Biochemical Engineering Journal 199 (2023) 109029

Contents lists available at ScienceDirect

Biochemical Engineering Journal

journal homepage: www.elsevier.com/locate/bej

Regular article

Evaluation of the acidogenic fermentation potential of food industry

by-products
Angela Marchetti a, Gaia Salvatori a, Maria Luisa Astolfi a, b, Matteo Fabiani a, Joana Fradinho c, d,
Maria A.M. Reis c, d, Andrea Gianico e, David Bolzonella f, Marianna Villano a, b, *
a
Department of Chemistry, Sapienza University of Rome, P.le Aldo Moro 5, 00185 Rome, Italy
b
Research Center for Applied Sciences to the Safeguard of Environment and Cultural Heritage (CIABC), Sapienza University of Rome, P.le Aldo Moro 5, 00185 Rome,
Italy
c
Associate Laboratory i4HB – Institute for Health and Bioeconomy, NOVA School of Science and Technology, NOVA University Lisbon, 2829-516 Caparica, Portugal
d
UCIBIO – Applied Molecular Biosciences Unit, Department of Chemistry, NOVA School of Science and Technology, NOVA University Lisbon, 2829-516 Caparica,
Portugal
e
Water Research Institute, National Research Council (CNR-IRSA), Via Salaria km 29.300, Monterotondo, 00015 Rome, Italy
f
Department of Biotechnology, University of Verona, Via Strada Le Grazie 15, 37134 Verona, Italy

A R T I C L E I N F O A B S T R A C T

Keywords: Food industry annually generates huge amounts of waste and by-products, which represent untapped renewable
Acidogenic fermentation resources to produce platform chemicals (e.g., organic acids) through biological processes, such as acidogenic
Food industry by-products fermentation (AF). Here, the AF potential of nine food industry by-products has been evaluated in anaerobic
Mixed microbial cultures
batch experiments, with main attention to the spectrum of attained products and related conversion yields.
Circular bioeconomy
Farinaceous by-products, especially Bread Crust (BC) and Reground Pasta (RP), exhibited the highest conversion
yield of the initial Chemical Oxygen Demand (COD) into AF products, up to values of 55 ± 2% and 75 ± 9% (for
RP and BC, respectively). Substantially lower values, ranging between 6 ± 1% (for Spinach, SP) and 22 ± 1%
(for Grape pomace Lees, GL) were achieved with by-products from the vegetable and winery industry. Further
experiments with GL, SP, and WP (White exhausted grape Pomace) in the presence of glucose as reference
substrate, revealed the coexistence of both recalcitrant and inhibitory compounds in GL and WP, whereas no
inhibition on microbial activity was displayed by SP. A sonication pre-treatment experiment (at a low frequency
of 20 kHz for 15 min and at room temperature) allowed increasing the conversion yield of GL by approximately
45%.

1. Introduction produced food is annually wasted throughout the whole value chain
from primary producers to consumers, and this corresponds to around
Food production and processing generate huge amounts of by- 1.3 billion tonnes of wasted food globally generated [4,5]. At present,
products and wastes. In particular, food wastes are typically consid­ the European Union is committed to meeting the Sustainable Develop­
ered organic residues deriving from the processing of raw materials into ment Goal (SDG) target to halve per capita food waste at the retail and
food, which are typically not utilized and end up in municipal landfills, consumer level by 2030 as well as to reduce food losses along the whole
often causing serious environmental concerns related to their manage­ food production and supply chains [6]. Although reduction is certainly
ment and final disposal [1,2]. On the other hand, by-products result the preferred option in the food waste and by-products hierarchy, other
from the manufacturing of the main product, and they can be reused for approaches such as their recovery and valorisation are becoming routes
industrial practices without further processing [3], thus allowing easier of ever-increasing relevance, particularly in the context of the circular
management and green valorisation. Approximately one-third of the bioeconomy development. Along this line, different strategies and (bio)

Abbreviations: AF, Acidogenic Fermentation; COD, Chemical Oxygen Demand; AD, Anaerobic Digestion; PHA, Polyhydroxyalkanoates; MMC, Mixed Microbial
Cultures; RP, Reground Pasta; BC, Bread Crust; WB, Wheat Bran; WP, White exhausted grape Pomace; RGP, Red exhausted Grape Pomace; FP, Fresh grape Pomace;
GL, Grape pomace Lees; GS, Grape Seed; SP, Spinach.
* Corresponding author at: Department of Chemistry, Sapienza University of Rome, P.le Aldo Moro 5, 00185 Rome, Italy.
E-mail address: [email protected] (M. Villano).

https://doi.org/10.1016/j.bej.2023.109029
Received 4 March 2023; Received in revised form 11 June 2023; Accepted 2 July 2023
Available online 4 July 2023
1369-703X/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

