Degradation of phenethoxybenzene in sodium hydroxide

Article abstract of DOI:10.1039/c6ra07122h

Simple lignin model compounds containing β-O-4 aryl ether linkages have been utilized as a means to understand lignin depolymerisation. The effects of reaction temperature, time, catalyst concentration, and initial phenethoxybenzene (PEB) concentration on

Full text of DOI:10.1039/c6ra07122h

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Degradation of phenethoxybenzene in sodium
hydroxide
Cite this: RSC Adv., 2016, 6, 57889
*
*
Dylan Cronin, Lalehvash Moghaddam, Darryn Rackemann, John Bartley
and William O. S. Doherty
Simple lignin model compounds containing b-O-4 aryl ether linkages have been utilized as a means to
understand lignin depolymerisation. The effects of reaction temperature, time, catalyst concentration,
and initial phenethoxybenzene (PEB) concentration on the degradation of PEB in NaOH were
investigated. Operating at 300 ꢀC for 1 h resulted in the highest combined yield of the primary products,
phenol and styrene, and also resulted in a reduced amount of degradation products formed. The
proportion of oligomeric and polymeric materials formed depended on the NaOH concentration, but
not on the initial PEB concentration for equal reaction time. The results were used to suggest probable
reaction pathways for PEB degradation.
Received 17th March 2016
Accepted 6th June 2016
DOI: 10.1039/c6ra07122h
www.rsc.org/advances
process can be catalysed by the presence of either acid or base,
which also tends to reduce the range of products obtained.12,13
1. Introduction
Under ambient basic conditions ethers will generally not
undergo hydrolysis, so the depolymerisation of lignin is
generally performed under moderate temperature (250–333 ꢀC)
and high pressure (1–240 bar), in a process known as hydro-
thermal liquefaction (HTL).14 The signicant reaction pathways
involved are, (a) hydrolysis and cleavage of the ether and C–C
bonds, (b) alkylation, (c) demethoxylation, and (d) condensa-
tion. Demirbas et al.15 suggested that when lignin is exposed to
temperatures in excess of 250 ꢀC, free phenoxyl radicals are
formed via thermal decomposition and that these radical
species tend to elicit condensation and repolymerisation reac-
tions, resulting in the formation of solid char material. This
char product is undesirable as it is very difficult to break-down
or utilise in any economically advantageous manner. Roberts
et al.16 demonstrated that, under basic conditions, the primary
products obtained in lignin depolymerisation are in fact
phenolic monomers, and that the oligomers observed result
from subsequent condensation reactions of the (highly reactive)
monomer intermediates.
It is thought that the problematic repolymerisation reactions
associated with lignin depolymerisation, are initiated by radical
formation. Therefore, preventing or limiting these radical
reaction pathways is pertinent to increasing the yield of the
desirable low molecular weight hydrolysis products. Radical
reactions (as opposed to ionic reactions), become less prevalent
as solvent density increases, since the radical species have less
space to diffuse and propagate the chain reaction and instead
are more likely to recombine with their composite reactant
species.17 Previous studies have shown that this can be aided by
the addition of an alkali salt,18,19 which increases the ionic
properties of the water and results in an increase in monomer
Lignin is a complex, amorphous and non-linear biopolymer, the
molecular mass of which is determined by the random cross-
linked polymerisation of three different phenolic moieties.
The approximate structure of lignin was rst proposed in 1977,1
and is known to include many hydroxyl, ether and phenolic
functional groups. As a result, this material has the potential to
be used as a feedstock for the production of phenol and other
closely related derivatives; such as benzene, toluene and xylene
(BTX), vanillin, guaiacol and eugenol.2,3
Studies into the structure of lignin indicate that the mono-
mer units are interconnected via ether and C–C bonds, in a ratio
of 2 : 1.4 The most predominant linkage, the b-O-4 bond, occurs
within the guaiacylglycerol b-aryl ether substructure with
a prevalence of 40–60%5 (Fig. 1). The prevalence of the various
types of inter-monomer linkages, as well as the three monomer
units themselves, vary with the type of lignocellulosic material
source.6 Not only is the b-O-4 bond present at signicantly
higher quantities than any other monomer linkage, it is also the
most easily chemically cleaved with the exception of the a-O-4
bonds.7 The other types of linkages present, both ether and
C–C, are more resilient to thermal and thermochemical
disruption.
Many studies have been conducted by pyrolysis on lignin and
its model compounds.8–11 A complex set of radical, rearrange-
ment and elimination reaction occur, which are inuenced by
processing conditions and lignin origin. However, when lignin
is heated in the presence of water it undergoes hydrolysis. This
Centre for Tropical Crops and Biocommodities, Queensland University of Technology,
GPO Box 2432, 2 George St, Brisbane, QLD 4001, Australia. E-mail: l.moghaddam@
qut.edu.au; d.cronin@qut.edu.au
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 57889–57901 | 57889

View Article Online
RSC Advances
Paper
Fig. 1 Structure of a lignin fragment with b-O-4 linkages labelled.
product yield. Miller et al.20 concluded that the more alkaline condensation reactions and the resultant formation of char and
salts were most effective not only in enhancing lignin depoly- other low-value materials.
merisation, but in preventing repolymerisation reactions and
the subsequent formation of polymeric solids.
