Lignin Depolymerization with Nitrate-Intercalated Hydrotalcite Catalysts

Article abstract of DOI:10.1021/acscatal.5b02062

Hydrotalcites (HTCs) exhibit multiple adjustable parameters to tune catalytic activity, including interlayer anion composition, metal hydroxide layer composition, and catalyst preparation methods. Here, we report the influence of several of these parameters on β-O-4 bond scission in a lignin model dimer, 2-phenoxy-1-phenethanol (PE), to yield phenol and acetophenone. We find that the presence of both basic and NO₃^- anions in the interlayer increases the catalyst activity by 2-3-fold. In contrast, other anions or transition metals do not enhance catalytic activity in comparison to blank HTC. The catalyst is not active for C-C bond cleavage on lignin model dimers and has no effect on dimers without an α-OH group. Most importantly, the catalyst is highly active in the depolymerization of two process-relevant lignin substrates, producing a significant amount of low-molecular-weight aromatic species. The catalyst can be recycled until the NO₃^- anions are depleted, after which the activity can be restored by replenishing the NO₃^- reservoir and regenerating the hydrated HTC structure. These results demonstrate a route to selective lignin depolymerization in a heterogeneous system with an inexpensive, earth-abundant, commercially relevant, and easily regenerated catalyst.

Full text of DOI:10.1021/acscatal.5b02062

Research Article
pubs.acs.org/acscatalysis
Lignin Depolymerization with Nitrate-Intercalated Hydrotalcite
Catalysts
Jacob S. Kruger,† Nicholas S. Cleveland,† Shuting Zhang,† Rui Katahira,† Brenna A. Black,†
Gina M. Chupka,‡ Tijs Lammens,§ Phillip G. Hamilton,∥ Mary J. Biddy,*,† and Gregg T. Beckham*,†
‡
†National Bioenergy Center and Hydrogen and Transportation Systems Center, National Renewable Energy Laboratory, 15013
Denver West Parkway, Golden, Colorado 80401, United States
§Shell Global Solutions, Inc., Shell Technology Center, Houston, Texas 77082, United States
∥Shell Global Solutions, Inc., Shell Technology Center, Amsterdam, Netherlands
S
* Supporting Information
ABSTRACT: Hydrotalcites (HTCs) exhibit multiple adjustable parameters to tune catalytic activity, including interlayer anion
composition, metal hydroxide layer composition, and catalyst preparation methods. Here, we report the influence of several of
these parameters on β-O-4 bond scission in a lignin model dimer, 2-phenoxy-1-phenethanol (PE), to yield phenol and
acetophenone. We find that the presence of both basic and NO3− anions in the interlayer increases the catalyst activity by 2−3-
fold. In contrast, other anions or transition metals do not enhance catalytic activity in comparison to blank HTC. The catalyst is
not active for C−C bond cleavage on lignin model dimers and has no effect on dimers without an α-OH group. Most
importantly, the catalyst is highly active in the depolymerization of two process-relevant lignin substrates, producing a significant
amount of low-molecular-weight aromatic species. The catalyst can be recycled until the NO3− anions are depleted, after which
−
the activity can be restored by replenishing the NO3 reservoir and regenerating the hydrated HTC structure. These results
demonstrate a route to selective lignin depolymerization in a heterogeneous system with an inexpensive, earth-abundant,
commercially relevant, and easily regenerated catalyst.
KEYWORDS: lignin valorization, lignin nitration, layered double hydroxide, catalyst recycle, catalyst regeneration
of aromatic molecules, making subsequent separation and
upgrading a challenge to produce fungible chemical feedstocks
in an economically viable manner.2,8
INTRODUCTION
■
Lignin is a primary component of terrestrial plant cell walls
responsible for structure, defense, and water transport.1,2
During lignin biosynthesis, the three primary lignin monomers
(p-coumaryl, coniferyl, and sinapyl alcohols; Scheme 1) are
connected through a variety of alkyl, aryl, and ether linkages,
likely via oxidative cross-coupling reactions.3 Coumaric and
ferulic acid derivatives connected through ester linkages can
also account for a significant fraction of lignin mass, especially
in grasses.4 The most common linkage in lignin, and one of the
most labile, is the β-O-4 ether formed between the β-carbon of
the propenyl group on one monomer and the hydroxyl oxygen
and 4-carbon of a second monomer.5−7 The highly aromatic
structure of lignin is seemingly conducive to production of
aromatic-based chemicals and high-octane fuel additives.
However, the diversity of the interaromatic ring linkages
presents a considerable technical barrier to the selective
depolymerization of lignin. Moreover, depolymerization of
natural lignin almost invariably produces a heterogeneous slate
In current biorefinery designs for lignocellulosic ethanol,
most byproduct lignin is slated to be used for process heat or
for cofiring in power plants.9,10 Going forward, however,
strategies for selective lignin depolymerization will be required
for upgrading lignin to higher-value end products, which is of
paramount importance for advanced biofuels processes.11
A
number of thermochemical routes have been explored for lignin
depolymerization, and low-temperature catalytic approaches
that preserve the aromatic ring structure and minimize
repolymerization reactions are especially promising. In this
vein, several catalytic scenarios have been explored.
Received: September 15, 2015
Revised: January 7, 2016
© XXXX American Chemical Society
1316
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
Scheme 1. Three Primary Lignin Monomers in Terrestrial
Plants
Scheme 2. General Layered Double-Hydroxide Structure,
with Edge-Sharing Mg(OH)6 Octahedra (Blue), in Which
Some of the Mg Atoms Are Replaced by Al (Red)
a
a
a
Between M(OH)2 layers are H2O molecules and anions, e.g., OH−.
a
The aromatic ring substructure (without the hydroxypropenyl group)
is shown on the second line.
determined, but the catalyst was hypothesized to operate
analogously to a homogeneous base catalyst with enhanced
redox activity from the deposited Ni.58 Given the need to
increase catalyst activity for lignin depolymerization and to
understand catalyst recyclability and regeneration, we were
motivated by these initial experiments to pursue a deeper
understanding of catalyst activity and reaction parameters by
investigating a series of HTC catalysts prepared by different
methods. For the purposes of these experiments, we have
focused on commercially relevant catalyst preparation methods,
such as incipient wetness deposition and ion-exchange
techniques. Additionally, we focused on a lignin model
compound to help elucidate the catalyst mechanism and
subsequently examined the catalytic activity toward several
other lignin model dimers and two process-relevant lignin
streams. As described below, we find that the key component
for catalytic activity is nitrate anions intercalated in the
interlayer space. The nitrate anions are active even in the
absence of transition metals but may benefit from neighboring
Heterogeneous metal catalysts and H2 have been employed
for many decades to simultaneously depolymerize and
hydrogenate lignin via hydrogenolysis.1,12−21 A few types of
solid catalysts that do not require high-pressure H2 for lignin
depolymerization have been recently reported, including a
limited number of Pd/C22 and Ni/C23 catalysts. In these
examples, the catalysts either require addition of a hydrogen
donor22,24 or are hypothesized to generate hydrogen in situ
from the solvent (methanol).23,25 Alternatively, some hetero-
geneous metal oxides can catalyze lignin depolymerization in
26,27
the presence of O2
or H2O2.28,29 Additionally, homoge-
neous metal catalysts, such as complexes and salts of Co, Mn,
and Ru, are able to depolymerize lignin and lignin model
compounds.1,30−41 A few metal-free approaches using homoge-
neous oxidants have been recently developed.42,43
Finally, alkaline catalysts, such as NaOH, have been
employed by Shabtai et al.,44−50 Roberts et al.,51 and others52,53
in a “base-catalyzed depolymerization (BCD)” scheme. BCD
effectively depolymerizes lignin to monomers under optimized
conditions, producing more than 85 wt % yield as an oil
comprised of low-molecular-weight species (monomers,
dimers, trimers, and borate capping agent) when repolymer-
izaition inhibitors are added. Some solid base catalysts have
been explored, including metal-doped porous metal oxides
derived from calcined hydrotalcites (HTCs)54−57 and metal-
doped, uncalcined HTCs.58 These heterogeneous base catalysts
are particularly advantageous because they are derived from
earth-abundant materials and do not generate high-salt-content
wastewater.
HTCs belong to a class of layered double-hydroxide (LDH)
minerals that exhibit activity as solid base catalysts. The
structure of these catalysts is derived from a Mg(OH)2
foundation, in which a plane of Mg atoms is situated between
two planes of OH groups to form a brucite layer. These brucite
layers are stacked, each separated by a layer of H2O molecules
in the interlayer space. The Mg atoms can be substituted by
other di- and trivalent atoms, such as Al. If the substituting
atom is trivalent, a positive charge is introduced into the brucite
layer, which is balanced by an anion in the interlayer space, as
illustrated in Scheme 2. A large number of materials exist with
varying types and ratios of metal atoms and charge-
compensating anions.59 In addition to their inherent catalytic
activity as solid bases, LDH materials can serve as supports for
other catalysts.60
2−
basic anions, such as CO3 and OH−. The catalytic activity is
heterogeneous and may result from a nitration of a benzylic
hydroxyl group. These results suggest a direct means to
synthesize, regenerate, and recycle nitrate-HTC catalysts for
lignin depolymerization applications. Furthermore, the catalyst
is active only when an α-OH group is present on the substrate,
and it does not break C−C bonds. Nonetheless, the catalyst is
effective for generating compounds with molecular weights
indicative of monomeric and dimeric species from real lignin
streams.