technological platforms have been investigated to valorise food waste using mixed microbial cultures, with the aim to steer their valorisation
and by-products, mostly towards bioenergy production via anaerobic towards organic acids which in turn represent valuable molecules to be
digestion (AD). Presently, AD is considered one of the most effective used for several applications. The conversion yields of the used by-
biotechnological routes for food waste and by-products management, products into organic acids were analysed and compared, as well as
because of its limited environmental footprint, high potential for energy ad-hoc experiments were carried out to detect the possible presence of
recovery via biogas (i.e., a mixture of methane and carbon dioxide) and inhibitory or recalcitrant compounds challenging their AF potential.
possibly biomethane generation, and production of a nutrient-rich Finally, the effect of sonication as a means to enhance the AF potential of
digestate which can be utilized as soil amendment in agriculture [7]. the hardest-to-degrade by-products among those investigated was also
Recently, much attention has been also paid to the possibility to obtain explored. Overall, the obtained results provide new information and
products of higher market value than biogas, such as organic acids, novel insights into the valorisation potential of food industry by-
amino acids, alcohols, solvents, vitamins, and other secondary metab­ products which could be relevant to the development of novel bio­
olites [8,9]. All these products can find valuable application in a variety refinery platforms.
of industrial sectors, including food, pharmaceutical, medical, and
chemical applications, as well as textiles, leather processing, and plastic 2. Materials and methods
production [10,11]. As an example, organic acids have broad and rele­
vant industrial applications as intermediate feedstock for the production 2.1. Food industry by-products used throughout the study
of chemicals [8,12] and, also, they are precursors for the biosynthesis of
polyhydroxyalkanoates (PHAs), which are fully biodegradable polymers Nine food industry by-products have been herein used as feedstock to
with a broad portfolio of industrial applications, including the pack­ test their potential to be converted into platform compounds (e.g.,
aging (e.g., food-packaging) and medical sectors [13–16]. organic acids) through acidogenic fermentation (AF) experiments, per­
The conversion of food waste and by-products into organic acids can formed in batch mode with mixed microbial cultures (MMC).
occur through acidogenic fermentation (AF), which is a biotechnological Three set of experiments were performed. The first set involved the
process sharing several biochemical steps with anaerobic digestion [17, use of three farinaceous by-products, from pasta and bread processing
18]. In both AD and AF, the feedstock typically consists of particulate residues: Reground Pasta (RP), Bread Crust (BC), and Wheat Bran (WB).
matter, which is initially hydrolysed and converted into soluble prod­ The second set was carried out using five winery by-products, consisting
ucts, and subsequently fermented into a mixture of acids, alcohols, and of wine residues from the different steps involved in wine production,
hydrogen. To prevent these intermediate compounds from being ulti­ and one vegetable by-product consisting of the residue obtained from
mately converted into biogas, the activity of methanogenic microor­ processing spinach; namely: White exhausted grape Pomace (WP), Red
ganisms must be hampered by properly tuning key process parameters exhausted Grape Pomace (RGP), Fresh grape Pomace (FP), Grape
(such as pH, temperature, hydraulic and solid retention time, organic pomace Lees (GL), Grape Seed (GS), and Spinach (SP). A third set of
loading rate) or combinations thereof [19–23]. experiments was run to investigate the effect of a physical pre-treatment
Over the past years, the acidogenic fermentation of food industry (i.e., sonication) on the AF performance.
wastes and by-products, of often heterogeneous and undefined compo­ All by-products were provided by food industries involved in their
sition, has received considerable scientific attention and the topic has production and were stored at − 20 ◦ C before their usage. Before the
been covered in different comprehensive review papers [9,24–26]. As an start of AF experiments, all by-products underwent a preliminary char­
example, dealing with the winery industry, 4–5 M tonnes of wine were acterisation consisting of measurements for the determination of mois­
produced during 2020 by Italy, France and Spain and approximately ture and COD (Chemical Oxygen Demand) contents. From these results a
0.13 tonnes of grape pomace, 0.06 tonnes of wine lees, 0.03 tonnes of conversion factor (i.e., expressed as gCOD per gram of substrate on a dry
grape seed, and 1.65 m3 of wastewater are typically generated from the weight basis) for each by-product was obtained, in this way it was
processing of one tonne of grapes [27]. The acidogenic fermentation of possible to determine the substrate quantity to add for the AF tests in
these wastes has been investigated by adopting different operating order to achieve a desired ratio between inoculum and substrate (I/S).
conditions. As an example, the fermentation process of a synthetic
winery wastewater has been studied in sequencing batch reactors 2.2. Operation of batch AF experiments
operated at 35 ◦ C and acid conditions (pH 5), resulting in an estimated
potential of fermentation products generation (mainly acetic, propionic Batch AF experiments were set up in serum bottles (250 mL total
and butyric acids) from this waste equal to 6220 tonnes per year referred volume), each one filled with a defined volume of mineral medium and
to the initial glucose and fructose concentrations, which are the main anaerobic sludge (as inoculum) to reach a working volume of the liquid
winery wastewater components with a concentration of about 1.5 g/m3 phase of 150 mL. A specific amount of by-products was also added to
[28]. Also, the influence of pH and operating temperature on volatile obtain an initial substrate concentration of 2 gCOD/L (corresponding to
fatty acids (VFA) production yield was assessed with real winery 0.30 gCOD per serum bottle). The anaerobic sludge used as inoculum
wastewater (Www), obtaining at alkaline pH of 10 and thermophilic was taken, at the start of each set of experiments, from a laboratory-scale
temperature (55 ◦ C) a yield of 0.40 gVFAtotal/gCOD, with the acetate as anaerobic digester treating food waste (i.e., household and canteen food
the main fermentation product [29]. Other types of widely produced waste). Before the usage, the withdrawn sludge was settled and the
wastes and by-products come from the farinaceous industry. Specif­ liquid phase was replaced with mineral medium to remove all the re­
ically, bakery wastes and by-products account for 35% of the total sidual organics therein contained, and then was maintained at room
production in Europe, North America and industrialised Asia [30]. A temperature and an acid pH value of 5.5 to inhibit the methanogenic
study conducted in 2019 highlights how these by-products can be widely activity till the start of the fermentation tests. Prior to starting the AF
exploited through different uses. For example, in Italy, farinaceous experiments the sludge concentration was measured, and the average
by-products are largely used as fodder for cattle (40.28%, wt/wt) fol­ value accounted for about 15.0 g/L as volatile suspended solids (VSS).
lowed by energy recovery (22.05%, wt/wt) with only about 6% destined According to this value, a volume of sludge required to achieve an initial
to landfill [31]. Another way to valorise them involves the use of I/S ratio of 0.20 gVSS/gCOD was added in all the tests. The mineral
fermentation processes, whereby these by-products play the role of medium used for both the sludge washing and the AF tests contained all
substrates for the industrial production of enzymes [32]. the components indispensable for microbial growth and was prepared in
Along this line, herein the acidogenic fermentation of nine food in­ distilled water with the pH buffered at 5.5. The mineral medium
dustry by-products, including farinaceous, winery, and vegetable resi­ composition was as follows (in g/L): NH4Cl (0.50), MgCl2⋅6 H2O (0.10),
dues, has been investigated by means of anaerobic batch experiments CaCl2⋅2 H2O (0.05), K2HPO4 (8.70), KH2PO4 (61.22), a trace of metals

2
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

(10 mL/L) and vitamin (1 mL/L) solutions. Once prepared, all serum [36]. The total Chemical Oxygen Demand (tCOD) of all by-products,
bottles were flushed with a N2/CO2 (70/30, v/v) gas mixture for about referred to the dry weight, was measured by using a commercial test
20 min to establish anaerobic conditions, sealed by a Teflon-faced grey (Tube test NANOCOLOR COD 600, Macherey Nagel) consisting of glass
butyl stopper (Wheaton, Millville, NJ, USA) and crimped by an cuvettes containing potassium dichromate, concentrated H2SO4
aluminum cap. All experiments were performed at room temperature (80–98%) and HgSO4. After samples digestion at 150 ◦ C for 2 h, the COD
(22 ± 1 ◦ C) to avoid possible activation of methanogenic bacteria and content was determined using a spectrophotometer (SHIMADZU Spec­
under a complete mixing of the liquid phase with a magnetic stirrer trophotometer UV-1800) at a wavelength of 605 nm. Samples (1 mL) for
(350 rpm) (Fig. 1). Each operating condition was carried out in triplicate the determination of fermentation products (e.g., organic acids and
(unless otherwise stated), for a total of over 40 tests. Control tests in the ethanol) were filtered (0.45 µm porosity), acidified with 100 µL of a
absence of by-products were also performed to assess the endogenous AF solution of oxalic acid (0.33 M), and injected (1 µL) into a
potential of the anaerobic sludge used as an inoculum. Approximately gas-chromatograph (GC Dani-Master, Milan, Italy) equipped with a
one month after the start of the experiments, a synthetic carbonaceous flame ionization detector (FID) and a glass column (2 m × 2 mm)
substrate (glucose) was added to the control tests at the same initial packed with Carbopack B 60/80 mesh. Helium (He) was used as carrier
concentration of food industry by-products used in the other tests, to gas at a flow rate of 25 mL/min, the oven temperature was set at 175 ◦ C,
assess the maximum metabolic activity of the inoculum. and the FID temperature at 200 ◦ C.
The operation of further experiments with three by-products (GL, Gas phase analysis from the headspace of all serum bottles was also
WP, and SP) chosen from the six used in the second set of experiments performed to assess the absence of methane generation during the
either augmented or not with glucose as reference synthetic substrate, operation of AF tests. The analysis was carried out by injecting 50 µL of
were also carried out. The aim of these additional experiments was to gaseous sample with a gas-tight Hamilton syringe into a “Dani Master”
identify the possible presence of inhibitory or recalcitrant compounds gas-chromatograph (glass column packed with Carbopack; He carrier
contained in the by-products on the microbial activity. Additionally, for gas 25 mL/min; oven temperature 50 ◦ C; FID temperature 200 ◦ C).
three winery by-products (namely GL, WP, and GS), AF tests were also
performed after the application of an ultrasonic pre-treatment (third set
2.4. Calculations
of experiments), applied using a low frequency (20 kHz) sonication bath
(Branson® 2210 Sonicator). Indeed, in presence of substrates with a
The fermentation yield was used as a main parameter to evaluate the
high cellulose content (such as GS), the application of pre-treatments
conversion potential of the used by-products into fermentation products.
might represent one of the most suitable options for the increase of
The yield was defined as the sum of each individual fermentation
the soluble COD, enhancing the AF process [33,34]. For these tests, each
product (expressed as COD equivalents) at the maximum concentration
by-product was placed inside sealed falcon tubes and immersed in the
data point (ΣPEND , typically corresponding to the last day of operation)
sonic bath for 15 min at room temperature. To analyse the presence of
minus its equivalent in the control test (ΣPCONTROL ) divided by the initial
inhibitory compounds as well as the effect of the sonication
total COD added for each assay (tCODi). However, since in this study
pre-treatment, statistical analysis was performed on obtained data by
fermentation products were typically not detected in the control tests
using the t-student test. For all the three sets of experiments, the liquid
performed in the absence of organic substrates (i.e., glucose), a simpli­
phase and the headspace of the serum bottles were periodically sampled
fied equation was used in which the second term of the equation was set
to measure the production of fermentation products and to verify the
to zero, as follows:
absence of methane production, respectively.
(ΣPEND )
Net maximum Fermentation Yield (%, COD/COD) = • 100
2.3. Analytical methods tCODi
Differently, for the tests performed with by-products either in pres­
The concentration of microorganisms in the anaerobic sludge used as ence or in absence of glucose augmentation (second set of experiments),
inoculum of the AF tests was determined as Volatile Suspended Solids the fermentation yield from glucose for a specific by-product was esti­
(VSS) according to standard methods [35]. The moisture content mated as the difference between the sum of products measured in the
(expressed as a percentage by dry weight) in the tested by-products was tests with both the by-product and glucose and the sum of products
determined according to the reference method described in ISO 712 deriving from the test with only the by-product, divided by the amount

Fig. 1. Scheme of acidogenic fermentation tests with the nine food-industry by-products (Created with Biorender.com).

