:(OCH3)]+-catalyzed methanolysis of carbonates: possible
application for the recycling of bisphenol A polycarbonates
Alexei A. Neverov, Leanne D. Chen, Sean George, David Simon, Christopher I. Maxwell, and R. Stan Brown
Abstract: The kinetics of the methanolysis of seven methyl aryl carbonates (3) and two methyl alkyl carbonates (4) promoted by
[12[ane]N3:Zn:(OCH3)]+ and [La2(OCH3)2]4+ catalysts (1 and 2, respectively) have been studied at 25.0 °C. Brønsted plots of the
s
s
log ko2bs values for methanolysis versus aryloxy and alkoxy leaving group (LG) pKHa OAr or pKaHOR values (the pKa values of the
s
s
s
parent ArOH or ROH in methanol) for substrates 3 and 4 show an apparent downward break at pKa ϳ16.6 and 15.2 with
s
[12[ane]N3:Zn:(OCH3)]+ and [La2(OCH3)2]4+, respectively. The breakpoint is not due to a change in rate-limiting step in a two-step
process involving metal ion delivery of a coordinated methoxide to a transiently associated substrate and the subsequent
breakdown of a tetrahedral intermediate to form product. The more satisfactory explanation is that the break arises when one
correlates the rate constants for two dissimilar sets of substrates, namely aryloxy- and alkoxy-substituted 3 and 4. DFT calcula-
tions for the 1-promoted reactions of methyl 4-nitrophenyl carbonate (3b), which has a good aryloxy leaving group, and methyl
isopropyl carbonate (4b), which has a relatively poor alkyl one, indicate that the catalyzed processes involve two steps. Accord-
ingly, the methanolysis of all 3 having sspKaHOAr values for the parent phenols ≤15.3 involves rate-limiting nucleophilic attack and
fast breakdown. For the isopropyl alkyl derivative (4b) having a sspKaHOR > 18.13, the rate-liming step is the metal ion promoted
breakdown of a tetrahedral intermediate. The catalytic system employing 2 has utility for the catalytic decomposition of
poly(bisphenol A carbonate). In a semi-optimized system where 1000 mg of poly(bisphenol A carbonate), treated at 100 °C for
30 min in 2 mL of 60:40 chloroform−methanol containing La(OTf)3:NaOMe (5:7.5 mmol L−1), the reaction gave an 84% yield of
bisphenol A, corresponding to >300 turnovers per catalyst.
Key words: methanolysis, catalysis, carbonate, polycarbonate, reaction mechanism, DFT computations.
Résumé : On étudie la cinétique de la méthanolyse de sept carbonates de méthylaryle (3) et de deux carbonates de méthylalkyle
(4) favorisée par les catalyseurs [12[ane]N3:Zn:(OCH3)]+ et [La2(OCH3)2]4+ (1 et 2 respectivement) a` 25,0 ° C. Les courbes de Brønsted
des valeurs du log ko2bs pour la méthanolyse en fonction des valeurs du sspKHa OAr ou du sspKHa OR du groupe partant aryloxy ou alkoxy
(les valeurs du pKa du ArOH ou du ROH parent dans le méthanol) pour les substrats 3 et 4 révèlent une apparente rupture a` la
baisse aux valeurs de ϳ16,6 et 15,2 du sspKa avec [12[ane]N3:Zn:(OCH3)]+ et [La2(OCH3)2]4+, respectivement. Le point de rupture n’est
pas attribuable a` un changement de l’étape limitant la vitesse dans un processus en deux étapes faisant intervenir la fourniture
de l’ion métallique d’un méthoxyde coordonné a` un substrat provisoirement associé et la décomposition subséquente d’un
intermédiaire tétraédrique pour former le produit. Une explication plus satisfaisante est que la rupture a lieu lorsqu’on établit
une corrélation entre les constantes de vitesse de deux ensembles dissemblables de substrats, nommément les composés 3 et 4
contenant des substituants aryloxy et alkoxy. Des calculs par la méthode de la théorie de la fonctionnelle de la densité (DFT) pour
les réactions favorisées par 1 du carbonate de méthyl-4 nitrophényle (3b), qui possède un bon groupe partant aryloxy, et du
carbonate de méthyl-isopropyle (4b), dont le groupe partant est relativement mauvais, indiquent que les processus catalysés
s
comprennent deux étapes. Par conséquent, la méthanolyse de tous les dérivés 3 dont les valeurs du pKHa OAr pour les phénols
s
parents sont ≤15,3 comporte une attaque nucléophile limitant la vitesse et une décomposition rapide. Pour le dérivé isopropyl
s
alkyle (4b) dont le pKHa OR > 18,13, l’étape limitant la vitesse est la décomposition de l’intermédiaire tétraédrique favorisée par
s
l’ion métallique. Le système de catalyse faisant intervenir 2 utilise la décomposition catalytique du poly(carbonate de bisphénol
A). Dans un système semi-optimisé où 1 000 mg de poly(carbonate de bisphénol A), traités a` 100 ° C pendant 30 min dans 2 mL
d’un mélange 60:40 de chloroforme−méthanol contenant La(OTf)3:NaOMe (5:7,5 mmol L−1), la réaction a donné du bisphénol A
avec un rendement de 84%, ce qui correspond a` >300 moles de produit par mole de catalyseur. [Traduit par la Rédaction]
Mots-clés : méthanolyse, catalyse, carbonate, polycarbonate, mécanisme de réaction, calculs DFT.
monomer and reused.1 Existing methods for chemically recycling
PC (by methanolysis under basic conditions) typically require
pressures of 2–25 MPa, temperature ranges of 75–180 °C, and the
presence of stoichiometric amounts of sodium hydroxide or other
strongly alkaline catalysts.2 These conditions are required, since
PC is not soluble in pure methanol under ambient conditions so
the surface of the polymer is highly resistant to penetration by
Introduction
Polycarbonate (PC), such as that made from bisphenol A and
phosgene, is widely used in glazing, eyewear, packaging contain-
ers, electronics, and optical media (DVD, Blu-ray) and is notable
for its impact resistance, clarity, and optical properties. Simple
methods for catalytic chemical recycling of these polymers are
continually sought, as waste material could be recycled back into
Received 13 June 2013. Accepted 1 August 2013.
A.A. Neverov, L.D. Chen, S. George, D. Simon, C.I. Maxwell, and R.S. Brown. Department of Chemistry, Queen’s University, Kingston, ON K7L 3N6, Canada.
Corresponding author: R. Stan Brown (e-mail: rsbrown@chem.queensu.ca).
Can. J. Chem. 91: 1139–1146 (2013) dx.doi.org/10.1139/cjc-2013-0270
Published at www.nrcresearchpress.com/cjc on 13 August 2013.