RESULTS
■
Metal and Precursor Screening. The initial set of
experimental studies focused on screening the activity of a set
of HTC catalysts with a model lignin dimer. To explore the
generality of the activity previously observed with Ni-HTC
catalysts,58 we conducted a survey of metal-loaded HTC-rh
catalysts, varying the transition metal, metal oxidation state, and
precursor salt. To gain further understanding of the roles of the
metals and anions, we also explored a series of metal-free, ion-
exchanged HTC as catalysts. The general preparation
procedure for these catalysts is shown in Scheme 3 (as = as-
2−
synthesized, with CO3 anions; ca = calcined, mixed-oxide
form, without brucite layers or interlayer anions; rh =
rehydrated, with OH− anions). Details of the materials used
in these experiments and methods for model compound and
catalyst syntheses are given in the Experimental Section and in
the Supporting Information.
The catalysts were screened in the conversion of a lignin
model compound, 2-phenoxy-1-phenethanol (PE), to phenol,
In our previous work, we reported on a Ni-HTC catalyst that
was active for β-O-4 bond cleavage in a lignin model dimer and
that reduced the molecular weight of two insoluble lignin
streams.58 The active site in the Ni-HTC catalysts was not
1317
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
anions (Figure S2 in the Supporting Information), which
showed a moderate decrease in phenol yield in comparison to
the partially exchanged HTC-as(NO3), as shown in Figure 1D.
The latter catalyst still shows relatively high activity in
comparison to HTC-rh and HTC-as, likely due to “XRD-
invisible” basic anions still present in nitrate-dominated catalyst
domains.62 Indeed, benzoic acid titration of the nitrate ion-
exchanged catalysts shows a slightly elevated content of basic
sites relative to HTC-as (Table S1 in the Supporting
Information).
Scheme 3. Catalyst Preparation Diagram and Naming
Conventions for (A) M-HTC-rh Catalysts and (B) Ion-
Exchanged HTC Catalysts
a
Remarkably, the difference in basic site strength between Ni-
HTC-as (with carbonates as basic anions) and hydroxide-form
Ni-HTC-rh (with hydroxides as basic anions) appears to make
little difference in the catalytic activity. We observe little change
in activity at 1 h reaction time between Ni-HTC-rh and Ni-
HTC-as (Figure S3 in the Supporting Information) or between
Ni-HTC-rh prepared without atmospheric exposure and the
same catalyst exposed to the atmosphere for 24 h, despite
significant differences in basic site concentration measured by
benzoic acid titration (Figures S4 and S5 and Table S1 in the
a
For M(A)x·yH2O in R-OH, M is a generic metal, A is a generic anion,
x = 2, 3, 0 ≤ y ≤ 6, and R = CH3CH2, CH3, H.
acetophenone, and 1-phenylethanol via β-O-4 bond cleavage, as
shown in Scheme 4. In all cases, the reaction solvent is methyl
isobutyl ketone, MIBK. Error bars in all figures represent the
standard deviation of at least two replicate experiments.
2−
Supporting Information), suggesting that CO3 is a strong
enough base to serve the catalytic function.
Recycle and Regeneration. The critical function of the
nitrate anion, along with the general phenomenon that nitrates
are converted to other species in reactions in which they
participate, suggests that the activity of these materials should
be maintained until the nitrate reservoir on the catalyst is
depleted and that activity can likely be restored by replenishing
the nitrate reservoir. To test this hypothesis, we performed a
recycle and regeneration study. As illustrated in Figure 2, the
catalyst activity of Ni-HTC-rh is maintained through two
cycles, after which activity decreases. The loss in activity is
accompanied by a loss of the nitrate-intercalated domains and a
partial loss of the HTC structure, as shown by XRD (Figure S6
in the Supporting Information). The partial loss of the HTC
structure is not surprising, as loss of interlayer H2O and partial
dehydroxylation of the brucite layers is known to occur below
300 °C.63 Attempts to restore the HTC structure and NO3− by
direct rehydration and direct ion exchange of the used catalyst
were unsuccessful, possibly due to a minor amount of
carbonaceous material deposited on the catalyst surface.
However, calcining the catalyst to fully produce the mixed-
oxide state and remove any carbonaceous deposits, followed by
rehydration and redeposition of an ethanolic Ni(NO3)2
solution, restored catalyst activity. The regeneration protocol
is described in the Experimental Section below.
Effect of Solvent and Substrate on Catalyst Activity.
In previous work, we investigated PE conversion in methyl
isobutyl ketone (MIBK) as a representative model system for
some organosolv pulping processes.64−67 However, with a basic
catalyst such as HTC, a carbonyl-containing solvent such as
MIBK can be deprotonated at the α-position from the ketone,
resulting in side products and solvent degradation. Thus, we
were motivated to find an alternative solvent to MIBK for lignin
depolymerization. We screened 10 additional solvents,
ultimately selecting 3-methyl-3-pentanol (3M3P) as an optimal
solvent to maximize catalyst activity and selectivity, solvent
stability, substrate solubility, and handling convenience. Further
details of the screening are given in Figures S7−S9 in the
Supporting Information and the accompanying text.
Scheme 4. Reaction of Lignin Model Compound 2-Phenoxy-
1-phenethanol (PE) to Phenol, Acetophenone, and 1-
Phenylethanol
a
a
The A and B ring and inter-ring carbon labeling convention for lignin
model dimers is shown for PE.
The results of the metal precursor screening are shown in
Figure 1A, represented by phenol yield. PE conversion and
acetophenone yield show similar trends (Figure S1 in the
Supporting Information). These results demonstrate that,
regardless of metal, all catalysts prepared from nitrate precursor
salts exhibit high activity, whereas those prepared from chloride
salts are no more active, or in some instances are less active,
than the blank HTC-rh.
To evaluate the impact that the salt precursor has on the
overall catalyst activity and gain insight into whether the nitrate
salts were unique in their high activity, we prepared a series of
Ni-HTC-rh catalysts from acetate, formate, and sulfate
precursors. As summarized in Figure 1B, the other precursor
salts also produce catalysts that are no more active, or are less
active, than the blank HTC-rh. However, a metal-free, ion-
2−
exchanged HTC-as, in which the CO3 anions were replaced
by OH− and NO3 anions (HTC-as(NO3)), is as active as the
−
Ni-loaded catalyst. Characterization of these catalysts by XRD
shows unique structural changes in the catalysts prepared from
nitrate precursors. In particular, catalysts prepared from nitrate
precursors display additional peaks at 2θ = 20, 38, 43° (Figure
1C), which is characteristic of HTC domains with intercalated
NO3 ions.61,62
−
These results demonstrate that a nitrate precursor is key to
enhancing catalyst activity beyond blank HTC. However,
retention of some basic anions appears to be beneficial for the
desired reaction. We prepared a fully nitrate exchanged catalyst
(HTC-as(Cl→NO3), with no significant XRD peak for basic
Similarly, although PE is a convenient lignin model
compound due to its ease of synthesis and simplified β-O-4
linkage, realistic lignin substrates will exhibit a more
1318
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
Figure 1. (A) Phenol yield (mol/mol of PE) in the metal-loaded HTC-rh screening study. The horizontal axis shows metal precursor salts. (B)
Phenol yield from PE over Ni-HTC-rh catalysts prepared from different precursor salts and a nitrate ion exchanged catalyst prepared from HTC-as.
(C) XRD traces for selected catalysts, showing additional peaks from nitrate intercalation. (D) Phenol yield from PE over ion-exchanged HTC
catalysts.
dimers using 3M3P as solvent. The additional model
compounds were selected to represent variations on the β-O-
4 linkage and the other predominant linkages found in lignin.
The model compounds selected were 1-phenyl-2-phenoxy-1,3-
propanediol (PPPD), 1-(3-methoxyphenyl)ethylene glycol-β-
guaiacyl ether (MGE), veratrylethylene glycol-β-guaiacyl ether
(VG), guaiacylglycerol-β-guaiacyl ether (GG), guaiacylethylene
glycol-β-guaiacyl ether (GG2), 1,2-diphenylethanol (DPEt), 4′-
benzyloxy-3′-methoxyacetophenone (BMA), diphenyl ether
(DPE), and biphenol (BP).
β-O-4 Dimers. The model dimers containing a β-O-4 linkage
and their reaction products are shown in Scheme 5 and Table 1.
The presence of substituents on the rings and inter-ring linkage
strongly influences the substrate activity and catalyst effect. In
particular, on comparison of PE with PPPD, and of GG2 with
GG, it is apparent that the presence of a CH2OH group
pendant on Cβ increases the reactivity of the substrate without
any catalyst present and results in a significant fraction of
unknown products. Additionally, on comparison of PE with VG
and MGE, the presence of OCH3 substituents on the rings
appears to open a second reaction pathway that produces a
compound with molecular weight and mass spectrometric
fragmentation pattern consistent with a dimeric alkene
(product 4 in Scheme 5), rather than an acyl-aryl ketone and
a phenolic compound from the A ring and B ring, respectively.
Fragmentation patterns of these compounds are shown in
Figure 2. Phenol yield (mol/(mol of PE)) during recycling and
regeneration of the Ni-HTC-rh catalyst, with regeneration by
calcination, rehydration, and redeposition of nitrate after cycle 3.
The regeneration protocol includes calcining at 450 °C, rehydrating,
and redepositing Ni(NO3)2 as an ethanolic solution, as described in
the Experimental Section.
heterogeneous suite of linkages and monomers. To further
explore the activity of nitrated HTC catalysts, we applied the
same reaction conditions used for PE to other lignin model
1319
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
Scheme 5. Reaction of β-O-4 Model Dimers over HTC
Table 1. Reactions of β-O-4 Model Dimers over HTC
a
b
name
PE
R1
R2
R3
R4
catalyst
no cat.
conv of 1
2
3
4
unknown
H
H
H
H
22.0
49.3
81.9
82.9
2.6
24.9
66.1
73.2
3.7
28.5
72.4
86.2
0.0
0.0
0.0
0.0
18.8
22.6
12.6
3.2
HTC-rh
Ni-HTC-rh
HTC-as(NO3)
PPPD
MGE
VG
H
H
CH2OH
H
no cat.