2. Materials and method
Despite NaOH catalysis system been widely studied with
lignin model compounds, such as phenethoxybenzene, PEB
(Fig. 2) and benzyl phenyl ether (BPE), the physico-chemical
reactions in these systems with lignin are still difficult to
predict, and how to limit the multiplicity of products formed is
still a big challenge. This has implication in product recovery
and purication. As such, the approach the authors have
adopted in the present study is to specically focus on the
optimization of the reaction conditions of PEB in NaOH. This
will provide pointers for the HTL of black liquor, which is of
particular interest to our group. This is achieved by investi-
gating the inuence of reaction conditions, such as reaction
time and reactant concentration on the product range obtained
from PEB using HTL. It is acknowledged that PEB is one of the
simplest lignin model compounds as it does not contain phenyl
hydroxyl groups besides the b-O-4 bond. Of particular interest in
the present study is the molecular weight distribution of the
products obtained. If the reaction conditions can be optimised
in such a way so as prevent or limit the polymerisation of
primary products, similar conditions may prove effective in the
depolymerisation of lignin (in black liquor) without subsequent
Phenethoxybenzene (>95%) was obtained from Ark Pharm. For
the processing of analytical results, phenol (Sigma-Aldrich,
>99%) and styrene (Fluka, >99%) were used as standards for
product quantication. Acetone (Sigma-Aldrich, >99%) was
used as the GC-MS solvent and 4-ethylphenol (Fluka, >98%) as
an internal standard. Organic extractions were performed using
diethyl ether (Sigma-Aldrich, >99%). All chemicals were used as
received.
2.1 Experimental
The HTL experiments were conducted in 316-stainless steel
tube reactors of 6.3 mm internal diameter and length 16 cm
(volume of 5 mL). An additional reference reactor was designed
with an internal thermocouple and pressure transducer
attachment, to allow for the real-time monitoring of the internal
reaction temperature and pressure. Experiments were con-
ducted within a uidised sand bath (SBL-2D, Techne Inc.,
Burlington, NJ) over a temperature range of 250–350 ꢀC and
reaction time of 30–120 min. This temperature range was
selected based on numerous previous studies on lignin depo-
lymerisation, wherein the temperature for effective hydro-
thermal cleavage of the b-O-4 bond was determined to be
approximately 280 C.21–23 The concentration of NaOH catalyst
ꢀ
solution was varied between 0 and 5 wt%. This range was uti-
lised due to both the approximate compositional similarity to
black liquor,24 and previous similar work performed in the
area.12 The concentration of PEB in solution was varied between
2.4 and 9.0 wt%. Experiments were conducted in duplicate,
though for the molecular weight measurements experiments
were in triplicate to provide errors for the data reported in the
gures.
A standard work-up method for analysing the reactant
mixture obtained in the HTL of PEB was developed. Aer the
Fig. 2 Lignin model compound phenethoxybenzene (PEB) with b-O-
4 linkage labelled.
57890 | RSC Adv., 2016, 6, 57889–57901
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
reaction was performed the reactors were removed from the recorded. For the experiments performed with no catalyst the
heated sand bath and immediately quenched in room- pH of the aqueous phase was not adjusted, as the pH was
temperature water. The contents of the reactors were then already slightly acidic (as opposed to highly basic), and as such
washed with 3 ꢁ 3 mL volumes of diethyl-ether. The biphasic a one-step organic extraction procedure was performed. Initial
mixture obtained was vigorously agitated and then centrifuged, testing indicated the importance of removing the solvent under
before collecting the organic phase. In order to remove the mild conditions, due to the relative volatility of styrene. A
small quantity of solid material present the aqueous phase was owchart of this workup procedure is illustrated in Fig. 3.
ltered through a pre-weighed sintered crucible. Once ltered,
the pH of the aqueous phase was adjusted to 2.5 with the use of
2.2 Analysis of products
2 M H2SO4. A second organic extraction was then performed
using 3 ꢁ 3 mL portions of diethyl ether. Once again the solu-
tion was vigorously agitated and centrifuged, and the organic
phase obtained was combined with the rst. The bulk of the
diethyl ether was removed from the organic phase via mild
vacuum distillation at 35 ꢀC. In order to remove any residual
ether, the oil was air-dried overnight and the nal dry mass
The organic/oil phase samples were analysed by GC-MS using
an Agilent 6890 Series Gas Chromatograph and a HP 5973 mass
spectrometer detector, employing helium as the carrier gas. The
installed column was an Agilent CP-wax 52 CB, 30 m ꢁ 0.25 mm
ꢁ 0.25 mm. Samples were injected with a split ratio of 10 : 1 into
the injection port set at 230 ꢀC. The temperature program
Fig. 3 Flowchart describing the work-up method used for preparing and characterising the organic, aqueous and solid phases for the HTL of
PEB.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 57889–57901 | 57891

View Article Online
RSC Advances
Paper
commenced at 90 ꢀC and heated at a rate of 3 ꢀC minꢂ1 to
The results indicate that at 250 C the majority of the PEB
ꢀ
a temperature of 230 ꢀC. Once the temperature was reached the does not degrade, and so the yields of styrene and phenol are
column was then held for another 5 min. Compounds were low. At 300 ꢀC (except at PEB concentration of 6.8 and 9.0 wt%)
identied by means of the Wiley library-HP G1035A and NIST and 350 ꢀC no residual PEB remain. It is at these temperatures
library of mass spectra and subsets-HP G1033A (a criteria that the phenol yields are highest, and reached ꢃ48%. The
quality value >80% was used). Acetone was used as the solvent styrene yields are highest at 300 ꢀC, and were hardly detected at
with approximately 15 mg of oil dissolved in 1.0 mL of acetone. 350 C. In the study by Eom et al.25 with use of 0.25% Na2CO3,
ꢀ
1H-NMR spectra of the organic phase samples were obtained PEB conversion reached 100% at 300 ꢀC with a reaction time of
using a Varian VNMRS 400 NMR and analysed using the ACD/ 2 h, and maximum yield of phenol (40%) was obtained at 400
NMR Processor Academic soware package. FT-IR spectra of ꢀC. Although the processing conditions between the present
the samples were collected using a Nicolet 870 FT-IR spec- study and those of Eom et al.25 are slightly different, the general
trometer equipped with a Smart Endurance single bounce dia- trends are similar in terms of PEB degradation and the increase
mond ATR accessory (Nicolet Instrument Corp., Madison, WI). in phenol yield with temperature, as well as phenol being the
Spectra were collected in the spectral range 4000–525 cmꢂ1
,
major product. The reduction in styrene concentration with
increasing temperature was obtained in both studies. However,
using 64 scans and 4 cmꢂ1 resolution.