1140
Can. J. Chem. Vol. 91, 2013
background reactions at neutral sspH, 7 and their ability to func-
tion at ambient temperature in light alcohols where the metal
alkoxides act as both buffers and catalysts. Recently, the use of
La(III) isopropoxide complexed with polyoxy ligands such as
2-(2-methoxyethoxy)ethanol has been shown to have very wide
application in the transesterification of esters, carbamates, and
dimethyl carbonate,8 which underscores the potential utility
and wide applicability of these and related catalysts.
solvent and lyoxide. However, two key reports3,4 disclosed that
methanolysis of PC in the presence of sodium hydroxide at
temperatures between 40 and 60 °C is greatly accelerated by
surface-softening cosolvents such as toluene, dioxane, chlori-
nated hydrocarbons, DMF, N-methylpyrrolidone, and THF.
Previously, we have reported methods for the metal ion pro-
moted methanolysis of carboxylate5 and phosphate6 esters
where the catalytic systems comprise complexes of certain
transition metal ions in methanol as well as lanthanide alkox-
ides formed in situ. Well studied in these laboratories are
(RO−):Zn(II):(1,5,9-triazacyclododecane) complexes (1) (which, in
methanol, is herein designated as [12[ane]N3:Zn(II):(OCH3)]+) or
simple catalysts containing lanthanide ions such as La(III), the
latter present in solution as a solvated dimeric form with one to
five associated methoxides ([La2(OCH3)n](6−n)+) depending on the
‘pH’. Generally, the dimer [La2(OCH3)2]4+ (2) is the most active
form exhibiting maximal effect at near neutral conditions of
sspH ϭ 8.8.7 These are highly active for promoting transesterifi-
cation of carboxylate esters, phosphate triesters, and phos-
phonate diesters in light alcohol solvents such as methanol and
ethanol. Attractive features are their ease of preparation in
situ, high turnover numbers, large rate enhancements over the
Understanding the reaction mechanisms for La(III) and transi-
tion metal ion transesterifications for a variety of substrates is an
important step for further development of their utility. The gen-
eral mechanisms for these catalyzed alcoholyses have been rea-
sonably well established with carboxylate5,9 and phosphate/
phosphonate esters,6,10 so our goal in the current study was (1) to
investigate, through kinetic study, the mechanisms for meth-
anolysis of some aryl and alkyl methyl carbonates (3 and 4) under
sspH-controlled conditions and (2) to provide DFT-derived compu-
tational details for the mechanistic pathway of 1-promoted meth-
anolysis of two selected carbonates (3b and 4b). In addition, we
report on some preliminary studies of the La(III)-containing cata-
lyst that show that it might be suitable for further development
into a practical method for the preparative methanolysis of
poly(bisphenol A carbonate).
+
H
O
O
H
N
N
Zn
N
H
Ar
CH3
R
CH3
OCH3
O
O
O
O
4
3
1
a R = CH3
b R = CH(CH3)2
a Ar = pentafluorophenyl
b Ar = 4-nitrophenyl
c Ar = 3-nitrophenyl
d Ar = 4- chlorophenyl
e Ar = 4-methoxyphenyl
f Ar = 3,5-di t-butylphenyl
g Ar = 2,3,5-trimethylphenyl
4+
CH3
La
O
La
O
H3C
2
1
had H NMR and mass spectra consistent with the structure (for
physical data for the new carbonates; see the Supplementary ma-
terial section).
Experimental
Materials
Zn(OTf)2 (98%), La(OTf)3 (99.999%), and dimethyl carbonate
(99%) were used as supplied (Aldrich) as were dimethylformamide,
1,2-dichloroethane, paraoxon (O,O-dimethyl, O-(p-nitrophenyl)
phosphate), sodium methoxide, methanesulfonic acid, trifluoro-
methanesulfonic acid, ethylene glycol, and paraformaldehyde.
Chloroform, tetrahydrofuran, acetone, and dichloromethane were
from Fisher Scientific and were used as received. Anhydrous meth-
anol was from EMD Chemicals (DriSolv) and La(OEt)3 was from
Alfa Aesar. PC (poly(bisphenol A carbonate) was obtained from
Aldrich as opaque beads (3 mm by 2 mm, batch No. MKAA3883).
4-Nitrophenol (99%), 3-nitrophenol (99%), 4-chlorophenol
(99+%), 4-methoxyphenol (99%), 3,5-di-tert-butylphenol, 2,3,5-
trimethylphenol, pyridine (99+%), and methyl chloroformate
(99%) were acquired from Aldrich and used as received for synthe-
ses of the corresponding carbonates. Methyl pentafluorophenyl
carbonate (97%), methyl phenyl carbonate (97%), and methyl iso-
propyl carbonate (97%) were purchased from Alfa Aesar.
The other aryl methyl carbonates were prepared following
slight modifications of literature procedures.11 In the general
proceedure, methyl chloroformate was added slowly to the appro-
priate phenol and pyridine, all in equimolar quantities with meth-
ylene chloride as the solvent, and the reaction mixture stirred at
room temperature for 12–24 h. Aqueous washings (basic, acidic,
neutral) were followed by sodium sulfate drying of the organic
layer and evaporation of the solvent. All of the above carbonates
1,5,9-Triazacyclododecane (12[ane]N3) was prepared according
to a literature description.12
General UV/vis kinetics
Solutions for kinetics were prepared using sodium methoxide
(0.5 mol dm−3 (Aldrich) or tetrabutylammonium hydroxide
(1.0 mol dm−3 (Aldrich) and La(OTf)3 (99.999%), formulating these
as stock solutions in anhydrous methanol at 50 mmol dm−3. Stock
solutions (5 mmol dm−3) of each substrate were prepared in 99.8%
anhydrous acetonitrile.
Buffer solutions (6.6−20 mmol dm−3) were prepared in anhydrous
methanol using triethylamine (sspKa ϭ 10.78),13 N-ethylmorpholine
(sspKa ϭ 8.6), N-methylimidazole (sspKa ϭ 7.6), and 2,6-lutidine
(sspKa ϭ 6.7) partially neutralized by triflic acid. Reaction solutions
were prepared by adding an appropriate amount of the La(OTf)3
stock solution to a UV cuvette containing buffer so that the total
volume was 2.5 mL. The reactions were initiated by the addition of
the above substrate stock solutions.
The kinetics of the 2-promoted methanolysis of methyl aryl
carbonates 3a−3g were followed by monitoring the rate of change
in absorbance of 5 × 10−5 to 2 × 10−4 mol dm−3 solutions of each at
274.9, 340, 400, 284, 292, 272, and 272 nm, respectively, using a
Cary 100 UV/visible spectrophotometer thermostated at 25 °C. The
absorbance versus time data were fit to a standard first-order
exponential equation to obtain the first-order rate constants (kobs
)
Published by NRC Research Press

Neverov et al.
