Steric effects of bulky tethered arylpiperazines on the reactivity of Co-Schiff base oxidation catalysts—a synthetic and computational study

Article abstract of DOI:10.1016/j.tet.2019.04.059

New C2-symmetric and C2-asymmetric Co-Schiff base catalysts tethered to arylpiperazine units were synthesized and used to oxidize phenolic lignin models to para-benzoquinones. Synthetic approaches to these catalysts were optimized to include fewer steps and broaden the types of catalyst structures available. In contrast to conventional Co-Schiff base catalysts, these systems induce phenolic oxidation in the absence of an external axial base, simplifying the process. Asymmetric catalysts bearing a phenylethylene or diphenylmethyl piperazine substituent display the highest catalytic activity observed to date for the conversion of S-models to 2,6-dimethoxybenzoquinone (DMBQ). Computational analysis shows that more reactive catalysts populate conformations that favor oxidation in preference to non-productive decomposition routes. This balance between catalyst reactivity and catalyst deactivation is optimized by inclusion of sufficient steric bulk around the periphery of the Schiff base ligand, reducing catalyst deactivation and allowing oxidations to proceed in the absence of an added axial ligand.

Full text of DOI:10.1016/j.tet.2019.04.059

Tetrahedron 75 (2019) 3118e3127
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Steric effects of bulky tethered arylpiperazines on the reactivity of Co-
Schiff base oxidation catalystsda synthetic and computational study
a
b
a, *
Rebecca E. Key , Thomas Elder , Joseph J. Bozell
a
Center for Renewable Carbon, Center for Direct Catalytic Conversion of Biomass to Biofuels (C3Bio), University of Tennessee, 2506 Jacob Drive, Knoxville,
TN 37996, USA
USDA Forest Service, Forest Operations Research, 521 Devall Drive, Auburn, AL 36849, USA
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 January 2019
Received in revised form
New C2-symmetric and C2-asymmetric Co-Schiff base catalysts tethered to arylpiperazine units were
synthesized and used to oxidize phenolic lignin models to para-benzoquinones. Synthetic approaches to
these catalysts were optimized to include fewer steps and broaden the types of catalyst structures
available. In contrast to conventional Co-Schiff base catalysts, these systems induce phenolic oxidation in
the absence of an external axial base, simplifying the process. Asymmetric catalysts bearing a phenyl-
ethylene or diphenylmethyl piperazine substituent display the highest catalytic activity observed to date
for the conversion of S-models to 2,6-dimethoxybenzoquinone (DMBQ). Computational analysis shows
that more reactive catalysts populate conformations that favor oxidation in preference to non-productive
decomposition routes. This balance between catalyst reactivity and catalyst deactivation is optimized by
inclusion of sufficient steric bulk around the periphery of the Schiff base ligand, reducing catalyst
deactivation and allowing oxidations to proceed in the absence of an added axial ligand.
© 2019 Elsevier Ltd. All rights reserved.
23 April 2019
Accepted 24 April 2019
Available online 30 April 2019
Keywords:
Phenolic oxidation
Lignin
Lignin models
Arylpiperazine base tethers
Co-Schiff base
Renewable carbon
1. Introduction
a feedstock. Recent improvements in catalytic hydrogenolysis
processes first developed in the 1940s give lignin-derived products
Recent years have seen a dramatic increase in efforts to under-
stand and control the reactivity of lignin as a low cost and excep-
tionally abundant source of renewable carbon. Nonetheless, lignin's
structural heterogeneity continues to frustrate development of
chemoselective catalysts for its conversion to high value biobased
chemicals [1e6]. These challenges are amplified when biorefining
converts native lignin (lignin as found in a raw material's ligno-
cellulosic matrix) into technical lignin (lignin isolated as a separate
process stream). This transformation introduces a wide range of
additional structural changes that depend on the methodology and
severity of the biorefining process [7e11]. For example, multiple
2 3
enriched in low molecular weight aromatics [23]. Ni/Al O cata-
lyzes the hydrogenation of lignin in birch wood to an oil containing
a high proportion of n-propanolguaiacol and -syringol [24]. Cata-
lytic systems based on Pd/C/Zn and Ni/C selectively afford a small
number of propylphenols at 50e60% conversion of model lignin
polymers or the lignin in Miscanthus [25,26]. Catalytic Ni/C gives
89% selectivity for the hydrogenolysis of the lignin in birch sawdust
to propylsyringol and propylguaiacol at 54% lignin conversion.
However, the failure of these and additional studies to improve
these yields suggests that a maximum conversion limitation may
exist for reductive lignin depolymerization processes [27].
catalytic processes have targeted cleavage of the
b
-O-4 unit, the
Moreover, the continuing focus on biofuel development in the
biorefining industry indicates that technical lignin isolated from
well-recognized biomass fractionation processes (e. g., dilute acid,
steam explosion, organosolv) will likely remain the standard [28].
Thus, catalytic conversion methodology for technical lignin must be
able to accommodate the structural changes that result from these
primary interunit linkage in native lignin [12e18]. We have
described how such transformations are not suitable for technical
lignin generated within the biorefinery, as typical lignin isolation
processes greatly reduce or eliminate this linkage [7,19e22]
Newer “lignin first” approaches have addressed this challenge
through the use of whole biomass and its contained native lignin as
types of processing. For example, cleavage of lignin's b-O-4 linkage
during biorefining simultaneously generates a markedly higher
proportion of free phenolic groups (7e70%) than are present in
native material [29]. Further, the substituted aromatics present in
*
Corresponding author.
E-mail address: jbozell@utk.edu (J.J. Bozell).
https://doi.org/10.1016/j.tet.2019.04.059
040-4020/© 2019 Elsevier Ltd. All rights reserved.
0

R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
3119
technical lignin retain the electron-rich character of native lignin,
making them susceptible to selective oxidation. Accordingly, we
have focused on catalytic oxidation of lignin-like phenols as the
best model of the lignin that will be available as a renewable carbon
source from the biorefinery.
Specifically, we have examined lignin and lignin model oxida-
tion catalyzed by Co-Schiff base complexes (e. g., Co(salen) 1,
Scheme 1) using molecular oxygen as the terminal oxidant
Fig. 1. S- and G-type lignin models and dimeric lignin models examined with second-
generation Co-Schiff base catalyst 7.
[
19,20,30,31]. These catalysts have known utility in the oxidation of
simple phenols unsubstituted in the para-position to the corre-
sponding para-benzoquinones [32e38]. We demonstrated that Co-
Schiff base catalyzed oxidation of a series of lignin-like para-
substituted phenols also afford para-benzoquinones [30], likely
following the mechanism shown in Scheme 1 [35,36,39]
synthetic and computational results for a family of new, sterically
hindered Co-Schiff base complexes.
2. Results and discussion
An external ligand, such as a substituted pyridine or imidazole,
binds to the axial position on Co and promotes the formation of Co-
superoxo complex 2 [35,40,41]. Complex 2 removes a phenolic
hydrogen from the substrate to give phenoxy radical 3, initiating a
process that ultimately leads to the production of para-benzoqui-
nones [35,40e47]. We reported the first examples of this trans-
formation and showed that syringyl (S) lignin models were
converted in good yield to dimethoxybenzoquinone (DMBQ, 4), but
the corresponding conversion of guaiacyl (G) models to mono-
methoxybenzoquinone (MMBQ, 5) proceeded in markedly lower
yields [30]. Subsequent work revealed that the yield of 5 from
vanillyl alcohol increased when the reaction was supplemented
with a sterically hindered, non-coordinating aliphatic base, such as
DIPEA. These results can be attributed to the hindered base
deprotonating the phenol and forming a more oxidizable phenolate
ion 6, giving a more rapid formation of the phenoxy radical 3 [31].
Based on these results, we developed a family of second-
2
.1. The role of sterics in M-Schiff base-catalyzed reactions
2
The accepted model for binding of O to a Co-Schiff base com-
plex postulates that the interaction of the catalyst and an external
base raises the energy of the metal dz orbital sufficiently to overlap
with appropriate orbitals on O [51e53]. Since our complexes
2
catalyzed oxidation of lignin model phenols to quinones in the
2
2
absence of base, we compared the relative dz orbital energies of
several Co-Schiff base complexes examined in this study [54] to
previous results for Co(salen) 1 in the presence and absence of
common ligands using the B3LYP density functional method and
the 6-31G(d) basis set (Fig. 2).
