Journal of Environmental Management 226 (2018) 278–288

Contents lists available at ScienceDirect

Journal of Environmental Management

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

Research article

Volatile fatty acids production from food wastes for biorefinery platforms: A T
review
Giuseppe Strazzera, Federico Battista∗, Natalia Herrero Garcia, Nicola Frison, David Bolzonella
Department of Biotechnology, University of Verona, Strada Le Grazie 15, 37134, Verona, Italy

A R T I C LE I N FO A B S T R A C T

Keywords: Volatile fatty acids (VFAs) are a class of largely used compounds in the chemical industry, serving as starting
Food wastes molecules for bioenergy production and for the synthesis of a variety of products, such as biopolymers, reduced
Volatile fatty acids chemicals and derivatives. Because of the huge amounts of food waste generated from household and processing
Operative parameters optimization industry, 47 and 17 million tons per year respectively only in the EU-28 Countries, food wastes can be the right
Applications
candidate for volatile fatty acids production. This review investigates all the major topics involved in the op-
Polyhydroxyalkanoates
timization of VFAs production from food wastes. Regarding the best operative conditions for the anaerobic
fermenter controlled pH in the neutral range (6.0–7.0), short HRT (lower than 10 days), thermophilic tem-
peratures and an organic loading rate of about 10 kgVS/m3d, allowed for an increase in the VFAs concentration
between 10 and 25%. It was also found that additions of mineral acids, from 0.5 to 3.0%, and thermal pre-
treatment in the range 140–170 °C increase the organic matter solubilisation. Applications of VFAs considered in
this study were biofuels and bioplastics production as well as nutrients removal in biological wastewater
treatment processes.

1. Introduction exploitation for the production of Volatile Fatty Acids (VFAs). These are
linear short-chain aliphatic mono-carboxylate compounds, having from
The more and more impacting problem of global warming, due to two (acetic acid) to six (caproic acid) carbon atoms. Due to their
the increase of greenhouse gases in the atmosphere, encouraged the functional groups, VFAs are extremely useful for the chemical industry:
emerging of the “biorefinery” concept. Biorefinery represents an in- carboxylic acids are precursors of reduced chemicals and derivatives
novative approach in the environmental management, where products (esters, ketones, aldehydes, alcohols and alkanes) in conventional or-
at the end of their service life or waste materials are seen as valuable ganic chemistry (Dahiya et al., 2015). Moreover, they are also well-
resources for the production of high added value bio-products or bio- known substrates for the production of biofuels like methane and hy-
fuels and are produced from renewable sources, possibly from organic drogen as well as biopolymers, such as polyhydroxyalkanoates (PHAs)
wastes (Nghiem et al., 2017). Among them, Food Wastes (FWs) re- (Raganati et al., 2014; Domingos et al., 2017).
present the optimal candidates for the biorefinery processes (Alibardi Nowadays, VFAs are mainly produced through oxidation or car-
and Cossu, 2015). Food and Agriculture Organization (FAO) estimated boxylation of chemical precursors, such as aldehydes and alkenes, de-
that one third of world food production is lost or wasted along the food riving from petroleum processing (Riemenschneider, 2000). The bior-
supply chain including the final steps like households, restaurants, and efinery approach for VFAs production contemplates the fermentation
canteens. EU-28 Member States produced approximately 89 million process provided both by pure culture of specific anaerobic bacterial
tons of food waste in 2012 (Braguglia et al., 2018), (Lucifero, 2016). In strains, and by Mixed Microbial Cultures (MMCs) (Dai et al., 2017).
particular, the sectors contributing to FWs production are households Although MMCs may lead to lower yields in terms of VFAs, they
(47 million tonnes ± 4 million tonnes) and processing industry (17 have several advantages, since non-sterile conditions are needed, and
million tonnes ± 13 million tonnes) (Stenmarck et al., 2016). Very risk of contamination is decreased (Bhatia and Yang, 2017). At the same
often these FWs are disposed of in landfill or sent to incineration with time, MMCs can metabolize a wide spectrum of organic molecules,
null or limited recovery of resources and high emission of greenhouse being able to face with the heterogeneous composition of FWs in terms
gases and toxic compounds in the atmosphere and in the soil in both the of carbohydrates, proteins and lipids (Jankowska et al., 2015). VFAs
cases (Ren et al., 2018). A sustainable alternative is the FWs production through MMCs is made possible by different metabolic

∗
Corresponding author.
E-mail addresses: federico.battista@univr.it, federico.battista@gmail.com (F. Battista).

https://doi.org/10.1016/j.jenvman.2018.08.039
Received 23 May 2018; Received in revised form 5 August 2018; Accepted 7 August 2018
Available online 16 August 2018
0301-4797/ © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

List of abbreviations: iso-HCa iso-caproic acid

iso-Hva iso-valeric acid
C/N carbon – nitrogen ratio MMC Mixed Microbial Cultures
COD total chemical oxygen demand OLR organic loading rate
HAc acetic acid PHA poly-hydroxy-alkanoates
HBu butyric acid sCOD soluble chemical oxygen demand
HCa caproic acid TKN total Kjeldahl nitrogen
HPr propionic acid TS total solids
HRT hydraulic retention time TVS total volatile solids
HVa valeric acid VFAs volatile fatty acids
iso-HBu iso-butyric acid

pathways, which depend essentially on the substrates (carbohydrates, methanogens. On the contrary, a pH in the range 7.0–7.5 and higher
lipid and proteins) (Garcia et al., 2018). HRT enable the survival and the activity of microorganisms producing
In order to strengthen the potential of MMCs fermentation in terms methane from VFAs (Baumann and Westermann, 2016). Several studies
of VFAs production, it is necessary to pay close attention to process emphasized that the two-stage anaerobic digestion (AD) process as the
parameters and operational conditions during the experimental setup. most promising way for VFAs production (Nagao et al., 2012; Jiang
In fact, different hydraulic retention time (HRT), organic loading rate et al., 2013; Browne and Murphy, 2014). Hydrolysis and acidogenesis
(OLR), temperature, and pH, influence the VFAs yield as far as the are the first steps of AD: firstly, the complex polymers present in or-
amount of other fermentation by-products, such as longer chain fatty ganic matter are reduced into soluble monomers (simple sugars, amino
acids, other carboxylic acids, alcohols, biohydrogen, biomethane, es- acids, glycerol) along hydrolytic phase and then fermented by MMCs
ters, and other intermediates (Mohan et al., 2016). The different into VFAs, carbon dioxide (CO2), and hydrogen (H2) along the acido-
pathways seem to have some preferential operative conditions: ther- genic phase, also known as dark fermentation (DF) (Fig. 1) (Parawira
mophilic temperatures, controlled pH at neutral values (6.0–7.0) and et al., 2004).
short HRT between 1 and 10 days, depending on the substrates This paper reviews the results obtained in several studies dealing
(Bolzonella et al., 2005). For example, these conditions favour the with the production of VFAs through acidogenic fermentation of FWs
glycolytic pathway allowing the acetic acid production, inhibiting the carried out by MMCs, in order to provide a view on the best FWs

Fig. 1. Process cascade of anaerobic digestion process. Red line indicates the phase separation to obtain VFAs (Adapted from Batstone et al., 2002).