1141
for the methanolysis of these substrates. The rates of reaction
were monitored in duplicate at four to seven different catalyst
concentrations from 0.4 to 1.4 mmol dm−3. Second-order rate con-
stants ko2bs) for thecatalyzed reactions were obtained from the lin-
ear portions of the plots of kobs versus [La(OTf)3].
the two latter cases, the residues dissolved completely in CD3OD
and gave 1H NMR results indicating 97% and 94% reaction comple-
tion. Under the reaction conditions described in the Methanolysis
of PC section above, homogeneous final solutions in deuterated
methanol indicated that the PC had substantially degraded to the
point that all of the crude product was soluble in CD3OD, giving a
high conversion of PC to products. The most upfield aromatic
hydrogen peak (␦ ≈ 6.69 ppm) arises from bisphenol A, while the
most downfield peak (␦ ≈ 7.2 ppm) is atttributable to unreacted
polymer/oligomers. The ratio of the integration of the bisphenol A
6.69 ppm aromatic peak to the sum of the bisphenol A and oli-
gomer peaks at 7.2 ppm was taken as a percentage of reaction
completion.
For 1-catalyzed reactions, stock solutions of ligand (12[ane]N3),
Zn(OTf)2, and sodium methoxide were prepared in 99.8% anhy-
drous methanol. A working stock solution was prepared by add-
ing each component in equimolar amounts. This was then diluted
in situ with methanol such that the final volume was 2.5 mL in
each cuvette. Stock solutions of carbonates 3a−3g were added to
initiate reaction and the kinetics of methanolysis were followed
by monitoring the rate of change in absorbance of 2.0 × 10−5 to
1.0 × 10−4 mol dm−3 of 3a−3g at 258, 320, 330, 284, 291, 272, and
275.9 nm, respectively, as above. The rates of reaction were mon-
itored in duplicate at six different catalyst concentrations from
0.175 to 1.75 mmol dm−3. Second-order rate constants ko2bs for the
catalyzed reaction were obtained from the linear plots of kobs
versus [1].
Computational details
Geometry optimizations and free energy calculations of ground- and
transition-state structures were completed at the B3LYP level of
theory with the IEFPCM14 methanol solvent model. The 6-31G(d,p)
basis set was used for carbon and hydrogen, while 6-31++G(d,p)
was used for oxygen and nitrogen. The LANL2DZ15 pseudopoten-
tial was used for Zn(II). Frequency calculations were conducted at
this level of theory to characterize stationary points as intermedi-
ates or transition states and as a basis for free energy determina-
tions at 298 K. All calculations were conducted using Gaussian
09.16
1H NMR kinetics
The kinetics of methanolysis of methyl alkyl carbonates 4a and
4b were followed by 1H NMR spectroscopy (Bruker Avance
500 MHz). Stock solutions of Zn(OTf)2, La(OTf)3, and sodium me-
thoxide were prepared in 99.8% deuterated methanol (Cambridge
Isotope Laboratories, Inc.). For the 2-catalyzed reactions, the cata-
s
Results and discussion
lytic complex was self-buffered at pH ϭ 8.85 in the presence of
0.76 equiv. of methoxide/La3+. For sthe 1-catalyzed reactions, the
catalytic complex for each kinetic run was formed in situ through
the addition of the Zn(OTf)2, 12[ane]N3, and sodium methoxide in
equimolar amounts. The rate of methanolysis of dimethyl carbon-
ate (4a) was followed by observing the rate of disappearance of the
starting materials (CH3-O-C(=O)-OCH3, ␦ = 3.75 ppm) referenced to
an acetonitrile standard (CH3CN, ␦ = 2.05 ppm) in CD3OD. The
methanolysis of methyl isopropyl carbonate (4b) was followed by
observing the rate of appearance of isopropanol ((CH3)2CHOH, ␦ =
1.17 ppm), its concentration being referenced to the acetonitrile
standard. The relative integration intensity versus time data were
fit to an initial rate expression to obtain the first-order rate con-
stants kobs. The gradients for the kobs versus [1] and kobs versus [2]
Kinetics
In Fig. 1 is an example of the plot of the kobs for the methanolysis
of p-nitrophenyl methyl carbonate (3b) as a function of [La(OTf)3]
at sspH ϭ 8.85. As is observed for the La(III)-catalyzed methanolysis
of carboxylate and phosphate esters,5,6,17 the plots of kobs versus
[La(OTf)3] at all sspH values have a significant x-intercept due
to the formation of active methoxide-bridged La(III) dimers,
[La2(OCH3)n](6−n)+, where n = 1 to 5. The plots are linear above
[La(III) total = 4 × 10−4 mol dm−3 where dimer formation is complete
]
in CH3OH. The gradient of the linear portion of these plots is
taken as the second-order rate constant = ko2bs for the La(III)
catalyzed methanolysis of 3 at a given sspH.
-
In Fig. 2 is a plot of log ko2bs determined for cleavage of 3b pro-
plots were calculated to give the overall k2obs
.
s
moted by [La(III) dimers] at various pH values; these kinetic data
s
are overlayed with the speciation diagram of [La2(OCH3)n](6−n)+ in
methanol that was previously determined by potentiometric titra-
tion.17 The plot has a skewed bell-shape profile maximizing at
sspHϳ8.8, suggesting that the the most active species is the dimer
[La(OCH3)2]4+, although [La2(OCH3)3]3+ also has good activity that is
lessened in accordance with its reduced concentration in solu-
tion. Similar experiments were performed for the methanolysis of
3 and 4 promoted by 1 at concentrations between 0.175 and
1.75 mmol dm−3; the gradients of the kobs versus [1] plots gave the
ko2bs values that are also in Table 1. A representative plot with 3b is
shown in Fig. 3.
Methanolysis of PC
Microwave heating was used to heat samples formulated from
200 mg of PC beads crushed to 1–2 mm in size and 2 mL of a
chloroform−methanol cosolvent mixture containing La(OTf)3 and
NaOMe (5 and 7.5 mmol dm−3, respectively). The weighed polymer
samples were placed in 2–5 mL of Biotage microwave vials (rated
to withstand a maximum pressure of 20 bar or 2000 kPa) equipped
with a magnetic stir bar followed by the addition of cosolvents
and stock solutions of the catalyst components. The microwave
vial was sealed and placed into a Biotage Initiator Robot Eight
microwave oven equipped with a proprietary IR temperature sen-
sor in the microwave cavity. The reaction was set to run for 10 min
at 100 °C, this temperature being attained in 30–60 s after the start
of the experiment; 1 min after cessation of heating, the tempera-
ture was 45 °C, at which time the vial was opened and the contents
analyzed. Three different chloroform−methanol cosolvent com-
positions were subjected to analysis: 50:50, 60:40, and 70:30.