2
As expected, the evaluation supports lower dz energies for
Co(salen) 1 in the presence of weaker ligands (e. g., DMF) and
significantly higher dz energies in the presence of strong ligands
2
(e. g., N-Me imidazole, N,N-diMe piperazine, DMAP). Moreover, the
2
generation, asymmetric (i.e., no C symmetry) Co-Schiff base cat-
2
evaluation is well ordered, in that the dz energy is lowest for
Co(salen) in the absence of a ligand and rises predictably in the
presence of increasingly stronger ligands. Accordingly, and
consistent with the literature, Co(salen) is also an excellent phenol
oxidation catalyst in the presence of strong ligands. Our early
studies showed that attempting oxidation of substituted phenols
with simple Co(salen) in the absence of an external ligand failed to
generate the quinone product and led to the formation of a cata-
lytically inactive complex [30].
alysts that placed the sterically hindered base (in the form of a
substituted piperazine) in closer proximity to the Co center by
incorporating it into the Schiff base ligand itself (e. g., 7 and 8,
Scheme 1). In particular, complex 7 effectively catalyzed the high
yield oxidation of S and G monolignol models, as well as S,S- and
S,G-dimeric models (Fig. 1) to the corresponding para-benzoqui-
nones, in less time and using half the catalyst loading of earlier
studies [19]. Of particular interest was that the increased reactivity
of the second generation catalysts occurred in the absence of an
external axial base, in marked contrast to oxidations carried out
with the structurally simpler 1 and many related catalysts. This
observation suggested that the increased steric bulk of catalysts 7
or 8 was a critical factor in improving the known balance between
catalyst reactivity and catalyst deactivation [48e50]. As part of our
effort to understand the fundamental features that control the
reactivity of Co-Schiff base oxidation catalysts, we wish to report
Of particular interest, however, is that some catalysts (S1, S2,
2
[
Supporting Information]; 12, 13) with low calculated dz orbital
energies still serve as oxidation catalysts for substituted phenols.
S2, 12, and 13 give moderate to good yields of DMBQ from syringyl
Fig. 2. Calculated relative dz2 orbital energies for a series of ligated and non-ligated
a
Scheme 1. Mechanism of the Co-Schiff base-catalyzed oxidation of para-substituted
Co-Schiff base catalysts; Modeling carried out with Co-Jacobsen catalyst
phenols with oxygen.
(Supplemental Information).

3120
R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
alcohol. Further, our earlier work showed that the piperazine
substituent in complex 7 was not within binding distance to the Co
center [20]. We therefore suspected that the introduction of steri-
cally bulky groups on the catalyst was reducing or slowing catalyst
deactivation pathways rather than promoting formation of Co-
superoxo complex 2.
The literature reports several examples of the influence of ste-
rics on the reactivity of metal-Schiff base catalysts. For example,
Pospisil et al. synthesized various Mn-Jacobsen catalysts and
examined their catalytic activity in olefin epoxidation. They
observed that higher ees were observed when bulky tert-butyl
groups were present on the catalysts, as these groups effectively
limit substrate approaches to a trajectory that favors formation of a
single enantiomer [55,56]. Higher catalytic activity in ring-opening
polymerization of rac- and L-lactide was observed when less ste-
rically hindered Al(salen) catalysts were used, while higher ster-
eoselectivities were observed with more sterically hindered
catalysts [57]. An increase in the steric bulk of Cr(salen) complexes
resulted in a higher yield and ee of cycloadducts in oxo-Diels-Alder
reactions when compared to the less hindered Cr-Jacobsen catalyst
Scheme 2. Synthesis of arylpiperazine-tethered Co-Schiff base catalysts 9e10 and
2e13.
1
[
58]. In our work, the introduction of tert-butyl groups on the
catalyst is thought to reduce catalyst deactivation that occurs via
formation of Co-peroxo dimers seen in simpler Co(salen) systems
phenethyl substituted piperazines give the advantage of simpli-
fying the synthesis of ligands 22e25. Metallation of 22 and 23 using
[
41,46,47]. Therefore, our catalyst design targeted optimizing the
Co(OAc).
2
2
4H O provided asymmetric catalysts 9 and 10, respec-
steric bulk environment of several piperazine-tethered-Co-Schiff
base catalysts to maximize interactions between catalyst, base,
and substrate while minimizing catalyst deactivation.
tively. The low yields for catalysts 9 and 12 are likely the result of
forcing bulky substituents into proximity as the ligand adopts a
square planar conformation upon insertion of the Co.
2.2. Syntheses of arylpiperazine-tethered Co-Schiff base catalysts
2.3. Arylpiperazine-tethered Co-Schiff base-catalyzed oxidation of
S-lignin models
The family of new catalysts synthesized for this study is shown
2 2
in Fig. 3. Both C symmetric and C asymmetric ligands were pre-
The reactivity of catalysts 9e13 was evaluated at different re-
action times in MeOH solvent for the oxidation of syringyl alcohol
pared, incorporating differing side chains at the piperazine moiety,
exemplifying both increased steric bulk (9, 11, 12 and 13) and
increased chain length (10 and 13).
Symmetric catalysts 12 and 13 (Scheme 2) were prepared by
condensing one equivalent of (±)-trans-1,2-diaminocyclohexane
and
to compound 7, the most active Co-Schiff base catalyst to date
Table 1, entries 1 and 2). Overall, the catalysts give good yields of
a-methylsyringyl alcohol to DMBQ. The results were compared
(
DMBQ, accompanied by smaller amounts (<10%) of the corre-
sponding aldehyde arising from oxidation of the benzylic alcohol.
In addition, the asymmetric catalysts give generally higher yields of
quinone than the symmetric catalysts, reflecting the impact of
greater steric bulk on the oxidation process. In all cases, the cata-
lysts operated successfully without the addition of an external axial
base.
(
1
19) and two equivalents of either arylpiperazine salicylaldehyde
7 or 18 to give ligands 24 and 25 in moderate and high yields,
.
2 2
respectively. Metallation of 24 and 25 with Co(OAc) 4H O provided
Co-Schiff base catalysts 12 and 13. Asymmetric catalysts 9 and 10
were synthesized in a similar manner but required protecting one
of the amine groups of 19 as the HCl salt 20 [59,60]. One-pot
condensation of the free amine in 20 with one equivalent of 21
followed by deprotonation and then condensation of the interme-
diate with one equivalent of either 17 or 18 afforded moderate
yields of Schiff base ligands 22 and 23, respectively [19,59].
The commercial availability of the diphenylmethyl and
Asymmetric catalyst 9, bearing
a single diphenylmethyl
(diPhMe) piperazine substituent, produced DMBQ in 70 and 81%
yield at 18 h and 1 h reaction times, respectively (Table 1, entries 3
and 4), comparable to catalyst 7 (Table 1, entries 1 and 2). Both 7
and 9 gave DMBQ in slightly lower yields at longer reaction times.
The reason for these lower DMBQ yields at longer reaction times is
not clear but may suggest that the quinone product is being
consumed in a slower reaction process. Related work in our group
has revealed that the quinone product may inhibit the oxidation
through the formation of intermediate Co-quinone complexes.
Over long reaction times, a Co-DMBQ complex may be forming
slowly [61]. Asymmetric catalyst 10, bearing a phenethyl side chain,
catalyzed the formation of DMBQ in comparable yields to 7 and 9
(Table 1, entries 5 and 6) when oxidations were run overnight, and
comparable yields (73%, Table 1, entry 6) when run for 1 h.
In some cases, a further increase in the steric bulk of the cata-
lysts resulting from the introduction of two large substituents onto
the Schiff base ligand decreased the quinone yield (Table 1, entries
9
(
e11). Surprisingly, introduction of a second, large diPhMe-group
catalyst 12) had little impact on the yield of DMBQ, with yields
of 73 and 74% at 18 h and 1 h reaction times, respectively (Table 1,
2 2
Fig. 3. C Symmetric and C asymmetric Co-Schiff base catalysts evaluated in this
study.

R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
3121
Table 1
Arylpiperazine-tethered Co-Schiff base-catalyzed oxidation of S models in MeOH.
Entry
Catalyst
R
Reaction Time (h)
DMBQ yield (%)
71a
Aldehyde yield (%)
a
1
2
3
4
5
6
7
8
9
7, asym-NBn
7, asym-NBn
H
H
H
H
H
H
H
H
H
H
H
CH
CH
CH
CH
CH
18
1
18
1
18
1
18
1
4
1
6
4
7
7
3
4
4
9
9
b
a
75
70
81
80
73
73
74
67
64
52
73
76
77
80
b,c
a,c
a,c
b,c
a-c
a,c
a,c
a,c
a,c
a,c
9, asym-diPhMe
9, asym-diPhMe
10, asym-Phenethyl
10, asym-Phenethyl
12, sym-diPhMe
12, sym-diPhMe
11, sym-NBn
11, sym-NBn
13, sym-Phenethyl
7, asym-NBn
b,c
b,c
a-c
a,c
a,c
a-c
18
1
a,c
10
11
12
13
14
15
16
a,c
18
16
16
16
16
16
d
3
3
3
3
3
e
d
9, asym-diPhMe
10, asym-Phenethyl
11, sym-NBn
e
b
a,c
1
c,d
e
c.d
12, sym-diPhMe
73
e
Mol % is calculated relative to the substrate; diPhMe: 1-diphenylmethyl; Phenethyl: 1-phenethyl.
a
Yields obtained via 1H NMR integration.
b
c
1
Combined yield from isolated and H NMR integration.