279
 G. Strazzera et al.

Table 1
a
Physical and chemical characterizations of FWs. analysis carried out on dry basis.
Characteristics References

Origin pH TS TVS TKN tCOD C/N N–NH4 carbohydrates proteins lipids total VFAs

−1
university 3.9 12.9 (%) 12.5 (%) 3.33 (g/L) 166.18 (g/L) 49.9 < 10 (mg × L ) 69.3 (%TS) 16.1 (%TS) 10.6 (%TS) 3.6 (g/L) Zhang et al.
canteen (2005)
university 5.9 847.0 (g/Kg) 724.1 (g/Kg) 14.9 (g/L) 472.0 (g/L) 37.0 0.13 (g/L) – – – 0.49 (gHAc/L) Forster-
canteen Carneiro et al.
(2008)
synthetic 4.72 97.8 (g/L) 91.8 (g/L) 2500 (mg/L) 108.4 (g/L) – 54.0 (mg/L) 45.3 (g glucose/ 16.2 (g – – Lee et al. (2008)
L) albumin/L)
synthetic 4.9 16.0 (%) 94 (%)a – – – – – – – 816.3 (mg/kg)a Komemoto
et al. (2009)
synthetic 6.7 16.5 (%) 15.5 (%) – 250.5 (g/L) – – – – – – Izumi et al.
(2010)
university – 18.28 (%) 87.48 (%)a – – 63.62 – 35.47 (%)a 14.42 (%)a 24.11 (%)a – He et al. (2012)
canteen
garbage 3.77 10.3 (%) 9.2 (%) – 152 (g/L) – – – – – – Nagao et al.
collection (2012)

280
company
synthetic 4.59 ± 0.17 20.53 ± 2.04 (%) 19.95 ± 2.21 – – 13.45 760.56 ± 30.33 – – – 829.53 ± 103.16 Jiang et al.
(%) (mg/L) (mg/L) (2013)
university – 22.61 (%) 17.90 (%) 2.63 (%)a 30.25 (%)a 11.50 – – – – – Shen et al.
canteen (2013)
university 3.9 ± 0.1 20 ± 3.5 (%) 19 ± 2.3 (%) – – 24.03 – 70 ± 2.1 (%TS) 13 ± 2.1 10 ± 1.5 – Bo and Pin-jing
canteen (%TS) (%TS) (2014)
catering 4.85 ± 0.05 31.5 ± 0.25 (%) 91.2 ± 0.94 – – 14.4 – 34.7 ± 2.8 29.6 ± 1.6 26.7 ± 0.4 – Browne and
premises (%)a (%TS) (%TS) (%TS) Murphy (2014)
university 5.6 150 (g/L) 90 (% TS) – – 34 0.11 (g/L) – – – 0.84 (g/L) Gou et al.
canteen (2014)
university 6.1 24.0 (%) – 1.8 (%) – – – 39.5 (%) 11.0 (%) – – Wang et al.
canteen (2014)
waste treatment – 23.65 ± 0.38 (%) 21.99 ± 0.49 7.44 ± 0.02 998 ± 71 (g/kg) – – – – – – Yirong et al.
a
plant (%) (g/kg) (2015)
a a a a a
university – 25.7 (%) 98.2 (%)a – 1.1 (%) – – 0.6 (%) 15.1 (%) 0.03 (%) – Yin et al.
canteen (2016b)
university – 17.4 ± 1.6 (%) 16.1 ± 2.6 (%) 2.0 ± 0.2 229.7 ± 15.9 – – 61.5 ± 11.1 9.8 ± 2.6 9.2 ± 1.2 – Zhang et al.
canteen (%TS) (g/L) (%TS) (%TS) (%TS) (2016)
university 4.8 ± 0.1 3.96 ± 0.08 (%) 90.23 ± 1.95 – 124.2 ± 1.2 (g/ – – 8.95 (%TS) 5.32 (%TS) – – Zhen et al.
canteen (%TS) L) (2016)
Journal of Environmental Management 226 (2018) 278–288
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

pretreatments and of the best operational conditions for maximizing 2017). Chemical pretreatments include mainly acid, alkali or organic-
VFAs production. In addition, this work summarizes the most inter- aqueous solvent mixtures (ethanol, benzene, ethylene glycol, or bu-
esting VFAs applications focusing the attention on some innovative tanol) addition at different concentration to favour the solubilisation of
examples. the FWs. In this way, the surface area is increased and the lignin
structure is altered. The drawback of chemical pretreatment is the high
2. The feedstock: food wastes amount of reagents, which makes them not economically attractive
(Passos et al., 2017). Biological pretreatments involve enzymes adop-
As reported in the previous paragraph, huge amounts of FWs from tion, with the ability to degrade cellulose (cellulase) and to favour the
household and the assimilate sectors (canteen, restaurants, catering, hemicellulose solubilisation. Biological pretreatments require no ex-
vegetables and fruit residues from markets, supermarkets and food cessive energy and chemical additions are needed, thus are more en-
factories) are annually produced. All these streams present character- vironmental friendly than chemical ones. Enzymatic pretreatment in-
istics that, although variable and often seasonal, are in some way si- tensify the hydrolytic activity improving the sugars production and
milar (Traverso et al., 2000; Stenmarck et al., 2016). consequentially the VFAs production (Braguglia et al., 2018). Enzy-
Table 1 shows how chemical and physical characteristics are greatly matic hydrolysis is a very sensitive process, influenced by several
influenced by municipality, time of collection, social and economic boundary conditions: the way of enzymes addition within the reactor
aspects, seasonal food availability and different consumption patterns. (prior to digestion as pre-treatment or directly in the digester), the
For example, it is interesting to observe how University Canteens’ FWs operative mode of the reactor (batch, fed-batch or continuous) and the
present a very heterogeneous nature, despite their common origin. device assuring the mixing (Battista et al., 2019). However, enzymatic
Although these differences, it should be recognized that FWs, in pretreatments have slower kinetics than the chemical pretreatments
general, has some specific features: they contain high levels of organic (Battista and Bolzonella, 2018), which make them not very competitive.
matter (around 15–20% TVS), high nitrogen and phosphorous con-
centrations (2–15 g/kg and 0.5–1.0 g/kg, respectively), indispensable 3.2. Pretreatments to improve VFAs production from FWs
nutrients for the metabolic pathway of the microorganisms. These
chemical features make FWs optimal substrates for VFAs production In the last decades, a huge amount of research works has been
through dark fermentation. In particular, FWs, as all the organic sub- conducted on the optimization of FWs hydrolysis to increase the VFAs
strates, are essentially composed by three groups of macro-molecules content. In particular, chemical and thermal pretreatments and their
which are able to influence the kinetic of the VFAs production: carbo- combination emerged as the preferred methods for the improvements of
hydrates, proteins, and lipids according the concentration reported in VFAs production from FWs (Ma et al., 2011; Bolzonella et al., 2018).
Table 1 (Alibardi and Cossu, 2015; Shen et al., 2017). For example, Vavouraki et al. (2013) tested different acid and al-
The influence of these macromolecules will be analyzed along this kaline compounds at various concentrations (from 0.5 to 3%) on a re-
review. presentative sample of Greek household FWs. Chemical additions were
combined to thermal effect at different temperatures (from 50 °C to
3. Pretreatments to increase VFAs production 120 °C) acting for a variable time range, between 15 and 120 min. The
results were interesting because a different behaviour between acid and
3.1. Different pretreatments options for FWs hydrolysis alkaline compounds on acidogenic hydrolysis was observed. HCl and
H2SO4 have brought to higher soluble organic matter concentration in
Before any bioprocess is applied, the substrates pass from the pre- the reaction medium, attributable to carbohydrates conversion into
treatment stage, having the aim to prepare them to microorganisms’ simpler sugars. On the contrary, NaOH addition emerged in negative
activity (Dahiya et al., 2018). performances due to Maillard reactions, which occur at high tempera-
The main scopes of pretreatments are: ture and comport the formation of inhibiting and toxic compounds.
Regarding the best temperature range for pretreatment, Li et al. (2017)
i) the reduction of the substrates' size, found that the optimal operative temperature range was 140–170 °C.
ii) the extraction of smaller and simpler chemical compounds to im- Lower temperatures led to a partial FWs degradation, while higher
prove the fermentation stage, temperatures contributed to inhibiting compounds caused by a dena-
iii) the removal of inert material not suitable for the following bio- turation of the functional groups of the substrates.
processes (Li et al., 2017). More controversial is the effect of the thermal combination with
microwave pretreatment as opposing results on VFAs production were
Regarding FWs, the most recalcitrant organic fractions are the lig- recorded. Shahriari et al. (2013) concluded that only marginal benefits
nocellulosic materials. They include long polymers chains, constituted can be given by microwaves on VFAs increasing in presence household
by cellulose hemicellulose and lignin, whose solubilisation starts at FWs, which are rich in water content and poor in recalcitrant or com-
about 180 °C in neutral conditions and about 150 °C in an acid en- plex chemical compound. On the contrary, better results were observed
vironment (Veluchamy et al., 2018). on Organic Fraction of Municipal Solid Wastes (OFMSW), which may
Pretreatments are usually defined as: have a larger component of less readily biodegradable compounds.
When thermal, chemical or microwave pretreatments are adopted,
i) physical, an environmental and economic evaluation is recommended because
ii) chemical and they require a great energy demand, which is not always recovered
iii) biological, through an adequate increase of VFA yield (Bolzonella et al., 2018). The
interest in biorefinery emerges in an increasing in the research of new
or their combinations, all widely adopted for the pretreatment of and efficacy pretreatment methods, such as microwave. Although,
FWs. sometimes these pretreatment had better yields in the organic matter
Physical pretreatments allow the reduction of substrates dimen- hydrolysis, they did not achieve such improvements to cover their en-
sions, of the crystallinity and polymerization degrees, and, con- ergetic and economic costs. As consequence, they have not still im-
sequentially, they lead to a larger specific surface available to the en- plemented in pilot or industrial scale processes (Aguilar-Reynosa et al.,
zymes and microorganisms reactions. The most adopted machines are 2017). For this reason, more economic approaches are recommended.
cutters milling, wet disk milling, ball milling, vibratory ball milling, Yu et al. (2016), for example, tested a pre-fermentation stage consisting
compression milling, hammer milling and roll milling (Paudel et al., in a slight fermentation at an initial pH of 6.0, conducted by the