Brønsted plots
To probe the effect varying the carbonates’ leaving groups (LG),
we determined the ko2bs values for the 1-catalyzed methanolysis of
3a−3g and 4a and 4b. These constants are given in Table 1 and a
Brønsted plot using all data (log ko2bs versus sspKHa OAr,HOR) is given in
Fig. 4. The black solid and broken lines through the experimental
data points for the 1-promoted methanolyses are fits to equations
derived from models including all of the data (solid, from treat-
ment of data with equations in ref. 5c), as well as two dashed lines
separately treating the aryl carbonates and alkyl carbonates,
which appear to fall on two different plots; these will be discussed
1H NMR product analysis
The contents of the microwave vial were poured into a small
round bottom flask and the solvent removed under vacuum. The
crude product was then redissolved in deuterated methanol and
1
analyzed using H NMR spectroscopy. Three different chloroform−
later. Also given in the figure is a downward breaking red line that
methanol cosolvet compositions were subjected to analysis: 50:
50, 60:40, and 70:30. In the former case, the residue would not
dissolve completely in CD3OD, indicating incomplete reaction. In
s
describes the Brønsted plot for the ko2bs versus pKaHOAr data for
s
methoxide-promoted methanolysis of 3 determined by Schowen
Published by NRC Research Press

1142
Can. J. Chem. Vol. 91, 2013
Fig. 1. Plot of the first-order rate constant for the methanolysis of
4-nitrophenyl methyl carbonate (3b) catalyzed by added La(OTf)3 at
sspH = 8.85, T = 25 °C.
Fig. 3. Plot of the first-order rate constants for the methanolysis of
4-nitrophenyl methyl carbonate (3b) as a function of concentrate of
1 (formed in situ by addition of equimolar amounts of Zn(II)
,
methoxide, and ligand, where sspH varies from 9.1 to 10.0 with
0.006
0.004
0.002
0.000
increasing concentration) at T = 25 °C.
0.02
0.01
0.00
[La(OTf)3] (mol L–1)
Fig. 2. Plots of the speciation for La(III) dimers with one to five
-3
[12[ane]N3:Zn(II):(OCH3)]+ (mol dm )
attached methoxides as a function of sspH as computed for
[La(III) total = 2 × 10−3 mol dm−3 from fits of potentiometric titration
]
Fig. 4. Brønsted plot of the log ko2bs versus sspKaHOAr,HOR constants for
the 1-catalyzed methanolysis of carbonates 3a−3g and 4a and 4b
determined at T = 25 °C. The solid black line through all points was
computed from an NLLSQ fit of all of the data as described in ref. 5c,
yielding two gradients of –0.15 and –1.02; broken lines are the aryl
and alkyl carbonates treated separately (3a−3g gradient = –0.16,
4a and 4b gradient = –1.0). The red line is the Brønsted plot for
methoxide-promoted methanolysis of aryl methyl carbonates was
determined by Schowen et al.11a, corrected to T = 25 °C (color in the
online version only) (see Supplementary material section).
data.17 Superimposed on the plots as circles are the ko2bs data for the
La(III)-catalyzed methanolysis of p-nitrophenyl methyl carbonate
(3b) determined at various sspH values, T = 25 °C.
La2(O C H3
)
2
La2(O C H3
)
4
4
3
2
1
0
0 .0 0 0 8
0 .0 0 0 6
0 .0 0 0 4
0 .0 0 0 2
0 .0 0 0 0
2
La2(O C H3
)
5
La2(O C H3
)
1
La2(O C H3
)
3
1
5
6
7
8
9
1 0
1 1
1 2
0
s
s
p H
-1
-2
-3
-4
Table 1. Second-order rate constants (ko2bs) for the methanolysis of
carbonates 3a−3g and 4a and 4b promoted by 1 and 2 in methanol at
T = 25 °C.
ko2bs for 1
(dm3 mol−1 s−1
ko2bs for 2
(dm3 mol−1 s−1
)
Carbonate
sspHHa OAr,HOR
)
3a
3b
3c
3d
3e
3f
3g
4a
4b
8.84
11.18
12.41
13.59
14.77
14.80
15.30
18.13
19.79
15.83
9.42
4.92
3.38
2.86
1.87
1.09
0.05
0.001
5.92
4.80
4.56
3.64
2.39
0.89
0.63
0.08
0.002
8.0
10.5
13.0
15.5
18.0
sspKa (leaving group)
2-promoted methanolyses are fits to equations derived from
models5c including all the data (solid), as well as two broken lines
separately treating the aryl carbonates and alkyl carbonates, which
appear to fall on two different plots; these will be discussed later. The
downward breaking red line in Fig. 5 describes the Brønsted plot for
Note: ko2bs determined at sspH = 8.85 for reactions promoted by 2 and within a
sspH range of 10.4−11.5 for 1 formed in situ from simultaneous addition of 1:1:1
Zn(OTf)2, 12[ane]N3, and methoxide.
Schowen’s11a k2obs versus pKaHOAr data for CH3O−-promoted meth-
s
s
anolysis of 3 (see the Supplementary material section.
Analysis of the Brønsted plots
The solid black lines through the data in Figs. 4 and 5 exhibit
apparent downward breakpoints at pKa ϳ16.6 and 15.2 for cata-
et al.11a (for a tabulation of these data, see the Supplementary
material section).
s
s
In Fig. 5 is a similar plot for the methanolysis of 3 and 4 pro-
lysts 1 and 2, respectively. This behavior was previously observed
for metal ion catalyzed methanolysis of aryl and alkyl acetates5c
and rationalized at that time as being a result of a change in
s
s
moted by 2 at pH 8.85 (corresponding to the pH for maximum
s
activity attributed to catalysis by [La(OCH3)2]4+)s(Fig. 2). These con-
stants are given in Table 1 and a Brønsted plot using all data
rate-limiting step from nucleophilic attack on the C=O unit of
s
s
(log k2obs versus pKaHOAr,HOR) is given in Fig. 5. The black solid and
substrates with “good” LG (having lower pKHa OAr,HOR values for
s
s
broken lines through the experimental data points for the
their corresponding parent phenols/alcohols) to the departure of
Published by NRC Research Press

Neverov et al.
1143
Fig. 5. Brønsted plot of the log ko2bs versus sspKaHOAr,HOR for the
[La(OCH3)2]4+-catalyzed methanolysis of carbonates 3a−3g, and 4a
and b determined at sspH = 8.85, T = 25 °C. The solid black line is
the fit of all data as described in ref. 5c, yielding gradients of –0.03
and –0.68; broken lines are the aryl and alkyl carbonates treated
separately (3a−3g gradients = –0.26 and –1.0, 4a and 4b gradient =
–0.97). The red line is the Brønsted plot for methoxide-promoted
methanolysis of aryl methyl carbonates determined by Schowen
et al.11a (color in the online version only) (see Supplementary
material section).
“enforced concerted process”23,24 has been recently offered to
explain the downward break noted for the −OH-promoted hy-
drolyses of a series of aryl-substituted benzene sulphonates.25
Here, it is suggested that substrates with LG having a pKa >8.5 pro-
ceed via a stepwise process with a pentacoordiante intermediate,
while those having a pKa <8.5 proceed through a pentacoordinate
transition structure but the intermediate produced is too unsta-
ble to exist, so the reaction becomes concerted. A recent study of
the –OH-catalyzed hydrolysis of 10 phenyl aryl carbonates (pKHa OAr
in water from 5.42 to 9.95) concludes that the reactions are con-
certed or enforced concerted, with little cleavage of the departing
group in the rate-limiting transition state;26 this is also sug-
gested27 to be the case for for the ethoxide-promoted ethanolysis
of some phenyl aryl carbonates.
