From lignocellulosic biomass to lactic- and glycolic-acid oligomers: A gram-scale microwave-assisted protocol

Article abstract of DOI:10.1002/cssc.201403183

The conversion of lignocellulosic biomass into platform chemicals is the key step in the valorization of agricultural waste. Of the biomass-derived platform chemicals currently produced, lactic acid plays a particularly pivotal role in modern biorefineries as it is a versatile commodity chemical and building block for the synthesis of biodegradable polymers. Microwave-assisted processes that furnish lactic acid avoid harsh depolymerization conditions while cutting down reaction time and energy consumption. We herein report a flash catalytic conversion (2 min) of lignocellulosic biomass into lactic and glycolic acids under microwave irradiation. The batch procedure was successfully adapted to a microwave-assisted flow process (35 mL min^-1), with the aim of designing a scalable process with higher productivity. The C2 and C4 units recovered from the depolymerization were directly used as the starting material for a solvent and catalyst-free microwave-assisted polycondensation that afforded oligomers in good yields.

Full text of DOI:10.1002/cssc.201403183

DOI: 10.1002/cssc.201403183
Full Papers
From Lignocellulosic Biomass to Lactic- and Glycolic-Acid
Oligomers: A Gram-Scale Microwave-Assisted Protocol
Diego Carnaroglio,[a] Silvia Tabasso,[b] Beata Kwasek,[c] Dariusz Bogdal,[c] Emanuela
Calcio Gaudino,[a] and Giancarlo Cravotto*[a]
To Professor Bernd Ondruschka, a pioneer and a master of applied microwave chemistry.
The conversion of lignocellulosic biomass into platform chemi-
cals is the key step in the valorization of agricultural waste. Of
the biomass-derived platform chemicals currently produced,
lactic acid plays a particularly pivotal role in modern biorefiner-
ies as it is a versatile commodity chemical and building block
for the synthesis of biodegradable polymers. Microwave-assist-
ed processes that furnish lactic acid avoid harsh depolymeriza-
tion conditions while cutting down reaction time and energy
consumption. We herein report a flash catalytic conversion
(2 min) of lignocellulosic biomass into lactic and glycolic acids
under microwave irradiation. The batch procedure was suc-
cessfully adapted to
a microwave-assisted flow process
(35 mLminÀ1), with the aim of designing a scalable process
with higher productivity. The C2 and C4 units recovered from
the depolymerization were directly used as the starting materi-
al for a solvent and catalyst-free microwave-assisted polycon-
densation that afforded oligomers in good yields.
Introduction
A combination of environmental concerns and the diminishing
availability of fossil sources has prompted the scientific com-
munity to search for alternative energy and chemical sources.
Lignocellulosic biomass is the most abundant renewable
source of biofuels and biochemicals that is suitable for the bio-
refinery concept.[1] Lignocellulosic biomass (cellulose, hemicel-
lulose and lignin mixtures), in the form of plant and agricultur-
al waste, is an ideal and freely available source of raw sugars
for industrial processes as it does not affect food supplies nor
entail ethical issues. It is well known that plant biomass can be
converted into useful materials via hydrothermal treatment
under harsh conditions.[2,3] Indeed, product mixtures of varying
compounds,[4] such as monosaccharides,[5] 5-hydroxymethylfur-
fural,[6] and organic acids are obtained. For these reasons, cata-
lytic systems for the selective transformation of cellulose into
high value-added chemicals are becoming more and more at-
tractive. Lactic acid (LA) plays an important role in bio-refinery
management as it is a versatile commodity chemical widely
used in food processing as well as the pharma and cosmetic
industries.[7] LA is used to prepare ethyl lactate (a green sol-
vent),[8] 1,2-propanediol[9] and biodegradable plastics.[10] In par-
ticular, LA and glycolic acid (GA) copolymers such as PLGA
have been synthesized, via direct polycondensation and ring
opening polymerization, to improve the properties of polylac-
tic acid (PLA).[11] Their biocompatibility, biodegradability and ef-
ficient bioabsorption mean that PLGA and PLA have found
a number of applications in medical fields.[12,13] LA is normally
produced via monosaccharide fermentation,[7] however this
biological approach is not able to directly convert lignocellu-
lose to monosaccharides without pretreatment. The develop-
ment of fast, sustainable and cost effective chemical process-
es,[14,15] for LA production has therefore become one of the
hottest research topics in the last few years.[16] Some authors
have recently reported the use of alkaline hydrothermal pro-
cesses in the conversion of glucose or cellulose to LA and
other short chain carboxylic acids.[17,18] Other catalytic systems
involve the application of divalent and trivalent metal cat-
ions.[19] However, all these processes require high temperatures
and long reaction times.