281
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

indigenous microorganisms in FWs at room temperature. This stage has (1) Stickland reaction,
been stopped when pH dropped at 3.5; then the effective fermentation (2) oxidative deamination of amino acids, and
started at 30 °C for 21 days. The authors found a 12% increase of the (3) reductive deamination of amino acids.
VFA yield, essentially promoted by the improvements of carbohydrates
hydrolysis. Economic analysis indicated that pre-fermentation was an The first pathway allows at least 90% of protein degradation; the
earning-effective way approach for FWs to bioenergy such as VFAs. A second, that leads to hydrogen as unique end product, is not thermo-
variation of the pre-fermentation stage is the hydrothermal pre-fer- dynamically favourable and needs very low values of hydrogen partial
mentation where substrates are treated at 160 °C for 30 min before the pressure. Finally, the third reaction is energetically favourable and
pre-fermentation. In this way, the solubilisation of more complex sub- consume hydrogen (Alibardi and Cossu, 2016).
strates (cellulose and proteins) has also favoured, with a VFA yield Although fermentation of proteins leads to production of the same
improvement of 21%. VFAs obtained from sugar metabolism (HAc, HPr, and HBu), the re-
lative ratio among them is quite different. HAc was the main VFA ob-
4. The most relevant parameters influencing VFAs production tained from fermentation of peptone, as reported by Yin et al. (2016a),
accounting for the 70% of total VFAs produced, while HBu and HPr
4.1. Influence of carbon-source nature in FWs representing around the 10 and 15%. Furthermore, HVa production is
mainly associated with proteins fermentation, as a result of redox
The FWs composition is fundamental in VFAs biosynthesis, being Stickland reaction between couple of amino acids and of reductive
able to influence both their quantity and their chemical distribution deamination of sole amino acids (Parawira et al., 2004) but its con-
(HAc, HBu, HPr). In addition, the FWs nature determines also the centration was only the 5% of total VFAs produced (Yin et al., 2016a).
choice of the operative parameters. As for carbohydrates, VFAs production seems also to be affected by
Generally, lipids in FWs are less suitable for fermentation than origin of proteins. Lately, it was demonstrated that fermentation of
carbohydrates and proteins. Even if they give contribution to high COD animal or vegetal proteins lead to a different VFAs profiling. Shen et al.
levels into substrate, lipids have slower biodegradation kinetics carried out a fermentation of two different proteinaceous substrates
(Alibardi and Cossu, 2016). Furthermore, hydrolysis of lipids produces (i.e. proteins from tofu or eggs), finding out that metabolism of vegetal
glycerol and long chain fatty acids (LCFAs). Although glycerol can be proteins leads to production HAc, HPr, HBu, and HVa with a relative
used as a fermentation substrate, LCFAs are able to adhere to cellular ratio of 56.3:15.7:10.4:17.6, while fermentation of animal proteins
wall, affecting the transport of nutrients, and, consequentially, in- leads to an equal production of these acids (Shen et al., 2017). Likely,
hibiting the metabolism of anaerobic bacteria (Alibardi and Cossu, this different VFAs profiling can be related to a different amino acids
2016; Shen et al., 2017). composition of animal and plant proteins. The former, for example,
Carbohydrates are easily converted by microbial enzymes into usually shows a consistently lower concentration in lysine, sulphur-
glucose, which is immediately available for glycolysis and fermentation containing amino acids with respect to animal derived protein, and a
into VFAs (Shen et al., 2017). Moreover, in many cases it was observed similar difference can be found also in terms of threonine in cereal-
that fermentation of pure carbohydrates, such as glucose, leads mainly derived proteins (Sosulski and Imafidon, 1990; Young and Pellett,
to production of HBu, HPr and HAc, in this exact order as observed by 1994).
Yin et al. (2016a), who investigated the pure glucose fermentation in
batch reactor, at mesophilic range, under pH 6, using granular activated 4.2. Influence of operational parameters on VFAs production
sludge as seed. They obtained a maximal VFAs production of 38.2
gCOD/L, with HAc, HPr, HBu, and HVa accounting for 17, 30, 50, and Operational parameters, such as temperature, pH, hydraulic reten-
3% respectively. Authors assumed that, despite a higher theoretical tion time (HRT), and organic loading rate (OLR) show important effects
high conversion efficiency of glucose into HAc, it was not accumulated on VFAs production from FWs fermentation, both in terms of yield and
into reactor as consequence of its consumption for the production of H2 relative distribution among different products. Although these para-
or for microbial growth. However, another study on fermentation of meters have a synergetic effect on the microbial communities involved
starch coming from fresh potatoes in a packed-bed reactor operating at in fermentation processes as act on cellular metabolism, most of re-
mesophilic range and uncontrolled pH, led to a total VFAs production of searches investigated on their influence one at a time. The same ap-
around 18 g/L, with HAc, HBu and HPr representing about the 45, 28, proach has been followed along this work.
and 14% of total, respectively. Nevertheless, total VFAs and their pro-
filing changed doubling the loading rate; VFAs production reached a 4.2.1. pH
slightly lower value, of around 17 g/L, with HAc representing about Among the operational parameters, pH has a very strong effect on
70% of total (Parawira et al., 2004). Alibardi and Cossu (2016) found VFAs production from FWs fermentation. Jiang et al. (2013) studied the
the fermentation of synthetic substrates, composed by fresh food rich in influence of different pH values, comprised in slightly acidic range
carbohydrates, was achieved a concentration of HBu close to that of (5–7), on VFAs production from synthetic kitchen waste. They found
HAc, with a HBu/HAc ratio of around 0.8. All these studies carried out out that a pH value comprised between 6 and 7 brought to an increase
using real substrates confirmed HBu, HPr and HAc as main carbohy- of around 20% of hydrolysis rate, achieving a value for soluble COD of
drates fermentation products. The differences in terms of single per- 82 g/L. At lower uncontrolled pH, the observed sCOD was 60 g/L. This
centage in VFAs depend on the complexity and heterogeneity of the increase of solubilisation allowed to double VFAs production in the
substrates and of the different operative parameters, which can activate batch reactor with pH adjusted to a value of 6 and 7 and an increase of
or inactivate a specific metabolic pathway (Feng et al., 2009). fermentation products by tenfold compared to the uncontrolled pH, as
Instead, proteins are generally characterized by a lower biode- consequence of a higher hydrolytic enzymes activity and avoidance of
gradability, due to their tertiary and quaternary structure, which make inhibition due to acidification of medium. Furthermore, a pH close to
them less susceptible to protease action (Battista and Bolzonella, 2018). neutrality brought to a different distribution among VFAs, with HBu,
The result is that the efficiency of carbohydrates hydrolysis from FWs is HAc, and HPr accounting for around 50, 25, and 15% of total fer-
up to 80%, while from protein is in the range 40–70%. For this reason, mentation products. Zhang et al. (2005) carried out a study about pH
FWs proteins hydrolysis is considered as a rate-limiting step during effect on fermentation performance in a reactor working in continuous
acidogenic fermentation (Shen et al., 2017; Yin et al., 2016a). mode adjusting pH to a value of 5, 7, 9, and 11. This study confirms that
Degradation of proteins takes place essentially through three bio- pH close to neutrality leads to a better VFAs yield. In fact, under a pH
chemical pathways: value of 7, fermentative metabolism is favoured, and a VFAs yield of