2
1
Breaks in linear free energy relationships are reported for the
methoxide-promoted cleavages of carbonates and acetates with
aryloxy LG.11 Given as the red solid lines in Figs. 4 and 5 is the
0
Brønsted plot of the extant11a k2ϪOCH data for aryl carbonates,
3
which exhibit a downward break at sspKa ϳ14.1, far lower than the
expected quasi-symmetrical point of the conjugate acid of the
attacking methoxide (sspKa methanol 18.13). Schowen11a noted that
methoxide attacks on aryl acetates and carbonates are two-step
reactions with a rate-limiting addition that becomes progressively
less sensitive to the substituent as its electron-withdrawing power
increases. Such curvature is not related to a change in rate-
limiting step but rather to substrate variations that introduce
curvature into a linear free energy relationship, which compares
the ⌬G‡ of a kinetic process with the equilibrium ⌬Go of an acid
dissociation process. The nonlinear behavior results from greater
-1
-2
-3
8.0
10.5
13.0
15.5
18.0
s
spKa (leaving group)
importance of the resonance interaction in the equilibrium ion-
the LG in the case of substrates with the “poor” LG’s (higher
ꢀ
s
sspKHa OR values).
ization processes on which the pKa values are based than in the
s
kinetic processes where there is less charge development on the
aryloxy group during the nucleophilic attack step.
The solid black curved lines in the two figures can each be
mathematically deconvoluted5c to provide gradients of –0.15 and
s
s
We have recently addressed the question of breaks in the Brøn-
sted plots for the methanolysis of aryl and alkyl acetates pro-
moted by methoxide and complex 1 with DFT computations and
found that there are separate relationships for the aryl and alkyl
acetates9 thatarerelevanttothemethoxide-and1-promotedmeth-
anolyses of 3a−3g and 4a and 4b. The reactions catalyzed by 1
involve two steps with a Zn(II)-coordinated tetrahedral intermedi-
ate with all aryloxy acetates except those with the best LG such as
2,4-dinitrophenoxy, where the reaction is concerted or enforced
concerted. With a slightly poorer p-nitrophenoxy LG, the rate-
limiting step is attack, and the departure of the LG is fast, occur-
ring without metal ion assistance. But departure of even poorer
aryloxy LG does require assistance of the metal ion, although the
reactions of all substrates containing these LG still have rate-
limiting attack. With alkoxy acetates, the transesterifications
have two steps with metal ion assistance of LG departure, but the
latter is rate-limiting only for examples where the LG is worse
than methoxide. This work9 and our results on the methoxide-
and hydroxide-promoted cleavages of aryl and alkyl phosphates
and phosphorothioate triesters10 suggest that downward breaks
in their corresponding Brønsted plots arise from the incorpora-
tion of members (aryloxides and alkoxides) that adhere to sepa-
rate lines. In the case of carbonates, the available evidence is
consistent with a similar conclusion, but there are only two ex-
perimental examples for the metal ion catalyzed processes, which
weakens the argument.
–1.02 with a breakpoint at pKa ϳ16.6 for the plot in Fig. 4 and
gradients of –0.03 and = –0.68 with a breakpoint at sspKa ϳ15.2 in
Fig. 5. For both plots, the numerically small gradient in the region
corresponding to the methanolysis of the aryloxy carbonates is ratio-
nalized as resulting from a cancellation of the electronic effects on
substrate binding to the catalysts followed by a rate-limiting nucleo-
philic attack within the catalyst−substrate complex. Substrates with
electron-withdrawing OAr groups will bind to the catalyst more
poorly but suffer nucleophilic attack more rapidly, and vice versa.
This is similar to the acid-catalyzed hydrolysis of esters where the
relative insensitivity of the catalytic rate constant to substitution on
the ester was rationalized18 as a counterbalancing of the alkoxy/
aryloxy substituent’s electronic effect on equilibrium protonation
and the subsequent nucleophilic addition of water to C=O−H+.
In neither of the Figs. 4 and 5 plots does the breakpoint come at
s
a pKHa OAr,HOR corresponding to a quasi-symmetrical point where
s
there is a transition between rate-limiting attack to form a tetra-
hedral intermediate and rate-limiting breakdown of the interme-
diate. That point should occur where the incoming nucleophile
s
and outgoing LG have identical pKHa OAr,HOR values. Although it is
s
not clear what these sspKa values should be in a metal ion catalyzed
process, one can suggest about 9.1 if the reaction is considered to
involve a fully metal ion coordinated methoxide as in 1 and 8.1 if
it involves [La2(OCH3)2]4+ 19
.
Alternatively, if the nucleophile is con-
sidered to be methoxide, the quasi-symmetrical point should be at
18.13, the sspKaHOCH
.
3
Based on the above, we individually evaluate the experimental
data for aryl- and alkyl-containing carbonates in Figs. 4 and 5
(broken lines). The data in Fig. 5 for cleavage of 3 promoted by 2
We have considered other explanations for breaks in the
Brønsted plots that do not result from changes in rate-limiting
step20,21,22 for the behavior exhibited here. There are interesting
cases where members of a set of substrates follow the same gen-
eral two-step mechanism with rate-limiting attack, but small
changes in the substrate may reduce the stability of the addition
intermediate to the point it becomes too unstable to exist, imme-
diately decomposing without barrier to product. This sort of
s
exhibit a small break at pKa ϳ14.2, similar to what is observed
with the methoxide cleavsage of series 311a and are displaced from
the data obtained for 4, which form a (two-point) linear free en-
ergy relationship unrelated to that obtained with 3. The reasons
for this are not clear, as we were not able to probe this more
Published by NRC Research Press

1144
Can. J. Chem. Vol. 91, 2013
Fig. 6. DFT-calculated low-energy pathway for the 1-catalyzed methanolysis of carbonates 3b (blue) and 4b (red); free-energy values are
reported in kJ mol−1 and at 298 K and the diagram is to scale (color in the online version only).
complicated system using DFT calculations.28 In the case of the
1-promoted methanolysis (Fig. 4), substrates 3 are fit satisfactorily
with a linear plot, perhaps with the slightest hint of a break at the
highest sspKHa OAr values (although the linear relationship fits just as
well) and a two-point line for substrates 4 with a steeper gradient
of –1.0. The steeper gradient in the two-point line for 4 could be
indicative of a greater charge acruing to the departing alkoxide
group but may also be affected by a greater steric demand of the
isopropyl group in 4b. The experimental data do not indicate
whether the metal ion catalyzed reactions are concerted or step-
wise or, if the latter is the case, whether the metal ion assists in
both the attack step and LG departure. Further information is
provided by the computational study given below.
