C H A P T E R

14
Leveraging novel green solvents to drive
conceptual and practical biorefinery
innovation
Agnese Ciccia, Marco Bravib,*
a
Bio-P srl, via di Vannina, Roma, RM, Italy; bDipartimento Ingegneria Chimica Materiali Ambiente, via
Eudossiana, Roma, Italy
* Corresponding author. e-mail address: [email protected]

1. Introduction regulatory issues. However, material balance

closure constrains the amount of each fraction
The influence of oil and the omnipresence of that is produced by a given biorefinery process,
oil-derived products on the market can possibly the production rates of which are then in defined
be duplicated by biomass and bio-refined sub- ratios. Entirely covering the market capacity for
stances, respectively, only if biorefining technol- one product might lead to exceeding market ca-
ogy achieves a degree of optimization similar to pacity for a coproduct, thus calling for an inte-
that attained by oil refining technology, and if grated marketing strategy that is very new for
the adopted platform chemicals resemble to- most of the current players in biomass processing.
day’s ones in convenience. From a commercial In a traditional oil refinery, an essential
point of view, compounds, or fractions, which portion of the refining tasks relies on highly opti-
can be obtained from higher plants (complex mized networks of distillation columns served
vegetal organisms, such as vascular plants), mac- by a highly thermally integrated network of
roalgae, and especially microalgae may have heat exchangers. Therefore, distillation serves
multiple markets that may require different as the single most important unit operation.
specifications, impose different regulations, and Learning the lesson from oil refining history
consequently attach a different value to the begins with identifying a single (and ideally
very same semifinished product. Hence, a pro- “universal”) unit operation, which should be
cess should aim at “market-neutrality” as much flexible enough to be up to the hard challenge
as possible, to avoid putting restrictions on of addressing the fractionation of a diverse
the market placement of a product due to pool of biomass types, by efficiently separating

Catalysis, Green Chemistry and Sustainable Energy

https://doi.org/10.1016/B978-0-444-64337-7.00014-8 243 Copyright © 2020 Elsevier B.V. All rights reserved.
244 14. Leveraging novel green solvents to drive conceptual and practical biorefinery innovation

very different fractions [1]. As a route toward

that (admittedly extremely ambitious) objective,
the identified unit operation should target at
least those biomasses whose fractionation is ex-
pected to provide the maximum (specific, or cu-
mulative) revenue, and should preferably avoid
adopting fundamental steps that would
completely defeat the purpose for other impor-
tant biomass types. Inasmuch as the diversity
and complexity of biomasses is greatly higher
than the diversity among crude oils, biomasses
are solids, so the question arises as to whether,
and in which way, extraction mediated by a
liquid can be suitably operated to serve as the
single, analogous-to-distillation unit operation
set of the biomass era.
The present chapter strives to give a bleeding- FIGURE 14.1 Layout of cellulose, hemicellulose, and
edge account of sustainable techniques to perform lignin in lignocellulosic cell walls. US Department of Energy
a market-neutral biomass fractionation devel- Genome Programs image gallery, http://genomics.energy.gov.
oped in our and other leading laboratories,
with a skew toward techniques targeting a sub- solvation of the fragments. Hemicellulose and
stantial incorporation of the green chemistry lignin are not only physically interweaved but
principles [2] and explore paradigmatic shifts, also cross-linked by covalent bonds. Overcoming
and it argues about the role of green solvent tech- such cross-linking, to which cell walls owe their
nology in designing the future of biorefineries. rigidity and enzymatic recalcitrance, generally re-
quires chemical hydrolysis [5].
The biorefining of lignocellulosic materials
2. Classical biomass fractionation aims at the maximal recovery of both hexose
processes sugars, which are easily fermented, and address-
ing pentose sugars (especially xylose, which is
The structure of vegetal biomass is mainly frequently the most abundant member of this
composed of polysaccharides (starch, cellulose, class) recovery and utilization to improve the
hemicellulose), proteins, lipids, and lignin with a fractionation economy, especially in low-value
complex, irregular, and biomass type-dependent applications (e.g., biofuels) [6]. The recovery of
three-dimensional arrangement. Fig. 14.1 illus- nonfermentable fractions as coproducts (e.g.,
trates this arrangement between biopolymers lignin) multiplies value-added streams compared
for the case of lignin-containing cell walls. to the case where only fermentation products are
Lignin constitutes a major obstacle to energy- sought [7]. Many approaches are being explored
efficient biomass deconstruction [3] because to obtain this value-improving fractionation,
lignin prevents enzyme access to carbohydrates which resort to physical, chemical, and combined
and unproductively binds them [4]. Most chem- means [8]. One of these is the disassembling of
ical depolymerization approaches modify lignin, biomass by resorting to ionic liquids, as described
but only some remove it from the pulp (e.g., elsewhere in this chapter.
organosolv, sulfite, and Kraft pulping) usually The structure of biomasses can be frequently
by a combination of chemical fragmentation subdivided in recognizable areas characterized