282
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

0.27 gVFA/gTS was achieved, while it was of 0.15 gVFA/gTS in the how the choice of the optimal operative temperature is related to pH
control reactor, where the pH was not controlled. The positive effect of with respect to VFAs yields. They treated substrates in two different
slightly acid-neutral conditions on microbial metabolism and therefore fermentation steps. The first was at 55 °C with uncontrolled pH. Under
on fermentative production was demonstrated in another study in these conditions, total VFAs production achieved a value of around
which, besides VFAs production, carbohydrates and proteins utilization 12 g/L, with iso-HBu as quite unique fermentation product. In the
rates were monitored. Under a pH set up at a value of 6, VFAs pro- subsequent experimental run, fermentation was carried out at 65 °C,
duction and VFAs yield increased 17 and 7.5 times respectively, with both with uncontrolled pH and set up at a value of 7. The correction of
respect to pH 4 (Wang et al., 2014). A study carried out by Dahiya et al. pH allowed to obtain a VFAs production of around 18 g/L, twofold
(2015) emphasizes the relationship between pH and individual VFAs greater with respect to the experimental phase carried out under un-
obtained from acidogenic fermentation. They fermented FWs from controlled pH, but slightly greater than the production achieved during
canteen in bench-scale batch reactors, under the following conditions: the initial experimental phase at lower temperature. Nevertheless, the
28 °C, OLR 15 kgCOD/m3d, 10% w/w inoculum. pH was adjusted at the temperature increase led to a good relative distribution of fermentation
beginning of experiment to 5, 6, 7, 8, 9, 10, and 11. The maximal total products, with HAc, HPr, and HBu accounting for the 39-28-17%, re-
VFAs concentration was detected into reactor operating with an initial spectively. A further increase of temperature until 70 °C, with a pH
pH of 10, followed by pH 9 (5.17 g VFA/l), pH 6 (4.5 g VFA/L), pH 5 controlled at 7, brought to a higher total VFAs production of around
(4.2 g VFA/L), pH 7 (4.1 g VFA/L), pH 8 (3.8 g VFA/L) and pH 11 (3.5 g 35 g/L. Under these condition HAc was still the main fermentation
VFA/L). The lower VFAs production was achieved in correspondence of product, accounting for 55% of total VFAs, but HPr was produced in
the lowest and highest pH ranges considered, and can be explainable smaller quantitative representing the 14% of total production, while
because acidogenic bacteria cannot survive under extremely acidic (pH HBu was slightly more abundant, accounting for the 31% of total VFAs
3) or alkaline (pH 12) conditions. It must be taken into account that the obtained. A further increment of temperature until 80 °C led to a gra-
consumption of the produced VFAs occurred for methane production dually decrease of VFAs production, up to a value similar to that ob-
carried out by acetoclastic (pH, 6–8) or by hydrogenoclastic (pH, 9–10) tained at 65 °C.
archaebacteria. Therefore, it is not possible to conclude that alkaline pH Therefore, it is clear that a higher working temperature, around
is ideal for VFAs production as their concentrations were higher before thermo- and hyperthermophilic range (40–80 °C), leads to an increase
their conversion. With regard to VFAs distribution pattern, an alkaline of hydrolysis rate, giving hydroxylates theoretically available for fer-
pH (10) seemed to favour HAc production, which reached a maximal mentative microbial metabolism. Nevertheless, an increase of VFAs
value of 4.2 g/L after 36 h, and then it decreased up to around 3.6 g/L production is possible only according to optimal bacterial growing
after 48 h, probably as consequence of methane production. HBu was temperature, since many acidogens cannot survive at extreme tem-
the main fermentation product under a pH of 5 (1.8 g/L), and its con- perature range (Shin et al., 2004).
centration did not show a remarkable decrease between 36 and 48 h.
HPr achieved a concentration of around 1.4 g/L under all pH values 4.2.3. Hydraulic retention time (HRT)
tested, and it was not consumed even after its maximal concentration Hydraulic retention time (HRT) can be defined as the average length
was reached. of time that matter (both substrate and biomass) remains in a reactor.
Consequently, it is a very important parameter in a full-scale perspec-
4.2.2. Temperature tive, since it establishes the flow rate daily treated into the reactor. It
Along with pH, temperature is a key parameter during acidogenic should be long enough to allow solubilisation of complex organic
fermentation, due its direct involvement both in microbial growth and matter, thus favouring subsequent acidogenic fermentation of hydro-
metabolism. Every microbial taxon has an optimum range of tempera- lysates. At the same time, a too high HRT reduces the quantity of
ture for its replication, consequently a change of working temperature substrate manageable per day and favour methanogens at opportune
can alter the microbial structure of the microbial consortium involved pH values (6.5–7.5). As many researches on acidogenic FWs fermen-
in acidogenic fermentation. tation were carried out in batch reactors, few information are available
He et al. (2012) found out that an increase of operating temperature in literature on the effect of HRT on overall VFAs production.
from mesophilic (35 °C) to thermophilic range (55 °C) brought to a Lim et al. (2008) investigated the effect of increasing HRT on
decrease of total VFAs production from a maximum concentration of acidogenesis, starting from FWs, in a semi-continuous fed fermenter,
17 g/L to 11 g/L, under acidic uncontrolled pH. A further increase of under mesophilic conditions, with a pH adjusted to 5.5, and a OLR of 5
temperature toward hyperthermophilic range (70 °C) caused a lower gTS/Ld. They tested three different HRT: 4, 8, and 12 days. Total VFAs
reduction of total VFAs production, which reached a maximum value of concentration increased with HRT, from 5.5 g/L, to 13 g/L, and finally
about 13 g/L. Komemoto et al. (2009) explored the effect of tempera- to 22 g/L. Furthermore, with the highest HRT a change in relative VFAs
ture on acidogenic fermentation in a greater range, ranging from psy- distribution was observed. In general, HAc represented the main fer-
chrophilic (15, 20 °C) to hyperthermophilic (65 °C). They found out that mentation product under shorter HRTs, while under a HRT of 12 days
at 55 and 65 °C the sCOD into reactor remarkably increased at the be- HPr was the predominant VFA. Considering the yields obtained at the
ginning of experimental time up to a value of around 40 g/L, and after it three HRTs investigated, it increased as result of HRT increment, but no
dropped quickly to 30 g/L. Instead, in mesophilic range sCOD reached a significant difference between those at HRTs of 8 and 12 days was
similar value (30 g/L), but it remained stable until the end of experi- observed (from about 0.34 to 0.39 respectively). Thus, a HRT of 8 days
mental trial. This difference is due to microbial hydrolysis activity; at resulted adequate to allow an appropriate level of substrate degrada-
higher temperature, solubilisation is the result of chemical-physics ef- tion, at least in the OLR investigated. Another interesting work was
fect, while in mesophilic range there is an active action of microbial conducted by Han and Shin (2002), who studied the effect of different
enzymes. With regard to VFAs production HAc was produced at the HRT rate on fermentation performance (1.00, 0.50, 0.33 and 0.25 d).
beginning of the experiment, reaching a value of 1 g/L and 2 g/L at 35 The authors carried out fermentation of FWs from a cafeteria in a 2 L
and 45 °C respectively, and thereafter its concentration dropped as leach bed reactor (LBR), under mesophilic conditions, and uncontrolled
consequence of biogas production. Instead, HBu was obtained at the pH. The effect of two different inocula were also studied: ruminal
middle and late experimental time, and it showed a high concentration bacteria from the stomach of a cow and an anaerobic consortium taken
(6.2 and 5.7 g/L at 35 and 45 °C respectively) regardless of biogas from another reactor fermenting FWs. A maximal VFAs concentration
production. Under psychrophilic conditions (T = 20 °C), VFAs produc- of 202 and 181 mmol/L was achieved with a HRT of 1 with rumen and
tion was extremely lower and this range of temperature can be con- anaerobic bacteria, respectively. On the contrary, the lower VFAs ac-
sidered unsuitable for any applications. Lee et al. (2008) demonstrated cumulation was achieved with a HRT of 0.25 d, obtaining a VFA