containing a good LG, and a much larger barrier for reaction of 4b
(197.1 kJ mol−1), which contains the poor LG. To engender metal
ion assisted LG departure, a rearrangement must occur in which
the nucleophilic methoxide dissociates from the metal ion and
the LG is placed in a position where its departure can be facilitated
by Zn(II). A single transition state for this process could not be
found for either 3b or 4b; however, potential energy scans indi-
cate that there is only a small (less than ϳ4 kJ mol−1) barrier for
this process. The intermediate prior to metal-assisted departure
(INTLG) involves the LG oxygen not bound to the metal but situ-
ated 3.22 Å (3b) or 3.05 Å (4b) from it. As the C–O(LG) bond length
increases, the Zn(II)−O(LG) distance decreases such that in the tran-
sition state (TSLG), it is 1.70 Å (3b) or 1.80 Å (4b). The free energy
barrier for metal-assisted LG departure is lower and not rate-
limiting for 3b, but it is much higher for the alkyl substrate and
this transition state represents the rate-determining step for the
methanolysis of 4b. The nucleophilic attack step, TSNu, is limiting
for the substrate with the good LG, 3b.
Calculations of the reaction free energies for 1 plus 3g (not
shown) having the poorest aryloxy LG indicated the rate-limiting
step for this process was nucleophilic addition, with the depar-
ture of the LG being metal ion assisted, with a 5–7 kJ mol−1 lower
energy than the attack step. In view of this finding, 3g is consid-
ered to lie on the broken line linear relationship in Fig. 4. The
apparent break in the Brønsted relationship in Fig. 5 with the
La(III)-based catalyst, suggesting that 3h and 3g might be experi-
encing a change in rate-limiting step from attack to LG departure,
still remains a possibility, but we could not probe this using DFT
calculations.29
Density functional theory study
The mechanisms of the 1-catalyzed methanolysis of substrates
3b and 4b were modeled using DFT calculations. The calculated
initial binding processes are quite endergonic, although the com-
puted free energy of binding can be considered an overestimation
due to an exaggeration of translational and rotational entropic
contributions29 and solvation of the ionic species, which is differ-
ent in the presence of bound substrate. The free energy values for
the various reaction intermediates and transition states are re-
ported graphically in Fig. 6 relative to the computed ground states
for 1 plus 3b and 1 plus 4b. The resulting complexes, 1:3b and
1:4b, are trigonal bipyramidal at Zn(II) with methoxide and two
ligand nitrogens occupying the equatorial positions and the car-
bonate substrate and remaining ligand nitrogen occupying the
axial positions. Nucleophilic attack occurs through an attack of
the Zn(II)-coordinated methoxide oxygen on the carbonate carbon
(TSNu) and is associated with a free energy barrier of 120.5 kJ mol−1
for 3b and 121.3 kJ mol−1 for 4b. The nucleophilic attack leads to a
tetrahedral intermediate INTNu bound to the metal ion through
methoxide and the oxyanion. LG departure was modeled in two
ways, with and without metal ion assistance. The free energy
barrier for LG departure without metal ion assistance was mod-
eled as a lengthening of the carbonate carbon–LG oxygen bond of
INTNu and is associated with a small barrier (102.6 kJ mol−1) for 3b,
Catalytic methanolysis of poly(bisphenol A carbonate) (PC)
The initial catalytic system used for the depolymerisation of PC
was formulated in methanol adding La(OTf)3 and NaOMe (5 and
7.5 mmol dm−3). Three beads of commercial PC (ϳ53 mg) were
heated to 100 °C in 2 mL of catalyst solution for 10 min with no
observable changes to the beads. To determine whether the prob-
lem was due to poor catalytic activity or the catalyst transport into
the solid PC, the beads were ground and sieved to a particle size of
Published by NRC Research Press

Neverov et al.
1145
Fig. 7. Plot of pseudo-first-order rate constants for the methanolysis
of 1e catalyzed by 1 mmol dm−3 La(OTf)3 in the presence of
1.5 mmol dm−3 NaOMe as a function of chloroform−methanol
solvent composition, T = 25 °C. The line through the data is
presented as an aid to visualization but does not represent any fit.
action of 1 with 3 and 4 involves the equilibrium formation of a
transient complex between the metal catalyst (as its methoxide
form) and substrate within which a two-step reaction leading to
a set of intermediates with different lifetimes ensues. For aryl
ꢀ
carbonates with LG whose conjugate acids have sspKaHOR val-
ues <16, the rate-limiting step is metal ion assisted delivery of a
coordinated methoxide to the C=O of the transiently bound car-
bonate to form a tetrahedral intermediate with bidentate coord-
iantion to the metal ion. Breakdown of the intermediate is rapid
and occurs with metal ion assistance of LG departure for the
2
1
0
ꢀ
s
p-nitrophenoxy derivative and those with higher pKHa OR values.
s
HORꢀ
Carbonates with −OAr LG having sspKa values less than ϳ10 such
as 3a may decompose without metal ion assistance of LG depar-
ture.9 For methyl alkyl carbonates with LG whose conjugate acids
ꢀ
s
s
have pKHa OR values >18.13 (the pKa of methanol), metal assisted
s
s
departure of the LG is the rate-limiting step. Although we do not
have reliable computational data28 for the lanthanum catalytic
sytem 2 due to the difficulties in dealing with the dimeric, highly
charged catalyst, the general mechanism should be similar to that
0
25
50
75
100
%MeOH (v/v)
employed by 1, but there is some question whether the La(III)
-
promoted methanolysis of aryl carbonates does undergo a change
in rate-limiting step from metal ion assisted nucleophilic attack to
metal ion assisted departure of the LG.
250 ␮m and the experiment was repeated using this higher sur-
face area material. Under the same experimental conditions and
reaction time, it was observed that approximately half of the poly-
mer had disappeared; 1H NMR analysis of the solution confirmed
the presence of bisphenol A, indicating that the polymer was
partially depolymerized. This suggests that the limiting step in
the depolymerization of PC under these conditions is the mass
transfer of catalyst and solvent into the polymer structure, so
further effort was expended to optimize this process. Following
It is interesting that these two systems have similar activities
for all of the substrates; the maximum discrepancy is threefold,
with 2 being slightly more active than 1 for methanolysis of sub-
strates 3 having strong electron-withdrawing groups and 1 being
twofold more reactive than 1 for the cleavage of 4b. It is also
interesting that, in terms of second-order rate constants for their
catalytic reaction, both 1 and 2 have reactivities toward carbon-
ates similar to that of methoxide. For the three carbonates where
comparison can be made11a, the kϪ2 OMe rate constants for the
methoxide reactions of 3b, 3e, and 4a are 16.2, 0.47, and
0.0047 dm3 mol−1 s−1, respectively. These values are similar to the
metal ion catalyzed rate constants in Table 1 for the substrates
with good LG, but the metal-containing catalysts become progres-
sively less reactive than methoxide as the carbonate LG becomes
worse. The similarity of rate constant for the methoxide reaction
and 1- and 2-promoted methanolysis of carbon esters has been
seen before.5c Thus, the catalytic benefit for metal-promoted car-
bonate and ester methanolysis is that maximum activity is
reached at sspH ϳ9, close to neutrality under catalytic conditions,
while the methoxide-promoted reactions must be conducted at
high base concentration under stoichimetric conditions.