IV. Selected examples and case history

 2. Classical biomass fractionation processes 245
with a certain level of compositional uniformity which implies that more than half of the proteins
and a characteristic size. These structures have can be solubilized by a single extraction step.
specific names, relevant to their function (e.g., Protein extraction from green biomass has a
pyrenoid), shape (e.g., oil bodies), or structural major hindrance in cell walls, which are hard
aspect (e.g., starch granules). How abruptly the and insoluble. Usually, plant leaf proteins are
composition changes from one defined structure first extracted by mechanical pressing, and sub-
to another defined structure, the characteristic sequently recovered by precipitating coextracts
size of said structures, and the hardness of the (fiber, chlorophyll, and other components) by
biomass matrix (either because it is surrounded heat treatment, pH shift, flocculation, or mem-
by a hard shell or because it features a uniformly braneless osmosis. Only soluble proteins can be
hard structure) influences the accessibility of the obtained at high purity [13], while insoluble pro-
biologic fractions contained therein. teins are recovered in an impure fraction also
In storage tissues, biologic fractions are accu- containing pigments (chlorophylls), fiber-based
mulated in discrete entities distributed within a debris, and salts. To date, further processing is
millimeter size range (Fig. 14.2), which facilitates required to separate this fraction into cellulosic
fragmentation and the subsequent extraction. material, pigments, and proteins, which increases
Compared to storage tissues, the anatomy and overall processing costs. With current technol-
biologic function of green biomass tissues fea- ogy the most economically feasible approach
tures a much smaller-length scale, with a twofold is carrying out the design of leaf protein extrac-
implication: a finer mechanical fragmentation is tion and the design of the final product to be
required and the number of bonds/interactions obtained together, because the required nutri-
to be severed to obtain the separation of fractions tional and functional properties expected from
is much higher. In lack of a specifically tailored a pure compound may be equaled or even
(or sufficiently general and, at the same time, exceeded by the impure protein stream.
energy-efficient) “cutting technology,” the more As far as polysaccharides are concerned, no
intensive processing conditions required may general method is available for their extraction
lower the value of side streams and compromise from plant material. Moreover, the ratio between
the overall process sustainability. soluble and insoluble fractions is, for several
Green biomass and protein crops exhibit a polysaccharides, dependent on extraction con-
marked physiologic difference: green biomass ditions (e.g., physical pretreatment, enzymatic
is mostly comprised by structural and enzymatic treatment, temperature, time). Isolating cell-wall
proteins (carrying out the photosynthetic pro- polysaccharides generally follows this sequence:
cess), while seed or protein crops are composed removal of low-molecular-weight carbohydrates;
mainly by storage proteins (aimed at the next minimization of the activity of cell-wall-degrading
generation of the plant) generally without an enzymes (e.g., by hot aqueous ethanol); selec-
enzymatic activity, so they are more stable dur- tive dissolution of noncell-wall proteins (e.g.,
ing processing [12]. In both biomass types, pro- by phenoleacetic acidewater); delignification
teins have been classed making reference to (when required, e.g., by sodium chloriteeacetic
their solubility. In storage proteins, four protein acid); polysaccharides extraction (some pro-
classes or “Osborne fractions” are recognized: al- teins will also be extracted): pectic substances
bumins (soluble in water); globulins (soluble in can be extracted with an aqueous solutions of
saline solutions); prolamins (soluble in alcohol), chelating agents (e.g., EDTA). Other noncellu-
and glutelins (soluble in alkali). In each vegetal losic polysaccharides are preferentially extracted
species, usually one or two categories accounts using alkaline aqueous solutions. The extracted
for more than half of the total proteins (Table 14.1), polysaccharides may be recovered by

IV. Selected examples and case history

246 14. Leveraging novel green solvents to drive conceptual and practical biorefinery innovation

FIGURE 14.2 Vegetal materials domains and their average size. Duplicated with permission from A.T. Tenorio, K.E. Kyriako-
poulou, E. Suarez-Garcia, C. van den Berg, A.J. van der Goot, Understanding differences in protein fractionation from conventional crops,
and herbaceous and aquatic biomass-Consequences for industrial use, Trends in Food Science and Technology 71 (2018) 235e245. Orig-
inal 3D stacked thylakoids, based on A.V. Ruban, M.P. Johnson, Visualizing the dynamic structure of the plant photosynthetic membrane,
Nature Plants 1 (11) (2015) 15161. Algal cell of Chlorella vulgaris, based on C. Safi, B. Zebib, O. Merah, P.Y. Pontalier, C. Vaca-Garcia,
Morphology, composition, production, processing and applications of Chlorella vulgaris: a review, Renewable and Sustainable Energy Re-
views 35 (2014) 265e278.

precipitation with alcohol and other antisol- Starch is naturally found tightly and radially
vent substances, and then purified (by adsorp- packed into dehydrated granules with origin-
tion, ion-exchange, and partition specific shape and size, varying from few mi-
chromatographic methods) [14]. crons to 100 mm in size [15]. The size distribution

IV. Selected examples and case history

 2. Classical biomass fractionation processes 247
TABLE 14.1 Process protein extractions by vegetable species.

From A.T. Tenorio, K.E. Kyriakopoulou, E. Suarez-Garcia, C. van den Berg, A.J. van der Goot, Understanding differences in protein fractionation from
conventional crops, and herbaceous and aquatic biomass-Consequences for industrial use, Trends in Food Science and Technology 71 (2018) 235e245
with permission.

determines its swelling functionality with gran- starch concentration). However, a large granule
ules being generally either larger (A-starch), or size leads to higher swelling at equally high
smaller (B-starch) with a lesser tendency to amylose content [17]. Microalgae are novel
swelling [16]. Granules contain small blocks of highly efficient starch producers [18], and they
amylopectin that feature crystalline and amor- produce only small starch granules (size up to
phous areas. During the absorption of water, 2 mm) with a narrow size distribution range
they swell and diminish their crystallinity, so [19]. In view of its relevant technologic proper-
they leach amylose. A high amylose content ties (crystallinity, molecular weight, and thermal
leads to lower swelling and gel strength (at equal properties) resembling those of other sources,