283
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

concentrations of 53 and 47 mmol/L, for rumen and anaerobic meso- −200/-300 mV, corresponding to a limited O2 presence and strictly O2
philic bacteria respectively. According to the authors, this result was absence respectively. In order to find out the optimal fermentation
due to the microbial washout, which took place at low HRT (0.25 d) conditions, two kinds of inocula were used: anaerobic granular sludge
resulting in a consequent total VFAs concentration decrease. Further- and aerobic activated sludge. FWs were fermented in batch reactor, at
more, the optimal HRT was 0.33 d for each inoculum, in terms of fer- mesophilic condition, pH adjusted to 6, for 21 days. A maximal VFAs
mentation efficiencies, expressed as ratio between theoretical VFA of concentration of 29.4 g/L was obtained on day 17, in the reactor in-
substrate and actual VFA produced: 71.2 and 59.8% at 0.33 d for rumen oculated with anaerobic sludge, under an ORP of −100/-200 mV. In-
microorganisms and mesophilic acidogens, respectively. This indicated stead, VFAs accumulation increased at a slow rate operating with an
that rumen bacteria had an enhancement effect on FWs fermentation ORP of −200/-300 mV, and the highest VFAs production (18.36 g/L)
capability. was achieved on day 17, stressing the negative effect of anaerobic
conditions acidogenic fermentation. Among the VFAs produced, in all
4.2.4. Organic loading rate (OLR) the reactors tested, 80% of VFAs consisted of HAc and HBu, and HBu
Organic loading rate (OLR) indicates the amount of FWs fed into the accounted for up to 60%. Moreover, other two VFAs (iso-HBu, HVa)
reactor per day and per unit of working volume. Lim et al. (2008) were apparently found under limited aeration, but they were less than
studied the effect of OLR on acidogenesis, in a semi-continuous reactor. 1.0 g/L in strictly anaerobic conditions. These results showed ORP le-
They observed that total VFAs production increased with increase of vels also would affect both total VFAs production and relative VFAs
OLR, achieving a maximum concentration of around 14.0, 24.0, and composition.
30 g/L, with an OLR of 5.0, 9.0 and 13.0 gTS/Ld, respectively. Although
the higher concentration at 13.0 g/Ld, the VFAs yield was lower respect 5. Volatile fatty acids concentration and purity after anaerobic
to smaller OLR values. It was due to high viscosity of medium that fermentation
negatively affects fermentation (Battista et al., 2018). The analysis of
the VFAs distribution showed that HAc was the most abundant product, As previously commented, VFAs are fundamental precursors of
and its concentration increased when the OLR increased, while the biopolymers, biofuels and chemicals such as esters, ketones, aldehydes,
other products, HPr, HBu and HVa, decreased. Moreover, by increasing alcohols and alkanes. The increasing of the VFAs concentration, which
OLR, the increasing of HCa, succinate and lactate was observed. Among is usually around 15–30 g/L after the anaerobic production process, and
the three different OLRs tested, an OLR of 9.0 gTS/Ld is more suitable their separation from the reaction medium represent the two crucial
for an appropriate VFAs production, relatively to the operational challenges in the optimization of the VFAs production process
parameters set up. A similar result was obtained from synthetic FWs, (Rebecchi et al., 2016). In the last years, many techniques for the VFAs
fermented in semi-continuous with the subsequent operational set up: recovery have been proposed. High efficiencies have been found by
pH 6, T 35 °C, and HRT 5 days. By an increment of OLR from 5.0 to 11.0 liquid-liquid extraction, which is the oldest and well established tech-
gTS/Ld, total VFAs concentration increased from about 13.0 to 21.0 g/ nique. It adopts some anionic solvents (alcohols, ketones, ethers and
L. In both cases, a homogeneous VFAs distribution was achieved, but at aliphatic hydrocarbons) to separate VFAs. The separation yields depend
lower OLR HBu was the main product, while HAc was more abundant on the nature of the acid extracted, the concentration of the extracting
with the higher OLR. Anyway, the VFA yield (g VFA/g TS) was 13% agent, the type of diluent and pH (Alkaya et al., 2009). The main dis-
better when an OLR of 5.0 gTS/Ld was adopted. Lately, it is important advantages of liquid-liquid extraction are the economic costs of the
to remark that further increase of OLR from 11.0 to 16.0 gTS/Ld caused solvents and its environmental impact. Membranes separation con-
instability of rector, and a stable production rate was achieved de- stitutes a valid alternative. In particular, two different approaches are
creasing the OLR (Jiang et al., 2013). possible: the electro-dialysis, where the application of a voltage dif-
Gou et al. (2014) investigated on a possible relation between OLR ference created by two electrodes, promotes the VFAs passage through
and working temperature from a co-fermentation of WAS (Waste Ac- the membrane; and nanofiltration based on size and pressure selection.
tivated Sludge) and FWs (mixed in a 2:1 ratio in terms of TS) con- The high costs of the membranes, their fouling and the inability to
ducting in a semi-continuous mode. An increasing OLR was tested, from reach high VFAs concentration, limit these separation methods
1.0 up to 8.0 gTVS/Ld, in three identical CSTR, operating at 35, 45, and (Singhania et al., 2013). Another way for the VFA recovery seems to be
55 °C. They found out that a stable total VFAs is achieved with higher the VFA adsorption and ion exchange. The first is based on the physical
temperature (4 g/L), while under mesophilic conditions OLR must be interaction between the VFAs’ carboxylate groups and the active sites of
kept under 5 gTVS/Ld to obtain a stable total VFAs production of a solid matrix. On the contrary, in the ion exchange process ionic bonds
around 3.5 g/L. between the ionized acid and a cation, such as an ammonium salt
Therefore, the choice of an optimal OLR is fundamental for reaching contained in an anionic resin, occurs (López-Garzón and Straathof,
a stable VFAs production. To ensure a good yield in terms of VFAs, OLR 2014). In particular, Rebecchi et al. (2016) found interesting results on
should be abundant enough to provide an adequate amount of carbon- ion exchange technique which is strongly influenced by the interaction
source to fermentative metabolism, according to working temperature. of multiple variables, such as the resin typologies, the nature of the
In fact, under mesophilic conditions an OLR higher than 5 gTS/Ld may reaction medium and the pH. An optimal pH of about 3.0–4.5 has been
cause the apparent viscosity increasing of the medium, reducing the considered ideal for acetic acid separation, while higher pH (6.5) for
mass and heat transfers and consequentially the substrates conversion VFAs separation from real digestate. Anyway, few works have been
into VFAs. This limit can be overcome applying a thermo or hy- conducted for the optimization of VFAs recovery and all the above
perthermophilic conditions. presented technologies are still poorly tested on real reaction medium
from acidogenic effluents.
4.2.5. Effect of low oxygen concentration on anaerobic fermentation
It was demonstrated that low oxygen concentration leads to a high 6. Biotechnological conversion of VFAs into bioproducts
VFAs yield, due to the favourable effect on facultative acidogens, and
on production of high amounts of extracellular enzymes (Lim et al., The increasing interest in VFAs production is easily explainable as
2014; Liu et al., 2013). they are the main precursors of interesting added value compounds,
Yin et al. (2016b) tested the effect of strictly anaerobic conditions such as polyhydroxyalkanoates (PHAs), different biofuels and their use
and moderate aerated conditions on VFAs production from fermenta- as carbon source in nitrogen and phosphorus removal from wastewaters
tion of thermal pretreated synthetic FWs. They set up oxidation re- (Elbeshbishy et al., 2017). Some of these applications are discussed
duction potential (ORP) in two different range: −100/-200 mV and below.