3
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

of glucose initially provided, according to the following equations: the time of harvest as well as the vinification techniques used [43]. The
percentage of moisture in the pomace can vary between 55% and 75%
Net maximum Fermentation Yield (%, COD/COD)
( ) (wt/wt) depending also on the intensity of the pressing that the must
ΣP(glucose+by− products) END − ΣP(by− products) END undergoes [44], and this is fully in agreement with data reported in
= • 100
t(Glucose)i Table 1. However, the different pomace constituents can have different
characteristics, such as grape seeds which contain between 13% and
The concentration of glucose, organic acids and ethanol was con­
19% of oil (rich in essential fatty acids), about 11% of proteins, 60–70%
verted into COD in accordance with the following oxidation stoichi­
of non-digestible carbohydrates, and non-phenolic antioxidants such as
ometry: 1.07 gCOD/gGlucose, 1.07 gCOD/gAcetate, 1.51 gCOD/gPropionate,
tocopherols and beta-carotene [45–47]. Moreover, seeds are richer in
1.82 gCOD/gButyrate, 2.04 gCOD/gValerate, 2.21 gCOD/gHexanoate and 2.09
dietary fibers (ranging between 43% and 75%) and lipids (ranging be­
gCOD/gEthanol.
tween 14% and 17%) than the skin, and red wine pomace is richer in
All descriptive statistics values were calculated for each test using
fibers than white wine pomace. Additionally, pomace contains a
Microsoft Excel and error bars stand for the standard deviation of the
considerable proportion of tannins and proteins [43]. The protein con­
mean of tests conducted in triplicate or in duplicate. The statistical
tent of grape pomace may range between 6% and 15% (dry matter)
analysis was performed by analysing data with the t-student test.
depending on the grape variety and harvesting conditions. The pro­
portion of protein in the skins (for WP, RGP and FP) and seeds (for GS) is
3. Results and discussion
similar, but the skins from wine pomace are slightly richer than the seeds
separated from the wine pomace [43]. Differences in the characteristics
3.1. General characteristics of food industry by-products
of winery by-products are also related to the stage of the winemaking
process they are generated from. Indeed, the pomace differs between
Prior to performing batch fermentation tests, the moisture percent­
exhausted red and exhausted white, as the former remains in contact
age and COD content (on a dry matter basis) of the nine types of by-
with the grape must throughout the fermentation process, thus
products have been determined and related data are reported in Table 1.
absorbing more alcohol [48]. Residues collected from the bottom of the
All farinaceous by-products (RP, BC, WB) were characterized by a
containers after fermentation, during storage or after treatments, as well
relatively low moisture percentage (less than 9%, wt/wt), whereas
as the residues obtained after filtration or centrifugation are referred to
higher values (between ca. 50% and 70%, wt/wt) were observed for the
as GL. This mainly consists of insoluble residues such as solid parts of the
winery (WP, RGP, FP, GL, GS) and spinach (SP) by-products; except for
grapes (skin fragments and seeds), yeasts, colouring agents, ferments,
GS which showed values close to those of the farinaceous by-products.
and other impurities as well as microorganisms that remain suspended
The moisture content of GS differed from the other winery by-
in the wine and tend to settle [49,50]. Concerning polyphenols content,
products as it represents the driest lignocellulosic residue of the wine
the phenolic composition of wine pomace has been extensively
production process, made of a very resistant and hard-to-break shell.
described in the literature with notable qualitative and quantitative
In terms of COD content, farinaceous and winery by-products (except
differences [51–53]. In general, exhausted grape pomace differs from
for GL) resulted in values close to 1 (wt/wt) or slightly higher, whereas a
the fresh grape one for the presence of a residual alcohol content,
lower value was obtained for SP. These results were used to make cal­
minimal sugar content, less polyphenolic substances, organic salts, and
culations necessary to setup all AF tests at the same initial total COD
elements (such as Ce, Cr, La, Li, Pb, Sb, Sn, Sr, Ti, U, V, and Zr). This
concentration (i.e., 2 gCOD/L), except for control tests that were carried
gives GL very variable characteristics.
out in the same conditions but in the absence of by-products and in the
Based on these considerations, here the main objective was to
presence of glucose (as reference substrate).
qualitatively assess the acidogenic fermentation potential of different
As widely reported in the literature, carbohydrate and fiber content
food industry by-products mainly focusing on the composition of the
might be different for the tested farinaceous by-products and this, in
obtained products which ultimately governs their final application
turn, can affect the fermentation pattern [37]. In particular, the carbo­
within a circular bioeconomy model.
hydrate content (in terms of total starch) is usually higher in the case of
RP (around 70 g/100 g) than for BC (around 50 g/100 g), although in
3.2. Acidogenic fermentation tests with farinaceous by-products
the latter the content is highly dependent on the type of flour used [38].
However, RP and BC are more similar in terms of fiber content (around
The results of AF experiments with farinaceous by-products in terms
3 g/100 g and between 2 and 7 g/100 g, respectively) [39,40] if
of products concentration and distribution are reported in Fig. 2. In the
compared with the typical fiber content of the Wheat Bran, in which a
case of RP and BC (Figs. 2a and 2b, respectively), fermentation mainly
value up to around 43 g/100 g can be reached [41]. According to these
resulted in the production of acetic, propionic, and butyric (at low
literature data, it is also important to highlight the fact that the fiber
extent) acids, along with ethanol; whereas no butyric acid was observed
content has a major impact on the efficiency of the acidogenic fermen­
when WB was used as substrate (Fig. 2c). The assessment of the AF
tation process, being the hardest constituent to be biologically digested
potential was mainly evaluated by considering the conversion yield,
[42]. As for the winery by-products, the chemical composition of the
which represents the fraction of the initial total COD converted into
pomace varies with the season, the grape diversity, the place of origin,
fermentation products (Fig. 3b). As a main result, it was found a higher
conversion yield with RP and BC than with WB, with values (on COD
basis) accounting for 47 ± 5% (RP), 58 ± 10% (BC), and 20 ± 3% (WB),
Table 1
respectively, just after 10 days of operation. These values remained
Moisture percentage and COD content in all tested food industry by-products.
almost unchanged by day 36, but a further increase was observed of
BY-PRODUCT MOISTURE (%, wt/wt) COD (wt/wt) about 8% for RP and 17% for BC and WB, reaching a conversion yield
Reground Pasta (RP) 5.90 ± 0.50 1.14 ± 0.16 approximately equal to 55 ± 2% (RP), 75 ± 9% (BC), and 37 ± 6%
Bread Crust (BC) 7.20 ± 0.30 0.99 ± 0.11 (WB) by the end of the experiments (day 55), suggesting the presence of
Wheat Bran (WB) 8.60 ± 0.10 1.16 ± 0.08
slowly biodegradable COD in all the three by-products. Indeed, as pre­
White ex. grape Pomace (WP) 51.0 ± 2.70 1.06 ± 0.04
Red ex. Grape pomace (RGP) 54.1 ± 2.95 1.08 ± 0.04 viously discussed, the less biodegradable component contained in these
Fresh grape Pomace (FP) 47.7 ± 1.70 1.15 ± 0.01 by-products corresponds to the fiber content, whose total percentage is
Grape pomace Lees (GL) 62.8 ± 0.10 0.47 ± 0.03 typically much higher in WB compared to RP and BC, around 43% (wt/
Grape pomace Seed (GS) 7.10 ± 0.10 1.13 ± 0.10 wt) versus 3% and 2–7%, respectively [39–41]. This is fully in agree­
Spinach (SP) 71.6 ± 3.70 0.78 ± 0.04
ment with the lower observed acidogenic fermentation potential of WB

4
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

Fig. 2. Trends of obtained products in batch fermentation experiments with: Reground Pasta (a), Bread Crust (b), Wheat Bran (c), and in absence (control) of by-
products (d).