[a] D. Carnaroglio, E. C. Gaudino, G. Cravotto
Dipartimento di Scienza e Tecnologia del Farmaco and
Centre for Nanostructured interfaces and surfaces (NIS)
University of Turin, Via P. Giuria 9, 10125 Turin (Italy)
E-mail: giancarlo.cravotto@unito.it
The last three decades have seen the development of envi-
ronmentally friendly procedures which involve the generation
of high energy micro-environments by means of non-conven-
tional energy sources, such as microwaves (MW)[20] and ultra-
sound (US).[21] The peculiar properties of MW as a source of
volumetric and selective dielectric heating can be efficiently
exploited in the extraction of natural matrices[22] and the hy-
drolysis of biopolymers and their further conversion into plat-
form chemicals.[23] In particular, as reported by Fan et al. the
hydrolysis step is particularly driven by MW irradiation’s weak-
ening of the cellulose hydrogen bond network, which is in
[b] S. Tabasso
Dipartimento di Chimica
University of Turin
Via P. Giuria 7, 10125 Turin (Italy)
[c] B. Kwasek, D. Bogdal
Chair of Biotechnology and Physical Chemistry
Politechnika Krakowska
ul. Warszawska 24, 31-155 Krakow (Poland)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201403183.
ChemSusChem 0000, 00, 0 – 0
1
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
turn due to selective action on polar –CH2OH groups.[24] MW-
assisted protocols for the acid-catalysed flash conversion of
cellulose, chitosan and post-harvest tomato plants (PHTP) to
levulinic acid have recently been reported and have yielded
high conversions and selectivity.[25,26] Several catalytic systems
such as ionic liquids,[27.] zirconia,[28] TiO2,[29] large-pore mesopo-
rous tin phosphate,[30] have been used in the MW-assisted con-
version of polysaccharides into 5-hydroxymethylfurfural (HMF)
and other platform chemicals. Besides the efficient conversion
process, MW irradiation also played a pivotal role in the prepa-
ration of supported nanocatalysts.[31]
mainly lies in the direct conversion of lignocellulosic biomass
into platform chemicals, several investigations described the
simple conversion of monosaccharides. Scaling up is strictly re-
lated to process efficiency: conversion, selectivity and energy
consumption are the key points for industrialization. Esposito
et al.[18] have tested various alkaline systems (calcium, stronti-
um and barium hydroxide) for the conversion of glucose to LA.
Barium hydroxide was the most promising of these and afford-
ed LA in yields of up to 57%. The authors proposed a mecha-
nism for LA formation based on a retro-aldol reaction following
the isomerization of glucose to fructose. This reaction pathway
highlights the key role of divalent cations in the stabilization
of dicarbonyl intermediate. This feature has been extensively
investigated by Wang et al.,[19] who evaluated the effect of sev-
eral divalent and trivalent ions and showed that PbII gave the
highest LA selectivity. The authors used computational studies
to show that the presence of divalent ions significantly de-
creased the activation barrier for the retro-aldol fragmentation
of fructose thus changing the main reaction pathway from the
formation of HMF to the formation of LA.
Moreover, in the last few years, the innovative peculiar fea-
tures of most recent dedicated MW reactors for gram-scale
synthesis (from low vacuum to high pressure, fast heating and
cooling, etc.,) enable high efficiency and versatility.
This work investigates new MW-assisted processes for ligno-
cellulosic biowaste valorization. This goal has been achieved
over two steps: firstly, the conversion of cellulose and PHTP
into high value-added chemicals (LA and GA) and the use of
these platforms for the synthesis of bio-based co-polymers. A
continuous-flow MW process for the conversion of cellulose
into LA and GA using a dedicated tubular reactor has also
been developed. This method makes the valorization of agri-
cultural residues feasible and overcomes waste disposal issues.
MW technology dramatically enhances reaction rates while
also reducing energy consumption.
Using one of our previous works on the MW-assisted flash
conversion of cellulose and PHTP waste to levulinic acid[25] as
a foundation, we applied a similar protocol to divalent and tri-
valent metal ion catalyzed syntheses of LA in a high-pressure
stainless steel MW reactor (power 1.5 kW, pressure up to
200 bar). The catalytic activity of several salts, including
Pb(OAc)2, Pb(NO3)2, Zn(OAc)2, ZnCl2, AlCl3, Al(NO3)3 and SnCl2,
was tested. As shown in Figure 1, the ball-milling pre-treat-
ment of cellulose strongly influenced subsequent conversions.