284
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

6.1. Polyhydroxyalkanoates (PHAs) biodegradable organic compounds, which cause a quickly acidification
into digester, as result of rapid VFAs production. Generally, to avoid
Among the wide application of VFAs obtaining from acidogenic inhibition of methanogenic archaebacteria, due to VFAs accumulation,
fermentation, polyhydroxyalkanoates (PHAs) production is one of the two-stage digestion systems are employed. In this way, VFAs are par-
main attractive. In fact, PHAs are biodegradable plastic-like materials tially neutralized by the reaction medium recirculation, rich in am-
that can be used to replace petroleum-derived plastics. They are natural monia and other buffer agents, from second stage AD (Bolzonella et al.,
thermoplastic polyesters made of 3-, 4-, 5-, and 6-hydroxyalknoic acids, 2018). In addition, AD allows assuring optimal growth conditions for
which are synthesized by bacteria as granular intracellular inclusions the two microorganisms groups: fast-growing fermenting microorgan-
for energy storage (Arcos-Hernández et al., 2015). These inclusions isms (hydrolytic and acidogenic bacteria) are cultured under optimal
have a core made of polyester, surrounded by both phospholipids and condition for the VFAs production in the first stage. While, in the
proteins, and their size ranging from 0.2 to 0.5 μm (De Grazia et al., second stage (methanogenic phase), slow-growing acetogens and me-
2017). More than 90 genera of both Gram-positive and Gram-negative thanogens are fed using the VFAs obtained at the first stage. Although,
bacteria were classified as PHAs producers, under aerobic and anae- microorganisms involved in AD process are able to convert VFAs into
robic conditions. These microorganisms are able to use different carbon HAc, allowing their utilization as carbon source for downstream me-
sources for PHA production. VFAs are between the preferred ones, since thane production, some of these carboxylic acids are less available for
they are direct metabolic precursors of PHAs (Anjum et al., 2016; this bioconversion. For example, it was found out that butyric acid
Girotto et al., 2015). Worldwide research interest therefore is focusing accumulates into digester during AD of FWs, demonstrating the difficult
on reducing the cost of production using different waste materials as for microorganism to convert it toward biogas (Braguglia et al., 2018).
carbon source (Raza et al., 2018). The use of residual organic waste as
carbon substrates for accumulating-PHAs bacteria is a viable alternative 6.2.2. Biohydrogen
to reduce cost production. In this scenario, FWs can represent a good Another advantageous gas, the hydrogen (H2) can be obtained by
candidate for providing an inexpensive carbon source, due to its wide AD, during the acidogenic fermentation step. Because of its efficiency
continuous availability and high biodegradability (Nielsen et al., 2017). characteristics, H2 is one of the most desirable form of renewable en-
Another way to reduce production cost is the use of MMCs, namely ergy, and it is considered to be the future fuel (Hay et al., 2013; Dutta,
selected PHAs-storing cultures and PHAs accumulation obtained using 2014; Ghimire et al., 2015; Kim and Kim, 2013). Generally, hydrogen is
VFAs from acidogenenic fermentation of organic wastes as carbon produced from non-renewable hydrocarbons through steam reforming,
source. In fact, the use of pure cultures is intrinsically more expensive, which is the cause of high greenhouse-gas emission. In the last years,
because of major sterility demands, and higher requirements for many researches were carried out to develop biotechnological routes to
equipment and control devices in comparison to MMCs approaches produce H2 in an environmental friendly manner from renewable
(Koller et al., 2017), (Morgan-Sagastume et al., 2017). In this case, sources, such as residual biomass, by dark fermentation (DF) and
PHAs are produced by mean of a three-stages process, in which an in- photochemical fermentation (PF) (Elbeshbishy et al., 2017). DF in-
itial acidogenic fermentation phase is followed by selection, through a dicates the acidogenic step of AD which has hydrogen as main product
feast and famine cycle, and production of PHA-storing bacterial bio- and organic acids, alcohols and CO2 as byproducts (Yasin et al., 2013).
mass through a dynamic feeding scheme, and a third phase in which PF is a biochemical process, generally carried out by purple non-sul-
PHAs was accumulated in batch conditions. (Reis et al., 2011). phur bacteria (PNSB), which allow to produce H2 using the light-driven
It is important to remark that the distribution between VFAs pro- energy (Ghosh et al., 2017). Photofermentative bacteria, like PNSB,
duced during acidogenic fermentation of FWs strongly affects the kind conduct an anoxygenic photosynthesis, during which ATP is produced
of PHAs produced downstream. Reddy and Mohan (2012) carried out using light driven energy. In presence of molecular N2, PNSB use ATP
an experiment aimed at PHAs production, using FWs as starting ma- molecules photosynthetically produced to catalyse the N2-fixation,
terial. They used both fresh unfermented FWs and residual fermented namely the reduction of N2 into NH3 through the action of nitrogenase
FWs originating from a reactor set up for H2 production through dark enzyme. Thus, nitrogenase represents the biocatalyst that mediates the
fermentation. PHAs production was carried out both in anoxic and production of H2 in PNSBs. Yet, it requires reducing equivalents to
aerobic microenvironment. The experiment lasted at least for 72 h. In produce H2 from N2-fixation, which are generated through the catabolic
both case, the maximal PHAs production of 0.3 g PHA/gVFAs was breakdown of carbon compounds assimilated during photo-hetero-
achieved after 24 h, and total PHAs production ranging from around 30 trophic growth (Hay et al., 2013).
to 40% of cell dried weight. Furthermore, authors found that a high Metabolism of PNSBs exhibits a great metabolic versatility, that
amount of acetic acid and butyric acid are produced during fermenta- make them capable to use a wide range of carbon compounds for their
tion of FWs, and propionic acid and valeric acid are obtained in low metabolism, such as simple sugar, alcohols, glycerol, aromatic com-
amount, a higher PHB and a lower PHV composition of total PHAs pounds, and VFAs. (Ghosh et al., 2017). Rhodopseudomonas spp. were
produced is achieved. able to utilize different fatty acids, while Rhodobacter spp. show more
In conclusion, it is clear that a driving of VFAs production during difficult to use the heavier VFAs HVa or HCa, which have an inhibitory
acidogenic fermentation is a key aspect to take into account for the effect on Rhodobacter spp. growth (Okubo et al., 2005). Tao et al.
production of PHAs with optimal characteristics for their industrial (2008) found out that a specific isolated PNSB, ZX-5, is able to produce
application. H2 growing on VFAs, with a conversion efficiency of 69, 71.5, and
61.9% growing on HAc, HBu, and HPr, respectively. Furthermore,
6.2. Bioenergy conversation efficiency showed a remarkably decrease using HVa,
reaching only a value of 12%, while HCa resulted to be inhibitory on
6.2.1. Biogas cells growth, thus on H2 production. Surprisingly, conversion efficiency
Biogas is the final product of AD, a complex biological process was lower or zero using simple sugars. This demonstrate that VFAs are
which leads to a gaseous blend rich in methane and carbon dioxide. As extremely suitable for H2, using specific strain of PNSB bacteria.
result of the complex sequence of steps involved in the process, AD
shows a fragile equilibrium, which involves VFAs concentration within 6.3. Biological nutrient removal in wastewater treatment
the reactor. When VFAs concentration exceeds the buffering capacity of
system, they can cause inhibition of downstream methanogenesis pro- Another common application of VFAs is biological nutrient (ni-
cess (Braguglia et al., 2018). This is a common problem using FWs as trogen and phosphorous) removal from wastewaters. To implement a
feedstock for biogas production, due to its intrinsic abundance of complete denitrification step the addition of carbon sources is required,