Denotes the average yields between two runs; dIsolated yield.
entries 7 and 8). Interestingly, symmetric phenethyl catalyst 13
gave noticeably lower yields of DMBQ (52%, Table 1, entry 11) when
compared to all other catalysts tested, contrasting the results ob-
tained with its asymmetric counterpart 10. The presence of a single
phenethyl group appears to provide the steric bulk necessary to
generate an active catalyst, but the presence of two phenethyl
groups and their longer tethers may afford catalyst conformations
that impede access of the substrate to the Co center.
catalyst 7 (Table 2). DCM, a non-coordinating solvent, should not
bind to the Co-center, thus inhibiting the formation of the cata-
lytically active Co-superoxo complex, and reducing the yield of
DMBQ in comparison to reactions performed in MeOH. Indeed, all
catalysts evaluated for oxidative activity in DCM provided lower
yields of DMBQ when compared to their activity in methanol.
Specifically, catalyst 7 gave DMBQ in 53% yield (Table 2, entry 1).
Catalyst 10 gave a yield significantly lower yield of DMBQ in DCM
when compared to MeOH (26%, Table 2, entry 2). Catalyst 13, which
also gave lower yields in MeOH, gave a 38% yield of MMBQ in DCM
(Table 2, entry 3).
In DCM a decrease in DMBQ is accompanied by an increase in
syringaldehyde when compared to the oxidations in methanol.
Thus, the yield of DMBQ in methanol could be due, in part, to the
conversion of intermediate syringaldehyde to DMBQ. To further
investigate this, oxidations of syringaldehyde in both methanol and
DCM were compared using catalyst 7. The yield of DMBQ from
syringaldehyde is higher in methanol (29%) than in DCM (3%).
Therefore, it can be concluded that higher yields of DMBQ from
oxidations run in methanol result from the oxidation of both
syringyl alcohol and syringaldehyde, whereas the DMBQ yield from
the oxidations run in DCM result from the conversion of syringyl
alcohol [30].
We also evaluated the activity of catalysts 9e12 with a-meth-
ylsyringyl alcohol and compared the results to oxidations using
catalyst 7 (Table 1, entries 12e16). Overall, the trends were very
similar to oxidation of syringyl alcohol. Oxidation using catalyst 7
provided yields of DMBQ comparable to previously reported results
[
19] (73%, Table 1, entry 12). Catalyst 9 gave yields of DMBQ that
were similar to catalyst 7 (76%, Table 1, entry 13). Catalyst 10, the
most active catalyst for the oxidation of syringyl alcohol at long
reaction times, performed similarly, providing DMBQ in slightly
higher yield (77%, Table 1, entry 14) than catalyst 7. Catalyst 12
provided DMBQ in yields equivalent to catalyst 7 (73%, Table 1,
entry 16), again showing that the addition of a second diPhMe-
piperazine substituent has little effect on the oxidation. Since the
catalysts enable oxidation in the absence of a strong axial ligand,
we assume that the axial position is filled with a more weakly
binding ligand, such as the MeOH solvent used in the reaction.
While simple Co(salen) 1 preferentially deactivates in the absence
of an added base, the presence of large groups on the periphery of
the second generation catalysts induces an inhibition of catalyst
deactivation and enables reaction in the absence of an additional
axial ligand. Although the ability of MeOH to raise the energy of the
2.4. Arylpiperzine-tethered Co-Schiff base-catalyzed oxidation of G-
lignin models
The reactivity of the catalysts toward vanillyl alcohol as a model
of the G substructural units in lignin was also evaluated and again
compared to catalyst 7 in both MeOH and DCM (Table 3).
2
dz orbital on Co is much lower than pyridine, it may be sufficient to
generate enough catalytically active Co-superoxo intermediate to
carry the oxidation of S model substrates as long as deactivation
processes are inhibited.
To examine the effect of solvent on oxidation, we examined the
activity of catalysts 10 and 13 for the oxidation of syringyl alcohol
oxidation in dichloromethane (DCM) and compared the results to
Consistent with our earlier work [30], oxidation of vanillyl
alcohol using either symmetric or asymmetric catalysts in MeOH
gave quinone yields that were lower than those observed for the
corresponding S-model oxidations. Compared to the S-models,
vanillyl alcohol has only a single electron-donating -OMe group.
Therefore, the phenolic OH bond is stronger, resulting in a lower

3122
R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
Table 2
Arylpiperazine-tethered Co-Schiff base-catalyzed oxidation of syringyl alcohol in DCM.
Entry
Catalyst
Substrate
Reaction Time (h)
DMBQ yield (%)a
Syringaldehyde yield (%)a
1
2
3
7, asym-NBn
10, asym-Phenethyl
13, sym-Phenethyl
SyOH
SyOH
SyOH
18
18
18
53
26
38
17
30
24
Mol % is calculated relative to the substrate.
a
Yields obtained via 1H NMR integration.
Table 3
Arylpiperazine-tethered Co-Schiff base-catalyzed oxidation of vanillyl alcohol.
Entry
Catalyst
Solvent
Reaction Time (h)
MMBQ yield (%)a
2,5-DMBQ yield (%)a
Vanillin yield (%)b
Recovered vanillyl alcohol (%)a
1
2
3
4
5
6
7
8
9
7, asym-NBn
7, asym-NBn
DCM
MeOH
DCM
MeOH
DCM
MeOH
DCM
MeOH
DCM
DCM
17
18
17
18
17
18
17
18
17
17
18
59
34
48
32
42
28
16
1
e
5
e
e
e
e
e
e
2
32
10
22
33
30
38
39
15
c
c
c
c
9, asym-diPhMe
9, asym-diPhMe
10, asym-Phenethyl
10, asym-Phenethyl
13, sym-Phenethyl
13, sym-Phenethyl
11, sym-NBn
e
c
c
3
e
5
b,c
c
c
e
e
e
e
1
e
2c
e
e
c
c
c
c,d
48
48
13
10
1
12, sym-diPhMe
12, sym-diPhMe
e
c
c
1
MeOH
9
45
Mol % is calculated relative to the substrate; diPhMe: 1-diphenylmethyl; Phenethyl: 1-phenethyl.
a
Isolated yield.
b
Yields obtained via 1H NMR integration.
c
Denotes the average yields between two different runs.
Combined yield from isolated and H NMR integration.
d
1
rate of phenoxyl radical formation and thus, a lower yield of MMBQ
62e66]. The baseline oxidation using catalyst 7 provided MMBQ in
9% and 34% yield in DCM and methanol, respectively (Table 3,
entries 1 and 2). Asymmetric catalysts 9 and 10 followed the same
pattern (Table 3, entries 3e6). Catalyst 13 was the least effective,
giving only a 13% yield of MMBQ after 17 h in DCM and negligible
amounts in MeOH (Table 3, entries 7 and 8).
In contrast to S models, DCM is a better oxidation solvent than
MeOH for oxidation of G models. We speculate that while the
MeOH solvent may be sufficiently coordinating to promote oxida-
tion of the more reactive S models, the low reactivity of G models in
either solvent allows alternate deactivation reactions to occur, such
as the direct reaction of the catalyst with the MeOH solvent to
afford non-reactive Co complexes [67]. In DCM, this alternate re-
action is eliminated. The reactions still proceed in lower yield than
oxidations of S models, but the steric bulk of the complexes also
slows catalyst dimerization and deactivation, allowing at least
some MMBQ formation to occur.
Although the absolute yields for G model oxidations are lower
than the corresponding S model oxidations, there are interesting
parallels as a function of substituent size. Catalyst 7 bearing a single
N-Bn group gives higher yields with both S (Table 1, entries 1 and 2)
and G models (Table 3, entry 1) than the corresponding catalyst 11
bearing two N-benzyl groups (Table 1, entries 9 and 10 and Table 3,
entry 9). Both asymmetric and symmetric diPhMe catalysts 9 and
12 show little or no difference in reactivity, and also show reactivity
similar to symmetric catalyst 11 (Table 1, entries 9 and 15 and
Table 3, entry 9). These results suggest that 9, 11, and 12 present a
similar steric environment during the oxidation of either S or G
substrates. In contrast, the steric environment presented by sym-
metric catalyst 13 leads to markedly lower yields for oxidation of
both S and G models in comparison to all other catalysts.