Size exclusion chromatographic (SEC) analysis for molecular no ethyl benzene was obtained in the present study as the gas
weight measurement incorporated a GPC Water Breeze system used was air instead of hydrogen. Eom et al.25 observed the
model 151 with an isocratic HPLC pump. Eluted fractions were presence of ethyl benzene when the gas present in the reactor
detected with UV and refractive index Water model 2414 was hydrogen instead of helium, as the compound is produced
detector. Three Phenomenex phenogel columns (500, 104 and from the hydrogenation of styrene.
Table 1 shows that at 250 ꢀC as the initial PEB concentration
˚
106 A porosity; 5 mm bead size) were used for size exclusive
ꢀ
separation. The mobile phase was tetrahydrofuran (THF) with 1 increases the amount of phenol dropped, while at 350 C the
mL minꢂ1 ow rate at 30 ꢀC. A 100 mL sample having reverse is true. The styrene yield remains constant at 250 ꢀC and
a concentration of 5 mg mLꢂ1 in THF was injected into the SEC the yield is 0% at 350 ꢀC. At the lower temperature, as shown in
instrument aer ltration with a 0.4 mm Teon syringe type Table 1, with increasing initial PEB concentration the amount
lter.
of unconverted PEB increases. This does not fully account for
the reduced amount of phenol obtained; indicating that
perhaps some of the phenol is undergoing some side reactions.
At 350 ꢀC any styrene formed in situ would have participated in
some side reactions either on its own or with phenol.
3. Results and discussion
3.1 Inuence of reaction temperature
Table 1 also shows that with increasing reaction tempera-
ture, the cumulative yield of styrene, phenol and residual PEB
drops signicantly, indicating the formation of products other
than styrene and phenol. At 350 ꢀC there is little impact on
phenol yields, but a signicant effect on styrene is observed as
yield reduces to 0%. Fig. 4 shows the continuous conversion of
PEB to phenol and styrene, as well as increasing peak intensities
A series of HTL experiments were performed in 5 wt% NaOH
solutions at 250, 300 and 350 ꢀC for 120 min, and at PEB
concentrations of 2.4–9.0 wt%. The yields of phenol, styrene
and residual PEB obtained by quantitative GC-MS analysis of
the oil phase are presented in Table 1. Qualitative FT-IR and 1H-
NMR analysis were conducted on the oil phase samples to
support the results obtained by GC-MS.
Table 1 Products of PEB conversion (in 5 wt% NaOH) with temperature at 120 mina
Amount of styrene,
phenol & PEB
(wt% of PEB mass)
Oil phase accounted
for by styrene,
phenol & PEB (wt%)
Reaction
PEB conc.
(wt%)
Styrene yield
(wt% of PEB mass)
Phenol yield (wt%
of PEB mass)
Residual PEB (wt%
of PEB mass)
temperature (ꢀC)
250
300
350
2.4
4.7
6.8
9.0
2.4
4.7
6.8
9.0
2.4
4.7
6.8
9.0
3.4
4.2
3.6
3.4
13.4
9.6
10.3
7.9
0.0
0.0
0.0
0.0
20.9
9.1
6.6
37.3
60.7
63.8
70.2
0.0
0.0
6.5
4.0
0.0
61.7
73.9
74.0
77.9
43.6
33.4
43.5
34.3
43.2
35.0
47.6
48.1
82.3
84.1
83.1
84.3
55.3
57.9
57.7
40.6
48.2
47.5
45.4
57.6
4.3
41.9
48.3
40.9
28.7
43.2
35.0
47.6
48.1
0.0
0.0
0.0
a
Error of ꢄ4%.
57892 | RSC Adv., 2016, 6, 57889–57901
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Fig. 4 GC-MS chromatograms of oil samples produced at 120 min and 5 wt% NaOH for PEB concentration of 4.7 wt% at (a) 250 ꢀC, (b) 300 ꢀC, (c)
350 ꢀC and (d) styrene concentration of 4.7 wt% at 350 ꢀC.
with temperature associated with other compounds (Table 2). It reactions between styrene molecules which occurred at 350 ꢀC,
ꢀ
is noted that not all polymeric (high molecular weight, HMWM) but not 300 C. Fig. 4(d) compares the GC-MS chromatograms
ꢀ
material may be volatile and identied by GC-MS.
obtained for the styrene and PEB HTL experiments at 350 C.
A previous study on the HTL of PEB reported a lower than
In an attempt to elucidate the origin of these other
compounds, several HTL experiments were conducted in which expected yield of styrene, which was explained by the strong
PEB was substituted with styrene. The GC-MS analyses indicate reactivity of this species and its propensity towards oligomer-
that the additional compounds present are formed as a result of isation.26 A similar work by Roberts et al.18 investigated the
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 57889–57901 | 57893

View Article Online
RSC Advances
Paper
Table 2 Identified compounds from GC-MS analysis
3.1.1 High molecular weight material (HMWM). The
molecular weight distribution of the oil samples was deter-
mined by SEC. The intention was to identify the prevalence of
any aromatic oligomers and larger polymeric material which
may have been formed in the HTL of PEB.
In addition to styrene, phenol and PEB, several polystyrene
standards (PSS) were also analysed in an attempt to calibrate
various elution periods to approximate molecular weight
ranges. It is important to bear in mind that SEC does not
provide a true measurement of molecular weight. The results
are determined by the hydrodynamic volume of the eluted
species, which does not necessarily correlate with molecular
weight but is useful for comparative analysis.