The systems we report here may be effective for the larger scale
decomposition of polycarbonates such as PC, although this reac-
tion requires more severe conditions than where the mechanistic
study was undertaken to circumvent the problem of penetration
of the solution and catalyst into the polymer matrix. In one exam-
ple, 1000 mg of PC in 2 mL of a 60:40 chloroform−methanol solu-
tion containing La(OTf)3:NaOMe (5:7.5 mmol L−1) at 100 °C for
30 min gave an 84% yield of bisphenol A, corresponding
to >300 turnovers per molecule of catalyst. Further work may lead
to preferred solvent systems having better catalyst and product
recovery characteristics that enable solvent recycling.
the reports for the methoxide-promoted methanolysis of PC3,4
,
we investigated the use of a cosolvent to increase the homogene-
ity of the reaction mixture and swell the residual solid polymer.
Screening experiments were performed where approximantely
35 mg of PC beads was placed into 1 mL of such solvents
as dimethylformamide, 1,2-dichloroethane, chloroform, tetrahy-
drofuran, and dichloromethane at room temperature. While each
of the solvents visibly showed swelling and softening of the poly-
mer after 2 h, the beads placed in the chloroform were nearly
completely dissolved, leaving a clear, yet viscous, solution. Subse-
quent experiments were performed to determine the effect on
the rate of La(III)-promoted methanolysis of 1e as a function of
methanol−chloroform mixtures (see Fig. 7). As the reaction rate
is reduced with more chloroform, we chose 60:40 chloroform−
methanol as a solvent mixture that represents a balance between
the rate of reaction, reduced by a factor of ϳ9 relative to that in pure
methanol at 25 °C, and dissolution of polymer in the solvent which
essential for reaction.
A typical experiment comprised 200 mg of PC beads, crushed to
1–2 mm in size, and 2 mL of the 60:40 chloroform−methanol cosolvent
mixture containing La(OTf)3 and NaOMe (5 and 7.5 mmol dm−3, respec-
tively), which was then heated to 100 °C for 10 min in a microwave
reactor. Figure. S1 (see Supplementary material section) is an ex-
1
ample of the H NMR spectrum of a typical reaction mixture of
this composition after depolymerization having 97% bisphenol A
and 3% incompletely reacted PC or oligomers thereof. The effect of
increasing PC loading was tested in this solvent system and the high-
est amount of PC acommodated in an experiment was 1000 mg in
2 mL of 60:40 chloroform−methanol catalyzed by La(OTf)3:NaOMe
(5:7.5 mmol dm−3). After heating at 100 °C for 30 min, the reaction
reached approximately 84% completion as observed by 1H NMR. This
corresponds to 315 turnovers per equivalent of the catalyst.
Supplementary material
Supplementary material is available with the article through
the journal Web site at http://nrcresearchpress.com/doi/suppl/
10.1139/cjc-2013-0270. Electronic Supplementary information (ESI)
available: tables of Cartesian coordinates for DFT-optimized struc-
tures, physical data for characterization of new carbonates 3f and
3g, 1H NMR spectra for various decompositions of poly(bisphenol
A) carbonate, and table of second-order rate constants corrected
to 25 °C from ref. 11a and phenol sspKa values (adjusted to methanol
Conclusions
The two metal ion containing catalytic systems efficiently pro-
mote the transesterification of a series of methyl aryl (3) and two
methyl alkyl (4) carbonates. The computed mechanism of the re-
Published by NRC Research Press

1146
Can. J. Chem. Vol. 91, 2013
1021/cr9904009; (b) Tomasi, J.; Mennuccia, B.; Cances, E. THEOCHEM 1999,
464, 211. doi:10.1016/S0166-1280(98)00553-3.
from aqueous values according to the formula given in ref. 5c used
to prepare the Brønsted plots shown in Figs. 4 and 5.
(15) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270. doi:10.1063/1.448799;
(b) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284. doi:10.1063/1.448800.
(16) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.;
Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.;
Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.;
Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Dapprich, S.; Daniels, A. D.; Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.;
Cioslowski, J.; Fox, D. J. Gaussian 09; Gaussian, Inc.: Wallingford, CT, 2009.
(17) Tsang, J. S.; Neverov, A. A.; Brown, R. S. J. Am. Chem. Soc. 2003, 125, 7602.
doi:10.1021/ja034979a.
Acknowledgements
The authors gratefully acknowledge the financial contributions of
the Natural Sciences and Engineering Research Council of Canada
and Queen’s University for this research. In addition, D.S. acknowl-
edges the NSERC Undergraduate Summer Research Awards program
for support of his participation in this project and L.C. and S.G. ac-
knowledge the support of the Summer Work Experience Program at
Queen’s University.
References
(1) Bozzano, G.; Dente, M.; De Rosso, R. In Material Recycling-Trends and
Perspectives; Achilias, D., Ed.; ISBN 978-953-51-0327-1; InTech: Croatia, 2012;
pp. 115−132, and references therein.
(2) Pinero, R.; Garcia, J.; Cocero, M. J. Green Chem. 2005, 7, 380. doi:10.1039/
b500461f.
(3) Hu, L. C.; Oka, A.; Yamada, E. Polymer 1998, 39, 3841. doi:10.1016/S0032-
3861(97)10298-1.
(4) Liu, F. S.; Li, Z.; Yu, S. T.; Cui, X.; Xie, C.- X.; Ge, X. P. J. Polym. Environ. 2009, 17,
208. doi:10.1007/s10924-009-0140-0.
(5) (a) Neverov, A. A.; McDonald, T.; Gibson, G.; Brown, R. S. Can. J. Chem. 2001,
79 (11), 1704. doi:10.1139/v01-149; (b) Neverov, A. A.; Gibson, G.;
Brown, R. S. Inorg. Chem. 2003, 42, 228. doi:10.1021/ic0203769;
(c) Neverov, A. A.; Sunderland, N. E.; Brown, R. S. Org. Biomol. Chem. 2005, 3,
65. doi:10.1039/b414763d.
(6) (a) Lu, Z.-L.; Neverov, A. A.; Brown, R. S. Org. Biomol. Chem. 2005, 3, 3379.
doi:10.1039/b508917d; (b) Neverov, A. A.; Brown, R. S. Org. Biomol. Chem.