[
5
2.5. Computational studies for the arylpiperazine-tethered Co-Schiff
base catalysts
All catalysts and their oxygenated adducts were subjected to
computational evaluation. An initial Monte Carlo search was car-
ried out on structures in which the Co-Schiff base, cyclohexyl bridge
and t-butyl groups were frozen, while allowing the substituents
and oxygens (where appropriate) to move. A family of conformers
was generated, and the low energy conformation for each was
further refined via DFT in Gaussian 16 (see Supplemental Material).
These data were then used to examine possible correlations be-
tween the structure of low energy conformations and the effec-
tiveness of the catalyst. DFT results may have several local minima,
but the Boltzman distribution values for several of the catalyst
structures indicate that certain conformations are much more

R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
3123
highly populated than others, suggesting that such geometric ori-
entations may have greater impact on reactivity than others. This
assumption is supported by the experimental results from our
oxidation, both in this work and earlier work from our group
of the Co center in the low energy conformer may slightly reduce
access to the Co. Further experiments are underway to verify this
impact.
Interestingly, asymmetric phenethyl complex 10 minimizes in a
unique conformation (Fig. 5). The calculated low energy confor-
mation has the phenethyl group rotated roughly parallel to the face
of the catalyst as with complex 11, but is the result of the piperazine
adopting a boat conformation with the nitrogen substituents in an
axial orientation.
In contrast to 7 and 9, which give lower DMBQ yields at longer
reaction times, phenethyl complex 10 gives higher yields after 18 h.
These observations suggest that the location of the phenethyl un-
der the plane of the ligand provides greater hindrance to non-
productive catalyst dimerization or deactivation processes. The
source of stabilization for the low energy conformations adopted by
[
19,20]. Fig. 4 illustrates the low energy conformations calculated
2
for complexes 7, 9, 11 and 12. The dz orbital energies for these
complexes are comparable to other complexes in this series (Fig. 2).
For all complexes, we observed little or no change in conformation
upon oxygenation, and thus, discussion will focus on the non-
oxygenated adducts.
The piperazine adopts a chair conformation in each complex
with the substituents on the piperazine nitrogens in the equatorial
position. With this orientation, complexes 7, 9 and 12 display
substituents roughly perpendicular to the plane of the Schiff base
ligand (e. g., the diPhMe substituent in 9 is positioned at an angle of
ꢀ
1
12.95 to the Schiff base plane while the two diPhMe substituents
both 10 and 11 is not obvious. p-Type interaction between the
ꢀ
ꢀ
in 12 are at angles of 114.65 and 117.12 , respectively). The low
energy conformation for 12 also places the substituents on opposite
sides of the Schiff base ligand. Complex 11 minimizes in an unusual
manner, with one of the N-Bn substituents bending under the plane
of the Schiff base ligand.
phenethyl group and the aromatic rings of the Schiff base may be
possible, but the calculated distance between the aromatic rings
(slightly greater than 4 Å) is on the far edge of typical pi-pi stacking.
Symmetrically substituted phenethyl catalyst 13 gives the
lowest oxidation yields for either S or G models in both MeOH and
DCM. Computational evaluation of 13 (Fig. 6) reveals significant
differences from its asymmetrically substituted counterpart (Fig. 4).
In contrast to the asymmetrically substituted complexes 10 or
11, neither of the phenethyl groups are rotated toward the plane of
the Schiff base ligand. Instead, the calculated low energy confor-
mation orients the substituents away from and to opposite sides of
the ligand. Both piperazines retain the chair conformation, with
one piperazine placing the nitrogen substituents in the axial posi-
tion. Surprisingly, however, the other piperazine adopts a confor-
mation placing one nitrogen substituent in the equatorial position
and the other in an axial position. The result is that one phenethyl
group is approximately parallel to the plane of the Schiff base
ligand. We find that this catalyst is markedly less effective in con-
verting syringyl alcohol to DMBQ, affording only 52% yield after
18 h. Similarly, vanillyl alcohol gives only 16% MMBQ upon oxida-
tion in DCM. Based on the computational results, we speculate that
the presence of two substituents reduces nonproductive deactiva-
tion pathways. Simultaneously, the presence of a piperazine sub-
stituent bearing both axial and equatorial groups may allow
approach of the large substituent to the Co, further reducing access
to the catalytic center and lowering the yield.
In this group of catalysts, asym-NBn complex 7 catalyzes the
oxidation of syringyl alcohol to DMBQ in 71% yield after 18 h in
MeOH. Similarly, asymmetric complex 9 affords a 70% yield of
DMBQ after 18 h. The effect of adding a second diPhMe-piperazine
group appears to have minimal effect on the catalyst effectiveness
when compared to the asymmetric catalyst, as 12 converts syringyl
alcohol to DMBQ in an almost identical 73% after 18 h. a-Methyl-
syringyl alcohol gives DMBQ in 73% yield after 16 h. The repro-
ducibility of yields at both short and long reaction times, however,
suggests that any slow quinone consumption path (as suggested for
asymmetric catalysts) may be reduced with the simultaneous
presence of large substituents on both sides of the catalyst.
Based on these similar yields, we conclude that the NBn and
diPhMe groups provide a steric environment similar to the t-Bu
groups on the periphery of the catalyst. Because the reactions do
not require addition of an external axial base, the substituents also
provide a steric barrier to catalyst dimerization and subsequent
deactivation. Neither the addition of a second phenyl group to the
benzylic position of the substituted piperazine (i. e., a diPhMe
substituent) nor the presence of a second diPhMe group negatively
affects catalyst reactivity when methanol is used as the solvent.
Complex 11 exhibits a small, but reproducibly lower yield of DMBQ
suggesting that the presence of one N-Bn piperazine in the vicinity
3. Conclusions
The steric environment around the Schiff base ligand and the Co
have a strong effect on the ability of the catalysts to carry out
oxidation of lignin models. The ability of Co-Schiff base catalysts to
oxidize lignin models to the corresponding quinones is subject to a
balance between catalyst reactivity and catalyst deactivation. With
sufficient steric bulk around the periphery of the Schiff base ligand,
catalyst deactivation is reduced, allowing oxidations to proceed in
Fig. 4. Low energy conformations of catalysts 7, 9, 11 and 12.
Fig. 5. Low energy conformation of phenethyl complex 10.

3124
R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
hydroxybenzylidene)amino)cyclohexyl)imino)methyl)phenol, and
0
,6'-((1E,1 E)-((((±)-trans)-cyclohexane-1,2-diyl)bis(azaneylyli-
6
dene))bis(methaneylylidene))bis(2-((4-benzylpiperazin-1-yl)
methyl)-4-(tert-butyl)phenol) were synthesized according to pub-
lished procedures and matched data previously reported
[
19,30,59,60,69e72]. Co(OAc).
2 2
4H O was dried under vacuum at
room temperature for several days prior to use. Molecular sieves
ꢀ
(
4 Å) were activated in a 110 C oven prior to use. Analytical thin-
layer chromatography (TLC) was performed using glass-backed,
pre-coated silica gel 60 F254 plates. Flash chromatography was
®
conducted using a Teledyne ISCO CombiFlash
f
R 200 with Tele-
dyne RediSep®
R
f
or SiliCycle SiliaSep™ (230e400 mesh, 40e63
mm
particle size, and 60 Å pore size) silica cartridges. The eluents used
are reported in % (v/v). For compounds 22 and 24, the columns were
washed with five column volumes of 1) methanol (to remove water
from the column), 2) ethyl acetate, and 3) dichloromethane prior to
liquid sample loading.
Fig. 6. Low energy conformations of catalyst 13.
Melting points were determined using a Fisher-Johns melting
point apparatus and are uncorrected. Oxidation procedures were
performed in Fisher-Porter tubes under the respective pressure of
oxygen denoted in the studies. H and C NMR analyses were
performed on a either a Varian Unity 400 MHz or a Varian VNMRS
the absence of an added axial ligand, in marked contrast to oxi-
dations using structurally simpler complexes (e. g., 1).
The reason for this effect is the ability of large substituents to
shift the catalyst reactivity/deactivation balance toward the former.
At present, the computational results suggest that these effects may
be subtle, but important, based on the small energy differences
between low energy conformations of the effective catalysts. We
find that for S-models, the diPhMe- and NBn-piperazine groups
maintain the reactivity of the catalyst system and act similarly to t-
Bu groups (i. e., Jacobsen-like ligands) on the periphery of the Schiff
base ligand. Catalyst 10, bearing a single phenethyl group, gave
excellent yields of quinone from the oxidation of S-models and may
have unique qualities as a result of its conformation during reac-
tion. The oxidation of G-models remains more challenging, as these
new catalysts gave lower yields of MMBQ than oxidation of the
corresponding S-models. Our ongoing work in catalyst design will
continue to couple synthetic and computational evaluation to
develop systems that will oxidize S and G units with equal facility.