65.5
76.4
82.0
81.1
0.0
12.0
13.5
14.1
0.0
14.9
31.8
30.9
0.0
0.0
0.0
0.0
65.5
62.9
59.4
58.6
HTC-rh
Ni-HTC-rh
HTC-as(NO3)
H
OCH3
OCH3
OCH3
H
OCH3
OCH3
OCH3
OCH3
no cat.
38.7
45.0
45.7
55.0
7.3
30.5
33.3
47.1
12.5
28.5
24.9
35.5
14.4
13.4
13.1
10.6
14.3
2.1
HTC-rh
Ni-HTC-rh
HTC-as(NO3)
3.5
3.1
OCH3
OH
OH
H
no cat.
44.7
47.6
47.6
47.5
5.7
8.7
6.4
11.6
11.3
12.0
c
c
c
c
c
c
c
c
HTC-rh
Ni-HTC-rh
HTC-as(NO3)
10.2
10.7
GG
CH2OH
no cat.
93.8
83.8
89.7
83.3
0.0
0.0
0.0
0.0
55.9
39.2
35.7
56.2
8.0
57.8
37.0
52.8
39.3
HTC-rh
27.2
19.0
15.9
Ni-HTC-rh
HTC-as(NO3)
GG2
OCH3
H
no cat.
34.7
90.0
77.0
89.9
0.0
0.0
0.0
0.0
27.4
33.8
27.4
48.9
8.0
12.9
45.9
44.3
49.5
HTC-rh
Ni-HTC-rh
27.2
19.0
15.9
HTC-as(NO3)
a
b
c
Alkene quantified as stilbene. Calculated as conversn − 4 − (2 + 3)/2. VG decomposes to the alkene in the GC, precluding quantification of that
and unknown products.
Figures S10−S13 in the Supporting Information. Finally, on
comparison of PE with GG2 and PPPD with GG, the presence
of a phenolic OH on the A ring appears to preclude the
formation of the acyl-aryl ketone, instead forming an
unidentified product from the A ring. Although the apparent
catalyst activity is not as high for the other substrates as for PE,
there is a clear effect on all of the substrates except VG. That is,
PPPD and MGE produce significantly more of the A-ring
ketone and B-ring phenol in the presence of HTC, while GG
and GG2 produce significantly more of the alkene product in
the presence of HTC. Additionally, conversion of MGE and
GG2 are significantly higher over HTC. In general, the HTC
catalysts are able to activate substrates that contain β-O-4
linkages, but the monomer type and yields depend strongly on
the substituents of the substrate.
substrates were significantly activated by HTC. BMA was
reactive even in the absence of catalyst and produced ∼15%
yield each of toluene and acetovanillone. DPE showed ∼10%
conversion, but no reaction products could be identified, and
BP showed no detectable conversion. Thus, it appears that the
HTC catalysts are not able to activate α-O-4, 4-O-5, or 5-5
linkages, at least when Cα on the α-O-4 substrate has no other
substituents. It is worth noting that all of the activated
substrates, i.e., the β-O-4 models and DPEt, have a benzylic
OH group. Thus, it may be the case that the α-OH group is the
key reactive site over the HTC catalysts.
Depolymerization of Lignin-Enriched Substrates.
Most importantly, we explored the activity of nitrate-
intercalated HTC catalysts toward two process-relevant
lignin-enriched streams produced at the pilot scale: a
deacetylated, disk-refined, enzymatically hydrolyzed lignin
from corn stover (DDE lignin)68 and a dilute acid pretreated,
enzymatically hydrolyzed lignin from corn stover (DAP
lignin).69 The isolation procedures are given in the
Experimental Section, and the composition of these lignin-
enriched streams is shown in Table 3.
Other Model Dimers. We also examined HTC activity for
four additional dimers representing other common linkages in
lignin. These dimers and their reaction products are shown in
Scheme 6 and Table 2. DPEt (the model β-1 linkage) readily
underwent dehydration to produce trans-stilbene over the
nitrate-intercalated HTC catalysts, but none of the other
1320
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
model dimers, significant depolymerization occurs. Some
monomeric and dimeric species are produced without catalyst
and with HTC-rh, but the fraction is significantly increased with
HTC-as(NO3) in both 3M3P and H2O and with Ni-HTC-rh in
water. As shown in Figure 3, the most prominent peaks after
acetylation correspond to apparent molecular weights of 200−
300 Da. Thus, nitrate-intercalated catalysts are highly active for
depolymerization of biomass-derived lignin. For reference, Mw,
Mn, and Mp are reported in Figure S14 in the Supporting
Information.
Scheme 6. Reactions of β-1, α-O-4, 4-O-5, and 5-5 Model
Dimers over HTC
Yields of Detectable Monomers. We also attempted to
identify and quantify a subset of the monomeric species
produced by the HTC catalysts. The main detected monomeric
products from both DAP and DDE lignins in H2O were
phenol, guaiacol, and syringol, with smaller amounts of the
corresponding acyl ketones (acetosyringone and acetovanil-
lone) and minute amounts of the corresponding aldehydes
(syringaldehyde and vanillin). When the aqueous reaction
products were run on HPLC without extraction by dichloro-
methane (DCM), small amounts of 4-hydroxy-benzaldehyde
and 4-hydroxyacetophenone could also be detected. In total,
monomer yields detected by GC were 3.5−4.5% without
catalyst and ∼7% with catalyst, referenced to the lignin mass
content in the feedstock, as shown in Figure 4A,B.
Table 2. Reactions of β-1, α-O-4, 4-O-5, and 5-5 Model
Dimers over HTC
In contrast, the major product from both substrates in 3M3P
was 4-vinylphenol (4-VP; Figure 4C,D, and Figures S15 and
S16 in the Supporting Information), which is produced in 4−7
wt % yields from DDE lignin and 4−5 wt % yields from DAP
lignin. Minor products included guaiacol, syringol, 4-ethyl-
phenol, 4-vinylguaiacol, isoeugenol, acetovanillone, and allyl
syringol. Total quantified monomer yields over the HTC
catalysts were 5−6 wt % from DAP lignin and 7−9 wt % from
DDE lignin; the corresponding selectivity to 4-VP was 76−80%
from DAP lignin and 60−76% from DDE lignin.
Yields of the detected monomeric products are lower than
expected from the GPC traces, especially for the cases with
aqueous solvent. The monomers suggested to be present by
GPC, but not quantified by GC, could not be conclusively
determined. There were no large unidentified peaks on the GC,
and standards of several aromatic acids were run on both GC-
MS and LC-MS to match retention times and MS
fragmentation patterns. Similarly, because the lignin streams
under evaluation contain significant fractions of residual
carbohydrates, sugar-derived monomeric species, such as 5-
hydroxymethyl furfural, furfural, other furan derivatives, and
aliphatic acids, were run (Table S2 in the Supporting
Information), but could not be matched, by LC.
conv
name
DPEt
type
β-1
catalyst
no cat.
(%)
2
3
unknown
10.0
28.3
100
1.2
21.1
100
100
8.8
7.2
0.0
0.0
HTC-rh
Ni-HTC-rh
HTC-as(NO3)
100
BMA
DPE
BP
α-O-4
4-O-5
5-5
no cat.
85.0
75.7
79.6
73.3
13.8
12.4
16.1
11.7
14.1
71.9
59.9
66.6
59.1
HTC-rh
15.4
14.2
14.4
Ni-HTC-rh
HTC-as(NO3)
no cat.
7.8
10.0
12.0
9.9
7.8
10.0
12.0
9.9
HTC-rh
Ni-HTC-rh
HTC-as(NO3)
no cat.
0.0
0.0
0.0
0.0
HTC-rh
Ni-HTC-rh
HTC-as(NO3)
Table 3. Compositional Analysis of Lignin-Enriched Streams
composition (%)
DISCUSSION AND CONCLUSIONS
■
substrate lignin glucan xylan galactan arabinan acetate protein
Cost-effective lignin valorization strategies to chemicals and
fuels in a biorefinery will rely on the development of cheap,
recyclable catalysts. We previously reported that Ni-HTC
catalysts are effective for depolymerization of insoluble lignin
and active for β-O-4 bond scission in a lignin model
compound.58 Catalysts synthesized from HTC are advanta-
geous because they are heterogeneous, generally nontoxic, and
inexpensive. Thus, we were motivated to understand the source
of activity and principle of operation for the Ni-HTC catalyst in
order to develop a regeneration protocol and improve activity
and economics over our initial baseline. The correlation of high
catalyst activity with structural changes to the catalyst upon
nitrate intercalation suggests that the nitrate is acting
heterogeneously. To test this hypothesis, we examined the
DDE
DAP
35.3
61.3
25.4
13.0
20.8
1.8
1.6
0.0
3.5
0.4
1.1
0.8
6.3
6.0
We evaluated depolymerization of DDE and DAP corn
stover lignins in four catalytic scenarios (no catalyst, HTC-rh,
Ni-HTC-rh, and HTC-as(NO3)) and in two solvents (H2O and
3M3P). Despite finding that H2O was not a high-performing
solvent for β-O-4 bond cleavage in the model dimer PE, we
were still interested in exploring aqueous lignin depolymeriza-
tion over HTC to facilitate integration into upstream and
downstream biorefining processes.
Gel Permeation Chromatography. When these two lignin
streams are subjected to the same reaction conditions as for the
1321
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
Figure 3. GPC traces for the THF-soluble portion of DAP and DDE lignins before and after reaction in 3M3P and H2O solvents. Lignins were
acetylated for GPC analysis.
Figure 4. Yields of DCM-extracted monomers from DDE and DAP lignins reacted over HTC catalysts in H2O solvent (A, B) and yields of
monomers over HTC catalysts in 3M3P solvent (C, D).
potential contribution of homogeneous chemistry to our
observed results. We performed a hot filtration test to ascertain
if the reaction continues after the catalyst is removed, and
control experiments with homogeneous Ni(NO3)2·6H2O,
1322
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
NaNO3, and Na2CO3 to test if the homogeneous anions have
activity when not associated with the solid HTC structure. The
reaction stops when the catalyst is removed from the solution
(Figures S17 and S18 in the Supporting Information), and
homogeneous Ni, NO3 , and CO3 do not show significant
catalytic activity (Figure S19 in the Supporting Information).