This is a clear example of how a combination of different “ena-
bling technologies” can improve process efficiency in biomass
conversion. The milled cellulose was then used to evaluate
which catalytic system provided the best results under MW ir-
radiation in batch and flow reactors.
Results and Discussion
Catalyst effect on MW-assisted conversion
It is well known that cellulose can be converted into glucose
and other chemicals via alkaline hydrolysis[32] and then, in the
same conditions, to LA or HMF. Although industrial interest
Figure 1. Conversions of ball-milled and crystalline cellulose. Ball-milling conditions: 60 min, 550 rpm. Reaction conditions: cellulose (500 mg), catalyst
(35 mm), 1908C, 2 min, N2 (2 MPa).
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
2
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
between glucose and fructose is a crucial step in LA and GA
selectivity (Figure 2).
Table 1. MW-assisted conversion of milled cellulose.[a]
Entry
Catalyst
Yield [%][b]
LA
Method optimization was performed, on the basis of these
results, using Pb(NO3)2 and Al(NO3)3 catalysts. The reaction was
followed by purification via simple stirring in the presence of
a strong acidic cation-exchange resin.
GA
Others[c]
1
2
3
4
5
6
7
Pb(OAc)2
Pb(NO3)2
Zn(OAc)2
ZnCl2
AlCl3
Al(NO3)3
SnCl2
6
15
3
28
10
27
18
4
13
0
6
0
11
11
45
28
42
30[d]
9
Method optimization
25
3
Figure 3 compares the effect of ball-milling pretreatment con-
ditions (speed and time) on the conversion of cellulose. Almost
complete conversion was achieved after 60 min at 550 rpm
milling.
[a] Reaction conditions: milled cellulose (500 mg), catalyst (35 mm),
1908C, 2 min, N2 (2 MPa). [b] Quantified by GC-MS. [c] Mainly levulinic
acid. [d] Mainly monosaccharides.
LA and GA were obtained, after the flash process, in a mix-
ture with other minor products such as levulinic acid, HMF and
monosaccharides (Table 1). Conversion (Figure 1) and selectivi-
ty (Table 1) were strongly influenced by the anion used, proba-
bly because of different salt hydrolysis constants.
The best selectivity (Table 1; entries 2,6) in terms of LA and
GA were achieved with Pb(NO3)2 and Al(NO3)3. Using Al(NO3)3,
high amounts of monosaccharides were detected in a mixture
with LA and GA. This means that the conversion from mono-
saccharides to acids with Al3+ is slower than the conversion
with Pb2+ in these experimental conditions. Both acids can be
obtained via retro-aldol reactions, as shown in Figure 2. The
isomerization of glucose to fructose, followed by a C3–C4
retro-aldol reaction, leads to LA after several dehydration/hy-
dration steps and a 1,2 hydride shift. In contrast, a C2–C3
retro-aldol reaction affords C4 and C2 intermediates when
starting from glucose. Indeed, the C2 fragment undergoes a hy-
dration step followed by isomerization through a 1,2 hydride
shift before giving GA.
Figure 3. Effect of ball-milling treatment on the conversion of cellulose. Re-
action conditions: 2208C, 2 min, catalyst (35 mm), cellulose (500 mg), N2
(3 MPa); Blank: microcrystalline cellulose (without ball-milling treatment).
Higher temperature (from 190 to 2208C) dramatically affect-
ed conversion yield and selectivity. Conversion rose from 12 to
97% and LA and GA yields increased as well (Table 2, entry 2
and entry 4). In spite of respectable selectivity in the 170–
1908C range, conversions were too low (Table 2, entries 1,2)
even at longer reaction times (Table 2, entry 3).
The same reaction protocol was carried out using Al(NO3)3
and gave a good conversion. However, monosaccharides,
formed at 1908C (Table 1, entry 6), were not converted to LA
It is well known that PbII generates PbII–OH in water and
promotes the isomerization of glucose.[33,34] The equilibrium
Figure 2. Proposed reaction pathways for the conversion of cellulose into LA[a] and GA, as catalysed by PbII in water. [a] Confirmed by theoretical computa-
tions.[19]
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
3
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
The same method used on milled PHTP biomass gave 50%
conversion and 34 and 21% yields for LA and GA, respectively
(Figure 4b). Biomass conversion and product yields were calcu-
lated on the basis of the dry weight of starting biomass and its
organic content (the weight loss after sample calcination at
6508C for 5 h). These results prove the versatility of the
method; it is efficient not only on purified cellulose but also
when applied to a more complex matrix, such as lignocellulo-
sic biomass. The MW-assisted protocol dramatically improves
conversions and yields over conventional heating while also
cutting down reaction time from 4 h to 2 min.[19]
Table 2. Temperature and time screening using Pb(NO3)2 catalyst.[a]
Entry
T [8C]
t [min]
Conversion [%]
Yield [%][b]
GA
LA
Others[c]
1
2
3
4
5[e]
170
190
150
220
220
2
2
30
2
10
12
7
97
95
18
21
–
43
8
12
14
–
18
14
9
14
3[d]
4
50
2
[a] Reaction conditions: catalyst (35 mm), ball-milled cellulose (500 mg),
N2 (3 MPa). [b] Quantified by GC-MS. [c] Mainly levulinic acid. [d] Mainly
monosaccharides. [e] The reaction was performed using a Al(NO3)3 cata-
lyst.