Given that lignin composed almost entirely of S units can be ob-
tained from transgenic poplar with increased F5H expression [68],
future work will also involve the examination of catalysts 7 and
1
13
500 MHz instrument; the chemical shifts are given in ppm and are
referenced to the appropriate residual nuclei in deuterated sol-
vents. FT-IR spectra were acquired on either a PerkinElmer Spec-
ꢁ
1
trum One or PerkinElmer Spectrum Two spectrometer at 4 cm
ꢁ
1
resolution and are reported in cm . DART-TOF and high-resolution
mass spectrometry (HRMS) analyses were obtained on a JEOL
AccuTOF-DART™ and an Applied Biosystems QStar Elite HPLC-
QTOF mass spectrometer, respectively, at the University of Ten-
nessee Biological and Small Molecule Mass Spectrometry Core.
Values are reported as m/z (relative ratio). Accurate masses were
þ
determined for the molecular ion (M þ H) , an acceptable ion, or
the neutral species; the results are reported with an error of
<3 ppm for DART-TOF and <4 ppm for ESI-MS.
4.2. General procedure for the oxidation of lignin models
CAUTION: The oxidation reactions were performed in thick-walled,
9
e12 for the oxidative deconstruction of bioengineered, high-S
glass Fisher-Porter tubes under pressurized oxygen. Though no diffi-
culties were experienced, adequate precautions should be considered
when using organic compounds and oxygen above atmospheric
pressure. Lignin models were oxidized according to a previously
published procedure [19,30]. Briefly, to a Fisher-Porter tube was
added 1 equivalent of the lignin model, 0.05 equivalents of Co-
Schiff base catalyst, and the appropriate amount of methanol or
dichloromethane to make a 0.2 M solution relative to the lignin
model substrate. The tube was flushed three times and then pres-
surized to 50 psi using oxygen. Once pressurized, the reaction was
stirred at room temperature for times denoted in Tables 1e3. If
significant quantities of yellow solid (quinone) precipitated upon
completion of the reaction, the mixture was filtered, and the
mother liquor was concentrated under reduced pressure. This
residue was analyzed by NMR to determine the amount of any
additional quinone. If minimal or no yellow solid was present upon
completion of the reaction, the mixture was concentrated under
reduced pressure. Purification (silica gel; 0-2-5-10-100%
EtOAc:DCM) gave the respective para-benzoquinones in yields re-
ported in Tables 1e3
lignin. The result will be new methodology for the conversion of
lignin, which will streamline operation and offer new catalytic
routes to biobased chemicals and fuels from renewable carbon.
4
. Experimental
4.1. General Information
All reactions were performed under a nitrogen atmosphere
unless otherwise stated. Commercially available chemicals (com-
pounds 14, 15, 19, 21, benzyl bromide, 4-tert-butylphenol,
dichloromethane, diethyl ether, ethyl acetate, formaldehyde, hex-
anes, hydrochloric acid, methanol, methylmagnesium bromide,
paraformaldehyde, piperazine, potassium iodide, sodium bicar-
bonate, sodium chloride, syringaldehyde, syringyl alcohol, trie-
thylamine, vanillyl alcohol) were used as received unless otherwise
ꢀ
stipulated. Magnesium chloride was dried overnight in a 110 C
oven prior to use. Anhydrous solvents were obtained from either a
MBRAUN MB-SPS solvent purification system or used as received
from commercial vendors. Symmetric catalyst 11 was synthesized
based on our previously reported procedures [19]. Compounds 7,
4.3. Computational modeling methodology
2
3
6, 20, 1-benzylpiperazine, 5-tert-butyl-2-hydroxybenzaldehyde,
-((4-benzylpiperazin-1-yl)methyl)-5-tert-butyl-2-
All calculations for the study were conducted on the Alabama
Supercomputer Network. An initial conformational search was
done using a 1000 step Monte Carlo procedure with MMFF
hydroxybenzaldehyde,
2-((4-benzylpiperazin-1-yl)methyl)-4-
(tert-butyl)-6-((E)-(((±)-trans-2-((E)-(3,5-di-tert-butyl-2-

R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
3125
minimization at each step, as implemented in Spartan ‘16. The Co-
4.5. Synthesis of asymmetric Schiff base ligands
salen moiety, cyclohexyl bridge and t-butyl groups were frozen,
while the substituents and oxygens, where appropriate, were
allowed to move. The low energy conformation for each was
refined using the M06-L density functional method, the LANL2DZ
basis set for the cobalt and the 6-31G(d,p) basis set for carbon,
nitrogen, hydrogen and oxygen. The structures were optimized,
with frequency calculations for thermal corrections and to insure
the identification of a stationary point, and done using the SMD
solvation model for ethanol. All DFT calculations were done with
Gaussian 16, Revision A.03. These data were then used to examine
possible correlations between the structure of low energy confor-
mations and the effectiveness of the catalyst. Final renderings were
carried out using Mercury 3.10 (Build 156946).
4.5.1. Synthesis of 2-((4-benzhydrylpiperazin-1-yl)methyl)-4-(tert-
butyl)-6-((E)-((((±)-trans)-2-(((E)-3,5-di-tert-butyl-2-
hydroxybenzylidene)amino)cyclohexyl)imino)methyl)phenol (22)
To a round-bottom flask equipped with a stirbar was added
(±)-trans-1,2-diaminocyclohexane (19) (1.2 mL, 9.99 mmol) and
ꢀ
anhydrous ethanol (52 mL), and the flask was cooled to 0 C. Hy-
drochloric acid (2.0 M in diethyl ether, 5 mL, 10 mmol) was added
dropwise, and the solution was stirred for 1 h at room temperature.
A
solution of 3,5-di-tert-butyl-2-hydroxybenzaldehyde (21)
(2.4079 g, 10.3 mmol) in anhydrous ethanol (52 mL) was added,
along with 4 Å MS (3.65 g). After stirring 3.25 h at room tempera-
ture, a solution of compound 17 (4.55 g, 10.3 mmol) in anhydrous
ꢀ
ethanol (172 mL þ heating at 95 C for complete dissolution) and
triethylamine (2.9 mL, 20.8 mmol) was added, and the solution was
allowed to stir for 20 h at room temperature. Upon completion, the
reaction mixture was concentrated under reduced pressure and
purified (silica gel; 2-5-10-15-20-100 EtOAc:DCM; see General In-
formation for column prep instructions). Drying under vacuum at
4
4
.4. Syntheses of catalyst precursors
.4.1. Synthesis of 3-((4-benzhydrylpiperazin-1-yl)methyl)-5-(tert-
ꢀ
butyl)-2-hydroxybenzaldehyde (17)
To a round-bottom flask equipped with a stirbar was added
potassium iodide (0.1553 g, 0.93 mmol), a solution of compound 16
65 C for 18 h gave the final product as a yellow solid in 64% yield.
ꢀ
mp: 163e166 C; R
f
¼ 0.15 (10% EtOAc:DCM); IR (neat): 2953, 2860,
2806, 2763, 1627, 1451, 1362, 1273, 1174, 1136, 1096, 1031, 1007, 911,
878, 849, 827, 773, 757, 745, 732, 705, 641 cm ; H NMR (400 MHz,
CDCl ): 13.68 (s, 1H), 13.42 (br, 1H), 8.29 (s, 1H), 8.26 (s, 1H), 7.40
ꢁ
1 1
(
(
1.98 g, 8.73 mmol) in ethyl acetate (22 mL), and a solution of 1-
diphenylmethyl)piperazine (14) (3.3410 g, 13.2 mmol) in ethyl ac-
3
d
etate (82 mL). The reaction was stirred for 1.6 h at room tempera-
ture. Upon completion, the mixture was diluted with
dichloromethane, washed with equal volumes of saturated aqueous
(d, J ¼ 8 Hz, 4H), 7.29 (d, J ¼ 4 Hz, 1H), 7.27e7.22 (m, 5H), 7.15 (m,
J ¼ 8 Hz, 2H), 7.02 (d, J ¼ 4 Hz, 1H), 6.97 (d, J ¼ 4 Hz, 1H), 4.21 (s, 1H),
3.61 (d, J ¼ 12 Hz, 1H), 3.49 (d, J ¼ 12 Hz, 1H), 3.36e3.22 (m, 2H),
2.52 (br, 4H), 2.43 (br, 4H), 1.97e1.81 (m, 4H), 1.77e1.64 (m, 2H),
NaHCO
under reduced pressure. Purification (silica gel; 0-2-5-10-100%
EtOAc:DCM) gave 17 as an off-white solid in 43% yield. R
¼ 0.62
10% EtOAc:DCM); IR (neat) 2963, 2806, 2762, 1677, 1605, 1479,
3 2 4
(2X), dried over anhydrous Na SO , and concentrated
13
1.50e1.42 (m, 2H), 1.40 (s, 9H), 1.23 (s, 9H), 1.20 (s, 9H); C NMR:
(100 MHz, CDCl ): 165.7, 165.1, 158.0, 157.2, 143.0, 140.4, 139.9,
f
3
d
(
136.3, 128.4, 128.0, 126.77, 126.7, 126.65, 126.0, 124.4, 76.4, 72.8,
1
6
451, 1281, 1236, 1216, 1133, 1004, 971, 896, 850, 758, 744, 708, 693,
72.2, 56.3, 54.0, 52.0, 34.9, 34.0, 33.8, 33.4, 33.1, 31.43, 31.35, 29.4,
ꢁ
1
1
þ
05 cm
;
H NMR (400 MHz, CDCl
3
):
d
10.29 (s, 1H), 7.62 (d,
24.3; DART-TOF Calcd for C50
755.52542.