Thus, the nitrate activity is heterogeneous.
under the same conditions, but no phenol was formed. Thus,
ring nitration appears not to be the primary pathway, which
leads us to tentatively suggest nitration of the benzylic OH
group as a key step in the mechanism. The mechanism will be
explored in more detail in subsequent studies.
−
2−
Another important result of these experiments is that
precautions against atmospheric exposure, a concern for basic
catalysts originating from HTC,84−86 appear unnecessary for β-
There are several possible mechanisms by which heteroge-
neous nitrate may function:
−
O-4 bond scission in the presence of NO3 anions. Thus, as-
synthesized HTC, which is produced commercially, can be used
for β-O-4 bond scission after loading with NO3 . Furthermore,
• increased accessibility to basic sites, due to larger
interlayer spacing
−
we have shown that HTC-based catalysts can be prepared by
multiple methods, including simple precipitation−deposition
and ion-exchange techniques, both of which are scalable to an
industrial level. Additionally, it is possible to directly synthesize
NO3-HTC,62 which should also find limited barriers to scale-
up.
Regarding the results described for lignin-enriched biorefi-
nery substrates, the formation of 4-VP in 3M3P suggests that
the catalyst is also active for ester linkages, as the primary
source of 4-VP is likely p-coumaric acid. p-Coumaric acid is
known81,82 to decarboxylate to 4-VP readily at temperatures
below 200 °C, and the decarboxylation is enhanced by basic
conditions.83 Coumarate ester linkages are often considered to
be the bridge between lignin and hemicellulose but may also
take the place of the analogous alcohol monomer building
blocks (Scheme 1), especially in herbaceous biomass. As noted
by Ralph, coumarate-type monomers can account for up to
18% of biomass in corn stover, often as pendant groups on S-
type lignin.4 Thus, it may be possible to significantly improve
yields of 4-VP and/or p-coumaric acid under optimized
conditions.
Quantified monomer yields from two process-relevant lignin
streams reach 7 wt % in H2O solvent, although GPC suggests
that actual monomer yields may be much higher. Similarly, in
3M3P, monomer yields, comprised mainly of 4-VP, con-
servatively reach 5−9 wt %. While these yields are lower than
those reported in some other catalytic systems, they are not
optimized. However, they are achieved using a catalyst that is
inexpensive, easily regenerated, does not require pressurized
H2, and does not generate high-salt-content wastewater.
Furthermore, the selectivity to 4-VP in 3M3P solvent suggests
an opportunity to improve yields to 4-VP and/or p-coumaric
acid, which will facilitate integration of the depolymerization
step into downstream upgrading opportunities.8,87−89
• introduction of an easily displaced anion, which allows
the substrate access to basic catalytic sites and/or
coordination to the metal ions in the brucite layers
• oxidation of PE alcohol to a PE ketone intermediate,
which has a lower Cβ−O bond strength
• initiation of a radical mechanism
• nitration of the PE molecule, facilitating decomposition
While we have not conclusively determined the mechanism,
we have performed some preliminary experiments that suggest
that some of these functions are not active. First, the solely base
catalyzed mechanism for PE decomposition is well-known and
includes 1-phenylethylene glycol (PhEG) as an intermedi-
ate.70−72 However, when PhEG is used as a starting material
with these catalysts, little acetophenone is produced (Figure
S20 in the Supporting Information), suggesting that the
mechanism is not purely base catalyzed and, hence, easier
access to the basic sites is not the only role of the nitrate.
Similarly, while the nitrate ion is one of the most easily
displaced anions from HTC,61 if catalyst activity depended on
nitrate displacement by a PE molecule, intermediate activity
should be observed over the chloride-exchanged HTC, as Cl− is
61,73,74
−
easier to displace than CO32−, but not as easy as NO3 .
Figure 1D shows that the chloride-exchanged catalyst is actually
less active than the carbonate form.
Second, if oxidation of PE to PE ketone were part of the
reaction pathway, PE ketone should be observable as an
intermediate unless it is rapidly consumed. However, kinetic
measurements show that PE ketone is never present in
significant quantities (Figure S3 in the Supporting Informa-
tion), and when PE ketone is used as a starting material,
conversion after 1 h is only ∼50% (Figure S21 in the
Supporting Information). The kinetics are not limited by
external mass transfer (Figure S22 in the Supporting
Information), which, together with the other PE ketone results,
suggests that PE ketone is not an intermediate, rapidly
consumed or otherwise. Additionally, we observed no
significant decrease in activity upon adding the radical quencher
BHT to the reaction, suggesting that the primary reaction
pathway does not involve radical chemistry (Figure S23 in the
Supporting Information).
In summary, HTC-based materials are effective catalysts for
lignin depolymerization: namely, for cleavage of β-O-4 bonds
and ester linkages. We show here that HTCs containing
intercalated NO3− anions increase activity by 2−3-fold for some
model compounds but that some accompanying basicity is
required for maximum activity. Both the strongly basic OH−
2−
and the weakly basic CO3 appear to be active for base
catalysis under the conditions studied. Quantified monomer
yields from two process-relevant lignin streams are around 7−9
wt % in aqueous solvent and 5−9 wt % in 3M3P. Simple
replenishment of the NO3 “reservoir” on the catalyst and
regeneration of the HTC structure from the partially
dehydrated state is effective in restoring catalyst activity.
Finally, the mechanism most consistent with the results
involves nitration. While nitration from nitrate typically occurs
under conditions much different from those in the current
experiments,75−77 our reaction temperatures may be sufficient
to decompose the nitrate into NO and NO2 species,78,79 which
would be more likely to interact with the substrates.
Furthermore, Lindeberg and Walding found that nitration on
the B ring of a lignin model increased the reaction rate of base-
catalyzed β-O-4 cleavage by more than 3 orders of magnitude
and resulted in nitrophenols as reaction intermediates.80 We
reacted 2-nitrophenol, 4-nitrophenol, and 2,4-dinitrophenol
EXPERIMENTAL SECTION
■
Materials. All materials were used as received. Acetone
(HPLC grade), methanol (laboratory grade), diethyl ether
(99.5%), NaOH (99%), NaCl (99.9%), hexanes (ACS reagent
1323
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
grade), ethyl acetate (99.9%), (R)-(−)-1,2-diphenylethanol,
and methyl isobutyl ketone (MIBK, 4-methyl-2-pentanone,
reagent grade) were purchased from Fisher Scientific. Ethanol
(200 proof) was purchased from Pharmco-AAPER. 2-
Bromoacetophenone (98%), phenol (99%), K2CO3 (99%),
KI (99%), hydrotalcite (HTC), acetone (HPLC grade), THF
(anhydrous), methyl isobutyl ketone (MIBK, HPLC grade),
NaNO3 (99%), Ni(NO3)2·6H2O (19−21% Ni), Cr(NO3)3·
9H2O (99%), K2CrO4 (99%), NaH (60% in mineral oil),
butylated hydroxytoluene (BHT, 99%), VCl2 (95%), Ga-
(NO3)3·6H2O (99.5%), NiCl2·6H2O (98%), Ni(CH3COO)2·
4H2O (98%), 2,2′-biphenol (99%), heptane (99%), i-PrOH
(99.5%, HPLC grade), acetonitrile (99.8%), 2EH (99+%),
diphenyl ether (ReagentPlus, 99%), and 3M3P (≥99%) were
purchased from Sigma-Aldrich. NaBH4 (99%), Mn(NO3)2·
4H2O (99%), Fe(NO3)3·9H2O (98%), CrCl2 (97%), Cu-
(NO3)2·6H2O (99%), Zn(NO3)2·6H2O (99%), Ni(HCOO)2·
2H2O (98%), and BMA (98%) were purchased from Alfa
Aesar. MgSO4 (99%), HNO3 (69%), HCl (37%), Co(NO3)2·
6H2O (98%), Na2CO3 (anhydrous), and toluene (99.8%) were
purchased from J. T. Baker. VCl3 (95%) was purchased from
Strem Chemicals. t-BuOH (99.5%) was purchased from Acros
Organics. GG (97%) was purchased from TCI America. GG2
(purity >95% verified by NMR) was purchased from Synthons,
Inc.
DDE lignin was produced from 25−50 mm particles of corn
stover. The biomass was deacetylated in 0.1 M NaOH (8 wt %
total solids) at 80 °C for 2 h, then rinsed with 45 °C water until
the pH was 8.5, and dried under flowing air. In some cases, an
acid treatment step can be incorporated after deacetylation to
prevent microbial growth, but it is not necessary. In those cases,
the acid is typically H2SO4 at 0.8 wt % concentration (10 wt %
with respect to the lignin), and the temperature can range from
room temperature to 160 °C for up to 2 h. The resulting
material was disk-refined to 1−5 mm particles, but other
mechanical refining processes have a similar effect. The refined
material was enzymatically hydrolyzed at 15 wt % solids and 50
°C for 72 h in pH 5.1 NH4OH-citrate buffer using Novozymes
CTEC3 (cellulase) and HTEC3 (xylanase) enzymes at 60 mg
of protein/g of cellulose and 40 mg protein/g of cellulose,
respectively. The lignin content of the starting material was
about 15.2 wt %, and of the isolated DDE lignin was 35.3 wt %.
DAP lignin was produced at NREL from corn stover at 8 wt
% total solids in 0.8 wt % H2SO4 at 160 °C for 10 min and then
neutralized to pH 5.1 with NaOH and enzymatically hydro-
lyzed under the same conditions as for the DDE lignin. The
lignin content of the starting material was about 15.2 wt % and
of the isolated DAP lignin was 61.3 wt %.