The general prole of the SEC chromatograms obtained
through analysis of the organic phase samples is shown in
Fig. 5. This gure is divided into peaks observed as a result of
those species of hydrodynamic volume less than or equal to
PEB, and those larger than PEB. The rst peak (from right to
le), is caused by styrene, residual solvents and other small
compounds, the second is due to phenol and the third by PEB
(and any similar sized compounds present at low concentra-
tion). Via the analysis of the polystyrene standards, the
maximum molecular weight material present in the organic
phase samples is approximately 5000 g molꢂ1. This indicates
that the HTL reaction produces not only small aromatic oligo-
mers, but also signicant amounts of large chain polymeric
materials.
In order to determine the inuence of reaction temperature
on the amount of HMWM produced, the SEC chromatograms of
each of the organic phase samples were compared. This
involved integrating the plots and determining the relative
percentage of area in each region relating to materials of
hydrodynamic volume greater than PEB. The results of this
analysis are presented graphically in Fig. 6.
Retention
Peak
(min)
Name
1
2
3.46
4.30
Styrene
a-Methylstyrene
3
4
4.69
5.43
2-Pentanon-4-hydroxy-4-methyl
1-Propenyl-benzene
5
6
6.74
1-Methyl-2-(2-propenyl)-benzene
Acetophenone
11.06
15.88
18.89
19.01
21.89
22.04
24.12
27.28
28.57
28.82
29.29
29.56
30.74
31.07
32.01
33.14
33.15
33.34
33.72
34.16
34.86
34.98
35.67
36.76
36.99
40.61
44.50
46.41
46.82
7
8
9
a-Methyl-benzenemethanol
Phenylethyl alcohol
Butylated hydroxytoluene
Diphenylmethane
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Phenol
1,1-(1,2-Dimethyl-1,2-ethanediyl)bis-benzene
1,4-Benzenediol-2,5-bis-(1,1-dimethylethyl)
1,10-(1-Butene-1,3-diyl)dibenzene
1,3-Diphenylpropane
(E)-But-1-ene-1,3-diyldibenzene
Unidentied compound
1-Phenyl-2-(2-vinylphenyl)ethanone
Dihydro-5H-dibenzo cycloheptene
3-Phenethylbenzoic acid
2-Methyl-9,10-dihydroanthracene
Phenethoxybenzene (PEB)
1,2,3,4-Tetrahydro-1-1-phenyl napthalene
3-Phenyl-2H-chromene
Unidentied compound
1,2-Diphenylcyclopropane
1,10-(3-methyl-1-propene-1,3-diyl)bis benzene
1,10-(1,2-dimethyl-1,2-ethenediyl)bis(E)benzene
4-Phenyl-1,2-dihydronaphthalene
1,10-(1-butene-1,4-diyl)bis(Z)benzene
1-Phenyl-napthalene
9-(2-Propenyl)-anthracene
2-Phenyl-naphthalene
2-(1-Phenylethyl)-phenol
The data indicates that for a reaction temperature of 250 ꢀC
very little of the organic phase produced has a hydrodynamic
volume greater than that of PEB. This is not surprising, as the
vast majority of the PEB sample is not actually broken down at
this reaction temperature. When the temperature is increased
hydrolysis of benzyl phenyl ether (BPE) in water and several
metal carbonate bases. It was reported that with increasing
reaction severity (increased reaction temperature and reaction
time), there was an increase in the degree of secondary reac-
tions of the primary products (phenol and benzyl alcohol),
resulting in the production of HMWM. The authors were
surprised to observe that upon exposing a 5 wt% aqueous
ꢀ
to 300 C, the percentage of higher molecular weight material
increases to approximately 33% irrespective of the starting PEB
concentration. Surprisingly, when the reaction temperature is
further increased to 350 ꢀC the amount of HMWM reduces.
Given that this is an increase in reaction severity, it is expected
that this value would increase.18
ꢀ
solution of BPE to a temperature of 300 C for 1 h, that more
than 20% of the material was converted to compounds too large
to be detected by GC-MS. On this basis of the work of Roberts
and co-workers, a proportion of the phenol generated in the
present studies would have been involved in secondary reac-
tions producing HMWM.
This observation can be explained by reviewing the GC-MS
chromatograms of the oil phase and styrene test samples
produced at 300 ꢀC and 350 ꢀC. Numerous products were
observed in the chromatograms of the 350 ꢀC tests that were not
present at 300 ꢀC. What appears to be causing this result is that
at 350 ꢀC the entirety of the styrene produced not only
undergoes polymerisation, but is then subsequently thermally
The oil phase samples obtained when a reaction temperature
ꢀ
ꢀ
of 300 C and 350 C was applied, have a cumulative composi-
tion of styrene, phenol and residual PEB of only around 50%
(based on GC-MS analysis, see Table 1). It appeared likely that
the oil phase samples contained a signicant proportion of
material too large and non-volatile to be observed via GC-MS.
The oil phase samples were analysed via SEC in order to
determine the relative amount of material present that was
likely not to have been detected via GC-MS.
ꢀ
degraded to smaller volatile species. At 300 C a portion of the
styrene is polymerised, but this material does not appear to
then experience thermal degradation. This observation corre-
lates with that previously reported in the literature, in that
57894 | RSC Adv., 2016, 6, 57889–57901
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Fig. 5 General appearance of the SEC chromatograms of oil phase samples obtained via HTL of PEB.
simpler, since the most reactive species involved is the
hydroxide ion, and is therefore most likely the attacking species
(Fig. 8). The reaction produces styrene and the phenoxide ion
(later acidied to phenol), which provides a driving force for the
process due to the resonance stabilization this species experi-
ences. As these hydrothermal conditions favour radical reac-
tions, homolytic cleavage and hydrogen abstraction dominates.
The proposed mechanism is catalytic and so would not require
a mole of NaOH to produce a mole of the primary products as
stated in standard organic chemistry text books.