operation to verify the effectiveness of the bacterial activity. As shown in

Fig. 2d, the addition of glucose resulted in its fast and nearly complete
a conversion into fermentation products, mainly comprising of ethanol,
REGROUND PASTA BREAD CRUST WHEAT BRAN CONTROL Ini�al COD
2.5 butyric acid and, to lower extent, acetic acid, whereas no propionic acid
was produced. This finding indicates that the substrate concentration
Concentra�on (gCOD/L)

2.0 chosen to perform the batch tests was not inhibitory to the inoculum,
thus the complete conversion of the supplied COD into fermentation
1.5
products was not achieved likely due to the presence of
1.0 non-biodegradable constituents in the supplied by-products. An inter­
esting finding is the different distribution of fermentation products,
0.5
Glucose addition mainly in terms of propionic and butyric acid, between the control tests
and the tests performed with the by-products. The low, or completely
0.0
0 6 12 18 24 30 36 42 48 54 absent, conversion of farinaceous substrates into butyric acid could be
Time (day) due to the slightly acidic pH value (i.e., 5.5) of the tests, which has been
previously shown to favour the accumulation of acetic acid, over butyric
b acid, in butyrate-producing bacteria such as Clostridium [54,55]. How­
ever, being all AF tests performed at the same temperature and pH
conditions, the different profiles observed for fermented products are
Conversion yield (%, COD/COD)

100
more likely attributed to the different characteristics (in terms of car­
80
bohydrates, lipids, fibers etc.) of the three farinaceous by-products with
60 respect to the single glucose used in the control experiments.
Also, it is important to highlight that the propionic acid concentra­
40
tion represented a high fraction of the overall fermentation products in
20 Glucose addition all tests performed in presence of farinaceous by-products, reaching
values between 35% and 40% (on total COD basis) for RP and BC,
0
0 6 12 18 24 30 36 42 48 54 respectively; and over 50% for WB. This finding is particularly inter­
Time (day) esting because, as already mentioned, the acid composition of the fer­
mented mixture affects its final exploitation potential. As an example, in
Fig. 3. AF tests with farinaceous by-products: total concentration of fermen­ the general context of polyhydroxyalkanoates (PHA) production, pro­
tation products (a); conversion yield (b). pionic acid is the precursor for the hydroxy-valerate (HV) monomer
formation, which confers better processing properties to the biopolymer
compared to the other farinaceous by-products. However, it cannot be [56].
excluded that WB fermentation could be improved through a Overall, the main results obtained with the farinaceous by-products,
pre-treatment step aimed at hydrolysing fibers and increasing the sol­ in terms of both total concentration and yield of fermentation products,
uble COD available to bacteria. In addition, an easily biodegradable are summarized in Fig. 3, which highlights the occurrence of a very
substrate (glucose) was added in the control tests after 32 days of similar trend for RP and BC as well as the presence of not easily

5
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

fermentable compounds in all by-products, especially in WB. Overall, a by-products compared to the farinaceous ones, can be likely due to their
conversion yield of around (for RP) or over (for BC) 60% was reached, different intrinsic characteristics which are, in turn, linked to a variable
making them promising by-products to be valorised towards the pro­ content of compounds with recalcitrant or inhibitory effects on the mi­
duction of platform chemicals. crobial activity. However, the unexpectedly poor AF conversion yield
As expected, under the applied working conditions no methane observed with these by-products warrants further investigation. There­
production was observed in all the AF tests with farinaceous by- fore, to evaluate the presence of inhibitory or recalcitrant compounds,
products. another set of experiments was set up with the two by-products that
showed the highest conversion (i.e., GL and WP), and with the one that
presented the lowest degree of conversion (i.e., SP). The tests were
3.3. Acidogenic fermentation tests with winery and vegetable by-products
started at different times for GL and WP with respect to SP, and lasted 26
days and 40 days, respectively. Also in this case, each set of experiments
As for acidogenic fermentation tests with by-products from the
included control tests in the absence of substrate, to account for the
winery and vegetable industry, the overall obtained results are reported
endogenous activity of the inoculum. This further experimental activity
in Fig. 4. Totally, AF tests lasted more than 30 days which, however,
consisted of tests in which each by-product was supplied either alone (at
were not sufficient to obtain a substantial conversion of these by-
an initial concentration of 2 gCOD/L) or along with glucose (the latter at
products into fermentation products. Indeed, a negligible conversion
an initial concentration of 1 gCOD/L), a substrate easily fermented by
(less than 5%) was obtained with RGP, FP, GS, and SP, by the end of the
the used anaerobic sludge. Similarly, control tests (CL) in the absence or
experiments (Fig. 4b).
presence of glucose (at an initial concentration of 2 gCOD/L) were also
Higher yield values were achieved with WP and GL, accounting for
performed. The results of these experiments are reported in Fig. 5, in
about 9.4 ± 1.9% and 22 ± 1% of the overall supplied COD, respec­
terms of both the average value of the maximum attained total con­
tively. While, the lowest conversion yield was achieved by using SP as
centration of fermentation products (Fig. 5a) and their corresponding
substrate, approaching to a final value of around 4%, with a maximum
distribution (Fig. 5b).
of around 6% on day 12. In particular, GL was mostly fermented into
As for control tests performed in the absence of glucose, a very low
acetic acid (reaching a final concentration of about 250 mgCOD/L) and
production of acids was observed (mainly consisting of acetic, valeric,
ethanol (final concentration slightly higher than 70 mgCOD/L) along
and hexanoic acids). This suggests that bacteria in the inoculum prob­
with lower amounts of acids from C3 to C6. The trend of the sum of all
ably thrived on some residual COD which was, however, consumed
fermentation products obtained with the used by-products is reported in
during the first few days of the test. As expected, a much higher pro­
Fig. 4a.
duction was obtained in the control tests in the presence of glucose,
To rule out the possibility that the observed incomplete conversion of
reaching at the end of the test (on day 26) a concentration of 1.9 gCOD/L
these by-products was due to inherently slow microbial activity, a syn­
(Fig. 5a) in acids corresponding to about 89 ± 2% (COD/COD) of con­
thetic solution of glucose was added to the control experiments (on day
version yield with respect to the glucose consumed (Fig. 5b). Besides
30), at the same concentration as that used for tests with by-products. A
acids (mainly acetic, isobutyric, and butyric acid) also an alcoholic
fast and complete production of fermentation products was obtained,
fermentation was triggered resulting in ethanol production. In the case
thus confirming the occurrence of efficient microbial activity. From data
found in the literature [2,47,48], it can be stated that the different re­
sponses to the fermentation process of the winery and vegetable
a **
**
2.0
**
Concentra�on (gCOD/L)

1.5

1.0

0.5

0.0
CL CL + GL GL + WP WP + SP SP +
glucose glucose glucose glucose

b
Conversion yield (%, COD/COD)

100

80

60

40 Glucose addi on

20

0
0 6 12 18 24 30 36
Time (day) Fig. 5. AF tests with GL, WP, and SP with and without glucose: maximum
concentration of fermentation products (a); and their distribution (b). Student t-
Fig. 4. AF tests with winery and vegetable by-products: total concentration of test was adopted to statistically analyse the data (**: p < 0.01). Data reported
fermentation products (a); conversion yield (b). in the graphs are representative of duplicate experiments.