IV. Selected examples and case history

248 14. Leveraging novel green solvents to drive conceptual and practical biorefinery innovation

microalgal starch could prove a valid alternative found in the residue). Spray drying permits hav-
to plant starch in several fields [20]. ing dry fractions.
Fiber is a heterogeneous fraction, which has Oil extraction from oilseeds is generally car-
a unitary definition in the food context, but a ried out by cracking, dehulling, heat treatment,
multifaceted chemical nature. Codex Alimentarius flaking, extrusion, prepressing (when oil content
[21] defines fiber (Table 14.2) as carbohydrates warrants it), and solvent extraction. Pressing,
that cannot be digested and have beneficial especially at low temperature, provides better
functional effects in the human body. Fiber retention of chemical and nutritional characteris-
composition comprises cellulose, hemicelluloses, tics of oil with respect to extraction with organic
pectin, gums, inulin, resistant starches, and some (hexane) solvent [23]. This latter, in turn, permits
nonpolysaccharidic compounds (such as waxes, a thorough exhaustion of the seed matrix than
cutin, saponins, phytates, polyphenols, and the former; hence, it is used as a second oil
resistant proteins) that, in a biorefinery perspec- extraction step to ensure a thorough oil recovery
tive, are all but a uniform fraction. from the matrix. An alternative to hexane-aided
While the extraction method of choice used is extraction aiming at high-quality products is
dependent on the composition of the specific extraction mediated by supercritical CO2 (SCe
fiber fraction, fiber recovery can usually be effi- CO2). Although to date a number of studies
ciently obtained by the water-saving wet milling have investigated and optimized the operating
method, which involves finely grinding the conditions [24] and the flowsheet [25] of SCe
seeds, then extracting proteins at an alkaline CO2 extraction from oilseeds, vegetal and micro-
pH (usually by NaOH) and recovering them by algal matrices [26], supercritical fluid extraction
acidic precipitation (using HCl) [22]. Insoluble still remains an energivorous operation.
fiber is separated by exploiting the difference be- The downstream processing of bulk products
tween fractions swelling properties. For instance, obtained by industrial biotechnology is currently
fiber has a high swelling capacity at room tem- sensibly more expensive than biorefining of
perature, while the starch swelling is limited, microalgae into multiple products (about twice
and wet sieving is used to separate starch (mainly as much on a total production cost basis) [27], a
found in the supernatant) from fiber (mainly figure that clearly shows that usual technologies
do not scale down well to more finely distrib-
uted, entangled/bonded fractions.
TABLE 14.2 Chemical fractions composing the fiber
food fraction of vegetal matrices [22].

Component Solubility 3. Novel green solvents

Cellulose Insoluble
One further step in learning the lesson of oil
Pectin Soluble
refining may be taken by recalling analogies
Lignin Insoluble among unit operations from classical undergrad-
Hemicellulose Insoluble uate unit operations courses: in distillation, the
separation between the compounds that occurs
Gums Soluble
in the feed streams as a mixture requires supply-
Mucilage Soluble ing heat (which is the “operating cost item”) to
Inulin Soluble take place, and it is effected by relative volatility
(the “driving force” of the separation). This sep-
Resistant starch Insoluble
aration between compounds that form a solution

IV. Selected examples and case history

 3. Novel green solvents 249
in the feed takes place under physical hindrances ability, liquidesolid and liquideliquid extrac-
of diffusional nature. tion unit operations can be innovated [29]. Bio-
In solvent-mediated extraction, the separation refinery operations can also be made “greener”
requires a suitable solvent (which represents the because biomass can be processed without ne-
cost item) and is prompted by the solubility of cessity of prior drying, so pretreatment is
substances in that solvent (the driving force), un- simpler and cheaper. The high affinity of a
der physical (entanglement) and chemical switchable solvent in its hydrophobic state to-
bonding hindrances. Along oil refineries history, ward nonpolar compounds can be exploited to
optimizing the fractionation of a complex feed extract oil from a dense, water-containing sus-
where many chemical compounds occur in pension of microalgae. Afterward, when the
mixture has signified deploying an array of distil- SHS is transformed into its hydrophilic form
lation columns, requiring heat to be supplied and (by contact with CO2), following solventelipid
withdrawn at different temperature levels. Opti- phase separation, oil can be recovered. By
mizing an array of distillation column, thus, has removing the CO2 (which can be obtained by
implied optimizing the heat integration between stripping, with N2, and/or by applying heat)
needs of supplying heat and needs of subtracting [30], the switchable solvent system (SSS) returns
heat, in turn obtained by reusing the “same heat” to the hydrophobic state and the cycle can be
at “different temperature levels” and by closed. The SSS cycle is completed by transform-
designing the whole required heat exchange ing the SSS to the nonpolar form again.
network by using the thermal pinch technology. Differently from switchable solvents, NaDESs
The question arises, hence, as to whether use- feature a hydrophilicity that can be assigned by a
ful hints toward optimized biomass fraction- proper choice of the mixture components, and
ation can be found by taking the analogy route that can be sometimes and somewhat tuned by
and looking for “reusable solvents” across adjusting the amount of one extra component
“different solubility features” (i.e., solvents that (so far, only water). They belong to the class of
may be modified, and thus be adapted) for low transition temperature mixtures (LTTMs)
very different tasks in the extraction process. and are synthesized by a complex formation
Modifiable-solvent technology has been used in between a hydrogen-bond donor (HBD) and
biomass processing for quite a long time now, hydrogen-bond acceptor (HBA) chosen among
possibly starting from the introduction of super- natural substances (metabolites) and feature a
critical fluid extraction (SFE) technology. In more temperature-depressed solidification point with
recent times, this class of solvents has come to respect to their constituents, due to the delocal-
include gas (generally, CO2)-expanded liquids ization of charge (Fig. 14.3).
(GXLs), hydrophilicity- and polarity-switchable Table 14.3 shows various HBA and HBD com-
solvents (SHSs and SPSs), while the role of natu- ponents that can be combined to form NaDES.
ral deep eutectic solvents (NaDESs) (synthesized Hydrophilic NaDESs can either be prepared
by a complex formation between a hydrogen- from the aqueous solutions consisting of each
bond donor and a hydrogen-bond acceptor, compound in which a component is in the
both substances being natural metabolites) is molten state and the other component is dis-
still settling [28]. solved, and then evaporating the added water
Switchable solvents are in situ feature chang- (if necessary) or by mixing the two solid compo-
ing solvents (from hydrophobic to hydrophilic nents heated to a predetermined value.
and vice-versa, or from nonpolar to polar and The most commonly synthesized NaDES have
vice-versa) that belong to the more general class shown high miscibility in water. The synthesis of
of switchable materials (SMs). By deploying this hydrophobic DES is relatively recent, and the

IV. Selected examples and case history

250 14. Leveraging novel green solvents to drive conceptual and practical biorefinery innovation

TABLE 14.3 Commonly used HBA and HBD for

synthesis of NaDESs [31].