H
66
N
4
O
2
(M þ H) : 755.5264; found
J ¼ 4 Hz, 1H), 7.42e7.38 (m, 4H), 7.29e7.24 (m, 5H), 7.20e7.15 (m,
13
2
H), 4.27 (s, 1H), 3.78 (s, 2H), 2.76e2.39 (br, 8H), 1.29 (s, 9H);
C
NMR (100 MHz, CDCl
3
):
d
192.0, 159.2, 142.3, 141.8, 133.1, 128.5,
4.5.2. Synthesis of 2,4-di-tert-butyl-6-((E)-((((±)-trans)-2-(((E)-5-
(tert-butyl)-2-hydroxy-3-((4-phenethylpiperazin-1-yl)methyl)
benzylidene)amino)cyclohexyl)imino)methyl)phenol (23)
127.8, 127.0, 124.9, 122.5, 121.9, 75.9, 59.6, 52.8, 51.4, 34.0, 31.2;
þ
DART-TOF Calcd for C29
H
34
N
2
O
2
(M þ H) : 443.26985, found
4
43.26883.
To a round-bottom flask equipped with a stirbar was added
compound 19 (0.35 mL, 2.91 mmol) and anhydrous ethanol (15 mL),
ꢀ
and the flask was cooled to 0 C. Hydrochloric acid (2.0 M in diethyl
ether, 1.5 mL, 3.0 mmol) was added dropwise, and the solution was
stirred for 1 h at room temperature. A solution of compound 21
(0.6758 g, 2.88 mmol) in anhydrous ethanol (15 mL) was added.
After stirring 2.5 h at room temperature, a solution of compound 18
(1.1035 g, 2.90 mmol) in anhydrous ethanol (98 mL and then heated
4
.4.2. Synthesis of 5-(tert-butyl)-2-hydroxy-3-((4-
phenethylpiperazin-1-yl)methyl)benzaldehyde (18)
To a round-bottom flask equipped with a stirbar was added
potassium iodide (0.3712 g, 2.24 mmol), a solution of compound 16
ꢀ
(
5.02 g, 22.1 mmol) in ethyl acetate (56 mL), and a solution of 1-
phenethylpiperazine (15) (6.3 mL, 33.3 mmol) in ethyl acetate
56 mL). The reaction was stirred for 1.5 h at room temperature.
at 80 C to fully dissolve the salicylaldehyde) and triethylamine
(0.82 mL, 5.88 mmol) was added, and the solution was stirred for
21 h at room temperature. Upon completion, the reaction mixture
was concentrated under reduced pressure and purified (silica gel;
5-10-20-40-50-100% EtOAc:hexanes). Drying under vacuum at
(
Upon completion, the mixture was diluted with an equal volume of
dichloromethane, washed with equal volumes of saturated aqueous
ꢀ
NaHCO
methane (1X), dried over anhydrous Na
under reduced pressure. Purification (silica gel; 0-5-20-25-100%
EtOAc:DCM) gave 18 as an off-white solid in 30% yield. R
¼ 0.31,
20% EtOAc:hexanes); IR (neat) 2948, 2847, 2803, 2847, 2762, 1673,
3
(2X), back extracted with an equal volume of dichloro-
65 C for 5 h gave the final product as a solid in 60% yield. mp:
ꢀ
2
SO , and concentrated
4
102e104 C; R
f
¼ 0.18 (40% EtOAc:hexanes); IR (neat): 2950, 2862,
2808, 1627, 1456, 1362, 1272, 1173, 1133, 1095, 1034, 1012, 908, 878,
818, 772, 731, 699, 644 cm
ꢁ
1 1
f
; H NMR (400 MHz, CDCl ): d 13.70
3
(
(br, 1H), 13.44 (br, 1H), 8.31 (s, 1H), 8.27 (s, 1H), 7.33 (d, J ¼ 4 Hz, 1H),
7.31 (d, J ¼ 4 Hz, 1H), 7.30e7.25 (m, 2H), 7.21e7.16 (m, 3H), 7.05 (d,
J ¼ 4 Hz, 1H), 6.99 (d, J ¼ 4 Hz, 1H), 3.64 (d, J ¼ 16 Hz, 1H), 3.54 (d,
J ¼ 16 Hz, 1H), 3.39e3.23 (m, 2H), 2.83e2.76 (m, 2H), 2.62e2.54 (m,
1
8
1
7
8
604, 1474, 1363, 1292, 1220, 1154, 1132, 1114, 1006, 962, 930, 888,
ꢁ
1 1
46, 819, 754, 701, 605 cm ; H NMR (400 MHz, CDCl
H), 7.64 (d, J ¼ 2.4 Hz, 1H), 7.33 (d, J ¼ 4 Hz, 1H), 7.31e7.27 (m, 2H),
.22e7.18 (m, 3H), 3.75 (s, 2H), 2.84e2.77 (m, 4H), 2.68e2.59 (m,
H), 1.30 (s, 9H); 13C NMR (100 MHz, CDCl
): 191.8, 159.2, 141.9,
3
): d 10.34 (s,
10H), 1.98e1.82 (m, 4H), 1.79e1.64 (m, 2H), 1.51e1.42 (m, 2H), 1.41
13
3
d
(s, 9H), 1.24 (s, 9H), 1.23 (s, 9H); C NMR (100 MHz, CDCl
3
):
d
165.7,
1
5
40.1, 132.9, 128.7, 128.4, 126.1, 124.8, 122.8, 122.1, 60.2, 59.9, 52.9,
165.1, 158.0, 157.3, 140.5, 140.4, 140.0, 136.4, 130.7, 128.7, 128.3,
126.7, 126.0, 125.97, 117.84, 117.80, 72.8, 72.3, 60.6, 56.3, 53.2, 53.1,
35.0, 34.1, 33.8, 33.7, 33.4, 33.1, 31.44, 31.37, 29.4, 24.33, 24.27;
þ
2.6, 34.1, 33.6, 31.3; DART-TOF Calcd for C24
32
H N
2
O
2
(M þ H) :
3
81.2542; found 381.25444.

3126
R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
þ
þ
DART-TOF: Calcd for C45
H
64
N
4
O
2
(M þ H) : 693.51075; found
TOF: Calcd for C50
812.44539.
H
64CoN
4
O
2
(M þ H) : 812.44395; found
6
4
4
93.51178.
.6. Synthesis of sym-Schiff base ligands
4.7.3. Co-Schiff base complex 10
ꢀ
Drying under vacuum at 75 C for 4 h provided catalyst 10 as a
brick red solid in 45% yield. IR (neat): 2945, 2902, 2866, 2800, 1592,
1525,1426,1321,1253,1155, 1132,1043, 926, 873, 837, 787, 740, 695,
.6.1. General
Arylpiperazine salicylaldehyde (2 equivalents) was dissolved in
ꢁ
1
þ
anhydrous ethanol (with gentle heating for complete dissolution).
Compound 19 (1 equivalent) was added, and the reaction was
stirred for 24 h at room temperature. Upon completion, the reac-
tion mixture was concentrated under reduced pressure, taken up in
640 cm ; DART-TOF Calcd for C45H62CoN O (M þ H) : 750.42578;
4
2
found 750.42704.
4.8. Synthesis of symmetric Co-Schiff base complexes
4.8.1. General
dichloromethane, dried over anhydrous Na
under reduced pressure again.
2 4
SO , and concentrated
In
a pressure tube, equipped with a stirbar, symmetric
0
4
.6.2. 6,6'-((1E,1 E)-((((±)-trans)-cyclohexane-1,2-diyl)
arylpiperazine-tethered Schiff base ligand (1 equivalent) was dis-
solved in anhydrous methanol. Then, Co(OAc) 4H O (1 equivalent)
.
bis(azaneylylidene))bis(methaneylylidene))bis(2-((4-
2
2
benzhydrylpiperazin-1-yl)methyl)-4-(tert-butyl)phenol) (24)
Purification (silica gel; 12-50-100% EtOAc:DCM; see General
Information for column prep instructions), followed by drying
was dissolved in anhydrous methanol and added to the tube, the
cap was affixed, and the reaction was allowed to reflux for 4 h.