6
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

of GL, the concentration of fermentation products was slightly higher in observed with the farinaceous by-products, a negligible production of
the tests performed in the presence of glucose than in the tests carried propionic acid was observed that, however, was mainly detected (in the
out in its absence. To calculate the percentage of glucose that was range between 1% and 8%) in the tests performed in the presence of
converted into fermentation products, the difference between the total glucose. Finally, it should also be considered, for GL and WP, the pres­
amount of products in the presence and in the absence of glucose was ence of polyphenolic compounds, which typically represent a significant
considered, and the yield of fermentation products referred to glucose fraction of the initial COD contained in these by-products and which
accounted for approximately 37%. These results clearly suggest that GL likely contributed to the low observed conversion yield. In fact, poly­
contained both slowly biodegradable and recalcitrant compounds which phenols are substances with phytotoxic and antimicrobial effects which
did not allow a complete conversion of glucose into fermentation can negatively impact the AF performance [51]. Although the profile of
products in its presence. However, by the end of the tests, the yield of GL phenolic compounds characterising the pomaces varies depending on
conversion into fermentation products accounted for approximately 32 their composition and on the efficiency of oenological practices on
± 1% (on a COD basis), a value slightly higher than that obtained in the phenols extraction compounds during the winemaking process, in
previous set of experiments (i.e., 22 ± 1% by the end of the experiment accordance with the ranges reported in the literature, it can be assumed
and a maximum conversion of around 28% on day 13), possibly due to a that winery by-products typically present a polyphenols concentration
slight change in the activity of the used inoculum. Also, in the case of in the range between 20 and 25 mg polyphenols/g TS, with a higher
WP, the concentration of fermentation products was higher in the tests percentage of these compounds in exhausted pomace and in grape lees
performed in the presence of glucose than in its absence, throughout all than in pomace lees [57]. Finally, as for the SP residues, the scarcely
the experiment. More in detail, the conversion yield of WP was similar to biodegradable component concerning vegetable fibers, may have
that obtained in the previous set of experiments (ca. 10%), whereas in compromised the conversion of these substrates into fermentation
the presence of glucose the overall conversion yield accounted for about products.
25 ± 1% by the end of the tests. With specific reference to glucose, its
conversion into fermentation products was estimated to 58 ± 2%, which
is a higher value than that obtained in the presence of solely GL, but still 3.4. Effect of sonication pre-treatment on acidogenic fermentation
lower than the value of 89% registered in the control test. Therefore,
similarly to GL, WP presented compounds recalcitrant to the biological In this section, the main AF results obtained using three types of by-
activity, since only a little fraction of its initial COD was converted into products (i.e., GL, WP, and GS) pre-treated by means of a sonication
fermentation products, but also exhibited a partial inhibitory effect on process are reported. In order to establish the real ability of the ultra­
glucose fermentation, whose conversion into fermentation products was sonic pre-treatment to increase the soluble and more readily biode­
not complete. As for the tests with SP, its conversion into fermentation gradable fraction of these by-products, a comparison is shown with the
products remained very low in the experiments performed in the results obtained in the previous AF tests with the same untreated by-
absence of glucose, with a conversion yield very similar to that obtained products. The main results of these experiments are reported in Fig. 6,
in the previous set of experiments (6%). However, when SP was supplied in terms of the maximum reached acids concentration and the
along with glucose, the latter was completely degraded indicating a high
recalcitrance of SP to the biological action but the lack of inhibitory
effects of this by-product on microbial activity. Overall, taken as a
a 0.8

whole, the results obtained with the three by-products suggest that all of
*
them contained slowly biodegradable compounds (at higher amounts in
Concentra�on (gCOD/L)

0.6
WP and SP relative to GL) which caused an incomplete conversion into
fermentation products that was, however, also linked to a partial
inhibitory effect on the microbial activity, particularly evident with GL 0.4
and WP but completely absent for SP. Indeed, as for the presence of
recalcitrant compounds, the lower concentration of fermentation prod­
ucts in the tests only supplied with individual by-products than in the 0.2

control tests supplied with glucose (all tests at an initial COD concen­
tration of 2 gCOD/L) was very statistically significant in all cases
0.0
(Fig. 5a). GL GL_SN WP WP_SN GS GS_SN
Regarding the percentage of fermentation products obtained from
individual by-products without glucose, in all cases acetic, valeric and,
to a minor extent, hexanoic acids were obtained, while propionic and
butyric acids were found in lower concentrations in the case of SP, and in
the first days of operation in the presence of GL and WP. In addition,
when glucose was also present, the total COD supplied to the tests was
mainly converted into butyric acid and ethanol, reaching in terms of
butyric acid 32 ± 5%, 4 ± 1%, and 7 ± 1%, and in terms of ethanol 26
± 7%, 15.5 ± 0.5%, and 8.9 ± 0.1% for “control test + glucose”, “GL
+ glucose” and “WP + glucose”, respectively. While, when glucose was
added to the SP tests, acetic, propionic, and valeric acids were the main
constituents of the fermentation products, reaching conversion values of
11 ± 1%, 8 ± 1%, and 14.8 ± 0.4%, respectively.
Also, in the case of GL, the percentage of conversion into ethanol
decreased from 22.5 ± 0.2% (without glucose) to 15.5 ± 0.5% (with
Fig. 6. AF tests of GL, WP, and GS by-products without and with an ultrasonic
glucose), suggesting that in the presence of the synthetic substrate,
(SN) pre-treatment: concentration of fermentation products on the last day of
fermentation was more directed towards the production of acids rather operation (a); conversion yield and distribution of fermentation products on the
than alcohols. Indeed, an increase in acetic and butyric acids conversion last day of operation (b). Student t-test was adopted to statistically analyse the
in the presence of glucose occurred, resulting in an overall improvement data (*: p < 0.05). Data reported in the graphs are representative of duplicate
of GL fermentation degree into products. Differently from what was experiments.