Hydrogen-bond
acceptors (HBA) Hydrogen-bond donors (HBD)

Oxalic acid Ethylammonium chloride

Malonic acid Nicotinic acid
Xylitol Trimethylammonium chloride
Urea Betaine
Benzoic acid Choline nitrate
Citric acid Tetraethylammonium chloride

Imidazole Choline chloride

Cinnamic acid Histidine
Ethylene glycol Alanine
FIGURE 14.3 A NaDES0 solidification temperature is D-sorbitol Acetylcholine chloride
lower than either its constituents’ solidification temperature.
Duplicated from S. Sharma, A.K. Kumar, Role of natural deep Lactic acid Glycine
eutectic solvents (NADES) in the pretreatment of lignocellulosic
Levulinic acid Choline fluoride
biomass for an integrated biorefinery and bioprocessing concept.
In Recent Advancements in Biofuels and Bioenergy Utilization Glycerol Proline
(pp. 73e109). Springer, Singapor e2018, with permission.
Succinic acid Lidocaine
Hexanoic acid Methyltriphenylphosphonium chloride
appearance of hydrophobic NaDES has spurred Stearic acid Tetrabutylammonium chloride
studies concerning novel processing routes to
replace conventional organic solvents. A notable Oleic acid
example of such a NaDES is the object of one bio- Linoleic acid
refinery application described in the following. Adipic acid
Suberic acid

4. Biorefinery processes using alternative

solvents 4.1 “Circular extraction” with
Novel solvent systems (such as those that can
switchable solvents
be modified in situ and those capable of forming Process developers have deployed SHS only
multiple water-based, yet immiscible phases) in their hydrophobic form, using them to extract
and process technologies based on them are hydrophobic solutes from complex biomass. Any
increasingly developed to tackle the two major hydrophilic substances and debris contained in
challenges that biomass offers: the fractionation the raw biomass would remain undissolved. Hy-
of lignocellulosic materials, which make the drophilic solutes would be easily extracted by
largest biomass pool of today, and microalgae, treating the leftover of the hydrophobic solvent.
which represent the biomass with the largest However, by noting that, after the “switch,” the
(and richest) growth potential of tomorrow. SHS has acquired the hydrophilic character, it

IV. Selected examples and case history

 4. Biorefinery processes using alternative solvents 251
should be argued that the “switched” solvent
may be suitable for the extraction of hydro-
philic material from said leftover. During the
hydrophobic-to-hydrophilic switch, dissolved
hydrophobic solutes separate as a second (hy-
drophobic) phase that is easily collected. During
the hydrophilic-to-hydrophobic switch, the SHS
in hydrophobic form separates, leaving a hydro-
philic phase constituted by water and water-
loving solutes behind. These latter might be
recovered as a precipitate, be found enclosed in
micelles, or be dissolved as true solutes, and a
staged recovery could be performed based on
existing methods.
By noting that both hydrophilicity states of
the original switchable solvent can be used in
two specific and complementary extraction tasks,
it can be concluded that the biomass extraction
can begin with either the hydrophobic or the
hydrophilic form of the solvent, and it may
continue with the other form. For a natively
hydrophobic switchable solvent, the operation
sequence composed first by a hydrophobic
extraction, followed by a hydrophilic extraction FIGURE 14.4 Sequential two-stage extraction in forward
(after solvent switch) was denoted by the in- (first hydrophobic then hydrophilic, (A)) and backward (first
ventors as forward-mode extraction (FME) [32]. hydrophilic then hydrophobic, (B)) modes. Adapted from G. Sed,
Conversely, the opposite sequence (hydrophilic A. Cicci, P.G. Jessop, M. Bravi, A novel switchable-hydrophilicity,
natural deep eutectic solvent (NaDES)-based system for bio-safe bio-
extraction first, then hydrophobic extraction af- refinery, RSC Advances 8 (65) (2018) 37092e37097.
ter backward switch of the solvent) was called
backward-mode extraction (BME) (Fig. 14.4). It carried out at atmospheric pressure and moder-
should be noted that the solvent needed to carry ate temperature and are thus prone to be wise
out BME can be obtained by forward-switching in energy use. A number of chemicals exhibiting
pure switchable solvent in an experiment, but switchable behavior (such as DMCHA and
it is normally available as a result of solvent others) provide a specific additional perfor-
recycling after hydrophobic-to-hydrophilic switch mance as a cell-breaking means, thus bringing
and hydrophobic substances separation. a substantial simplification and reduction in
This dual use of SHSs, denoted as “circular the energy needed to access the individual bio-
extraction,” has been touted to provide a twofold logic fractions (proteins, carbohydrates, and
utility increase. Indeed, the solvent performs as a lipids). These solvents suffer from moderate to
second solvent and, across the switching pro- severe toxicity (expressed by the LD50 of the
cess, it can be used to perform a further, and switchable compound) that raises concerns
complementary, extraction task. The environ- with food-, feed-, cosmetic-, or pharma-bound
mental impacts due to the solvents are also sub- extracts. So far, the open literature list of food-
divided over two extractions per cycle. The safe switchable systems is very short. In this
operations comprised in circular extraction are short list an SHS (NaDES-Y) based on a dual