Diethyl ether was removed by evaporation to yield PE in 85−
1
90% mass yield. H and 13C-NMR spectra were acquired at 25
°C on a Bruker AVANCE 400 MHz spectrometer equipped
with a 5 mm BBO probe. Tetramethylsilane was used as a
reference. Data for PE: 1H NMR (CDCl3) δ 3.90 (1H, s, OH),
4.05 (1H, d, J = 2.96, Hβ), 4.07 (1H, s, Hβ), 5.07 (1H, dd, J =
4.84, 7.00, Hα), 6.90−7.45 (10H, m, aromatic H).
1-(3-Methoxyphenyl) Ethylene Glycol-β-guaiacyl Ether
(MGE). To a solution of guaiacol (1.92 g, 15.5 mmol) in
acetone (50 ml), 3′-methoxy-2-bromooacetophenone (2.96 g,
12.9 mmol), K2CO3 (2.67 g, 19.4 mmol) and KI (0.43 g, 2.58
mmol) were added. The solution was refluxed overnight. The
reaction mixture was filtered and the filtrate was concentrated
and recrystallized from ethyl acetate/n-hexane (1:2, v/v) to give
1-(3-methoxylphenyl)-2-(2-methoxyphenoxy)ethanone
(74.7%). To the ethanone derivative (0.98 g, 3.58 mmol) in
methanol (10 mL), NaBH4 (0.27 g, 7.15 mmol) was added
gradually. The mixture was stirred at room temperature for 3 h.
The reaction mixture was quenched by adding 10 mL of 10%
acetic acid, extracted three times with 50 mL of ethyl acetate,
washed with brine and dried over Na2SO4. Ethyl acetate was
removed by evaporation to yield MGE (99 %). MGE: 1H-NMR
(CDCl3): δ 3.58 (1H, s, OH), 3.81 (3H, s, OCH3), 3.88 (3H, s,
OCH3), 3.97 (1H, t, J = 9.6, Hγ), 3.97 (1H, t, J = 9.6, Hγ), 4.18
(1H, dd, J = 3.2, 10, Hβ), 3.97 (1H, t, J = 9.6, Hγ), 5.08 (1H, dd,
2.8, 9.2, Hα), 6.83−6.99 (8H, m, aromatic H). 13C-NMR
(CDCl3): δ 55.4 (OCH3), 56.1 (OCH3), 72.5 (Cα), 76.5 (Cβ),
111.9 (C2), 112.3 (C4), 113.9 (C3′), 116.2 (C6′), 118.8 (C6),
121.3 (C4′), 122.8 (C5′), 129.7 (C5), 141.4 (C1), 148.2 (C1′),
150.3 (C2′), 160.0 (C3).
Veratryl Ethylene Glycol-β-guaiacyl Ether (VG). To a
solution of guaiacol (0.70 g, 5.67 mmol) in acetone (20 ml),
3′,4′-dimethoxy-2-bromooacetophenone (1.0 g, 3.86 mmol),
K2CO3 (0.78 g, 5.67 mmol) and KI (0.062 g, 0.38 mmol) were
added. The solution was refluxed overnight. The reaction
mixture was filtered. Filtrate was diluted with deionized H2O
(50 mL) and then extracted two times with ethyl acetate (50
mL). Combined organic layer was washed with brine, dried
over Na2SO4, and evaporated, giving a viscous syrup. The
crystallization was conducted from ethyl acetate/n-hexane
(10:1, v/v) to give 1-(3,4-dimethoxylphenyl)-2-(2-
methoxyphenoxy)ethanone (78.9 %). To the ethanone
derivative (0.92 g, 3.0 mmol) in methanol/ethyl acetate (1:2,
v/v, 15 mL), NaBH4 (0.46 g, 12.1 mmol) was added gradually.
The mixture was stirred at room temperature for 2 h. The
reaction mixture was quenched by adding 13 mL of 10% acetic
acid, extracted three times with 40 mL of ethyl acetate, washed
with brine and dried over Na2SO4. Ethyl acetate was removed
1
Model Compound Synthesis. 2-Phenoxy-1-phenyletha-
nol (PE). 2-Phenoxy-1-phenylethanol (PE) was synthesized in a
two-step process as previously described.58 First, 2-bromoace-
tophenone (11.94 g, 60 mmol), phenol (7.06 g, 75 mmol),
K2CO3 (12.30 g, 89 mmol), KI (catalytic), and acetone (250
mL) were heated at reflux overnight, filtered, concentrated, and
recrystallized from cold ethanol (250 mL) to give 2-phenoxy-1-
phenethanone (PE ketone) in 85% mass yield. Second, PE
ketone (1.11 g, 5.2 mmol) was dissolved in methanol (35 mL),
and NaBH4 (0.35 g, 10.4 mmol) was added gradually, after
which the mixture was stirred at room temperature for 2 h to
reduce the ketone to the alcohol. The reaction mixture was
quenched by adding 30 mL of saturated aqueous NH4Cl,
extracted three times with 50 mL of diethyl ether, washed with
50 mL of saturated NaCl, dried over Na2SO4, and filtered.
by evaporation to yield VG (97.5 %). VG: H-NMR (CDCl3):
δ 3.84 (3H, s,OCH3), 3.86 (3H, s, OCH3), 3.88 (3H, s,
OCH3), 3.98 (1H, t, J = 9.6, Hβ), 4.14 (1H, dd, J = 2.8, 9.6,
Hβ), 5.05 (1H, dd, 2.8, 9.2, Hα), 6.83−7.01 (7H, m, aromatic
H). 13C-NMR (CDCl3): δ 55.9 (OCH3), 55.97 (OCH3), 56.04
(OCH3), 72.2 (Cα), 76.3 (Cβ), 109.6 (C2), 111.2 (C3′), 112.1
(C5), 115.9 (C6′), 118.7 (C6), 121.2 (C4′), 122.5 (C5′), 132.5
(C1), 148.2 (C1′), 148.9 (C4),149.2 (C2′), 150.1 (C3).
1-Phenyl-2-phenoxy-1,3-propanediol (PPPD). To a solu-
tion of 1-phenyl-2-phenoxyethanone (PE-ketone, 1.98 g, 9.33
mmol) in acetone (50 mL), formaldehyde (36 wt % aqueous
solution), 0.765 mL, 10 mmol) and K2CO3 (0.84 g, 6.08
mmol) were added. The solution was stirred at room
temperature for 20 h. After reaction, the reaction mixture was
filtered. Solvent in the filtrate was removed. The crude reaction
1324
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
mixture was purified by column chromatography using 4:1
hexanes:ethyl acetate as eluent. Purified PPPD was recovered in
68% yield. PPPD (Erythro/threo mixture): 1H-NMR (CDCl3):
δ 2.92 (1H, s,OH), 3.56 (1H, dd, J = 4.0, 12.4, Hβ), 3.84 (1H,
dd, J = 4.0, 12.0, Hβ), 4.42 (1H, m, Hβ), 5.04 (1H, d, J = 6.4,
Hα), 6.96−7.43 (10H, m, aromatic H). 13C-NMR (CDCl3): δ
61.3 (Cγ), 74.2 (Cα), 83.2 (Cβ), 116.8, 122.2, 126.5, 127.1,
128.5, 128.7, 129.8, 129.9, 139.9, 158.3 (Arom-C). H-NMR
spectra of PE, MGE, VG, and PPPD are shown in Figures S24,
S25, S26,and S27, respectively.
Catalyst Preparation. The base case catalyst was prepared
by calcining commercial HTC (HTC-as) at 450 °C for 16 h to
generate the mixed-oxide form (HTC-ca). The calcined form
was rehydrated by stirring in deionized water for 1.5 h and
sonicating in the same water for 5 min to generate the
meixnerite form (HTC-rh). A solution of Ni(NO3)2·6H2O in
ethanol was added to give a nominal Ni loading of 5 wt %; the
slurry was stirred for 5 min and dried overnight to give Ni-
HTC-rh. In this preparation, all calcining, drying, and storage of
prepared catalyst were done in an open atmosphere with no
effort to exclude CO2. For comparison, uncalcined HTC was
washed with hexanes to remove residual carbonaceous material
from the manufacturing process and loaded with 5 wt % Ni;
this catalyst is labeled Ni-HTC-as. Other metal catalysts were
prepared as for Ni-HTC-rh, substituting the appropriate
precursor for Ni(NO3)2·6H2O. Because anhydrous VCl2,
VCl3, and CrCl2 were not soluble in ethanol, these were
prepared from aqueous solution.
glass vials that were then purged with N2 and sealed with a
rubber septum and Parafilm to study the adsorption of CO2
onto the nickel-loaded, rehydrated form as a function of
atmospheric exposure time. These samples are labeled Ni-
HTC-xh, where x is the number of hours exposed to the
atmosphere.
To understand the role of the anion, Ni-HTC-rh catalysts
were prepared with different Ni precursor salts. NiCl2·6H2O
and NiSO4·6H2O were prepared from ethanolic solutions as for
the Ni(NO3)2·6H2O precursor. Due to the low solubility of
Ni(HCOO)2·2H2O and Ni(CH3COO)2·4H2O in ethanol, the
Ni(HCOO)2·2H2O catalyst (NiFm2-HTC-rh) was prepared
from an aqueous precursor solution and the Ni(CH3COO)2·
4H2O catalyst (NiAc2-HTC-rh) was prepared from a
methanolic solution.