The two major aryl dimer species observed in the GC-MS
results can be explained mechanistically as the product of two
styrene molecules. Two mechanisms have been proposed for
dimerization of styrene: the Flory mechanism;28 and the Mayo
mechanism.29 However, these relate to self-polymerisation
under low temperatures. Under hydrothermal and radical
conditions we propose that styrene is likely to dimerise in one of
two ways (Fig. 9), the more stable of which being the formation
of a head to tail linkage (H–T) with the tertiary radical. The
second process involves the less stable primary radical and the
formation of a tail to tail linkage (T–T). Both dimer species will
exhibit cis/trans isomerism. Given that the reaction mecha-
nisms described are competing, the relative proportion of each
species produced will be highly dependent upon reaction
kinetics and thermodynamics. These species (but-1-ene-1,3-
diyldibenzene and but-1-ene-1,4-diyldibenzene), would be
likely to undergo cyclisation (Fig. 10) and produce their
respective tetrahydrophenol naphthalene isomers (1,2,3,4-
Fig. 6 Percentage of HMWM in PEB HTL oil phase samples (based on
integration of respective SEC chromatograms) for reactions per-
formed at 250 ꢀC, 300 ꢀC and 350 ꢀC (120 min reaction time and 5 wt%
NaOH solvent).
polystyrene begins to experience signicant thermal degrada-
tion at approximately 350 C.27
ꢀ
3.1.2 Reaction pathways. In their recent publication, Eom
et al.25 proposed reaction mechanisms for the hydrolysis of PEB
to phenol and styrene, both in the presence and absence of
Na2CO3 (Fig. 7). It is the opinion of the authors of the present
work that the mechanisms suggested could be viewed as an
unlikely explanation for the results obtained. Each of the
mechanisms described involves the coordination of the sodium
cation with the oxygen atom of PEB. Sodium is a highly ionic
species and as such is considered incapable of coordinating in
this manner. Furthermore, these species have not previously
been shown to exist in free solution. They are observed in the
electrospray plasma of MS sources where droplet evaporation
forces sodium ions to form such complexes. Soer metals like
magnesium and transition metals such as iron and cobalt
frequently exhibit this kind of behaviour, but not hard bases
like sodium. It is likely that the reaction is actually much
tetrahydro-2-phenyl-napthalene
and
1,2,3,4-tetrahydro-1-
phenyl-napthalene, respectively), both of which have been
identied in the GC-MS chromatograms as the two most prev-
alent cyclic aryl dimers. Rather than terminate at the point of
dimerization, these radical mechanisms could potentially
propagate to form oligomers and solid polymeric materials. It is
likely that this is a signicant contributor to the HMWM
produced in the HTL of PEB. It should be noted that these
proposed reaction pathways are tentative though based on
sound organic chemistry knowledge, as high temperature
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 57889–57901 | 57895

View Article Online
RSC Advances
Paper
Fig. 7 Proposed reaction mechanisms of PEB pathways in the presence and absence of Na2CO3.25
57896 | RSC Adv., 2016, 6, 57889–57901
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
concentrations of PEB material within 120 min of reaction time.
ꢀ
However, at 350 C the styrene was completely consumed and
converted to HMWM, which then underwent partial thermal
degradation. At 300 ꢀC the styrene product did experience
a reduction in yield due to self-polymerisation, however this was
ꢀ
signicantly less pronounced than when reacted at 350 C. As
a result all further experiments were conducted at 300 ꢀC. A
second series of HTL experiments were performed for 120 min
at 300 ꢀC, utilising a NaOH solution of concentration 0.0–5.0
wt%, as solvent. These tests were performed in an attempt to
determine the inuence of NaOH concentration on product
selectivity. Based on the quantitative GC-MS analysis of the oil
phase obtained in each experiment, the relative yields of
phenol, styrene and residual PEB were determined and are
presented in Table 3. The GC-MS chromatograms also showed
the same minor compounds were produced for each of the
different NaOH catalyst concentrations tested.
With the addition of only 0.25 wt% NaOH to the reaction
solvent, the PEB is completely consumed. The phenol yield
tends to experience a slight increase with increasing NaOH
concentration above 0.25 wt%, however the styrene yield expe-
riences a positive inuence. The most signicant of these
occurs at a PEB concentration of 2.4 wt%, where the styrene
yield increased from 0.0 to 21.0%, for the 0.25 and 1.0 wt%
NaOH tests, respectively.
Fig. 8 Probable mechanism proposed for the base-catalysed hydro-
lysis of PEB.
electron paramagnetic resonance spectroscopy and isotopic
labelling will be required to conrm the depolymerisation and
repolymerisation processes.
3.2 Inuence of NaOH catalyst concentration
Given that PEB was found to remain predominantly intact when
ꢀ
exposed to a reaction temperature of 250 C, this temperature
was concluded to be insufficient for further experimentation.
Both 350 ꢀC and 300 ꢀC were effective at converting the various
Fig. 9 Probable mechanisms for the production of styrene dimers via radical polymerisation.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 57889–57901 | 57897

View Article Online
RSC Advances
Paper
Fig. 10 Mechanisms for the cyclisation of styrene dimers.
Table 3 Styrene, phenol & residual PEB yields for PEB HTL experiments of varying NaOH solvent and PEB concentration (120 min reaction time &
300 ꢀC reaction temperature)a
Yield of styrene,
phenol & PEB
(wt% of PEB mass)
Oil phase accounted
for by styrene,
phenol & PEB (wt%)
NaOH conc.
(wt%)
PEB conc.