ꢀ
under vacuum at 70 C for 5 h, gave the final product as a yellow
4.8.2. Synthesis of Co-Schiff base complex 12
ꢀ
solid in 48% yield. mp: 135e138 C; R
(
1
f
¼ 0.12 (50% EtOAc:DCM); IR
Following reflux, the orange precipitate was collected via
neat): 2955, 2933, 2900, 2862, 2807, 2762, 1627, 1450, 1272, 1221,
136, 1076, 1006, 850, 825, 745, 704, 639 cm ; H NMR (400 MHz,
filtration and washed with cold methanol. Drying under vacuum at
ꢁ
1 1
ꢀ
6
5 C for 5 h provided catalyst 12 as an orange solid in 59% yield. IR
3
CDCl ): d 13.38 (br, 2H), 8.26 (s, 2H), 7.43e7.39 (m, 8H), 7.31 (d,
(
1
neat): 2959, 2810, 1614, 1594, 1537, 1451, 1326, 1265, 1222, 1150,
ꢁ1
J ¼ 4 Hz, 2H), 7.28e7.22 (m, 8H), 7.18e7.13 (m, 4H), 7.03 (d, J ¼ 4 Hz,
132, 1078, 1046, 1007, 937, 847, 749, 697, 641, 609 cm ; HRMS
þ.
2
3
H), 4.22 (s, 2H), 3.63 (d, J ¼ 16 Hz, 2H), 3.49 (d, J ¼ 16 Hz, 2H),
.31e3.26 (m, 2H), 2.53 (br, 8H), 2.44 (br, 8H), 1.93e1.82 (m, 4H),
(ESI-MS): Calcd for
C
64
H76CoN
6
O
2
(M) : 1019.5361; found
1019.5392.
13
1.74e1.63 (m, 2H), 1.49e1.40 (m, 2H), 1.21 (s, 18H). C NMR:
(
1
100 MHz, CDCl ): 165.0, 157.2, 143.0, 140.4, 130.7, 128.4, 127.9,
3
d
4
.8.3. Synthesis of Co-Schiff base catalyst 13
Following reflux and cooling to room temperature, the brick red
26.8, 124.5, 117.1, 76.3, 72.7, 56.3, 53.5, 52.0, 33.8, 33.3, 31.4, 24.2;
þ
DART-TOF: Calcd for C64
9
78
H N
6
O
2
(M þ H) : 963.62395; found
solution was concentrated under reduced pressure to obtain a
brown viscous liquid, which was taken up in Et O and concentrated
63.62520.
2
under reduced pressure once again. The brown solid was isolated
0
.6.3. 6,6'-((1E,1 E)-((((±)-trans)-cyclohexane-1,2-diyl)
bis(azaneylylidene))bis(methaneylylidene))bis(4-(tert-butyl)-2-((4-
phenethylpiperazin-1-yl)methyl)phenol) (25)
4
via filtration and washed with hexanes. Drying under vacuum at
ꢀ
6
5 C for 4 h provided catalyst 13 as a dark brown solid in 96% yield.
IR (neat): 2953, 2871, 1570, 1395, 1319, 1044, 699 cm-1; HRMS (ESI-
Purification (silica gel; 50e100% EtOAc:hexanes), followed by
MS): Calcd for C54H72CoN6O2 (M þ H)þ: 896.51024; found
ꢀ
drying under vacuum at 60 C for 20 h gave the final product as a
896.5110.
ꢀ
yellow solid in 93% yield. mp: 141e144 C; R
f
¼ 0.12 (100% EtOAc);
IR (neat): 2935, 2808, 1628, 1459, 1362, 1262, 1133, 1093, 1012, 818,
Acknowledgements
ꢁ1
1
8
8
6
2
01, 730, 698 cm
; H NMR (400 MHz, CDCl ): d 13.42 (br, 2H),
3
.29 (s, 2H), 7.35 (d, J ¼ 4 Hz, 2H), 7.30e7.25 (m, 4H), 7.20e7.16 (m,
H), 7.07 (d, J ¼ 4 Hz, 2H), 3.66 (d, J ¼ 16 Hz, 2H), 3.54 (d, J ¼ 12 Hz,
H), 3.33e3.28 (m, 2H), 2.83e2.77 (m, 4H), 2.63e2.54 (m, 20H),
We would like to thank the Center for Direct Catalytic Conver-
sion of Biomass to Biofuels (C3Bio), an Energy Frontier Research
Center (EFRC) that is funded by the United States Department of
Energy, Office of Science, Basic Energy Sciences, award DE-
SC0000997. We would also like to thank Mr. Ernesto C. Zuleta for
thoughtful discussion regarding the mechanistic pathways for the
phenolic oxidation and the UT Biological and Small Molecule Mass
Spectrometry Core for DART-TOF and HRMS measurements.
1
.94e1.83 (m, 4H), 1.74e1.65 (m, 2H), 1.50e1.40 (m, 2H), 1.25 (m,
13
18H); C NMR (100 MHz, CDCl
3
):
d 164.9, 157.2, 140.5, 140.4, 130.7,
1
3
28.7, 128.3, 126.7, 126.0, 124.3, 117.8, 72.7, 60.6, 56.2, 53.2, 53.1,
3.9, 33.7, 33.3, 31.4, 24.2; DART-TOF: Calcd for C54
H
74
N
6
O
2
þ
(
M þ H) : 839.59515; found 839.59710.
4.7. Synthesis of asymmetric Co-Schiff base complexes
Appendix A. Supplementary data
4
.7.1. General
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tet.2019.04.059.
In a pressure tube equipped with a stirbar, asymmetric
arylpiperazine-tethered Schiff base ligand (1 equivalent) was dis-
solved in anhydrous methanol. Then, Co(OAc).
4H O (1 equivalent)
was dissolved in anhydrous methanol and added to the tube. The
cap was affixed, and the reaction was heated at reflux for 4 h.
Following reflux, the precipitate was collected via filtration and
washed using cold methanol.
2
2
References
[1] R. Rinaldi, R. Jastrzebski, M.T. Clough, J. Ralph, M. Kennema, P.C.A. Bruijnincx,
B.M. Weckhuysen, Angew. Chem. Int. Ed. 55 (2016) 8164e8215.
[
[
2] O. Gordobil, R. Moriana, L.M. Zhang, J. Labidi, O. Sevastyanova, Ind. Crops Prod.
3 (2016) 155e165.
3] J.J. Bozell, J.E. Holladay, D. Johnson, J.F. White, Pacific Northwest, National
8
4
.7.2. Co-Schiff base complex 9
Laboratory, Richland, WA, 2007.
ꢀ
[4] R.E. Key, J.J. Bozell, ACS Sustain. Chem. Eng. 4 (2016) 5123e5135.
5] J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, B.M. Weckhuysen, Chem. Rev. 110
2010) 3552e3599.
[6] R. Vanholme, B. Demedts, K. Morreel, J. Ralph, W. Boerjan, Plant Physiol. 153
Drying under vacuum at 80 C for 4 h provided catalyst 9 as an
orange solid in 21% yield. IR (neat): 2951, 2904, 2867, 2809, 1594,
[
(
ꢁ1
1528, 1429, 1315, 1255, 1135, 1004, 837, 748, 705, 557 cm ; DART-

R.E. Key et al. / Tetrahedron 75 (2019) 3118e3127
3127
(
2010) 895e905.
[36] B.B. Corden, R.S. Drago, R.P. Perito, J. Am. Chem. Soc. 107 (1985) 2903e2907.
[37] G.T. Musie, M. Wei, B. Subramaniam, D.H. Busch, Inorg. Chem. 40 (2001)
3336e3341.
[
[
7] J.J. Bozell, C.J. O'Lenick, S. Warwick, J. Agric. Food Chem. 59 (2011) 9232e9242.
8] G. Vazquez, G. Antorrena, J. Gonzalez, S. Freire, J. Wood Chem. Technol. 17
(
1997) 147e162.
[38] E.E. Podlesny, P.J. Carroll, M.C. Kozlowski, Org. Lett. 14 (2012) 4862e4865.
[39] A. Nishinaga, H. Tomita, J. Mol. Catal. 7 (1980) 179e199.
[40] P.G. Cozzi, Chem. Soc. Rev. 33 (2004) 410e421.
[
9] P. Prinsen, J. Rencoret, A. Gutierrez, T. Liitia, T. Tamminen, J.L. Colodette,
M.A. Berbis, J. Jimenez-Barbero, A.T. Martinez, J.C. del Rio, Ind. Eng. Chem. Res.
5
2 (2013) 15702e15712.
[41] D. Chen, A.E. Martell, Y.Z. Sun, Inorg. Chem. 28 (1989) 2647e2652.