IV. Selected examples and case history

252 14. Leveraging novel green solvents to drive conceptual and practical biorefinery innovation

fatty acid-based NaDES together with a lean 4.2 Aqueous two-phase systems
amine (Jeffamine) water solution represents the extraction
first NaDES-based switchable solvent system
ever [33]. Being that Jeffamine is used in very Aqueous two-phase systems (ATPSs) are
limited amount, and being of very low toxicity formed by two solutions, the former originated
in itself, the toxicity of the whole solvent system by dissolving a polymer in water above its crit-
is practically nil. The capability of this system ical concentration and the latter by dissolving
to extract microalgal biomass fractions was either a second polymer (immiscible in the
tested by resorting to the previously shown “cir- former one, and again with a concentration
cular extraction” scheme. From a process simpli- exceeding the critical value) or an inorganic
fication point of view, NaDES-Y by itself was salt. The two described solutions are immiscible,
found incapable of breaking cells: a supplemen- so they form two distinct phases. Generally, the
tary agent must be applied simultaneously to polymer-based solution lays above the inorganic
achieve this result, such as bead beating, micro- salt-based solution [34]. ATPSs contain much
waving, and possibly ultrasounds, electrical more water (80%e99% by weight) than conven-
fields, and others. However, with such a simulta- tional aqueous organic extraction systems, and
neous complement, NaDES-Y exhibits nearly they exhibit an exceptionally low interfacial ten-
equal performance to DMCHA for all the exam- sion (105 N$m). ATPSs can also be produced
ined biologic fractions (results reported for beads by deploying ionic liquids and short chain alco-
beating) and even higher than that of DMCHA hols, and micellar and reverse micellar forms of
for proteins and carbohydrates (results reported ATPSs can be formed by ionic and/or nonionic
for microwave-assisted extractions) (Tables 14.4 surfactants.
and 14.5). Therefore, it must be observed that The partitioning of crude proteins and carbo-
the hydrophobic NaDES-Y solvent, together with hydrates contained in microalgal crude extract
the choice between the forward- or backward- by an ionic liquid-based aqueous two-phase sys-
circular extraction mode gives the biochemical tem has been demonstrated [35]: proteins and
process engineer a powerful bio-safe tool. The sugars are selectively fractionated in the top
separation of the whole biomass into individual and in the bottom phases respectively. This
compounds or fractions involves (1) overcoming result is clearly extremely interesting for the bio-
the physical hindrance and (2) dissolving the refinery process developer, although the authors
fraction(s) of interest. Step (2) in turn depends warn that effective strategies for recycling the
on the actual solubility of each fraction, which ionic liquid after protein extraction are lacking
may depend on its hydrophilic/hydrophobic and that the functionality of the obtained pro-
character, pH, ionic strength, and temperature. teins should be assessed. Another extremely
In real life, a fraction may have a hybrid char- interesting experimental evidence [36] shows
acter (e.g., a lipoprotein, a glycolipid, a phospho- that an ionic liquid (PEG400-Ch DHP)-based
lipid, or something else) and will be separated system can be very selective for separating pig-
along with the other fractions it most resembles, ments (concentrating in the top phase) and pro-
or it will dissolve in structured form together teins (concentrating at the interface and in the
with other substances (micelles). By choosing a bottom phase): lutein was almost quantitatively
condition (e.g., pH), or by distributing an array (97%) extracted by the top phase, with very
of conditions over time, extraction and/or recov- limited contamination from proteins. Proteins,
ery can be sequenced. in turn, moved preferentially to the interface

IV. Selected examples and case history

 TABLE 14.4 Fractional extraction efficiency in forward-mode with DMCHA and NaDES-Y. SE: simply agitated extraction; BB: bead beating-
assisted extraction; MW: microwave-assisted extraction; p.d.: pigments degraded.

4. Biorefinery processes using alternative solvents

IV. Selected examples and case history

Proteins Carbohydrates Lipids (neutral)

Forward-mode DMCHA NaDES-Y NaDES-Y NaDES-Y DMCHA NaDES-Y NaDES-Y NaDES-Y DMCHA NaDES-Y NaDES-Y NaDES-Y
dual extractions SE SE BB MW SE SE BB MW SE SE BB MW

Stage I 7 4 p.d. 42 8 p.d. 96 88 p.d.

extraction:
hydrophobic

Stage II 34 32 n.a. 9 25 n.a. 0 0 n.a.

extraction:
hydrophilic
Total extraction 41 2 36 n.a. 51 5 33 n.a. 96 0.1 88 n.a.

Adapted from A. Holland, D. Wechsler, A. Patel, B. M. Molloy, A. R. Boyd, P. G. Jessop, Canadian Journal of Chemistry 90 (2012) 805e810; Y. Du, B. Schuur, S.R.A. Kersten, D.W.F. Brilman,
Algal Research 11 (2015) 217e283.

253
 254
14. Leveraging novel green solvents to drive conceptual and practical biorefinery innovation
TABLE 14.5 Fractional extraction efficiency in backward-mode with DMCHA and NaDES-Y. BB, bead beating-assisted extraction; MW,
microwave-assisted extraction; SE, simply agitated extraction.
IV. Selected examples and case history

System Metabolites (% DW)

Proteins Carbohydrates Lipids (neutral)

Backward-mode dual DMCHA NaDES-Y NaDES-Y DMCHA NaDES-Y NaDES-Y DMCHA NaDES-Y NaDES-Y
extractions SE BB MW SE BB MW SE BB MW

Stage I extraction: hydrophilic 50 17 60 47 8 18 0 0 0

Stage II extraction: 3 28 12 4 17 41 93 89 93
hydrophobic
Total extraction 53 45 72 51 25 59 93 89 93