(wt%)
Styrene yield
(wt% of PEB mass)
Phenol yield (wt%
of PEB mass)
Residual PEB (wt%
of PEB mass)
0.00
0.25
1.00
5.00b
2.4
4.7
6.8
9.0
2.4
4.7
6.8
9.0
2.4
4.7
6.8
9.0
2.4
4.7
6.8
9.0
1.3
1.3
1.4
0.5
0.0
1.7
2.2
3.5
21.0
15.7
8.8
9.6
13.4
9.6
1.3
1.8
1.8
98.6
92.6
87.3
92.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
6.5
4.0
101.2
95.7
90.5
94.7
32.4
38.0
38.9
41.9
63.1
65.9
44.3
51.0
55.3
57.9
57.7
40.6
74.0
80.7
79.7
83.4
58.5
52.9
56.1
50.5
39.8
39.1
38.7
41.0
43.6
33.4
43.5
34.3
1.7
32.4
36.3
36.7
38.4
42.1
50.2
35.5
41.4
41.9
48.3
40.9
28.7
10.3
7.9
a
Associated error range of ꢄ4%. b Data previously reported in Table 1.
It is clear from these data that PEB does not experience to 350 ꢀC is unlikely to have a highly signicant inuence on the
signicant degradation when reacted in water (no catalyst) at reaction rate observed, given that 300 ꢀC is already in excess of
300 ꢀC for 120 min. Further increasing the reaction temperature the 280 ꢀC b-O-4 degradation temperature.21–23 A second test was
57898 | RSC Adv., 2016, 6, 57889–57901
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
performed, this time increasing the reaction time to 10 h. Two degradation achieved was comparable to that of water, indi-
PEB concentrations were trialled, 2.4 and 4.7 wt%, and the cating that the catalytic threshold is between 0.04 and 0.25 wt%.
amounts of PEB remaining were 84 and 85%, respectively. Both
The NaOH concentration utilised in the remainder of this
tests yielded approximately 1 wt% styrene and 7 wt% phenol. study is 5 wt%. This value has been selected due to the fact that
The conversion of PEB with no catalyst (ꢃ15%) is slightly lower it is similar to the previously reported concentration of NaOH in
than the ꢃ30% conversion achieved under hydrogen atmo- black liquor, produced via the NaOH pulping process.24
sphere at 290 ꢀC and 10 h reaction.26 This conrms that radical
mechanisms are involved in PEB conversion as the hydrogen
will quench radicals and stop them from reforming into PEB.
3.3 Inuence of PEB concentration & reaction time
3.2.1 High molecular weight material (HMWM). The SEC A nal series of HTL experiments were performed at 300 ꢀC,
chromatograms of each of the organic phase samples were utilising a 5 wt% NaOH catalyst solution and reaction times of
obtained and the amount of HMWM for each determined. 30, 60 or 90 min and the results are presented in Table 4. For all
These data were grouped so as to elucidate any inuence of experiments performed in water, the PEB concentration had no
NaOH catalyst concentration on the amount of HMWM signicant effect on its conversion rate (see Table 3). It appears
produced (Fig. 11). The results indicate that the amount of therefore that the thermal degradation of PEB in water is a zero
HMWM produced was lowest when utilising a 1 wt% NaOH order reaction. It is likely however that the rate constant for this
solution and a PEB concentration of 2.4 wt%. A PEB concen- non-catalysed degradation causes the reaction to proceed so
tration of 4.7 wt% also produced a signicantly lower amount of slowly that the rate order cannot be determined over the time
HMWM at this catalyst concentration (1 wt% NaOH). For the period tested. PEB concentration probably does inuence the
conditions tested, HMWM was not signicantly inuenced by reaction rate, however it may require a reaction time of several
the concentration of NaOH catalyst solution.
days to properly observe this inuence. In a study on the similar
What is most surprising about these data is that a NaOH diaryl ether (BPE) it was determined that this compound
concentration of 0.25 wt% did not produce a signicantly degrades much faster in water than PEB, and that the reaction is
greater amount of HMWM, as expected. Considering the rst order with regards to BPE concentration.30
signicantly reduced yield of styrene obtained in these experi-
For the experiments which utilised NaOH as a catalyst, both
ments it was expected that this material had undergone self- the reaction time and the starting concentration of PEB had
polymerisation to produce HMWM. As previous stated, phenol a signicant inuence on the percentage of material le
will also participate in the formation of these products.
unconverted. Upon analysing the interrelationship of these
3.2.2 Selected NaOH concentration. The results of this area three factors, it is apparent that the NaOH catalysed reaction is
of work are inconclusive in determining which NaOH concen- pseudo-rst order.
tration is most effective in producing monomeric products from
Over the range of experimental conditions tested, the
PEB. What is clear however is that NaOH is very effective at maximum yield of primary products (phenol and styrene), was
catalysing the degradation of PEB in solution at 300 ꢀC. The obtained at both the lowest concentration of PEB (2.4 wt%) and
lower limit of efficacy for NaOH concentration has not been the shortest reaction time (60 min). Increasing reaction time
determined; however it has been shown to be less than 0.25 above 60 min did not generally increase the yield of phenol, and
wt%. An additional experiment was performed using 0.04 wt% tended to negatively inuence the yield of styrene. This is
(0.001 M) NaOH solution, however the degree of PEB despite the fact that a signicant quantity of PEB oen
remained aer 60 min (up to 34 wt%). When the reaction time
was increased the cumulative quantity of styrene, PEB and
phenol in the organic phase dropped signicantly. This effect is
explained by a coinciding increase in the amount of polymeric
material observed in the oil, most likely caused by the self-
polymerisation of styrene.
3.3.1 High molecular weight material (HMWM). The SEC
chromatograms of each of the organic phase samples were
obtained and compared. On analysis of these data several
trends are evident. Firstly (and as predicted), for each different
PEB concentration analysed, there is an increase in the
production of HMWM with increasing reaction time.
