Organic Letters
Letter
acquired NMR relaxation data and fitting of the data). T1/T2
values are summarized in Table 3.
Table 2. TOF Ratio between Heterogeneous and
Homogeneous Oxidative Coupling by NHC
solvent
THF
DMF
DCM
toluene
cyclohexane
TOFhetero/TOFhomo
Table 3. Values of T1/T2 Ratio of Solvents in the
Polystyrene-Supported NHC Catalyst Used in This Work
0.97 0.03
0.83 0.02
0.90 0.03
0.59 0.02
0.50 0.02
solvent
THF
DCM
DMF
toluene
cyclohexane
T1/T2
2.39 0.07
2.94 0.09
3.11 0.09
3.97 0.12
4.61 0.14
proticity, polarizability, or basicity that could be involved in the
stabilization or destabilization of the transition states in the
mechanism).
From the data in Table 2, it is possible to observe that a
similar reactivity is observed for polar solvents (THF, DMF,
DCM) when comparing the heterogeneous reaction with the
homogeneous one. This finding is of high relevance as it is
often the case that the reactivity of heterogeneous organo-
catalysts is much lower than that of the homogeneous
counterpart, whereas in this case, for those solvents, a
comparable reactivity is observed. The drop in TOF ratio
appears more remarkable for the slightly polar solvent toluene
and for the nonpolar solvent cyclohexane. Scheme 1 shows the
For fast tumbling bulk liquids, such as the solvents used in
this work, single values of T1 and T2 are higher than for liquids
confined in porous materials, and it is well-known that T1/T2
≈ 1, as also suggested by the theory.20 Aksnes and Gjerdaker
̊
have reported a T1 significantly different from T2 for bulk
cyclohexane, although this was measured at 400 MHz.21 We
note here that while the T1 of bulk liquids at room temperature
does not depend on frequency,22 the behavior of T2 is more
complex as it is affected by frequency as well as scalar
coupling.23 Previous experimental results at frequencies below
400 MHz have shown that for cyclohexane T1 = T2 = 2.9 s,16a
which is in agreement with the value of 2.8 s reported here (see
Supporting Information, Table SI4). We also note that in our
case, NMR relaxation times of the bulk liquid are not necessary
for our analysis, which is based on the T1/T2 of solvents
confined in the porous matrix of the catalyst. It is important to
highlight here that while single values of relaxation times can
be affected by several factors other than the surface influence,
such as amount of bulk liquid, degree of pore saturation, and
specific surface area of contact, the T1/T2 ratio is to a large
extent not affected by these factors and is a more direct and
robust indication of the influence of surface interactions.13a In
particular, as mentioned above, the T1/T2 ratio value is related
to the surface−molecule energy interaction. Higher T1/T2
values indicate a stronger affinity of the solvent for the surface,
and in our case this occurs for the less polar solvents; this is
expected due to the hydrophobic nature of the matrix, which
leads to preferential interactions with the less polar solvents.
The results reported here give experimental evidence of such
effects and provide quantitative metrics to assess them.
Diffusion measurements were also performed in order to
assess the influence of the pore diffusion on reaction rate, in
addition to solvent adsorption effects. The Weisz−Prater
criterion was applied to estimate the presence of diffusion
limitation. In all cases, the calculated values were below 1;
hence pore diffusion limitation can be excluded (see
Scheme 1. Generally Accepted Mechanism for the Oxidative
NHC Catalyzed Reaction5c
general accepted mechanism for the oxidation of aldehydes
promoted by NHC and an external oxidant. The Breslow
intermediate is the key species involved in the redox step, and
it is generated through the nucleophilic addition of the in situ
formed catalyst act-PS1. The catalytic species involved in the
process is act-PS1, whereas PS1 is the protonated form in
which usually the catalyst is stored for stability purposes. To
switch from PS1 to act-PS1 in situ, a base such as DBU is
required. For the NMR experiments, as we were interested in
investigating the physical behavior of act-PS1, because it is the
active form, we treated the precatalyst with a strong base in
excess (NaH) before the preparation of the NMR samples (see
The activated polystyrene-supported catalyst was soaked in
anhydrous solvent for 24 h to allow full saturation of the pores.
After this time, the catalyst particles were put onto a presoaked
filter paper, the external surface of the catalyst particles was
dried to remove any excess bulk solvent around the particles,
and the solid was then transferred into a 5 mm NMR tube. The
spin−lattice relaxation time, T1, was measured using an
inversion recovery pulse sequence, and the transverse
relaxation time, T2, was measured with the CMPG pulse
A plot of the TOF ratio versus T1/T2 of the solvent is shown
in Figure 1, and it clearly depicts a remarkable trend showing
that solvents with higher affinity for the surface result in lower
catalytic activity. This strongly suggests that the decrease in
catalytic activity is closely related to a stronger surface affinity
of the solvent, which inhibits reactivity by preventing access of
reactant molecules to the catalytic sites over the surface. A
similar effect has been reported for metal/support catalysts
used in oxidation of diols using NMR spectroscopic analysis,25
and here, for the first time, we observe a similar effect on
immobilized organocatalysts. This suggests that solvent
selection is a key parameter to consider when optimizing
and developing such materials. It is important to highlight that
C
Org. Lett. XXXX, XXX, XXX−XXX