285
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

considering the low carbon concentration in wastewaters. Usually me- production with a production yield of 0.3 gPHA/gVFA after 24 h and
thanol, ethanol and acetate are used to increase denitrification effi- the total PHAs production between 30 and 40% of cell dried weight.
ciency (Bolzonella et al., 2001). However, to reduce the overall cost of VFAs constitute also the main intermediates for bioenergy production,
biological nitrogen removal, VFAs obtained from acidogenic fermen- in particular biohydrogen from both DF during acidogenic phase of AD
tation of residual biomass can represent a suitable organic carbon and PF exploiting solar light.
source for the process (Kim et al., 2017; Zhang et al., 2016). Further-
more, it was proven that VFAs as denitrification carbon source can Acknowledgment
improve denitrification rate, and reduce nitrite accumulation in deni-
trification process, with respect to other chemicals such as methanol This study was financially supported in the framework of the
and ethanol (Liu et al., 2016; Kim et al., 2016). Elefsiniotis and European Horizon 2020 programme through the H2020-IND-CE-2016-
Wareham (2007) investigated the VFAs employment pattern during 17 RES URBIS project (id 730349).
denitrification process, using both synthetic and biologically produced
VFAs, obtained from a lab-scale fermenter. Nitrates completely dis- References
appeared and the average specific denitrification rate for natural and
synthetic VFAs was extremely similar (around 0.01 g NO3-N/g TVS d). Aguilar-Reynosa, A., Romaní, A., Rodríguez-Jasso, R.M., Aguilar, C.N., Garrote, G., Ruiz,
With regard to the different VFAs, using HAc denitrification rate was H.A., 2017. Microwave heating processing as alternative of pretreatment in second-
generation biorefinery: an overview. Energy Convers. Manag. 36, 50–65.
more than double that of propionate reflecting a preference for HAc. Alibardi, L., Cossu, R., 2015. Composition variability of the organic fraction of municipal
These results can be understood considering that denitrifying popula- solid waste and effects on hydrogen and methane production potentials. Waste
tions use preferentially acetic acid as carbon source, which was ex- Manag. 36, 147–155.
Alibardi, L., Cossu, R., 2016. Effects of carbohydrate, protein and lipid content of organic
hausted before the other VFAs, even though it occurred in the largest waste on hydrogen production and fermentation products. Waste Manag. 47, 69–77.
quantity. Only when HAc concentrations began to decline, micro- Alkaya, E., Kaptan, S., Ozkan, L., Uludag-Demirer, S., Demirer, G.N., 2009. Recovery of
organisms seem to use the other two most abundant VFAs, butyric acid acids from anaerobic acidification broth by liquid–liquid extraction. Chemosphere
77, 1137–1142.
and propionic acid. The final choice was represented by valeric acid, Anjum, A., Zuber, M., Zia, K.M., Noreen, A., Anjum, M.N., Tabasum, S., 2016. Microbial
confirming that denitrifying bacteria have a sequential preference for production of polyhydroxyalkanoates (PHAs) and its copolymers: a review of recent
VFAs. (Elefsiniotis et al., 2004). Concerning phosphorus, it can be re- advancements. Int. J. Biol. Macromol. 89, 161–174.
Arcos-Hernández, M., Montaño-Herrera, L., Murugan Janarthanan, O., Quadri, L.,
moved through a process known as enhanced biological phosphorus
Anterrieu, S., Hjort, M., Alexandersson, T., Karlsson, A., Karabegovic, L., Magnusson,
removal (EBPR), which requires an alternation of anaerobic and aerobic P., Johansson, P., Bengtsson, S., Morgan-Sagastume, F., de Vegt, O., Laycock, B.,
conditions. This process is widely used in wastewater treatment plants Pratt, S., Halley, P., Lant, P., Werker, A., 2015. Value-added bioplastics from services
and allows to remove 80–90% of soluble P from the effluent (Mehta of wastewater treatment. Water Sci. Technol. 10, 546–555.
Batstone, D.J., Keller, J., Angelidaki, I., Kalyuzhnyi, S.V., Pavlostathis, S.G., Rozzi, A.,
et al., 2015). VFAs can be used efficiently as carbon source in EBPR Sanders, W.T., Siegrist, H., Vavilin, V.A., 2002. The IWA anaerobic digestion model
process, since organic substrate are necessary during anaerobic treat- no 1 (ADM1). Water Sci. Technol. 45, 65–73.
ment, and then the resulting mixed liquor is alternated between anae- Battista, F., Bolzonella, D., 2018. Some critical aspects of the enzymatic hydrolysis at high
dry-matter content: a review. Biofuel. Bioprod. Bior 12, 711–723.
robic and aerobic conditions. EBPR requires a carbon/P ratio of 5 or Battista, F., Gomez Almendros, M., Rousset, R., Boivineau, S., Bouillon, P.A., 2018.
higher (Rashed and Massoud, 2015). It was demonstrated that VFAs Enzymatic hydrolysis at high dry matter content: the influence of the substrates'
obtained from acidogenic fermentation are more effective in promotion physical properties and of loading strategies on mixing and energetic consumption.
Bioresour. Technol. 250, 191–196.
of P removal with respect to synthetic acetic acid. Tong and Chen Battista, F., Gomez Almendros, M., Rousset, R., Bouillon, P.A., 2019. Enzymatic hydro-
(2007) observed that the phosphorus removal efficiency was around lysis at high lignocellulosic content: optimization of the mixing system geometry and
98% with the fermentative, and only about 71% using pure acetic acid. of a fed-batch strategy to increase glucose concentration. Renew. Energy 131,
152–158.
Furthermore, different VFAs affect in a different way the overall EBPR Baumann, I., Westermann, P., 2016. Microbial production of short chain fatty acids from
process. However, authors obtained different EBPR performances using lignocellulosic biomass: current processes and market. BioMed Res. Int. https://dx.
different VFAs, achieving unambiguous conclusion about the optimal doi.org/10.1155/2016/8469357.
Bhatia, S.K., Yang, Y.H., 2017. Microbial production of volatile fatty acids: current status
acid among VFAs for EBPR process. Hood and Randall (2001) found
and future perspectives. Rev. Environ. Sci. Biotechnol. 16, 327–345.
that HAc is the best substrate among other VFAs such as Hbu, HVa, and Bo, Z., Pin-Jing, H., 2014. Performance assessment of two-stage anaerobic digestion of
HPr in terms of yielding the highest phosphorus removal per unit COD kitchen wastes. Environ. Technol. 35, 1277–1285.
of VFA. According to Oehmen et al. (2005), acetate is commonly used Bolzonella, D., Battista, F., Cavinato, C., Gottardo, M., Micolucci, F., Lyberatos, G., Pavan,
P., 2018. Recent developments in biohythane production from household food
as carbon source for EBPR process because it is typically the main acid wastes: a review. Bioresour. Technol. 257, 311–319.
among VFAs found in the influent to wastewater treatment plants. Bolzonella, D., Fatone, F., Pavan, P., Cecchi, F., 2005. Anaerobic fermentation of organic
municipal solid wastes for the production of soluble organic compounds. Ind. Eng.
Chem. 44, 3412–3418.
7. Conclusions Bolzonella, D., Innocenti, L., Pavan, P., Cecchi, F., 2001. Denitrification potential en-
hancement by addition of anaerobic fermentation products from the organic fraction
This review focused the attention on the VFAs production from FWs, of municipal solid waste. Water Sci. Technol. 44, 187–194.
Braguglia, C.M., Gallipoli, A., Gianico, A., Pagliaccia, P., 2018. Anaerobic bioconversion
one of the most abundant organic substrates annually produced around of food waste into energy: a critical review. Bioresour. Technol. 248, 37–56.
the world. Browne, J.D., Murphy, J.D., 2014. The impact of increasing organic loading in two phase
It has been evidenced that FWs present high organic matter content digestion of food waste. Renew. Energy 71, 69–76.
Dahiya, S., Kumar, A.N., Sravan, J.S., Chatterjee, S., Sarkar, O., Mohan, S.V., 2018. Food
(100–150 g COD/L) and nitrogen and phosphorous concentrations
waste biorefinery: sustainable strategy for circular bioeconomy. Bioresour. Technol.
(2–15 g/L and 0.5–1.0 g/L, respectively) which make them adequate for 248, 2–12.
VFAs. The analysis of the most promising pretreatments indicates that Dahiya, S., Sarkar, O., Swamy, Y.V., Mohan, S.V., 2015. Acidogenic fermentation of food
waste for volatile fatty acid production with co-generation of biohydrogen. Bioresour.
0.5–3% HCl and H2SO4 additions and thermal pretreatment in the
Technol. 182, 103–113.
range 140–170 °C increase the organic matter's solubilisation. The best Dai, K., Wen, J.L., Zhang, F., Zeng, R.J., 2017. Valuable biochemical production in mixed
operative conditions which emerged from the different research works culture fermentation: fundamentals and process coupling. Appl. Microbiol.
on VFAs production are a slightly controlled pH (6.0–7.0), short HRT Biotechnol. 101, 6575–6586.
De Grazia, G., Quadri, L., Majone, M., Morgan-Sagastume, F., Werker, A., 2017. Influence
(1–7 days depending on the substrates complexity), thermophilic tem- of temperature on mixed microbial culture polyhydroxyalkanoate production while
peratures and OLR of about 10 gTS/Ld. All these conditions are de- treating a starch industry wastewater. J. Environ. Eng. 5, 5067–5075.
pending on each other, allowing an increasing of the VFAs concentra- Domingos, J.M.B., Martinez, G.A., Scoma, A., Fraraccio, S., Kerckhof, F.M., Boon, N., Reis,
M.A.M., Fava, F., Bertin, L., 2017. Effect of operational parameters in the continuous
tion of about 10–25%, according to their combinations. anaerobic fermentation of cheese whey on titers, yields, productivities, and microbial
Between the VFA applications, the most interesting is PHAs