MW-assisted continuous flow procedure on milled cellulose
A flow-mode procedure was investigated with the aim of scal-
ing-up the optimized batch protocol and the professional pres-
surized reactor FlowSYNTH–Milestone/MLS was used (see the
Figure S2 in Supporting Information). This multimode MW re-
actor (70 mL, 1.6 kW) can operate at high temperature (up to
2608C) and pressure (up to 3 MPa). The feeding pump can
work at flow rates of between 10 and 200 mLminÀ1, that is, up
and GA even when the temperature was increased to 2208C
(Table 2, entry 5). Our experimental results, according to the
two competitive mechanisms described by Wang et al., prove
that Pb2+ favors the retro-aldol reaction while Al3+ promoted
dehydration into levulinic acid (Table 2, entry 5).[19]
The optimization study also analyzed the best catalyst con-
centration (5, 15, 35, 75 mm) under the same temperature and
pressure conditions. The final MW-assisted protocol was as fol-
lows: 2208C, 3 MPa (N2) using 20 mL water solution of
Pb(NO3)3 (35 mm) for 2 min irradiation, affording 97% conver-
sion with 43 and 18% yield for LA and GA, respectively (Fig-
ure 4a).
to more than two column volumes per min or 12 LhÀ1
.
300 LdayÀ1 can be processed, assuming 24 h operation. Pre-
liminary cellulose sonication in water (immersion horn,
21.1 kHz, 150 W, 30 min) afforded a stable colloidal suspension
which was well suited for the flow reactor, according to instru-
ment specifications (see Figure S3 in Supporting information).
Various catalyst concentrations
(35 and 75 mm) and flow rates
were tested, namely 35 mLminÀ1
(2 min residence time) and
18 mLminÀ1 (4 min residence
time). As shown in Figure 5,
a steady state (constant yields
and variables) was reached after
3 column volumes in all experi-
ments. There were contrasts
with the batch experiments;
a
catalyst concentration of
35 mm ensured good conversion
(>80%) but low yields in LA
(<20%) even at longer resi-
dence times (Figure 5a, c). The
optimal catalyst concentration in
flow experiments was 75 mm
(Figure 5b, d), however longer
reaction times increased the
amount of other chemicals pro-
duced (mainly levulinic acid), re-
ducing the yields of LA and GA.
Similar results to batch experi-
ments (Table 2, entry 4) were ob-
tained at a 75 mm catalyst con-
centration and
flow (Figure 5b). Under these
conditions, productivity of
2.1 LhÀ1 (50.4 LdayÀ1) was ob-
a
35 mLminÀ1
Figure 4. Effect of different catalyst concentrations on the MW-assisted conversion of milled cellulose (a) and
PHTP (b). Reaction conditions: ball-milled cellulose/PHTP (500 mg), 2208C, 2 min, N2 (3 MPa); data from three ex-
periment replications. Others: Mainly levulinic acid.
a
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
4
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 5. Continuous MW-assisted flow conversion of milled cellulose. All flow experiments were performed at 2208C, 3 MPa, at the following catalyst concen-
trations and flow rates: a) 35 mm, 35 mLminÀ1. b) 75 mm, 35 mLminÀ1. c) 35 mm, 18 mLminÀ1. d) 75 mm, 18 mLminÀ1. Others: mainly levulinic acid. Column
volume: each unit corresponds to a complete reaction mixture cycle in the reactor; data from three experiment replications.
tained which corresponds to an energy consumption of 0.55
kWh. This is a promising result considering the further applica-
tions on negative cost feedstock.
based on direct polycondensation without catalysts, solvents,
and initiators has been reported for the production of PLA.[41]
Water was removed under vacuum, but the process required
high temperatures (>1608C) and very long reaction times (up
to 95 h).