Adapted from A. Cicci, G. Sed, P.G. Jessop, M. Bravi, Circular extraction: an innovative use of switchable solvents for the biomass biorefinery, Green Chemistry 20 (2018) (17) 3908e3911;
G. Sed, A. Cicci, P.G. Jessop, M. Bravi, A novel switchable-hydrophilicity, natural deep eutectic solvent (NaDES)-based system for bio-safe biorefinery, RSC Advances 8 (65) (2018)
37092e37097.
 4. Biorefinery processes using alternative solvents 255
(92% of proteins) and to the bottom phase. This hemicellulose, leaving solid cellulose behind.
interfacial partitioning is to be considered an Milder process conditions, wider choice of an-
advantage because it combines separation and ions, and significant moisture tolerance favor
concentration of proteins that do not need to be this latter approach [38]. Distinct features of ionic
extracted from the phase forming components liquid-based pretreatment include moderate
and, seemingly, appears to preserve the native (low pressure) process conditions and novel
protein conformation. product separations [39]. Recently, [TEA]
[HSO4] (triethylammonium hydrogen sulfate),
whose water solutions can be produced at bulk
scale at a price tag comparable to that of conven-
4.3 Ionic liquid-based lignocellulosic
tional organic solvents such as acetone and
biorefinery toluene [40], was shown [41] to exhibit the best
Ionic liquids are salts that are liquid at, or lignocellulose pretreatment performance. The
close to, room temperature. Modern ionic liquids workflow schematic diagram for the operation
contain organic cations (usually quaternized aro- (Fig. 14.5) was applied to Miscanthus giganteus
matic or aliphatic ammonium) and are either biomass and resulted in the following fraction-
inorganic or organic. Their negative charge is ated streams: (1) a cellulose rich material, (2) a
often distributed over several atoms. Anions precipitated lignin, and (3) a recovered ionic
often determine the ionic liquid solvation chem- liquid solution containing a range of solutes.
istry, but the cation can also be crucial in deter- Lignin yield, delignification, and saccharification
mining the dissolution effectiveness. yield loosely tracked each other during the time
Ionic liquids feature comparatively moderate course of the operation.
(low pressure) lignocellulose pretreatment pro- Recyclability of the ionic liquid solution was
cess conditions. Ionic liquid-based processing demonstrated and a technoeconomic analysis
may be used to either entirely dissolve the ligno- predicted that capital and operating cost would
cellulose [37] or dissolving just lignin and be lower than for the dilute acid pretreatment.

FIGURE 14.5 Workflow of bench-scale lignocellulose fractionation utilizing 80 wt% [TEA][HSO4] and 20 wt% water at a 1:
10 biomass:liquid ratio. A. Brandt-Talbot, F.J. Gschwend, P.S. Fennell, T.M. Lammens, B. Tan, J. Weale, J.P. Hallett, An economically
viable ionic liquid for the fractionation of lignocellulosic biomass, Green Chemistry 19 (13) (2017) 3078e3102. Published by The Royal
Society of Chemistry and reproduced with permission.

IV. Selected examples and case history

256 14. Leveraging novel green solvents to drive conceptual and practical biorefinery innovation

4.4 Natural deep eutectic solvent-based and the latter obtained by adding acetonitrile
lignocellulosic biorefinery as an antisolvent (xylan is obtained as a solid
precipitate). The process is depicted in Fig. 14.6.
Choline chloride-based NaDESs are of hydro- It can be observed that using acetonitrile taints
philic nature and exhibit higher polarity than the “intrinsically bio-safe” concept that could be
water. Their polarity can be tuned by adjusting claimed by the authors. However, it should
water content in the NaDES formulation itself. be noted that conceptual developments should
Because of their polarity, they are able to sol- be pursued on a parallel track with the develop-
ubilize lignocellulose. It has been shown [42] that ment of bio-safe equivalents thereof. Researchers
hydrophilic NaDES could replace other harsh will hopefully come across a bio-safe equivalent
chemical agents such as hydrochloric acid and of acetonitrile in its current role shortly.
sodium hydroxide for the pretreatment of ligno-
cellulosic agro-residues. A thorough process
scheme has been devised [43] that permits the 5. The market trap
fractionation of cellulose, hemicellulose, and
lignin; the process is based on a first extraction, Oil refineries fractionate crude oil providing
which solubilizes lignin and hemicellulose, fol- energetic goods and primary commodities. In-
lowed by two selective precipitations, the former dustry as a whole has developed and evolved
obtained by tuning NaDES polarity by adding the split to generate commodities to the market
water (lignin is obtained as a solid precipitate), with different requirements and values. The

FIGURE 14.6 Scheme of the NaDES-based lignocellulosic biorefinery (first step). Simplified from A.K. Kumar, S. Sharma, E.
Shah, A. Patel, Technical assessment of natural deep eutectic solvent (NADES) mediated biorefinery process: a case study, Journal of Mo-
lecular Liquids 260 (2018) 313e322.

IV. Selected examples and case history

 References 257
key point to determine the breakeven point for a simplification of process design would be
sustainable biomass biorefinery is the constraint possible beyond that of easing physical biomass
implied by the material balance, which requires fractionation. A number of optimizations might
high volumes of low-value products, implying a thus come, along the route of distillation trains
corresponding (according to the material balance and refinery plants evolution. A possible example
split) high volume of potentially high valued co- regards the development of conceptual exten-
products. However, providing large amounts of sions to current implementations of “mass trans-
these coproducts to specialized markets (such as fer pinch” technology, which could provide
the nutraceutical or cosmeceutical) lowers their powerful tools to adapt mass transfer networks
market value. It has been estimated [44] that if to biomass fractionation and help duplicate
30% of biofuels marked were to be covered by the ubiquitary deployment of thermal pinch
microalgae a 175 Mt/y feedstock of microalgal technology.
biomass would be required. If this biomass were
fractionated in all the energetic and functional
products envisaged, it would easily saturate the List of abbreviations and acronyms
market of these latter and negate the very same
sustainability from the biorefining hypothesis. ATPS(s) Aqueous two-phase system(s)
One can bet that high-valued product may BME Backward-mode extraction
develop a bit as a consequence of the price drop Ch DHP Cholinium dihydrogenphosphate
DES Deep eutectic solvent
and because other market opportunities (e.g., DMCHA N,N-dimethylcyclohexylamine
phytosanitary products) may become viable EDTA Ethylene diamine tetraacetic acid
again because of the price drop, and that this FME Forward-mode extraction
will reduce the saturation effect. However, esti- GXL(s) Gas expanded liquid(s)
mating these effects (“market elasticity”) is very HBA Hydrogen bond acceptor
HBD Hydrogen bond donor
hard and, consequently, business plans to sup- LD50 Lethal dose (for 50% of exposed subjects)
port decisions toward extensive biorefining are LTTM(s) Low transition temperature mixture(s)
difficult to set up. NaDES(s) Natural deep eutectic solvent(s)
PEG Polyethylene glycol
SC Supercritical
SFE Supercritical fluid extraction
6. Conclusions and future trends SHS Switchable hydrophilicity solvent
SM Switchable material
Fractionating a complex biomass containing SPS Switchable polarity solvent
different substances exhibiting different polarities/ SSS Switchable solvent system
TEA Triethylammonium
hydrophilicities traditionally requires deploy-
ing several solvents, thus complicating the full
exploitation of complex biomass (among which Acknowledgments
are microalgae) and impairing the biorefining
The authors of this contribution are indebted to Giorgia Sed
economy. for sharing, in good and bad times, the struggle for a unified
By resorting to novel solvents that can be view of biorefining, and whose constructive criticism was
made to change intrinsic features (such as hydro- essential in drafting this chapter.
philicity and polarity) at the process designers
will the “one-solvent-fits-all” target come into
range, especially if the adopted solvent also facil- References
itates chemical reactions (e.g., by promoting [1] A.D. Gonzalez-Delgado, V. Kafarov, Microalgae based
enzymatic activity): indeed, in this way a further biorefinery: evaluation of several routes for joint