2004, 2, 2245. doi:10.1039/b404740k; (c) Brown, R. S.; Lu, Z.-L.; Liu, C. T.;
Tsang, W. Y.; Edwards, D. R.; Neverov, A. A. J. Phys. Org. Chem. 2010, 23, 1.
doi:10.1002/poc.1584; (d) Brown, R. S.; Neverov, A. A. Adv. Phys. Org. Chem.
2008, 42, 271. doi:10.1016/S0065-3160(07)42006-8; (e) Liu, T.; Neverov, A. A.;
Tsang, J. S. W.; Brown, R. S. Org. Biomol. Chem. 2005, 3, 1525. doi:10.1039/
b502569a.
(18) (a) Taft, R. W., Jr. J. Am. Chem. Soc. 1952, 74, 2729. doi:10.1021/ja01131a010;
(b) Taft, R. W., Jr. J. Am. Chem. Soc. 1953, 75, 4231. doi:10.1021/ja01113a027.
These estimates are made on the basis of the measured sspH at half neutral-
(19)
ization of the [12andN3]:Zn(II)(HOCH3)]2+ ¢¡ [12aneN3:Zn(II)−(OCH3)]+ + H+
system5c and via potentiometric titration of La3+ in methanol; Gibson, G.;
Neverov, A. A.; Brown, R. S. Can. J. Chem. 2003, 81 (6), 495. doi:10.1139/v03-
035.
(20) (a) Chaw, Z. S.; Fischer, A.; Happer, D. A. R. J. Chem. Soc. (B) 1971, 1818.
doi:10.1039/J29710001818; (b) Kirsch, J.; Clewell, A.; Simon, A. J. Org. Chem.
1968, 33, 127. doi:10.1021/jo01265a023; (c) Humffray, A. A.; Ryan, J. J. J. Chem.
Soc. (B) 1967, 468. doi:10.1039/J29670000468.
(21) (a) Ryan, J. J.; Humffray, A. A. J. Chem. Soc. (B) 1966, 842. doi:10.1039/
J29660000842; (b) Bruice, T. C.; Mayahi, M. J. Am. Chem. Soc. 1960, 82, 3067.
doi:10.1021/ja01497a023; (c) Kirsch, J. F.; Jencks, W. P. J. Am. Chem. Soc. 1964,
86, 837. doi:10.1021/ja01059a019.
(22) (a) Guthrie., J. P. J. Am. Chem. Soc. 1991, 113, 3941. doi:10.1021/ja00010a040.;
(b) Stefanidis, D.; Jencks, W. P. J. Am. Chem. Soc. 1993, 115, 6045. doi:10.1021/
ja00067a020.
(7) We use the nomenclature recommended by the IUPAC, Compendium of
Analytical Nomenclature. Definitive Rules 1997; 3rd ed.; Blackwell: Oxford, UK,
1998, for the designation of pH in non-aqueous solvents. The pH meter
reading for an aqueous solution determined with an electrode calibrated
with aqueous buffers is designated as wwpH; if the electrode is calibrated in
water and the ‘pH’ of the neat buffered methanol solution then measured,
the term swpH is used; if the electrode is calibrated in the same solvent
and the ‘pH’ reading is made, then the term sspH is used. In methanol,
(23) (a) Jencks, W. P. Acc. Chem. Res. 1980, 13, 161. doi:10.1021/ar50150a001;
(b) Jencks, W. P. Chem. Soc. Rev. 1981, 10, 345. doi:10.1039/cs9811000345;
(c) Williams, A. Acc. Chem. Res. 1989, 22, 387. doi:10.1021/ar00167a003.
(24) IUPAC. Compendium of Chemical Terminology, 2nd ed. (the “Gold Book”); Compiled
by A. D. McNaught and A. Wilkinson; Blackwell Scientific Publications:
Oxford, 1997; XML on-line corrected version: http://goldbook.iupac.org
2006, created by M. Nic, J. Jirat, B. Kosata; updates compiled by A. Jenkins.
ISBN 0-9678550-9-8. doi:10.1351/goldbook.
(25) Babtie, A. C.; Lima, M. F.; Kirby, A. J.; Hollfelder, F. Org. Biomol. Chem. 2012, 10,
8095. doi:10.1039/c2ob25699a.
(26) Kim, S.-I.; Hwang, S.-J.; Jung, E.-M.; Um, I.-H. Bull. Korean Chem. Soc. 2010, 31,
2015. doi:10.5012/bkcs.2010.31.7.2015.
(27) Um, I.-H.; Seo, J. Y.; Kang, J. S.; An, J. S. Bull. Chem. Soc. Jpn. 2012, 85, 1007.
(28) Attempted calculations with [La(OCH3)2]4+ were inconclusive due to pre-
sumed difficulties in providing good basis sets for the lanthanide ions and
the large charges involved.
(29) The large suppression of translational and rotational degrees of freedom by
bulk solvent is not captured, as substrates are treated as ideal gases. Bimo-
lecular processes, such as this substrate−complex formation, have high
entropic requirements and resulting free energies barriers tend to deviate
from those calculated experimentally. See (a) Sumimoto, M.; Iwane, N.;
Takahama, T.; Sakaki, S. J. Am. Chem. Soc. 2004, 126, 10457. doi:10.1021/
ja040020r; (b) Tamura, H.; Yamasaki, H.; Sato, H.; Sakaki, S. J. Am. Chem. Soc.
2003, 125, 16114. doi:10.1021/ja0302937.
swpH −(−2.24) = sspH. Since the autoprotolysis constant of methanol is 10−16.77
,
neutral sspH is 8.4. The methods for measurement of pH in methanol can be
found in refs. 5, 6, and 17.
(8) Hatano, M.; Ishihara, K. Chem. Comm. 2013, 49, 1983. doi:10.1039/
C2CC38204K, and references therein.
(9) Maxwell, C. I.; Neverov, A. A.; Mosey, N. J.; Brown, R. S. J. Phys. Org. Chem.
Submitted for publication.
(10) Maxwell, C. I.; Liu, C. T.; Neverov, A. A.; Mosey, N. J.; Brown, R. S. J. Phys. Org.
Chem. 2012, 25, 437. doi:10.1002/poc.1938.
(11) (a) Mitton, G.; Schowen, R. L.; Gresser, M.; Shapley, J. J. Am. Chem. Soc., 1969, 91,
2036. doi:10.1021/ja01036a029; (b) Carpino, L. A.; Collins, D.; Gowecke, S.;
Mayo, J.; Thatte, S. D.; Tibbetts, F. Org. Syn. 1964, 44, 22. doi:10.1002/
0471264180.os044.08.
(12) Alder, R. W.; Mowlam, R. W.; Vachon, D. J.; Weisman, G. R. J. Chem. Commun.
1992, 507. doi:10.1039/C39920000507.
The term spK refers to the pKa of an acid measured in, and referenced to, an
(13)
s
a
organic solvent, in this case methanol; see refs. 5, 6, and 7.
(14) (a) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev, 2005, 105, 2999. doi:10.
Published by NRC Research Press