Bio-based polymers and oligomers
This work’s MW-assisted polycondensation dramatically re-
duced temperature and reaction times using a catalyst- and
solvent-free protocol to obtain PLGA oligomers. Indeed, the
oligomers were synthesized in our MW-assisted process in only
3 h at 1308C, in vacuum conditions (70 mbar), which allowed
for efficient water elimination via a dedicated MW-reactor. The
increasing power intensity (see Figure S12 in Supporting Infor-
mation) throughout the reaction was a good indicator of oligo-
mer formation. The decreasing polarity of the system during
reaction progress required an increase in MW emissions to
keep the reaction temperature stable. Three oligomers with
different LA/GA ratios were prepared. These reflect the ratios
obtained from biomass conversion in the first part of the work
(Table 3).
The selectivity achieved in this flash-MW-assisted protocol is
a good starting point for the synthesis of bio-based polymers
and oligomers. As already pointed out, co-polymers of LA and
GA (PLGA), are endowed with properties which make them
suitable for medical applications.[35–38] For example, high mo-
lecular-weight PLGA co-polymers have been used in prolonged
drug delivery systems because they degrade slowly to nontox-
ic monomers (LA and GA). In contrast, low molecular-weight
co-polymers are suitable for short-term drug release applica-
tions and, furthermore, their drug matrices do not need to be
retrieved from the body. High molecular weight PLGA has
been synthesized via the ring-opening polymerization of the
cyclic LA and/or GA diester. However, the direct polycondensa-
tion of LA and GA afforded PLGA oligomers.[39] Even if the in-
termediate step for making the
cyclic diester was avoided, yields
Table 3. MW-assisted direct polycondensation of LA and GA.[a]
were affected by removal of
water in the condensation of
carboxylic and hydroxyl groups.
Several procedures involve the
use of catalytic systems based
on Sn, Zn, or Sb salts[40] to syn-
thesize co-polymers with differ-
[c]
[c]
Entry
Feed [mol%]
LA/GA
Yield [%]
Polymer composition
LA/GA[b]
Mn
Mw/Mn
Td [8C][d]
1
2
3
100:0
50:50
70:30
80
75
77
100:0
53:47
69:31
2229
2923
2510
1.53
1.47
2.00
362
365
359
[a] 1308C, 3 h, 70 mbar. [b] Estimated from the integral height of hydrogen in 1H NMR spectra. [c] Determined
by GPC analysis. [d] Determined by TGA.
ent LA/GA ratios.
A process
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
5
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
results come from the polycon-
densation step, in which fast
polymerization (compared to
conventional methods) occurs in
solvent- and catalyst-free condi-
tions. Working in a batch pres-
surized MW system, we have de-
veloped a selective flash proto-
col which gives LA and GA at
high conversion values, from
both cellulose (97%) and PHTP
(50%), using a dilute catalyst so-
lution (35 mm). This method is
able to produce LA and GA
along two different proposed re-
Figure 6. TGA spectrum of PLGA 69:31 oligomer (heating rate of 108CminÀ1).
action pathways; C3–C4 or C2–
C3 retro-aldol reactions. The
batch protocol was converted
Thermogravimetric analysis (TGA, Figure 6) shows that all
samples degraded completely into volatile species at approxi-
mately 3608C (Table 3). Considering the LA/GA ratio obtained
from the best PHTP MW-assisted-conversion protocol (Fig-
ure 4b), the oligomer synthesized at a 70:30 (LA/GA) feed
(Table 3, entry 3) shows a number average molecular weight
which is comparable to the PLGA used for short term drug de-
livery.
into a flow process in which yields and conversions were main-
tained, productivity increased, and energy consumption low
1H NMR spectra confirmed oligomer formation and provided
a good estimate of the LA/GA ratio by integrating LA and GA
multiplet areas (d=5.62 and 5.28 ppm) (see Figure 7 and fig-
ures S4–S5 in the Supporting Information).
The FTIR spectrum of PLGA 69:31 (Figure 8) shows a sharp,
well-defined absorption at about 1752 cmÀ1 for the character-
istic ester carbonyl group of the oligomer. This feature, togeth-
er with the clearly diminished presence of the acidic hydroxyl
stretch, confirms that LA and GA have been polymerized. All
other co-oligomer sample IR spectra were of similar informa-
tive value (See Supporting Information).
Figure 7. 1H NMR spectrum of PLGA 69:31 oligomer.
PLA and PLGA 69:31 SEM images (Figure 9) show different
morphologic surfaces. The glob-
ular zones on the PLGA surface
could be ascribed to the inser-
tion of GA into the oligomer
chain.