[42] H.M. van Dort, H.J. Geursen, Recl. Trav. Chim. Pays-Bas 86 (1967) 520e526.
[43] L.H. Vogt, J.G. Wirth, H.L. Finkbeiner, J. Org. Chem. 34 (1969) 273e277.
[44] D.L. Tomaja, L.H. Vogt, J.G. Wirth, J. Org. Chem. 35 (1970) 2029 (&).
[45] R.A. Sheldon, J.K. Kochi, Metal Catalyzed Oxidations of Organic Compounds,
Academic Press, New York, 1981.
[46] C. Floriani, F. Calderazzo, J. Chem. Soc. A (1969) 946e953.
[47] D. Chen, A.E. Martell, Inorg. Chem. 26 (1987) 1026e1030.
[48] A. Pui, J.P. Mahy, Polyhedron 26 (2007) 3143e3152.
[49] A. Pui, C. Policar, J.P. Mahy, Inorg. Chim. Acta 360 (2007) 2139e2144.
[50] P. Zanello, R. Cini, A. Cinquantini, P.L. Orioli, J. Chem. Soc. Dalton (1983)
2159e2166.
[
[
10] J.B. Li, G. Gellerstedt, K. Toven, Bioresour. Technol. 100 (2009) 2556e2561.
11] J. Tao, O. Hosseinaei, L. Delbeck, P. Kim, D.P. Harper, J.J. Bozell, T.G. Rials,
N. Labb eꢀ , RSC Adv. 6 (2016) 79228e79235.
[
12] A. Wu, B.O. Patrick, E. Chung, B.R. James, Dalton Trans. 41 (2012)
1
1093e11106.
13] J.M. Nichols, L.M. Bishop, R.G. Bergman, J.A. Ellman, J. Am. Chem. Soc. 132
2010) 12554e12555.
[
(
[
[
[
14] A.G. Sergeev, J.F. Hartwig, Science 332 (2011) 439e443.
15] J.Y. He, C. Zhao, J.A. Lercher, J. Am. Chem. Soc. 134 (2012) 20768e20775.
16] J. Zakzeski, P.C.A. Bruijnincx, B.M. Weckhuysen, Green Chem. 13 (2011)
6
71e680.
[
[
17] S.K. Hanson, R.T. Baker, Acc. Chem. Res. 48 (2015) 2037e2048.
18] A. Rahimi, A. Azarpira, H. Kim, J. Ralph, S.S. Stahl, J. Am. Chem. Soc. 135 (2013)
[51] C. Daul, J. Weber, Helv. Chim. Acta 65 (1982) 2486e2497.
[52] E.G. Samsel, J.K. Kochi, Inorg. Chem. 25 (1986) 2450e2457.
[53] F. Basolo, B.M. Hoffman, J.A. Ibers, Acc. Chem. Res. 8 (1975) 384e392.
[54] T. Elder, J.J. Bozell, D. Cedeno, Phys. Chem. Chem. Phys. 15 (2013) 7328e7337.
[55] P.J. Pospisil, D.H. Carsten, E.N. Jacobsen, Chem. Eur. J. 2 (1996) 974e980.
[56] W. Zhang, J.L. Loebach, S.R. Wilson, E.N. Jacobsen, J. Am. Chem. Soc. 112 (1990)
2801e2803.
[57] B. Gao, X. Li, R.L. Duan, Q. Duan, Y.H. Li, X. Pang, H.J. Zhuang, X.S. Chen, RSC
Adv. 5 (2015) 29412e29419.
[58] W. Chaładaj, P. Kwiatkowski, J. Jurczak, Tetrahedron Lett. 49 (2008)
6810e6811.
[59] M. Holbach, X.L. Zheng, C. Burd, C.W. Jones, M. Weck, J. Org. Chem. 71 (2006)
2903e2906.
[60] E.J. Campbell, S.T. Nguyen, Tetrahedron Lett. 42 (2001) 1221e1225.
[61] Zuleta, E.; Bozell, J.; Elder, T., manuscript in preparation.
[62] J.A. Howard, K.U. Ingold, Can. J. Chem. 41 (1963) 1744e1751.
[63] A. Borghese, R. Mer eꢀ nyi, J. Penelle, P. Pigeolet, H.G. Viehe, J. Prakt. Chem. 334
6415e6418.
[
[
[
[
19] B. Biannic, J. Bozell, J. Org. Lett. 15 (2013) 2730e2733.
20] B. Biannic, J.J. Bozell, T. Elder, Green Chem. 16 (2014) 3635e3642.
21] J.J. Bozell, Top. Curr. Chem. 353 (2014) 229e256.
22] M. Wang, L.H. Li, J.M. Lu, H.J. Li, X.C. Zhang, H.F. Liu, N.C. Luo, F. Wang, Green
Chem. 19 (2017) 702e706.
[
[
23] J.M. Pepper, H. Hibbert, J. Am. Chem. Soc. 70 (1948) 67e71.
24] S. Van den Bosch, T. Renders, S. Kennis, S.F. Koelewijn, G. Van den Bossche,
T. Vangeel, A. Deneyer, D. Depuydt, C.M. Courtin, J.M. Thevelein, W. Schutyser,
B.F. Sels, Green Chem. 19 (2017) 3313e3326.
[
25] T.H. Parsell, B.C. Owen, I. Klein, T.M. Jarrell, C.L. Marcum, L.J. Haupert,
L.M. Amundson, H.I. Kenttamaa, F. Ribeiro, J.T. Miller, M.M. Abu-Omar, Chem.
Sci. 4 (2013) 806e813.
[
[
[
26] H. Luo, I.M. Klein, Y. Jiang, H.Y. Zhu, B.Y. Liu, H.I. Kenttamaa, M.M. Abu-Omar,
ACS Sustain. Chem. Eng. 4 (2016) 2316e2322.
27] Q. Song, F. Wang, J.Y. Cai, Y.H. Wang, J.J. Zhang, W.Q. Yu, J. Xu, Energy Environ.
Sci. 6 (2013) 994e1007.
28] A.J. Ragauskas, G.T. Beckham, M.J. Biddy, R. Chandra, F. Chen, M.F. Davis,
B.H. Davison, R.A. Dixon, P. Gilna, M. Keller, P. Langan, A.K. Naskar,
J.N. Saddler, T.J. Tschaplinski, G.A. Tuskan, C.E. Wyman, Science 344 (2014)
(1992) 563e569.
[64] E.F. Caldin, S.P. Dagnall, M.K.S. Mak, D.N. Brooke, Faraday Discuss. Chem. Soc.
74 (1982) 215e228.
[65] F.G. Bordwell, J. Cheng, J. Am. Chem. Soc. 113 (1991) 1736e1743.
[66] M.M. Suryan, S.A. Kafafi, S.E. Stein, J. Am. Chem. Soc. 111 (1989) 4594e4600.
[67] A. Nishinaga, T. Kondo, T. Matsuura, Chem. Lett. 14 (1985) 905e908.
[68] S.K. Huntley, D. Ellis, M. Gilbert, C. Chapple, S.D. Mansfield, J. Agric. Food.
Chem. 51 (2003) 6178e6183.
7
09e719.
29] D.R. Robert, M. Bardet, G. Gellerstedt, E.L. Lindfors, J. Wood Chem. Technol. 4
1984) 239e263.
[
(
[
[
[
[
[
[
30] J.J. Bozell, B.R. Hames, D.R. Dimmel, J. Org. Chem. 60 (1995) 2398e2404.
31] D. Cedeno, J.J. Bozell, Tetrahedron Lett. 53 (2012) 2380e2383.
32] I. Sasaki, D. Pujol, A. Gaudemer, Inorg. Chim. Acta 134 (1987) 53e57.
33] A. Pui, C. Dobrota, J.-P. Mahy, J. Coord. Chem. 60 (2007) 581e595.
34] A. Pui, I. Berdan, A. Cascaval, Rev. Roum. Chem. 45 (2000) 331e335.
35] A. Zombeck, R.S. Drago, B.B. Corden, J.H. Gaul, J. Am. Chem. Soc. 103 (1981)
[69] D.E. Bergbreiter, P.L. Osburn, C.M. Li, Org. Lett. 4 (2002) 737e740.
[70] P. Meier, F. Broghammer, K. Latendorf, G. Rauhut, R. Peters, Molecules 17
(2012) 7121e7150.
[71] Q. Wang, C. Wilson, A.J. Blake, S.R. Collinson, P.A. Tasker, M. Schroder, Tet-
rahedron Lett. 47 (2006) 8983e8987.
[72] M.W. Fennie, E.F. DiMauro, E.M. O'Brien, V. Annamalai, M.C. Kozlowski, Tet-
rahedron 61 (2005) 6249e6265.
7580e7585.