IV. Selected examples and case history

258 14. Leveraging novel green solvents to drive conceptual and practical biorefinery innovation

production of biodiesel, chlorophylls, phycobilipro- starches, Journal of Food Engineering 50 (3) (2001)
teins, crude oil and reducing sugars, Chemical Engi- 141e148.
neering Transactions 29 (1) (2012) 607e612. [18] I. Branyikova, B. Marsalkova, J. Doucha, T. Branyik,
[2] P. Anastas, N. Eghbali, Green chemistry: principles and K. Bisova, V. Zachleder, M. Vítova, Microalgaednovel
practice, Chemical Society Reviews 39 (1) (2010) 301e312. highly efficient starch producers, Biotechnology and
[3] N. Mosier, C. Wyman, B. Dale, R. Elander, Y.Y. Lee, Bioengineering 108 (4) (2011) 766e776.
M. Holtzapple, M. Ladisch, Bioresource Technology [19] O.U.M. Tanadul, J.S. VanderGheynst, D.M. Beckles,
96 (2005) 673. A.L. Powell, J.M. Labavitch, The impact of elevated
[4] X. Pan, D. Xie, N. Gilkes, D.J. Gregg, J.N. Saddler, CO2 concentration on the quality of algal starch as a
Strategies to enhance the enzymatic hydrolysis of pre- potential biofuel feedstock, Biotechnology and Bioengi-
treated softwood with high residual lignin content, neering 111 (7) (2014) 1323e1331.
in: Twenty-Sixth Symposium on Biotechnology for Fuels [20] I. Gifuni, G. Olivieri, I.R. Krauss, G. D’Errico, A. Pollio,
and Chemicals, Humana Press, 2005, pp. 1069e1079. A. Marzocchella, Microalgae as new sources of starch:
[5] J.H. Grabber, R.D. Hatfield, J. Ralph, Journal of the Sci- isolation and characterization of microalgal starch
ence of Food and Agriculture 77 (1998) 193. granules, Chemical Engineering Transactions 57 (2017)
[6] Y.H. Zhang, Journal of Industrial Microbiology and 1423e1428.
Biotechnology 35 (2008) 367. [21] C. Alimentarius, Nutrition and Foods for Special Die-
[7] A.J. Ragauskas, C.K. Williams, B.H. Davison, tary Uses; Report on the 30th Session of the Codex
G. Britovsek, J. Cairney, C.A. Eckert, W.J. Frederick, Committee, Codex Alimentarius Commission, Rome,
J.P. Hallett, D.J. Leak, C.L. Liotta, J.R. Mielenz, 2009, p. 46.
R. Murphy, R. Templer, T. Tschaplinski, Science 311 [22] Y. Maphosa, V.A. Jideani, Dietary fiber extraction for
(2006) 484. human nutritionda review, Food Reviews Interna-
[8] L. da Costa Sousa, S.P.S. Chundawat, V. Balan, B.E. Dale, tional 32 (1) (2016) 98e115.
Current Opinion in Biotechnology 20 (2009) 339. [23] D.R. Erickson (Ed.), Edible Fats and Oils Processing:
[9] A.T. Tenorio, K.E. Kyriakopoulou, E. Suarez-Garcia, Basic Principles and Modern Practices: World Confer-
C. van den Berg, A.J. van der Goot, Understanding ence Proceedings, The American Oil Chemists Society,
differences in protein fractionation from conventional 1990.
crops, and herbaceous and aquatic biomass-consequences [24] M. Bravi, R. Bubbico, F. Manna, N. Verdone, Process
for industrial use, Trends in Food Science and Technol- optimisation in sunflower oil extraction by supercriti-
ogy 71 (2018) 235e245. cal CO2, Chemical Engineering Science 57 (14) (2002)
[10] A.V. Ruban, M.P. Johnson, Visualizing the dynamic 2753e2764.
structure of the plant photosynthetic membrane, Na- [25] M. Bravi, F. Spinoglio, N. Verdone, M. Adami,
ture Plants 1 (11) (2015) 15161. A. Aliboni, A. d’Andrea, A. De Santis, D. Ferri,
[11] C. Safi, B. Zebib, O. Merah, P.Y. Pontalier, C. Vaca-Garcia, Improving the extraction of a-tocopherol-enriched oil
Morphology, composition, production, processing and from grape seeds by supercritical CO2. Optimisation
applications of Chlorella vulgaris: a review, Renewable of the extraction conditions, Journal of Food Engineer-
and Sustainable Energy Reviews 35 (2014) 265e278. ing 78 (2) (2007) 488e493.
[12] A.L. Tan-Wilson, K.A. Wilson, Mobilization of seed [26] C.H. Cheng, T.B. Du, H.C. Pi, S.M. Jang, Y.H. Lin,
protein reserves, Physiologia Plantarum 145 (1) (2012) H.T. Lee, Comparative study of lipid extraction from
140e153. microalgae by organic solvent and supercritical CO2,
[13] A. Tamayo-Tenorio, J. Gieteling, G.A. De Jong, Bioresource Technology 102 (21) (2011) 10151e10153.
R.M. Boom, A.J. Van Der Goot, Recovery of protein [27] M.H. Vermuë, M.H.M. Eppink, R.H. Wijffels, C. Van
from green leaves: overview of crucial steps for Den Berg, Multi-product microalgae biorefineries: from
utilisation, Food Chemistry 203 (2016) 402e408. concept towards reality, Trends in Biotechnology 36 (2)
[14] A.C. Eliasson, Carbohydrates in Food, CRC press, 2006. (2018) 216e227.
[15] S. Jobling, Improving starch for food and industrial [28] L. Di Paola, A. Cicci, M. Bravi, Toward an efficient bio-
applications, Current Opinion in Plant Biology 7 (2) refining of microalgae and biomass alike: a unit opera-
(2004) 210e218. tion view on how to mimic the optimisation history of
[16] Z. Ao, J.L. Jane, Characterization and modeling of the crude oil refining industry, Chemical Engineering
the A-and B-granule starches of wheat, triticale, and Transactions 43 (2015) 1321e1325.
barley, Carbohydrate Polymers 67 (1) (2007) 46e55. [29] A. Holland, D. Wechsler, A. Patel, B.M. Molloy,
[17] J.Y. Li, A.I. Yeh, Relationships between thermal, rheo- A.R. Boyd, P.G. Jessop, Canadian Journal of Chemistry
logical characteristics and swelling power for various 90 (2012) 805e810.