Conclusions
This work describes the first in-
stance of a fast protocol for the
preparation of bio-based oligo-
mers/polymers from lignocellulo-
sic biomass that is fully driven
by MW irradiation. In addition,
the batch procedure has been
redesigned, with an eye on pro-
cess scale-up, to create flow-
through conditions in a commer-
cial MW flow reactor. Impressive Figure 8. FTIR spectrum of PLGA 69:31 oligomer.
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
6
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
(MLS GmbH, Milestone Srl) reactor using a high pressure dosing
pump, DMH 28x (Allodos pump, Grundfos). Energy consumption
was measured on a Gossen-Metrawatt SECUTEST SIII instrument.
NMR spectra were recorded on a Bruker Avance 300 (300 MHz and
1
75 MHz for H and 13C, respectively) at 258C; chemical shifts were
calibrated to the residual solvent proton and carbon resonances.
GC–MS analyses were performed in a GC Agilent 6890 (Agilent
Technologies - USA) which was fitted with a mass detector Agilent
Network 5973, using a HP-5 column (30 m long capillary column,
i.d of 0.25 mm and film thickness 0.25 mm). The molecular weights
of the polymers were estimated on a gel permeation chromato-
graph (GPC) in THF (0.8 mLminÀ1), stabilized using BHT using Shi-
madzu UFLC chromatograph equipment with a CTO-10AS vp
Column Oven and refractive index detector RID-10 A. The calibra-
tion curve was obtained using polystyrene standards. TGA was per-
formed on a TGA 4000 (PerkinElmer), using a heating ramp of
108CminÀ1, operating under N2 with alumina crucibles. Infrared
spectra (FTIR) were recorded as a KBr pad using the Spectrum BX
FT-IR System (PerkinElmer). Sample morphology was investigated
by means of a Zeiss Evo50 SEM equipped with an energy disper-
sive detector and a LaB6 source.
Methods
Biomass conversions: In
a typical batch experiment, samples
(500 mg) were suspended in a water solution of Pb(NO3)3 (35 mm)
and the mixture was heated for 2 min at 2208C under MW irradia-
tion (average power 960 W) in a N2 (3 MPa) atmosphere and mag-
netic stirrer (450 rpm). The reactor was cooled after 2 min.
In the flow experiments, a preliminary calibration of the flow rate
was required. The flow was adjusted using the pump settings
which adjusted both pump speed and stroke volume to achieve
the desired flow rate. The backpressure valve regulator was set at
3 MPa to largely counterbalance the vapour pressure generated by
the reaction mixture at the operation temperature (2208C). The
stirrer was set to 80% of its maximum speed and chiller tempera-
ture was set at 708C.
Figure 9. SEM images of PLA and PLGA 69:31 oligomers.
The reactor was first filled with water and the inlet tube was then
transferred to the reaction mixture to begin processing at the de-
sired temperature and flow rate. The first column volume from any
given run should be ignored because inevitable priming solvent
back-mixing caused by imperfect plug-flow will have occurred.
Ball-milled cellulose (22.5 g) and water (900 mL) were sonicated to-
gether (21.1 kHz, 150 W, 30 min) at a controlled temperature
(308C) before the flow experiments. Pb(NO3)3 (10.43 or 22.36 g)
was then added to the colloidal suspension and the reaction mix-
ture was pumped into the MW-flow-reactor and samples were col-
lected after each column cycle, for a total of six cycles. Total
energy consumption (kWh) was measured for the MW reactor, stir-
rer functions and pump.
(0.55 kWh). The LA and GA mixture was a good starting materi-
al for the synthesis of bio-based oligomers (PLA and PLGA) via
direct polycondensation. A rotovap-like MW reactor working in
vacuum was used to develop a solvent- and catalyst-free pro-
tocol under mild conditions (1308C, 3 h). MW-assisted proce-
dures certainly play a fundamental role in the design of sus-
tainable methods which address the biorefinery concept.
Experimental Section
Reagents and equipments
Reaction mixtures were passed through laboratory filter paper to
remove insoluble residues. This solid was washed three times with
distilled water (3ꢁ5 mL) and all filtrates were combined. The solu-
tion was stirred together with a strongly acidic cation-exchange
resin (Amberlite IR 120 resin; Rohm and Haas, exchange capacity=
5.05 mequiv/g) for 15 min at room temperature to remove the cat-
alyst. A cloudy oil was obtained after freeze drying (À808C,
0.2 mbar). 1H NMR analyses were performed in D2O. GC-MS chro-
matograms were recorded after preliminary product derivatization
(10 mg) in pyridine (1 mL) using BSTFA (N,O-bis(trimethylsilyl)tri-
fluoroacetamide) (300 mL). The mixture was heated for 45 min at
608C under magnetic stirring and the product was directly ana-
lysed by GC-MS. GC conditions were: injection split 1:20, injector
All chemicals and solvents were purchased from Sigma–Aldrich
and used without further purification. Cellulose and PHTP were
ball milled in a planetary ball mill (PM100 Retsch GmbH), which
was equipped with a single milling beaker (stainless steel, 125 mL)
which used 25 stainless steel milling balls (Ø=10 mm). In the flow
experiments, ball-milled cellulose was suspended in water using
a dedicated ultrasonic device: a titanium immersion horn (Danaca-
merini, Italy) working at 21.1 kHz under temperature control. Batch
MW-assisted reactions were carried out in a SynthWAVE (MLS
GmbH, Milestone Srl) reactor, which houses a closed MW-cavity.