7
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

corresponding conversion yield and products distribution, obtained in [49,50], and are naturally able to convert sugars into alcohols [61].
correspondence to the end of the experiments. Regarding the concen­ Moreover, at pH of around 5–6, yeasts could be more competitive in the
tration of AF products at the end of the experiment (Fig. 6a), no sub­ fermentation process for ethanol production [62]. In addition, with GL
stantial improvements were observed as a consequence of the pre- an increment in propionic acid production with the sonicated
treatment, with concentrations being almost the same, except for GL. by-product was obtained. The change in the distribution of the
Comparing the results for WP and WP_SN (i.e., the sonicated by- fermentation products spectrum, as a consequence of sonication, is
product), a final concentration of 0.19 ± 0.04 gCOD/L (corresponding particularly interesting and deserves further investigation.
to 9.4 ± 1% of conversion yield on a COD basis) and 0.16 ± 0.01 gCOD/
L (8.0 ± 0.4% of conversion yield) were obtained, respectively. The 4. Conclusions
same behaviour was observed for GS and GS_SN, with a final concen­
tration equal to 0.08 ± 0.01 gCOD/L (4.0 ± 0.6% of conversion yield on Nine food industry by-products have been herein tested in batch
COD basis) and 0.09 ± 0.01 gCOD/L (4.6 ± 0.5% of conversion yield), experiments to assess their potential as renewable resources for the
respectively. This evidence further confirms that the WP and GS by- production of platform chemicals through acidogenic fermentation.
products were scarcely prone to the fermentation process, and that Among them, BC and RP displayed the highest conversion yield of the
bioconversion of their recalcitrant constituents was not improved by the initially supplied COD into fermentation products up to values of 75
applied pre-treatment. However, it could also be possible that the son­ ± 9% and 55 ± 2% by the end of the experiments (day 55), respectively.
ication treatment was not appropriate for such by-products, and future With winery and vegetable by-products, the potential of transformation
investigations in this perspective may be useful to confirm this hy­ into fermentation products was substantially lower than with the fari­
pothesis. A different trend was observed for GL, whereby an increase of naceous by-products, achieving values of conversion yield between 6
the COD converted into fermentation products was detected upon son­ ± 1% and 22 ± 1% (on a COD basis) with SP and GL, respectively.
ication. The overall percentage conversion of the total supplied COD In particular, tests performed with GL, WP, and SP in the presence of
increased from 22 ± 1% (data without pre-treatment) to 31.8 ± 0.1% glucose as reference substrate, suggested the existence of not readily
(data after pre-treatment) and, importantly, these results were obtained biodegradable substances in these three by-products (to a higher extent
with inocula taken from the lab-scale methanogenic digester in a very in WP and SP than in GL), as well as the coexistence of partial inhibitory
close time frame. This increase (approximately 45%) in the conversion effects, particularly evident with GL and WP but completely absent for
yield, statistically relevant (as reported in Fig. 6a), could be related to SP. These findings open interesting perspectives on the possibility to
the intrinsic characteristics of grape lees which, as already described in adopt appropriate pre-treatment strategies to enhance the biological
Section 3.1, is the wine by-product with the most diverse components transformation of such by-products into valuable compounds. A low
and extremely variable characteristics depending on the season. Indeed, frequency sonication pre-treatment (20 kHz for 15 min and at room
GL consists of both solid and liquid fractions which are typically sepa­ temperature) resulted in a considerable increase (approximately 45%)
rated by centrifugation followed by distillation and extraction proced­ in the conversion yield of GL, whereas no relevant changes were
ures. The liquid fraction usually contains organic acids (succinic and observed with WP and GS. As for the profile of fermentation products
acetic acid), sugars (glucose and xylose), polyols (1,2,3-propanotriol and obtained in all performed tests, acids were the main components and,
2,3-butanediol), phenolic compounds and aromatic alcohols (phenyl­ among them, acetic acid was the predominant product, whereas the
ethyl alcohol) [50]. Furthermore, other studies also reported the pres­ proportion of propionic, butyric, valeric, and hexanoic acids varied
ence of other valuable compounds in grape lees such as yeast, tartaric depending on the nature of the fermented by-products. In particular,
acid, amino acids, and phenolic compounds [58–60]. The remaining with farinaceous by-products the concentration of propionic acid
solid fraction is mainly composed of enriched yeast cells, that represent accounted for a high fraction of the overall fermentation products (with
the principal portion and can be exploited as fermentation feedstock values between 35% and over 50% on a COD basis), and this provides an
[50]. Therefore, the pre-treatment by sonication probably played an exciting opportunity for their exploitation in specific applications,
important role in making this component more available for the bio­ including the production of the poly(3-hydroxybutyrate/3-hydrox­
logical degradation. However, there are no studies in the literature yvalerate) (PHBV) copolymer.
confirming this hypothesis and a deeper investigation remains neces­ Overall, acidogenic fermentation can be regarded as a sustainable
sary. Regarding the distribution of fermentation products obtained in process to valorize the tested food industry by-products through the
correspondence to the end of the experiment (Fig. 6b), in the tests production of key platform chemicals in agreement with the principles
conducted with sonicated by-products ethanol was produced in lower of green chemistry. However, further investigations on the possibility to
concentrations than in the tests with the same not sonicated boost the fermentation degree of by-products deriving from the winery
by-products. In particular, for WP ethanol production was very low and and vegetable industry are required.
remained almost unchanged after pre-treatment, whereas in the case of
GL_SN and GS_SN the occurrence of alcoholic fermentation was less CRediT authorship contribution statement
evident than in the tests performed without pre-treatment. Probably, the
applied pre-treatment caused an increase of the most rapidly biode­ Angela Marchetti: Formal analysis, Investigation, Writing – original
gradable components, such as monosaccharides as glucose, and this draft, Visualization. Gaia Salvatori: Formal analysis, Investigation.
changed the rate of ethanol production by accelerating it in the first few Maria Luisa Astolfi: Formal analysis, Investigation, Writing – original
days of operation. This was particularly evident if comparing the draft, Visualization. Matteo Fabiani: Formal analysis, Investigation.
maximum ethanol conversion yield achieved by GL without Joana Fradinho: Validation, Writing – review & editing. Maria A.M.
pre-treatment (equal to 11.2 ± 0.5% COD/COD, on day 23), with the Reis: Conceptualization, Writing – review & editing. Andrea Gianico:
highest ethanol conversion yield obtained in the presence of GL upon Validation, Writing – review & editing. David Bolzonella: Conceptu­
sonication on day 10 of the test operation, which accounted for 7.2 alization, Writing – review & editing. Marianna Villano: Conceptuali­
± 0.8% (COD/COD), followed by a decrease to almost 2% in corre­ zation, Writing – review & editing, Supervision, Funding acquisition.
spondence to the end of the test (39th day of operation) (data not
shown). Furthermore, analyzing the fermentation products obtained Declaration of Competing Interest
with the three different winery substrates, it can be stated that when GL
was used as substrate, regardless of sonicated pre-treatment, higher The authors declare that they have no known competing financial
ethanol production was obtained. This result is probably due to the interests or personal relationships that could have appeared to influence
presence of yeasts, which are more likely to be found in this by-product the work reported in this paper.