IV. Selected examples and case history

 References 259
[30] Y. Du, B. Schuur, S.R.A. Kersten, D.W.F. Brilman, Algal [37] N. Sathitsuksanoh, K.M. Holtman, D.J. Yelle, T. Morgan,
Research 11 (2015) 217e283. V. Stavila, J. Pelton, H. Blanch, B.A. Simmons,
[31] S. Sharma, A.K. Kumar, Role of natural deep eutectic A. George, Green Chemistry 16 (2014) 1236e1247.
solvents (NADES) in the pretreatment of lignocellu- [38] A. Brandt, M.J. Ray, T.Q. To, D.J. Leak, R.J. Murphy,
losic biomass for an integrated biorefinery and bio- T. Welton, Green Chemistry 13 (2011) 2489e2499.
processing concept, in: Recent Advancements in [39] C.-Z. Liu, F. Wang, A.R. Stiles, C. Guo, Applied Energy
Biofuels and Bioenergy Utilization, Springer, 92 (2012) 406e414.
Singapore, 2018, pp. 73e109. [40] A. George, A. Brandt, K. Tran, S.M.N.S. Zahari,
[32] A. Cicci, G. Sed, P.G. Jessop, M. Bravi, Circular extrac- D. Klein-Marcuschamer, N. Sun, S. Singh, et al.,
tion: an innovative use of switchable solvents for the Design of low-cost ionic liquids for lignocellulosic
biomass biorefinery, Green Chemistry 20 (17) (2018) biomass pretreatment, Green Chemistry 17 (3) (2015)
3908e3911. 1728e1734.
[33] G. Sed, A. Cicci, P.G. Jessop, M. Bravi, A novel [41] A. Brandt-Talbot, F.J. Gschwend, P.S. Fennell,
switchable-hydrophilicity, natural deep eutectic sol- T.M. Lammens, B. Tan, J. Weale, J.P. Hallett, An
vent (NaDES)-based system for bio-safe biorefinery, economically viable ionic liquid for the fractionation
RSC Advances 8 (65) (2018) 37092e37097. of lignocellulosic biomass, Green Chemistry 19 (13)
[34] B.M. Khan, K.L. Cheong, Y. Liu, ATPS:“Aqueous two- (2017) 3078e3102.
phase system” as the “answer to protein separation” [42] A.K. Kumar, B.S. Parikh, M. Pravakar, Natural deep
for protein-processing food industry, Critical Reviews eutectic solvent mediated pretreatment of rice straw:
in Food Science and Nutrition (2018) 1e14. bioanalytical characterization of lignin extract and
[35] E. Suarez-Garcia, C.A.S. Ruiz, T. Tilaye, M.H. Eppink, enzymatic hydrolysis of pretreated biomass residue,
R.H. Wijffels, C. van den Berg, Fractionation of pro- Environmental Science and Pollution Research 23 (10)
teins and carbohydrates from crude microalgae ex- (2016) 9265e9275.
tracts using an ionic liquid based-aqueous two phase [43] A.K. Kumar, S. Sharma, E. Shah, A. Patel, Technical
system, Separation and Purification Technology 204 assessment of natural deep eutectic solvent (NADES)
(2018) 56e65. mediated biorefinery process: a case study, Journal of
[36] C.A. Suarez Ruiz, D.P. Emmery, R.H. Wijffels, Molecular Liquids 260 (2018) 313e322.
M.H. Eppink, C. van den Berg, Selective and mild frac- [44] G. Markou, E. Nerantzis, Microalgae for high-value
tionation of microalgal proteins and pigments using compounds and biofuels production: a review with
aqueous two-phase systems, Journal of Chemical Tech- focus on cultivation under stress conditions, Biotech-
nology and Biotechnology 93 (9) (2018) 2774e2783. nology Advances 31 (8) (2013) 1532e1542.

IV. Selected examples and case history