RotoSYNTH (MLS GmbH, Milestone Srl) was used for polycondensa-
tion reactions. Flow experiments were carried out in a FlowSYNTH
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
7
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[16] M. Dusselier, P. Wouwe, A. Dewaele, E. Makshina, B. Sels, Energy Environ.
Sci. 2013, 6, 1415–1442.
temperature 2508C, detector temperature 2808C. Gas carrier:
helium (1.2 mLminÀ1), temperature program: from 708C (2 min) to
3008C at 58CminÀ1
.
[17] L. Yan, X. Qi, ACS Sustainable Chem. Eng. 2014, 2, 897–901.
[18] D. Esposito, M. Antonietti, ChemSusChem 2013, 6, 989–992.
[19] Y. Wang, W. Deng, B. Wang, Q. Zhang, X. Wan, Z. Tang, Y. Wang, C. Zhu,
Z. Cao, G. Wang, H. Wan, Nat. Commun. 2013, DOI: 10.1038/
ncomms3141.
[20] Microwaves in Organic Synthesis 3rd Edition 1 (Eds.: A. De La Hoz, A.
Loupy), Springer Science, New York, 2012, pp. 541–562.
[21] G. Cravotto, P. Cintas, Chem. Eur. J. 2007, 13, 1902–1909.
[22] Microwave-assisted extraction for bioactive compounds: Theory and prac-
tice, Food Engineering Series, Vol. 4 (Eds.: F. Chemat, G. Cravotto), Spring-
er Science, New York, 2013.
MW-assisted polycondensation: The required amounts of lactic and/
or glycolic acids were added to a dedicated 50 mL RotoSYNTH
flask and the reaction mixture was kept in rotation. The MW pro-
gram was as follows: 10 min from 258C to 1308C, then 3 h at
1308C. Pressure was gradually decreased from 1000 to 70 mbar
during the ramp time. The resulting oligomers were dissolved in
CH2Cl2 after cooling to room temperature and precipitated in an
excess of methanol. Purified products were dried under vacuum.
[23] S. Tabasso, D. Carnaroglio, E. Calcio Gaudino, G. Cravotto, Green Chem.
2015, DOI: 10.1039/C4GC01545B.
[24] J. Fan, M. De bruyn, V. L. Budarin, M. J. Gronnow, P. S. Shuttleworth, S.
Breeden, D. J. Macquarrie, J. H. Clark, J. Am. Chem. Soc. 2013, 135,
11728–11731.
[25] A. Szabolcs, M. Molnꢂr, G. Dibꢃ, L. Mika, Green Chem. 2013, 15, 439–
445.
[26] S. Tabasso, E. Montoneri, D. Carnaroglio, M. Caporaso, G. Cravotto,
Green Chem. 2014, 16, 73–76.
Acknowledgements
This work was funded by the University of Turin (fondi ricerca
locale 2013). Part of this work was carried out by Giorgio Grillo
during his master thesis in Chemistry.
[27] Y. Q, Q. We, H. Li, P. Oleskowicz-Popiel, C. Huang, J. Xu, Bioresour. Tech-
nol. 2014, 162, 358–364.
[28] X. Qi, M. Watanabe, T. M. Aida, R. L. Smith, Jr., Catal. Commun. 2009, 10,
1771–1775.
Keywords: lactic and glycolic acids · lignocellulosic biomass ·
microwaves · PLGA · waste valorization
[29] S. Dutta, S. De, A. K. Patra, M. Sasidharan, A. Bhaumik, B. Saha, Appl.
Catal. A 2011, 409–410, 133–139.
[30] A. Dutta, D. Gupta, A. K. Patra, B. Saha, A. Bhaumik, ChemSusChem
2014, 7, 925–933.
[31] D. Wang, W. Niu, M. Tan, M. Wu, X. Zheng, Y. Li, N. Tsubaki, ChemSus-
Chem 2014, 7, 1398–1406.