8
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

Data Availability from dark fermentation acidogenic effluents: a path to an integrated bio-refinery,
Polymers 13 (2021) 4297, https://doi.org/10.3390/POLYM13244297.
[17] M. Ramos -Suarez, Y. Zhang, V. Outram, Current perspectves on acidogenic
Data will be made available on request. fermentaton to produce volatle faty acids from waste, Rev. Environ. Sci.
Biotechnol. 20 (2021) 439–478, https://doi.org/10.1007/s11157-021-09566-0.
[18] D. Dionisi, I.M.O. Silva, Producton of ethanol, organic acids and hydrogen: an
Acknowledgements opportunity for mixed culture biotechnology? Rev. Environ. Sci Biotechnol. 15
(2016) 213–242, https://doi.org/10.1007/s11157-016-9393-y.
This work has been financially supported by the USABLE Packaging [19] G. Strazzera, F. Battista, N.H. Garcia, N. Frison, D. Bolzonella, Volatile fatty acids
production from food wastes for biorefinery platforms: a review, J. Environ.
Project (Call: H2020- BBI-JTI-2018; G.A. No 836884). Prof. Mauro Manag. 226 (2018) 278–288, https://doi.org/10.1016/J.JENVMAN.2018.08.039.
Majone (Department of Chemistry, Sapienza University of Rome) is [20] M. Zhou, B. Yan, J.W.C. Wong, Y. Zhang, Enhanced volatile fatty acids production
gratefully acknowledged for his insightful discussions on the experi­ from anaerobic fermentation of food waste: a mini-review focusing on acidogenic
metabolic pathways, Bioresour. Technol. 248 (2018) 68–78, https://doi.org/
mental work and his valuable contribution to this study. Barilla G. e R.
10.1016/J.BIORTECH.2017.06.121.
Fratelli S.p.A. and Cooperative Agricole Viti-frutticoltori Italiani Riuniti [21] U. Jayakrishnan, D. Deka, G. Das, Regulation of volatile fatty acid accumulation
Organizzati (CAVIRO) are greatly acknowledged for providing most of from waste: Effect of inoculum pretreatment, Water Environ. Res. (2021)
the by-products used in the study. 1019–1031, https://doi.org/10.1002/wer.1490.
[22] L. Rocha-Meneses, Rawan Zannerni, A. Inayat, M. Abdallah, A. Shanableh,
C. Ghenai, Mohammed Kamil, T. Kikas, Current progress in anaerobic digeston
References reactors and parameters optmizaton, Biomass Convers. Biorefn. (2022) 1–24,
https://doi.org/10.1007/s13399-021-02224-z.
[23] A.F. Soomro, I.A. Abbasi, Z. Ni, L. Ying, J. Liu, Influence of temperature on
[1] C. Torres-León, N. Ramírez-Guzman, L. Londoño-Hernandez, G.A. Martinez-
enhancement of volatile fatty acids fermentation from organic fraction of
Medina, R. Díaz-Herrera, V. Navarro-Macias, O.B. Alvarez-Pérez, B. Picazo,
municipal solid waste: synergism between food and paper components, Bioresour.
M. Villarreal-Vázquez, J. Ascacio-Valdes, N. Aguilar Cristóbal, Food Waste and
Technol. 304 (2020), 122980, https://doi.org/10.1016/j.biortech.2020.122980.
Byproducts: An Opportunity to Minimize Malnutrition and Hunger in Developing
[24] D. Bolzonella, M. Papa, C. Da Ros, L. Anga Muthukumar, D. Rosso, Critcal Reviews
Countries, Front. Sustain. Food Syst. 2 (2018), https://doi.org/10.3389/
in Biotechnology Winery wastewater treatment: a critcal overview of advanced
fsufs.2018.00052.
biological processes, Crit. Rev. Biotechnol. 39 (2019) 489–507, https://doi.org/
[2] J.M. Ueda, M.C. Pedrosa, S.A. Heleno, M. Carocho, I.C.F.R. Ferreira, L. Barros,
10.1080/07388551.2019.1573799.
Food Additives from Fruit and Vegetable By-Products and Bio-Residues: A
[25] R. Ravindran, A.K. Jaiswal, Exploitation of food industry waste for high-value
Comprehensive Review Focused on Sustainability,, Sustainability 14 (2022) 5212,
products, Trends Biotechnol. 34 (2016) 58–69, https://doi.org/10.1016/J.
https://doi.org/10.3390/SU14095212.
TIBTECH.2015.10.008.
[3] 19 October 2007 M e m o r a n d u m Directive 2002/47/EC on Financial Collateral
[26] F. Battista, G. Strazzera, F. Valentino, M. Gottardo, M. Villano, M. Matos, F. Silva,
Arrangements (the “FCD”) – the Commission, (n.d.). 〈http://scholar.googleusercon
M.A.M. Reis, J. Mata-Alvarez, S. Astals, J. Dosta, R.J. Jones, J. Massanet-Nicolau,
tent.com/scholar?q=cache:7_hIC3Xx6_IJ:scholar.google.
A. Guwy, P. Pavan, D. Bolzonella, M. Majone, New insights in food waste, sewage
com/&hl=it&as_sdt=0,5&scioq=DIRECTIVE,+2008.+Directive+2008/98/EC+of
sludge and green waste anaerobic fermentation for short-chain volatile fatty acids
+The+European+Parliament+And+of+The+Council+on+Waste+and+Repealin
production: a review, J. Environ. Chem. Eng. 10 (2022) 108319, https://doi.org/
g+Certain+Directives.+Official+Journal+of+the+European+Union,+L312,+pp.
10.1016/J.JECE.2022.108319.
+3%E2%80%9330〉. (accessed November 15, 2022).
[27] M. Oliveira, E. Duarte, Integrated approach to winery waste: waste generation and
[4] C. Fritsch, A. Staebler, A. Happel, M. Angel Cubero Marquez, I. Aguilo-Aguayo,
data consolidation, Front Environ. Sci. Eng. 10 (2016) 168–176, https://doi.org/
M. Abadias, M. Gallur, I. Maria Cigognini, A. Montanari, M. Jose Lopez, F. Suarez-
10.1007/s11783-014-0693-6.
Estrella, N. Brunton, E. Luengo, L. Sist, M. Ferri, G. Belot, Processing, Valorizaton
[28] F.J. Fernández, D. Infantes, I. Buendía, J. Villaseñor, Volatile fatty acids production
and Applicaton of Bio-Waste Derived Compounds from Potato, Tomato, Olive and
from winery wastewaters by acidogenic fermentation, WIT Trans. Ecol. Environ.
Cereals, A Review, Sustainability 9 (2017) 1492, https://doi.org/10.3390/
111 (2008) 529–535, https://doi.org/10.2495/WP080521.
su9081492.
[29] M. Esteban-Gutiérrez, J. Garcia-Aguirre, I. Irizar, E. Aymerich, From sewage sludge
[5] P. Sharma, V.K. Gaur, S.H. Kim, A. Pandey, Microbial strategies for bio-
and agri-food waste to VFA: individual acid production potential and up-scaling,
transforming food waste into resources, Bioresour. Technol. 299 (2020) 122580,
Waste Manag. 77 (2018) 203–212, https://doi.org/10.1016/J.
https://doi.org/10.1016/J.BIORTECH.2019.122580.
WASMAN.2018.05.027.
[6] V. Amicarelli, G. Lagioia, C. Bux, Global warming potential of food waste through
[30] L. Xue, G. Liu, J. Parfitt, X. Liu, E. Van Herpen, Å. Stenmarck, C. O’connor,
the life cycle assessment: an analytical review, Global warming potential of food
K. Östergren, S. Cheng, Missing Food, Missing Data? A Critical Review of Global
waste through the life cycle assessment: an analytical review 91 (2021) 106677,
Food Losses and Food Waste Data, Environ. Sci. Technol. 51 (2017) 6618–6633,
https://doi.org/10.1016/J.EIAR.2021.106677.
https://doi.org/10.1021/acs.est.7b00401.
[7] C.M. Braguglia, A. Gallipoli, A. Gianico, P. Pagliaccia, Anaerobic bioconversion of
[31] L. Principato, L. Ruini, M. Guidi, L. Secondi, Adopting the circular economy
food waste into energy: a critical review, Bioresour. Technol. 248 (2018) 37–56,
approach on food loss and waste: the case of Italian pasta production, Resour.
https://doi.org/10.1016/J.BIORTECH.2017.06.145.
Conserv. Recycl. 144 (2019) 82–89, https://doi.org/10.1016/j.
[8] M.T. Agler, B.A. Wrenn, S.H. Zinder, L.T. Angenent, Waste to bioproduct
resconrec.2019.01.025.
conversion with undefined mixed cultures: The carboxylate platform, Trends
[32] C.M. Galanakis, Sustainable applications for the valorization of cereal processing
Biotechnol. 29 (2011) 70–78, https://doi.org/10.1016/J.TIBTECH.2010.11.006.
by-products, Foods 11 (2022) 241, https://doi.org/10.3390/foods11020241.
[9] M. Gottardo, D. Bolzonella, G. Adele Tuci, F. Valentino, M. Majone, P. Pavan,
[33] A. Gianico, C.M. Braguglia, D. Mescia, G. Mininni, Ultrasonic and thermal
F. Battista, Producing volatile fatty acids and polyhydroxyalkanoates from foods
pretreatments to enhance the anaerobic bioconversion of olive husks, Bioresour.
by-products and waste: a review, Bioresour. Technol. 361 (2022) 127716, https://
Technol. 147 (2013) 623–626, https://doi.org/10.1016/J.
doi.org/10.1016/J.BIORTECH.2022.127716.
BIORTECH.2013.08.054.
[10] J.W. Lee, H.U. Kim, S. Choi, J. Yi, S.Y. Lee, Microbial production of building block
[34] A. Cesaro, S. Velten, V. Belgiorno, K. Kuchta, Enhanced anaerobic digestion by
chemicals and polymers, Curr. Opin. Biotechnol. 22 (2011) 758–767, https://doi.
ultrasonic pretreatment of organic residues for energy production, J. Clean. Prod.
org/10.1016/j.copbio.2011.02.011.
74 (2014) 119–124, https://doi.org/10.1016/j.jclepro.2014.03.030.
[11] L.L. Jiang, J.J. Zhou, C.S. Quan, Z.L. Xiu, Advances in industrial microbiome base
[35] APHA, AWWA, WEF, Standard Methods for the Examination of Water and
d on microbial consortum for biorefnery, Bioresour. Bioprocess 4 (2017) 11,
Wastewater, Washingt. Am. Public Heal. Assoc. (2012).
https://doi.org/10.1186/s40643-017-0141-0.
[36] International Organization for Standardization, International Organization for
[12] S. Wainaina, Lukitawesa, M. Kuma r Awasthi, M.J. Taherzadeh, Bioengineering of
Standardization. ISO 712:2009 - Cereals and cereal products — Determination of
anaerobic digeston for volatle faty acids, hydrogen or methane producton: a critcal
moisture content — Reference method, Geneva, Switzerland. (2009). 〈htt
review, Bioengineered 10 (2019) 437–458, https://doi.org/10.1080/
ps://www.iso.org/standard/44807.html〉 (accessed June 1, 2023).
21655979.2019.1673937.
[37] Darwin, R. Cord -Ruwisch, W. Charles, Ethanol and lactc acid producton from
[13] J. Medeiros Garcia Alcântara, F. Distante, G. Stort, D. Moscatelli, M. Morbidelli,
sugar and starch wastes by anaerobic acidifcaton, Eng. Life Sci. 18 (2018)
M. Sponchioni, Current trends in the producton of biodegradable bioplastcs: the
635–642, https://doi.org/10.1002/elsc.201700178.
case of polyhydroxyalkanoates, Biotechnol. Adv. 42 (2020), 107582, https://doi.
[38] M.C. Casiraghi, M.A. Pagani, D. Erba, A. Marti, C. Cecchini, M. Grazia D’egidio,
org/10.1016/j.biotechadv.2020.107582.
International Journal of Food Sciences and Nutrition Quality and nutritional
[14] Z.A. Raza, S. Abid, I.M. Banat, Polyhydroxyalkanoates: Characteristics, production,
properties of pasta products enriched with immature wheat grain, Int. J. Food Sci.
recent developments and applications, Int Biodeterior. Biodegrad. 126 (2018)
Nutr. 64 (2013) 544–550, https://doi.org/10.3109/09637486.2013.766152.
45–56, https://doi.org/10.1016/j.ibiod.2017.10.001.
[39] A. Pasqualone, B. Laddomada, I. Centomani, V.M. Paradiso, D. Minervini,
[15] F. Valentino, F. Morgan-Sagastume, S. Campanari, M. Villano, A. Werker,
F. Caponio, C. Summo, Bread making aptitude of mixtures of re-milled semolina
M. Majone, Carbon recovery from wastewater through bioconversion into
and selected durum wheat milling by-products, LWT 78 (2017) 151–159, https://
biodegradable polymers, N. Biotechnol. 37 (2017) 9–23, https://doi.org/10.1016/
doi.org/10.1016/J.LWT.2016.12.032.
j.nbt.2016.05.007.
[40] S.P. Cauvain, Bread – the Product, in: S.P. Cauvain, L.S. Young (Eds.), Technology
[16] R.G. Saratale, S.K. Cho, G.D. Saratale, M. Kumar, R.N. Bharagava, S. Varjani, A.
of Breadmaking, Springer, Boston, MA, 2007, pp. 1–19, https://doi.org/10.1007/
A. Kadam, G.S. Ghodake, R.R. Palem, S.I. Mulla, D.S. Kim, H.S. Shin, An overview
0-387-38565-7_1.
of recent advancements in microbial polyhydroxyalkanoates (PHA) production