Catalyst Screening. Catalyst activity was evaluated by
reacting a solution of 33 mM PE in MIBK over the different
catalysts at 275 °C. A 3 mL portion of the PE/MIBK was
loaded into stainless steel batch reactors (nominal volume
capacity 3 mL), 40 mg of catalyst was added, and the reactors
were sealed and weighed. With Ni-HTC catalysts, this loading
scheme gives a Ni:PE mass ratio of 0.1. The reactors were then
placed in a steel holder inside a wire mesh basket and immersed
in a fluidized sand bath (Techne SBL-2D). Reactors were
removed from the sand bath at the desired time and quenched
in water. After reaction, the reactors were weighed to ensure
their mass had not decreased during the reaction and the
contents were analyzed by GC-FID and GC-MS as described
below. Other model dimers and the lignin streams were run
under the same conditions, except that in cases of limited
solubility the solid material and 3 mL of solvent were added
separately to give the equivalent of a 33 mM concentration (for
dimers) or 20 mg of substrate (for lignin).
1
A second set of catalysts was prepared by ion exchange
methods, following the general approach of Iyi et al.73,74 Briefly,
a known amount of HTC-as was immersed in a solution of acid
(HCl or HNO3) and sodium salt (NaCl or NaNO3). The acid
concentration was such that the molar ratio H+:CO3 was 2,
2−
and the salt concentration was generally >2 M. To obtain an
XRD-pure Cl−-form catalyst, two steps with concentrated NaCl
were required; the second step did not incorporate acid. From
the Cl−-exchanged catalyst, an additional ion exchange was
performed with NaNO3 to generate a fully nitrate exchanged
catalyst.
As the fluid dynamics inside the reactors while in the sand
bath were unknown, a separate set of experiments was
conducted to evaluate potential mass transfer limitations. In
these experiments, the reactors were placed inside a hot, sand-
filled cup and the contents stirred at a controlled rate using a
Parr 5000 MRS apparatus. From these experiments, an
“equivalent minimum stir rate” for the fluidized sand bath
could be obtained.
The potential for catalyst leaching and homogeneous
chemistry was evaluated by running the reaction with PE
with inline gas filters as reaction vessels, which were reacted
with controlled orientation and flipped at a certain reaction
time to hot-filter the catalyst from the reaction solution. These
reactors contain a cup-shaped 0.5 μm filter basket that could
effectively retain catalyst particles while allowing hot solvent to
pass through. Due to the potential for the gas-phase reaction of
PE with the separated catalyst, these reactors were flipped while
hot to filter the catalyst from the solution, the reaction mixtures
were quenched, and the catalyst was physically removed from
the reactor. The reaction solution was then allowed to react
further in the absence of catalyst. Although these reactors had a
nominal volume of 12 mL, the heating curves for these reactors
were similar to those for the 3 mL reactors, generally reaching
the reaction temperature (275 °C) from room temperature
within 7 min.
Two series of catalysts were prepared by varying the time
exposed to the atmosphere. For the first set, commercial HTC
was calcined in air at 450 °C for 16 h to generate the mixed-
oxide form and removed from the furnace hot (∼250 °C). This
HTC-ca was placed in a round-bottom flask and immediately
flushed with N2. Deionized water was added to the flask, and
the slurry was stirred for 1 h under flowing N2, the flask was
sealed with a rubber septum, and the slurry was sonicated for 5
min. This catalyst was filtered by vacuum filtration under a N2
atmosphere and dried overnight under flowing N2. When dry,
this catalyst was transferred to an open-top vessel exposed to
air. One gram samples were taken periodically from this HTC-
rh, placed in glass vials that were then purged with N2 and
sealed with a rubber septum and Parafilm to study the
adsorption of CO2 onto the rehydrated form as a function of
atmospheric exposure time. These samples are labeled HTC-rh-
xh, where x is the number of hours exposed to the atmosphere.
For the second set, HTC-rh was prepared without exposure
to CO2 as described above for the first set, and a solution of
Ni(NO3)2·6H2O in ethanol was added to give a nominal Ni
loading of 5 wt %. The slurry was stirred for 5 min and dried
overnight under flowing N2 to give the Ni-HTC. When dry, this
catalyst was transferred to an open-top vessel exposed to air.
One-gram samples were taken from this Ni-HTC and placed in
Catalyst Recycle and Regeneration. For recycle studies,
larger Parr batch reactors (Parr Series 5000 Multiple Reactor
System) were used to more efficiently generate sufficient
catalyst for characterization after each reaction cycle. These
reactions were scaled up by a factor of 10 from those for the 3
1325
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
mL reactors, and the contents were stirred at 400 rpm. After
each reaction cycle, the catalyst from each reactor was
combined to produce a pool of catalyst samples from which
samples were withdrawn for characterization; the remaining
catalyst was distributed to reactors for the next reaction cycle.
For catalyst regeneration, the spent catalyst was calcined for 16
h at 450 °C and then rehydrated and reloaded with an ethanolic
solution of Ni(NO3)2 by the same procedure used to generate
fresh Ni-HTC-rh.
Product Analysis. For dimers except GG, GG2, PPPD, and
VG, starting materials and reaction products were analyzed by
GC-FID and GC-MS on an HP-5MS capillary column. For
GC-FID, the injection volume was 5 μL, the split ratio was 8:1,
and the oven program was 100 °C for 3 min, ramp at 20 °C/
min to 160 °C, hold for 1.5 min, ramp at 20 °C/min to 250 °C,
and hold for 4.3 min. For GC-MS, the column was the same,
the injection volume was 1 μL, the split ratio was 25:1, and the
oven program was 100 °C for 1 min, ramp at 10 °C/min to 114
°C, hold for 0 min, ramp at 20 °C/min to 160 °C, hold for 1.5
min, ramp at 20 °C/min to 250 °C, and hold for 4.3 min.
For benzoic acid titration experiments, the GC column was
the same as above, the injection volume was 1 μL, the system
was operated as splitless with the purge valve open at 0.2 min,
and the oven program was the same as for GC-FID analysis
above.
For DPE, DPEt, BP, and BMA, a second method was
developed. The method was the same as the GC-MS method
described above, except the oven program was 35 °C for 3.5
min, ramp at 10 °C/min to 114 °C, hold for 0 min, ramp at 20
°C/min to 160 °C, hold for 1.5 min, ramp at 20 °C/min to 250
°C, and hold for 4.3 min.
For GG, GG2, PPPD, and VG, the starting material was
quantified by HPLC, and products were quantified by GC-FID.
For GC analysis, the injection volume was 1 μL, the split ratio
was 10:1, the system was operated under constant flow mode of
1 mL/min, and the oven program was 50 °C for 1 min, ramp at
10 °C/min to 250 °C, hold for 0 min, ramp at 25 °C/min to
300 °C, and hold for 5 min.
For HPLC, analysis of samples was performed on an Agilent
1100 LC system equipped with a G1315B diode array detector
(DAD) and an Ion Trap SL (Agilent Technologies, Palo Alto,
CA) mass spectrometer (MS) with in-line electrospray
ionization (ESI). Each sample was injected undiluted at a
volume of 50 μL into the LC/MS system.
Phenolic standard compounds were separated using reverse-
phase chromatography on an YMC C30 Carotenoid 0.3 μm,
4.6 × 150 mm column (YMC America, Allentown, PA). The
chromatography consisted of a solvent regime of water
modified with 0.03% formic acid (eluent A) and 9:1 acetonitrile
and water also modified with 0.03% formic acid (eluent B) at a
constant oven temperature of 30 °C and a solvent flow rate of
0.7 mL min−1, with a gradient as follows: 0−3 min, 0% B; 16
min, 7% B; 21 min, 8.5% B; 34 min, 10% B; 46 min, 25% B;
51−54 min, 30% B; 61 min, 50% B; and last 64−75 min, 100%
B before equilibrium.
Flow from the HPLC-DAD was directly routed to the ESI-
MS ion trap. The DAD was used to monitor chromatography at
210 nm for quantitation and a direct comparison to MS for
identification data. Source and ion trap conditions were
calibrated with Agilent ESI-T tuning mix (P/N:G2431A),
while tuning parameters were optimized under negative ion
mode by direct infusion of standards for major contributing
compounds. MS and MS/MS tuned parameters are as follows:
smart parameter setting with target mass set to 165 Da,
compound stability 10%, trap drive 50%, capillary at 3500 V,
fragmentation amplitude of 0.75 V with a 30−200% ramped
voltage implemented for 50 ms, and an isolation width of m/z 2
(He collision gas). The ESI nebulizer gas was set to 60 psi, with
a dry gas flow of 11 L min−1 held at 350 °C. MS scans and
precursor isolation−fragmentation scans were performed across
the range of 40−350 Da. This program was used to identify
monomers in lignin solutions and to quantify both identified
monomers and the model dimer GG.
Gel Permeation Chromatography (GPC) Analysis. Each
substrate and reaction product (20 mg) was acetylated in a
mixture of pyridine (0.5 mL) and acetic anhydride (0.5 mL) at
40 °C for 24 h with stirring. The reaction was terminated by
addition of methanol (0.2 mL). The acetylation solvents were
then evaporated from the samples at room temperature under a
stream of nitrogen gas. Addition of methanol and nitrogen
flushing were repeated until the vapors from pyridine and acetic
anhydride had dissipated. The samples were further dried in a
vacuum oven at 40 °C overnight. The dried, acetylated lignin
samples were dissolved in tetrahydrofuran (THF, Baker HPLC
grade). The dissolved samples were filtered (0.45 μm nylon
membrane syringe filters) before GPC analysis. The acetylated
samples were completely soluble in THF. GPC analysis was
performed using an Agilent HPLC with 3 GPC columns
(Agilent, PLgel, 300 × 7.5 mm) packed with polystyrene−
divinylbenzene copolymer gel (10 μm beads) having nominal
pore diameters of 104, 103, and 50 Å. The eluent was THF and
the flow rate 1.0 mL/min. An injection volume of 25 μL was
used. The HPLC was attached to a diode array detector
measuring absorbance at 260 nm (bandwidth 80 nm).
Retention time was converted into molecular weight (MW)
by applying a calibration curve established using polystyrene
standards of known molecular weight (1 × 106 to 580 Da) plus
toluene (92 Da).