[32] F. Hu, A. Ragauskas, Bioenerg. Res. 2012, 5, 1043–1066.
[33] K. McQuinn, F. Hof, J. McIndoe, X. Chen, G. Wu, A. J. Stace, Chem.
Commun. 2009, 4088–4090.
[34] T. Shi, G. Oriove, J. Guo, D. Bohme, A. C. Hopkinson, K. W. M. Siu, J. Am.
Chem. Soc. 2004, 126, 7975–7980.
[35] D. K. Gilding, A. M. Reed, Polymer 1979, 20, 1459–1464.
[36] M. Vert, S. M. Li, H. Garreau, Clin. Mater. 1992, 10, 3–8.
[37] D. J. Mooney, D. F. Baldwin, N. P. Suh, J. P. Vacanti, R. Langer, Biomaterials
1996, 17, 1417–1422.
[38] J. L. Cleland, O. L. Johnson, S. Putney, A. J. S. Jones, Adv. Drug Delivery
Rev. 1997, 28, 71–84.
[39] N. Wang, X. Shen Wu, L.-U. Hannia, E. Donahue, A. Siddiqui, J. Biomat.
Sci. Polym. Ed. 1997, 8, 905–917.
[1] A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J. Cairney,
C. A. Eckert, W. J. Frederick Jr., J. P. Hallett, D. J. Leak, C. L. Liotta, J. R.
Mielenz, R. Murphy, R. Templer, T. Tschaplinski, Science 2006, 311, 484–
489.
[2] M.-M. Titirici, M. Antonietti, Chem. Soc. Rev. 2010, 39, 103–116.
[3] J. Gonzꢂlez-Rivera, I. R. Galindo-Esquivel, M. Onor, E. Bramanti, I. Longo,
C. Ferrari, Green Chem. 2014, 16, 1417–1425.
[4] S. K. Hoekman, A. Broch, C. Robbins, Energy Fuels 2011, 25, 1802–1810.
[5] W. He, G. Li, L. Kong, H. Wang, J. Huang, J. Xu, Resour. Conserv. Recycl.
2008, 52, 691–699.
[6] T. M. Aida, Y. Sato, M. Watanabe, K. Tajima, T. Nonaka, H. Hattori, K. Arai,
J. Supercrit. Fluids 2007, 40, 381–388.
[7] R. Datta, M. Henry, J. Chem. Technol. Biotechnol. 2006, 81, 1119–1129.
[8] C. S. M. Pereira, V. M. T. M. Silva, A. E. Rodrigues, Green Chem. 2011, 13,
2658–2671.
[9] C. O. Tuck, E. Perez, I. T. Horvath, R. A. Sheldon, M. Poliakoff, Science
2012, 337, 695–699.
[10] G.-Q. Chen, M. K. Patel, Chem. Rev. 2012, 112, 2082–2099.
[11] M. Ajioka, H. Suizu, C. Higuchi, T. Kashima, Polym. Degrad. Stab. 1998,
59, 137–143.
[12] N. Wang, X. Shen Wu, H. Lujan-Upton, E. Donahue, A. Siddiqui, J. Bio-
mater. Sci. Polym. Ed. 1997, 8, 905–917.
[40] Z.-Y. Wang, Y.-M. Zhao, F. Wang, J. Wang, J. Appl. Polym. Sci. 2006, 99,
244–252.
[41] F. Achmad, K. Yamane, S. Quan, T. Kokugan, Chem. Eng. J. 2009, 151,
342–350.
[13] N. Wang, X. Shen Wu, C. Li, M. Fang Feng, J. Biomater. Sci. Polym. Ed.
2000, 11, 301–318.
[14] A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411–2502.
[15] M. Bicker, S. Endres, L. Ott, H. Vogel, J. Mol. Catal. A 2005, 239, 151–
157.
Received: October 26, 2014
Revised: November 24, 2014
Published online on && &&, 0000
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
8
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
D. Carnaroglio, S. Tabasso, B. Kwasek,
D. Bogdal, E. C. Gaudino, G. Cravotto*
&& – &&
From Lignocellulosic Biomass to
Lactic- and Glycolic-Acid Oligomers: A
Gram-Scale Microwave-Assisted
Protocol
Not drowning but microwaving: A
flash catalytic conversion (2 min) of
lignocellulosic biomass into lactic and
glycolic acids under microwave irradia-
tion is reported. The batch procedure is
successfully adapted to a microwave-as-
sisted flow process, with the aim of de-
signing a scalable process with higher
productivity. Good yields and conver-
sions are maintained, productivity is in-
creased and energy consumption is low.
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
9
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