9
A. Marchetti et al. Biochemical Engineering Journal 199 (2023) 109029

[41] R. Renato Balandrán-Quintana, J. Nemesio Mercado-Ruiz, A. María Mendoza- health promoters: a review, Int. J. Mol. Sci. 15 (2014) 15638–15678, https://doi.
Wilson, Wheat bran proteins: a review of their uses and potential, Food Rev. Int. 31 org/10.3390/ijms150915638.
(2015) 279–293, https://doi.org/10.1080/87559129.2015.1015137. [53] Á. Peralbo-Molina, M.D. Luque deCastro, Potential of residues from the
[42] W.L. Chow, S. Chong, J.W. Lim, Y.J. Chan, M.F. Chong, T.J. Tiong, J.K. Chin, G. Mediterranean agriculture and agrifood industry, Trends Food Sci. Technol. 32
T. Pan, Anaerobic co-digestion of wastewater sludge: a review of potential co- (2013) 16–24, https://doi.org/10.1016/J.TIFS.2013.03.007.
substrates and operating factors for improved methane yield, Processes 8 (2020) [54] C. Zhang, H. Yang, F. Yang, Y. Ma, Current progress on butyric acid production by
1–21, https://doi.org/10.3390/pr8010039. fermentation, Curr. Microbiol. 59 (2009) 656–663, https://doi.org/10.1007/
[43] J. García-Lomillo, M.L. Gonzalez-Sanjosé, Applicatons of Wine Pomace in the Food s00284-009-9491-y.
Industry: Approaches and Functons, Compr. Rev. Food Sci. F. 16 (2017) 3–22, [55] Y. Zhu, S.T. Yang, Effect of pH on metabolic pathway shift in fermentation of
https://doi.org/10.1111/1541-4337.12238. xylose by Clostridium tyrobutyricum, J. Biotechnol. 110 (2004) 143–157, https://
[44] M.R. González-Centeno, C. Rosselló, S. Simal, M.C. Garau, F. López, A. Femenia, doi.org/10.1016/J.JBIOTEC.2004.02.006.
Physico-chemical properties of cell wall materials obtained from ten grape [56] A. Marchetti, L. Lorini, G. Salvatori, A. Gianico, M. Majone, M. Villano,
varieties and their byproducts: grape pomaces and stems, LWT - Food Sci. Technol. Polyhydroxyalkanoates production by mixed microbial cultures in sequencing
43 (2010) 1580–1586, https://doi.org/10.1016/J.LWT.2010.06.024. batch reactors operated under different feeding conditions, Chem. Eng. Trans. 93
[45] P.U. Rao, Nutrient composition of some less-familiar oil seeds, Food Chem. 50 (2022) 163–168, https://doi.org/10.3303/CET2293028.
(1994) 379–382, https://doi.org/10.1016/0308-8146(94)90208-9. [57] L. Iglesias-Carres, A. Mas-Capdevila, F.I. Bravo, G. Aragonès, A. Arola-Arnal,
[46] M. Bravi, F. Spinoglio, N. Verdone, M. Adami, A. Aliboni, A. D’Andrea, A. de B. Muguerza, A comparatve study on the bioavailability of phenolic compounds
Santis, D. Ferri, Improving the extraction of α-tocopherol-enriched oil from grape from organic and nonorganic red grapes, Food Chem. 299 (2019), 125092, https://
seeds by supercritical CO2. Optimisation of the extraction conditions, J. Food Eng. doi.org/10.1016/j.foodchem.2019.125092.
78 (2007) 488–493, https://doi.org/10.1016/J.JFOODENG.2005.10.017. [58] A. De Iseppi, G. Lomolino, M. Marangon, A. Curioni, Current and future strategies
[47] A. Llobera, J. Cañellas, Dietary fibre content and antioxidant activity of Manto for wine yeast lees valorization, Food Res. Int. 137 (2020) 109352, https://doi.
Negro red grape (Vitis vinifera): pomace and stem, Food Chem. 101 (2007) org/10.1016/J.FOODRES.2020.109352.
659–666, https://doi.org/10.1016/J.FOODCHEM.2006.02.025. [59] C. Dimou, N. Kopsahelis, A. Papadaki, S. Papanikolaou, I.K. Kookos, I. Mandala, A.
[48] J. Yu, M. Ahmedna, Functonal components of grape pomace: their compositon, A. Koutinas, Wine lees valorization: Biorefinery development including production
biological propertes and potental applicatons, Int. J. Food Sci. Technol. 48 (2013) of a generic fermentation feedstock employed for poly(3-hydroxybutyrate)
221–237, https://doi.org/10.1111/j.1365- 2621.2012.03197.x. synthesis, Food Res. Int. 73 (2015) 81–87, https://doi.org/10.1016/J.
[49] B. Ahmad, V. Yadav, A. Yadav, M.U. Rahman, W.Z. Yuan, Z. Li, X. Wang, FOODRES.2015.02.020.
Integrated biorefinery approach to valorize winery waste: A review from waste to [60] K. Filippi, H. Papapostolou, M. Alexandri, A. Vlysidis, E.D. Myrtsi, D. Ladakis,
energy perspectives, Sci. Total Environ. 719 (2020) 137315, https://doi.org/ C. Pateraki, S.A. Haroutounian, A. Koutinas, Integrated biorefinery development
10.1016/J.SCITOTENV.2020.137315. using winery waste streams for the production of bacterial cellulose, succinic acid
[50] S.P. Ioannidou, A.G. Margellou, M.D. Petala, K.S. Triantafyllidis, Pretreatment/ and value-added fractions, Bioresour. Technol. 343 (2022) 125989, https://doi.
fractionation and characterization of winery waste streams within an integrated org/10.1016/J.BIORTECH.2021.125989.
biorefinery concept, Sustain Chem. Pharm. 27 (2022), 100670, https://doi.org/ [61] A.M. Erian, M. Sauer, Utilizing yeasts for the conversion of renewable feedstocks to
10.1016/J.SCP.2022.100670. sugar alcohols - a review, Bioresour. Technol. 346 (2022) 126296, https://doi.org/
[51] D. Kammerer, A. Claus, R. Carle, A. Schieber, Polyphenol Screening of Pomace 10.1016/j.biortech.2021.126296.
from Red and White Grape Varietes (Vits vinifera L.) by HPLC -DA D -MS /MS, [62] F.A.F. Antunes, K. Rajan, A. Djioleu, T.M. Rocha, L.P. Brumano, Y.C. de Souza
J. Agric. Food Chem. 52 (2004) 4360–4367, https://doi.org/10.1021/jf049613b. Melo, J.C. dos Santos, C.A. Rosa, D.J. Carrier, S.S. da Silva, Sustainable second-
[52] A. Teixeira, N. Baenas, R. Dominguez-Perles, A. Barros, E. Rosa, D.A. Moreno, generation ethanol production from switchgrass biomass via co-fermentation of
C. Garcia-Viguera, Natural bioactive compounds from winery by-products as pentoses and hexoses using novel wild yeasts, Bioenergy Res. 15 (2022)
1157–1168, https://doi.org/10.1007/s12155-021-10302-3.

10