Catalyst Characterization. Basic sites were quantified by
benzoic acid titration using a modified procedure of Parida and
Das.90 Briefly, 0.1 g of catalyst was stirred for 1.5 h in a solution
of benzoic acid in hexane, with benzoic acid concentrations
ranging from 0.5 to 50 mM. A time of 1.5 h was found to be
sufficient for equilibration of adsorption. The change in benzoic
acid concentration was measured by GC-FID, and the number
of basic sites was determined from the intercept of the
linearized Langmuir equation
C
x
1
αxm
C
xm
=
+
where C is the concentration of benzoic acid in solution (mol/
L), x is the amount of benzoic acid adsorbed (mol/gcat), xm is
the number of basic sites (mol/gcat), and α is a constant.
Catalysts were also characterized by XRD, ICP, and N2
physisorption. XRD was performed on a Rigaku Ultima IV X-
ray diffraction system using Cu Kα radiation, with operating
voltage and current of 40 kV and 44 mA, respectively, scan
speed of 5°/min, and point spacing of 0.02°. N2 physisorption
was carried out with a Quantichrome Quantisorb SI four-
station instrument, with outgassing at 110 °C for 2 h before
analysis, and and equilibration time of 30 s for both adsorption
and desorption branches. ICP analysis was conducted on a
Spectro Acros FHS12 instrument at a plasma power of 1425 W.
Ni was analyzed using the 231.6 nm line.
1326
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
Renewable Energy Laboratory (NREL): Golden, CO, 2013; No.
NREL/TP-5100-60223.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b02062.
■
S
(12) Wang, X.; Rinaldi, R. ChemSusChem 2012, 5, 1455−1466.
(13) Zaheer, M.; Hermannsdorfer, J.; Kretschmer, W. P.; Motz, G.;
̈
Kempe, R. ChemCatChem 2014, 6, 91−95.
(14) Song, Q.; Wang, F.; Xu, J. Chem. Commun. 2012, 48, 7019−
7021.
Accompanying conversion and yield data for Figure 1,
experimental data of reactions with homogeneous
catalysts, radical inhibitors, varying basic site concen-
trations, and nitrite-intercalated HTC, correlations of
phenol yield with solvent parameters, conversions and
yields from BMA, DPE, and PhEG, identification of 4-
VP, monomer candidates not detected by GC, stir rate
experiments, kinetic measurements, XRD traces of
selected HTC catalysts, nitrate ester decomposition
pathways, and catalyst characterization data (PDF)
(15) He, J.; Zhao, C.; Lercher, J. A. J. Am. Chem. Soc. 2012, 134,
20768−20775.
(16) He, J.; Zhao, C.; Mei, D.; Lercher, J. A. J. Catal. 2014, 309, 280−
290.
(17) Harris, E. E.; D’Ianni, J.; Adkins, H. J. Am. Chem. Soc. 1938, 60,
1467−1470.
(18) Pepper, J. M.; Hibbert, H. J. Am. Chem. Soc. 1948, 70, 67−71.
(19) Brewer, C. P.; Cooke, L. M.; Hibbert, H. J. Am. Chem. Soc. 1948,
70, 57−59.
(20) Van den Bosch, S.; Schutyser, W.; Vanholme, R.; Driessen, T.;
Koelewijn, S.-F.; Renders, T.; De Meester, B.; Huijgen, W.; Dehaen,
W.; Courtin, C. Energy Environ. Sci. 2015, 8, 1748−1763.
(21) Parsell, T.; Yohe, S.; Degenstein, J.; Jarrell, T.; Klein, I.; Gencer,
E.; Hewetson, B.; Hurt, M.; Kim, J. I.; Choudhari, H.; Saha, B.; Meilan,
R.; Mosier, N.; Ribeiro, F.; Delgass, W. N.; Chapple, C.; Kenttamaa, H.
I.; Agrawal, R.; Abu-Omar, M. M. Green Chem. 2015, 17, 1492−1499.
(22) Galkin, M. V.; Sawadjoon, S.; Rohde, V.; Dawange, M.; Samec,
J. S. M. ChemCatChem 2014, 6, 179−184.
AUTHOR INFORMATION
Corresponding Authors
*M.J.B.: fax, +01-303-384-6363; tel, +01-303-384-7806; e-mail,
mary.biddy@nrel.gov.
*G.T.B.: e-mail, gregg.beckham@nrel.gov.
Notes
The authors declare no competing financial interest.
■
(23) Song, Q.; Wang, F.; Cai, J.; Wang, Y.; Zhang, J.; Yu, W.; Xu, J.
Energy Environ. Sci. 2013, 6, 994−1007.
ACKNOWLEDGMENTS
(24) Wang, X.; Rinaldi, R. Angew. Chem., Int. Ed. 2013, 52, 11499−
11503.
(25) Klein, I.; Saha, B.; Abu-Omar, M. M. Catal. Sci. Technol. 2015, 5,
■
This work was funded by NREL CRADA 13-513 with Shell
Global Solutions, Inc., and by the U.S. Department of Energy
Bioenergy Technologies Office. The authors are grateful to Dr.
Steven Chmely for synthesis of deuterated PE, to Kelsey
Ramirez for assistance with HPLC analysis, to Melvin Tucker,
Michael Resch, Xiaowen Chen, and Erik Kuhn for preparation
of lignin substrates (DDE and DAP), and to Seonah Kim and
Roberto Rinaldi for discussions regarding potential mecha-
nisms.
3242−3245.
(26) Zhang, J.; Deng, H.; Lin, L. Molecules 2009, 14, 2747−2757.
(27) Sales, F. G.; Maranhao, L. C. A.; Filho, N. M. L.; Abreu, C. A. M.
̃
Chem. Eng. Sci. 2007, 62, 5386−5391.
(28) Crestini, C.; Pro, P.; Neri, V.; Saladino, R. Bioorg. Med. Chem.
2005, 13, 2569−2578.
(29) Crestini, C.; Caponi, M. C.; Argyropoulos, D. S.; Saladino, R.
Bioorg. Med. Chem. 2006, 14, 5292−5302.
(30) Biannic, B.; Bozell, J. J. Org. Lett. 2013, 15, 2730−2733.
(31) Lu, F.; Ralph, J. J. Agric. Food Chem. 1997, 45, 4655−4660.
(32) Partenheimer, W. Adv. Synth. Catal. 2009, 351, 456−466.
(33) Nichols, J. M.; Bishop, L. M.; Bergman, R. G.; Ellman, J. A. J.
Am. Chem. Soc. 2010, 132, 12554−12555.
REFERENCES
■
(1) Zakzeski, J.; Bruijnincx, P. C. A.; Jongerius, A. L.; Weckhuysen, B.
M. Chem. Rev. 2010, 110, 3552−3599.
(2) Ragauskas, A. J.; Beckham, G. T.; Biddy, M. J.; Chandra, R.;
Chen, F.; Davis, M. F.; Davison, B. H.; Dixon, R. A.; Gilna, P.; Keller,
M.; Langan, P.; Naskar, A. K.; Saddler, J. N.; Tschaplinski, T. J.;
Tuskan, G. A.; Wyman, C. E. Science 2014, 344, 1246843.
(3) Boerjan, W.; Ralph, J.; Baucher, M. Annu. Rev. Plant Biol. 2003,
54, 519−546.
(34) vom Stein, T.; Weigand, T.; Merkens, C.; Klankermayer, J.;
Leitner, W. ChemCatChem 2013, 5, 439−441.
(35) Nagy, M.; David, K.; Britovsek, G. J. P.; Ragauskas, A. J.
Holzforschung 2009, 63, 513−520.
(36) Chen; Capanema, E. A.; Gracz, H. S. J. Agric. Food Chem. 2003,
51, 1932−1941.
(4) Ralph, J. Phytochem. Rev. 2010, 9, 65−83.
(37) Son, S.; Toste, F. D. Angew. Chem., Int. Ed. 2010, 49, 3791−
3794.
(5) Azadi, P.; Inderwildi, O. R.; Farnood, R.; King, D. A. Renewable
Sustainable Energy Rev. 2013, 21, 506−523.
(38) Sedai, B.; Díaz-Urrutia, C.; Baker, R. T.; Wu, R.; Silks, L. A. P.;
Hanson, S. K. ACS Catal. 2013, 3, 3111−3122.
(39) Harms, R. G.; Markovits, I. I. E.; Drees, M.; Herrmann, h. c. m.
(6) Parthasarathi, R.; Romero, R. A.; Redondo, A.; Gnanakaran, S. J.
Phys. Chem. Lett. 2011, 2, 2660−2666.
(7) Kim, S.; Chmely, S. C.; Nimlos, M. R.; Bomble, Y. J.; Foust, T.
D.; Paton, R. S.; Beckham, G. T. J. Phys. Chem. Lett. 2011, 2, 2846−
2852.
W. A.; Cokoja, M.; Kuhn, F. E. ChemSusChem 2014, 7, 429−434.
̈
(40) Nguyen, J. D.; Matsuura, B. S.; Stephenson, C. R. J. J. Am. Chem.
Soc. 2014, 136, 1218−1221.
(8) Linger, J. G.; Vardon, D. R.; Guarnieri, M. T.; Karp, E. M.;
Hunsinger, G. B.; Franden, M. A.; Johnson, C. W.; Chupka, G.;
Strathmann, T. J.; Pienkos, P. T. Proc. Natl. Acad. Sci. U. S. A. 2014,
111, 12013−12018.
(9) Chundawat, S. P. S.; Beckham, G. T.; Himmel, M. E.; Dale, B. E.
Annu. Rev. Chem. Biomol. Eng. 2011, 2, 121−145.
(10) Himmel, M. E.; Ding, S.-Y.; Johnson, D. K.; Adney, W. S.;
Nimlos, M. R.; Brady, J. W.; Foust, T. D. Science 2007, 315, 804−807.