The data also suggests that the initial concentration of PEB
in the reactant mixture does not inuence the production of
higher molecular weight material for equal reaction times. The
proportion of such material produced appears to be equal
across each of the four PEB loading concentrations, for each
distinct reaction time tested. The only time when this is not the
case is for experiments where the combination of PEB and
reaction time results in little of the PEB material being
Fig. 11 Percentage of HMWM in PEB HTL oil phase samples (based on
integration of respective SEC chromatograms) for reactions per-
formed using 0.25, 1 and 5 wt% NaOH (120 min reaction time & 300 ꢀC
reaction temperature).
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 57889–57901 | 57899

View Article Online
RSC Advances
Paper
Table 4 Styrene, phenol & residual PEB yields for PEB HTL experiments of varying reaction time and PEB concentration (5 wt% NaOH solvent &
300 ꢀC reaction temperature)a
Yield of styrene,
phenol & PEB
(wt% of PEB mass)
Oil phase accounted
for by styrene,
phenol & PEB (wt%)
Reaction time
(min)
PEB conc.
(wt%)
Styrene yield
(wt% of PEB mass)
Phenol yield (wt%
of PEB mass)
Residual PEB (wt%
of PEB mass)
30
2.4
4.7
6.8
9.0
2.4
4.7
6.8
9.0
2.4
4.7
6.8
9.0
2.4
4.7
6.8
9.0
1.1
5.8
8.3
33.4
28.3
27.0
15.0
55.3
44.0
38.3
30.1
29.2
43.3
25.6
28.5
41.9
48.3
40.9
28.7
0.0
14.0
18.6
49.0
0.0
34.5
48.1
53.9
72.1
73.8
67.8
81.6
81.3
39.7
57.5
47.7
55.7
55.3
57.9
57.7
40.6
69.3
71.5
70.4
85.3
69.4
67.1
73.4
79.1
34.3
54.4
58.4
52.4
43.6
33.4
43.5
34.3
8.0
60
18.6
18.2
20.9
17.4
10.3
13.3
10.5
10.9
13.4
9.6
5.6
22.3
33.8
0.2
90
0.9
11.6
16.4
0.0
0.0
6.5
120b
10.3
7.9
4.0
a
Associated error range of ꢄ4%. b Data previously reported in Tables 1 and 3.
converted (such as when a 9.0 wt% PEB solution was reacted for insoluble in all solvents tested (including; diethyl ether, tetra-
30 min).
hydrofuran, acetone and hexane) and is highly likely to be
comprised of polymeric solids, as has been reported previ-
ously.25 Due to the very low density and quantity of this material
produced (generally <1 mg), the experimental error involved
made it difficult to accurately quantify the yield. It is approxi-
mated that up to 4 wt% was produced with NaOH catalyst, and
a negligible amount for the control (water) tests.
3.4 Analysis of aqueous phase
The aqueous phase samples from these experiments, as well as
standard solutions of PEB, phenol, styrene and diethyl ether
were analysed via HPLC. The intention of this work was to
determine the identity and prevalence of any observable species
which remained in the aqueous phase aer organic extraction.
It was determined that the immiscibility of styrene and PEB in
water prevented the retention of any detectable quantity of
these materials. Phenol and diethyl ether were clearly detected
in the samples and the quantity of phenol present was deter-
mined through the analysis of calibration standards. The
results indicated that the aqueous phase samples contain
between 0 and 3 wt% of the initial mass of PEB detected.
In addition to phenol and diethyl ether, two unidentied
peaks were observed. It was determined in the GC-MS analysis
of the organic phase samples that the aromatic alcohols 2-(1-
phenylethyl)-phenol and phenylethyl alcohol were always
present to some degree. It is likely that one of these chemical
species is responsible for one of the unknown HPLC chro-
matogram peaks. The second peak resembles a solvent peak, in
that it is intense and broad in nature and eluted early in the run.
This peak was only present for those HTL experiments where
signicant breakdown of the PEB occurred (i.e. not for those
experiments performed in water). Given the unexpected posi-
tion and shape of this peak it is possible that it is caused by
small quantities of polymeric material in solution.
4. Conclusions
To limit the proportion of HMWM formed (without the use of
capping agents) in lignin depolymerisation, the ratio of feed to
catalyst concentration as well as reaction time (preferably 1 h)
must be taken into consideration. Solvent density and ionic
strength should also be considered as these aspects will also
inuence the reaction mechanism. A reaction temperature
around 300 ꢀC gives high product yield, and reduces the
multitude of side products formed. The information obtained
from this study will be used in the investigation of the conver-
sion of black liquor to chemicals.
Acknowledgements
The authors would like to thank the Centre for Tropical Crops
and Biocommodities and nancial support from the Australian-
Indian Strategic Research Fund. Thanks also go to Queensland
University of Technology Science and Engineering Faculty for
the use of their research facilities.
References
3.5 Analysis of solid and gaseous phases
The solid phase of the reactant mixture was obtained upon
ltering the aqueous phase. This material was found to be
1 E. Adler, Lignin chemistry—past, present and future, Wood
Sci. Technol., 1977, 11, 169–218.
57900 | RSC Adv., 2016, 6, 57889–57901
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
2 C. Amen-Chen, H. Pakdel and C. Roy, Production of 17 B. C. Wu, M. T. Klein and S. I. Sandler, Inuence of
monomeric phenols by thermochemical conversion of
biomass: a review, Bioresour. Technol., 2001, 79, 277–299.
3 Y. Gao, H.-p. Chen, J. Wang, T. Shi, H.-P. Yang and
supercritical uid solvent density on benzyl phenyl ether
pyrolysis: indications of diffusional limitations, Energy
Fuels, 1991, 5, 453–458.
X.-H. Wang, Characterization of products from 18 V. Roberts, S. Fendt, A. A. Lemonidou, X. Li and J. A. Lercher,
hydrothermal liquefaction and carbonation of biomass
model compounds and real biomass, J. Fuel Chem.
Technol., 2011, 39, 893–900.