(11) Davis, R., Tao, L., Tan, E., Biddy, M., Beckham, G., Scarlata, C.,
Jacobson, J., Cafferty, K., Ross, J., and Lukas, J. Process design and
economics for the conversion of lignocellulosic biomass to hydro-
carbons: Dilute-acid and enzymatic Deconstruction of biomass to
sugars and biological conversion of Sugars to Hydrocarbons; National
́ ́
(41) Chmely, S. C.; Kim, S.; Ciesielski, P. N.; Jimenez-Oses, G.;
Paton, R. S.; Beckham, G. T. ACS Catal. 2013, 3, 963−974.
(42) Gao, Y.; Zhang, J.; Chen, X.; Ma, D.; Yan, N. ChemPlusChem
2014, 79, 825−834.
(43) Rahimi, A.; Azarpira, A.; Kim, H.; Ralph, J.; Stahl, S. S. J. Am.
Chem. Soc. 2013, 135, 6415−6418.
(44) Shabtai, J., Zmierczak, W., and Chornet, E. Conversion of lignin
to reformulated gasoline compositions. 1. Basic processing scheme.
Proceedings of the 3rd Biomass Conference of the Americas, Montreal,
Canada; Elsevier: Amsterdam, 1997; pp 1037−1040.
(45) Shabtai, J., Zmierczak, W., Chornet, E., and Johnson, D. K.
Conversion of lignin. 2. Production of high-octane fuel additives.
1327
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328

ACS Catalysis
Research Article
Preprints of the Symposium-American Chemical Society, Division of Fuel
Chemistry; American Chemical Society, Division of Fuel Chemistry:
Washington, DC, 1999; pp 267−272.
(75) Sergeeva, L. L.; Shorygina, N. N. Bull. Acad. Sci. USSR, Div.
Chem. Sci. 1965, 14, 1590−1595.
(76) Borsche, W. Ber. Dtsch. Chem. Ges. 1917, 50, 1339−1355.
(77) Chuksanova, A. A.; Sergeeva, L. L.; Shorygina, N. N. Bull. Acad.
Sci. USSR, Div. Chem. Sci. 1959, 8, 2114−2119.
(46) Shabtai, J., Zmierczak, W., Chornet, E., and Johnson, D. K.
Conversion of lignin. 2. Production of high-octane fuel additives.
Abstracts of Papers, 217th ACS National Meeting, Anaheim, CA;
American Chemical Society: Washington, DC, 1999; pp. FUEL-056.
(47) Shabtai, J., Zmierczak, W., Kadangode, S., Chornet, E., and
Johnson, D. K. Lignin conversion to high-octane fuel additives.
Biomass: A Growth Opportunity in Green Energy and Value-Added
Products, Proceedings of the 4th Biomass Conference of the Americas,
Oakland, CA; Elsevier Science: Amsterdam, 1999; pp 811−818.
(48) Shabtai, J. S.; Zmierczak, W. W.; Chornet, E. U.S. Patent US 09/
136,336, 1999.
(78) Stern, K. H. J. Phys. Chem. Ref. Data 1972, 1, 747−772.
́ ́
(79) Cseri, T.; Bekassy, S.; Kenessey, G.; Liptay, G.; Figueras, F.
Thermochim. Acta 1996, 288, 137−154.
(80) Lindeberg, O.; Walding, J. Tappi J. 1987, 70, 119−123.
(81) de Groot, M.; Buma, W. J. J. Phys. Chem. A 2005, 109, 6135−
6136.
̌
(82) Terpinc, P.; Polak, T.; Segatin, N.; Hanzlowsky, A.; Ulrih, N. P.;
Abramovic,
̌
H. Food Chem. 2011, 128, 62−69.
(83) Nomura, E.; Hosoda, A.; Mori, H.; Taniguchi, H. Green Chem.
2005, 7, 863−866.
(49) Shabtai, J. S.; Zmierczak, W. W.; Chornet, E. U.S. Patent US 09/
376,864, 2001.
(50) Shabtai, J.; Zmierczak, W.; Chornet, E.; Johnson, D. U.S. Patent
US 09/972,461, 2003.
(84) Abello,
2008, 342, 119−125.
́ ́
S.; Vijaya-Shankar, D.; Perez-Ramírez, J. Appl. Catal., A
(85) Debecker, D. P.; Gaigneaux, E. M.; Busca, G. Chem. - Eur. J.
2009, 15, 3920−3935.
(51) Roberts, V. M.; Stein, V.; Reiner, T.; Lemonidou, A.; Li, X.;
Lercher, J. A. Chem. - Eur. J. 2011, 17, 5939−5948.
(52) Miller, J. E.; Evans, L.; Littlewolf, A.; Trudell, D. E. Fuel 1999,
78, 1363−1366.
(53) Vigneault, A.; Johnson, D. K.; Chornet, E. Can. J. Chem. Eng.
2007, 85, 906−916.
(54) Barta, K.; Matson, T. D.; Fettig, M. L.; Scott, S. L.; Iretskii, A.
V.; Ford, P. C. Green Chem. 2010, 12, 1640−1647.
(55) Matson, T. D.; Barta, K.; Iretskii, A. V.; Ford, P. C. J. Am. Chem.
Soc. 2011, 133, 14090−14097.
(56) Barta, K.; Warner, G. R.; Beach, E. S.; Anastas, P. T. Green
Chem. 2014, 16, 191−196.
(86) Di Cosimo, J. I.; Díez, V. K.; Xu, M.; Iglesia, E.; Apesteguía, C.
R. J. Catal. 1998, 178, 499−510.
(87) Vardon, D. R.; Franden, M. A.; Johnson, C. W.; Karp, E. M.;
Guarnieri, M. T.; Linger, J. G.; Salm, M. J.; Strathmann, T. J.;
Beckham, G. T. Energy Environ. Sci. 2015, 8, 617−628.
(88) Johnson, C. W.; Beckham, G. T. Metab. Eng. 2015, 28, 240−
247.
(89) Salvachua, D.; Karp, E. M.; Nimlos, C. T.; Vardon, D. R.;
́
Beckham, G. T. Green Chem. 2015, 17, 4951−4967.
(90) Parida, K.; Das, J. J. Mol. Catal. A: Chem. 2000, 151, 185−192.
(57) Warner, G.; Hansen, T. S.; Riisager, A.; Beach, E. S.; Barta, K.;
Anastas, P. T. Bioresour. Technol. 2014, 161, 78−83.
(58) Sturgeon, M. R.; O’Brien, M. H.; Ciesielski, P. N.; Katahira, R.;
Kruger, J. S.; Chmely, S. C.; Hamlin, J.; Lawrence, K.; Hunsinger, G.
B.; Foust, T. D.; Baldwin, R. M.; Biddy, M. J.; Beckham, G. T. Green
Chem. 2014, 16, 824−835.
́
(59) Mills, S. J.; Christy, A. G.; Genin, J.-M. R.; Kameda, T.;
Colombo, F. Mineral. Mag. 2012, 76, 1289−1336.
(60) Debecker, D. P.; Gaigneaux, E. M.; Busca, G. Chem. - Eur. J.
2009, 15, 3920−3935.
(61) Miyata, S. Clays Clay Miner. 1983, 31, 305−311.
(62) Bontchev, R. P.; Liu, S.; Krumhansl, J. L.; Voigt, J.; Nenoff, T.
M. Chem. Mater. 2003, 15, 3669−3675.
(63) Zhang, J.; Xu, Y. F.; Qian, G.; Xu, Z. P.; Chen, C.; Liu, Q. J.
Phys. Chem. C 2010, 114, 10768−10774.
(64) Katahira, R.; Mittal, A.; McKinney, K.; Ciesielski, P. N.;
Donohoe, B. S.; Black, S. K.; Johnson, D. K.; Biddy, M. J.; Beckham, G.
T. ACS Sustainable Chem. Eng. 2014, 2, 1364−1376.
(65) Bozell, J. J.; Black, S. K.; Myers, M.; Cahill, D.; Miller, W. P.;
Park, S. Biomass Bioenergy 2011, 35, 4197−4208.
(66) Bozell, J. J.; O’Lenick, C.; Warwick, S. J. Agric. Food Chem. 2011,
59, 9232−9242.
(67) Resch, M. G.; Donohoe, B. S.; Ciesielski, P. N.; Nill, J. E.;
Magnusson, L.; Himmel, M. E.; Mittal, A.; Katahira, R.; Biddy, M. J.;
Beckham, G. T. ACS Sustainable Chem. Eng. 2014, 2, 1377−1387.
(68) Chen, X.; Shekiro, J.; Pschorn, T.; Sabourin, M.; Tao, L.;
Elander, R.; Park, S.; Jennings, E.; Nelson, R.; Trass, O. Biotechnol.
Biofuels 2014, 7, 98.
(69) Chen, X.; Shekiro, J.; Franden, M. A.; Wang, W.; Zhang, M.;
Kuhn, E.; Johnson, D. K.; Tucker, M. P. Biotechnol. Biofuels 2012, 5, 8.
(70) Gierer, J.; Kunze, I. Acta Chem. Scand. 1961, 15, 803−807.
́
(71) Gierer, J.; Noren, I. Acta Chem. Scand. 1962, 16, 1976−1988.
(72) Huo, W.; Li, W.; Zhang, M.; Fan, W.; Chang, H.-m.; Jameel, H.
Catal. Lett. 2014, 144, 1159−1163.
(73) Iyi, N.; Matsumoto, T.; Kaneko, Y.; Kitamura, K. Chem. Mater.
2004, 16, 2926−2932.
(74) Iyi, N.; Okamoto, K.; Kaneko, Y.; Matsumoto, T. Chem. Lett.
2005, 34, 932−933.
1328
DOI: 10.1021/acscatal.5b02062
ACS Catal. 2016, 6, 1316−1328