Inuence of alkali carbonates on benzyl phenyl ether
cleavage pathways in superheated water, Appl. Catal., B,
2010, 95, 71–77.
4 J. Haluk and M. Irmouli, The xed polymer constituents in 19 L. A. Torry, R. Kaminsky, M. T. Klein and M. R. Klotz, The
cooperage oak: cellulose, hemicelluloses and lignin, J. Sci.
Tech. Tonnellerie, 1998, 4, 1–82.
5 E. Dorrestijn, L. J. Laarhoven, I. W. Arends and P. Mulder,
effect of salts on hydrolysis in supercritical and near-
critical water: Reactivity and availability,, J. Supercrit.
Fluids, 1992, 5, 163–168.
The occurrence and reactivity of phenoxyl linkages in 20 J. E. Miller, L. Evans, A. Littlewolf and D. E. Trudell, Batch
lignin and low rank coal, J. Anal. Appl. Pyrolysis, 2000, 54,
153–192.
6 M. T. Klein and P. S. Virk, Model pathways in lignin
microreactor studies of lignin and lignin model compound
depolymerization by bases in alcohol solvents, Fuel, 1999,
78, 1363–1366.
thermolysis. 1. Phenethyl phenyl ether, Ind. Eng. Chem. 21 T. C. Drage, C. H. Vane and G. D. Abbott, The closed system
Fundam., 1983, 22, 35–45.
7 J. Zakzeski, P. C. Bruijnincx, A. L. Jongerius and
pyrolysis of b-O-4 lignin substructure model compounds,
Org. Geochem., 2002, 33, 1523–1531.
B. M. Weckhuysen, The catalytic valorization of lignin for 22 X. Erdocia, A. Toledano, M. A. Corcuera and J. Labidi,
the production of renewable chemicals, Chem. Rev., 2010,
110, 3552–3599.
Organosolv Black Liquor Hydrolysis to Obtain Low
Molecular Weight Phenolic Compounds, in 15th
International Conference on Process Integration, Modelling
and Optimisation for Energy Saving and Pollution Reduction,
ed. P. S. Varbanov, H. L. Lam, J. J. Klemes, S. Pierucci and
J. J. Klemes, 2012, pp. 535–540.
8 M. W. Jarvis, J. W. Daily, H.-H. Carstensen, A. M. Dean,
S. Sharma, D. C. Dayton, D. J. Robichaud and
M. R. Nimlos, Direct Detection of Products from the
Pyrolysis of 2-Phenethyl Phenyl Ether, J. Phys. Chem. A,
2011, 115, 428–438.
23 S. Kang, X. Li, J. Fan and J. Chang, Classied Separation of
Lignin Hydrothermal Liqueed Products, Ind. Eng. Chem.
Res., 2011, 50, 11288–11296.
9 A. Beste and A. C. Buchanan, Role of Carbon–Carbon Phenyl
Migration in the Pyrolysis Mechanism of b-O-4 Lignin Model
Compounds: Phenethyl Phenyl Ether and a-Hydroxy 24 B. Doherty and T. Rainey, Bagasse Fractionation by the Soda
Phenethyl Phenyl Ether, J. Phys. Chem. A, 2012, 116, 12242–
12248.
10 A. Beste and A. C. Buchanan Iii, Kinetic simulation of the
thermal degradation of phenethyl phenyl ether, a model
Process, in Proceedings of the Australian Society of Sugar Cane
Technologists, ed. D. Hogarth, Australian Society of Sugar
Cane Technologists Ltd/Scribe Consulting, Australia,
Queensland, Mackay, 2006, pp. 545–554.
compound for the b-O-4 linkage in lignin, Chem. Phys. 25 H.-J. Eom, D.-W. Lee, Y.-K. Hong, S.-H. Chung, M.-g. Seo and
Lett., 2012, 550, 19–24.
11 V. B. F. Custodis, P. Hemberger, Z. Ma and J. A. van
Bokhoven, Mechanism of Fast Pyrolysis of Lignin: Studying
K.-Y. Lee, Effects of a sodium carbonate (Na2CO3) additive on
the conversion of phenethyl phenyl ether (PPE) in high-
temperature water, Appl. Catal., A, 2014, 472, 152–159.
Model Compounds, J. Phys. Chem. B, 2014, 118, 8524–8531. 26 X. Wu, J. Fu and X. Lu, Kinetics and mechanism of
12 R. W. Thring, Alkaline degradation of Alcell® lignin, Biomass
Bioenergy, 1994, 7, 125–130.
hydrothermal decomposition of lignin model compounds,
Ind. Eng. Chem. Res., 2013, 52, 5016–5022.
13 O. Bobleter, Hydrothermal degradation of polymers derived 27 L. Fernandes, H. Gaspar and G. Bernardo, Inhibition of
from plants, Prog. Polym. Sci., 1994, 19, 797–841.
14 R. Sanghi and V. Singh, Green Chemistry for Environmental
Remediation, Wiley, 2012.
thermal degradation of polystyrene by C60 and PCBM: A
comparative study, Polym. Test., 2014, 40, 63–69.
28 F. R. Mayo, The dimerization of styrene, J. Am. Chem. Soc.,
1968, 90, 1289–1295.
15 A. Demirbas, Mechanisms of liquefaction and pyrolysis
reactions of biomass, Energy Convers. Manage., 2000, 41, 29 P. J. Flory, Mechanism of vinyl polymerizations, J. Am. Chem.
633–646. Soc., 1937, 59, 241–253.
16 V. M. Roberts, V. Stein, T. Reiner, A. Lemonidou, X. Li and 30 V. M. Roberts, Homogeneous and Heterogeneous Catalyzed
J. A. Lercher, Towards quantitative catalytic lignin
depolymerization, Chemistry, Weinheim an der Bergstrasse,
Germany, 17 (2011) pp. 5939–5948.
Hydrolysis of Lignin, 2008.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 57889–57901 | 57901