286
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

community structures. ACS Sustain. Chem. Eng. 5, 1400–1407. biomethanation of kitchen waste by different pre-treatments. Bioresour. Technol.
Dutta, S., 2014. A review on production, storage of hydrogen and its utilization as an 102, 592–599.
energy resource. J. Ind. Eng. Chem. 20, 1148–1156. Mehta, C.M., Khunjar, W.O., Nguyen, V., Tait, S., Batstone, D.J., 2015. Technologies to
Elbeshbishy, E., Dhar, B.R., Nakhla, G., Lee, H.S., 2017. A critical review on inhibition of recover nutrients from waste streams: a critical review. Crit. Rev. Environ. Sci.
dark biohydrogen fermentation. Renew. Sustain. Energy Rev. 79, 656–668. Technol. 45, 385–427.
Elefsiniotis, P., Wareham, D.G., 2007. Utilization patterns of volatile fatty acids in the Mohan, S.V., Butti, S.K., Amulya, K., Dahiya, S., Modestra, J.A., 2016. Waste biorefinery:
denitrification reaction. Enzym. Microb. Technol. 41, 92–97. a new paradigm for a sustainable bioelectro economy. Trends Biotechnol. 34,
Elefsiniotis, P., Wareham, D.G., Smith, M.O., 2004. Use of volatile fatty acids from an 852–855.
acid-phase digester for denitrification. J. Biotechnol. 114, 289–297. Morgan Sagastume, F., Valentino, F., Hjort, M., Zanaroli, G., Majone, M., Werker, A.,
Feng, L., Chen, Y., Zheng, X., 2009. Enhancement of waste activated sludge protein 2017. Acclimation process for enhancing polyhydroxyalkanoate accumulation in
conversion and volatile fatty acids accumulation during waste activated sludge activated-sludge biomass. Waste Biomass Valoriz. https://doi.org/10.1007/s12649-
anaerobic fermentation by carbohydrate substrate addition: the effect of pH. Environ. 017-0122-8.
Sci. Technol. 43, 4373–4380. Nagao, N., Tajima, N., Kawai, M., Niwa, C., Kurosawa, N., Matsuyama, T., Yusoff, F.M.,
Forster-Carneiro, T., Pérez, M., Romero, L.I., 2008. Influence of total solid and inoculum Toda, T., 2012. Maximum organic loading rate for the single-stage wet anaerobic
contents on performance of anaerobic reactors treating food waste. Bioresour. digestion of food waste. Bioresour. Technol. 118, 210–218.
Technol. 99, 6994–7002. Nghiem, L.D., Koch, K., Bolzonella, D., Drewes, J.E., 2017. Full scale co-digestion of
Garcia, N.H., Strazzera, G., Frison, N., Bolzonella, D., 2018. Volatile fatty acids from wastewater sludge and food waste: bottlenecks and possibilities. Renew. Sustain.
household food wastes. Chem. Eng. Trans. 64, 103–108. Energy Rev. 72, 354–362.
Ghimire, A., Valentino, S., Frunzo, L., Trably, E., Escudié, R., Pirozzi, F., Lens, P.N.L., Nielsen, C., Rahman, A., Rehman, A.U., Walsh, M.K., Miller, C.D., 2017. Food waste
Esposito, G., 2015. Biohydrogen production from food waste by coupling semi-con- conversion to microbial polyhydroxyalkanoates. Microb. Biotechnol 10, 1338–1352.
tinuous dark-photofermentation and residue post-treatment to anaerobic digestion: a Oehmen, A., Zeng, R.J., Yuan, Z., Keller, J., 2005. Anaerobic metabolism of propionate by
synergy for energy recovery. Int. J. Hydrogen Energy 40, 16045–16055. polyphosphate-accumulating organisms in enhanced biological phosphorus removal
Ghosh, S., Dairkee, U.K., Chowdhury, R., Bhattacharya, P., 2017. Hydrogen from food systems. Biotechnol. Bioeng. 91, 43–53.
processing wastes via photofermentation using Purple Non-sulfur Bacteria (PNSB)–A Okubo, Y., Futamata, H., Hiraishi, A., 2005. Distribution and capacity for utilization of
review. Energy Convers. Manag. 141, 299–314. lower fatty acids of phototrophic purple nonsulfur bacteria in wastewater environ-
Girotto, F., Alibardi, L., Cossu, R., 2015. Food waste generation and industrial uses: a ments. Microb. Environ. 20, 135–143.
review. Waste Manag. 45, 32–41. Parawira, W., Murto, M., Read, J.S., Mattiasson, B., 2004. Volatile fatty acid production
Gou, C., Yang, Z., Huang, J., Wang, H., Xu, H., Wang, L., 2014. Effects of temperature and during anaerobic mesophilic digestion of solid potato waste. J. Chem. Technol.
organic loading rate on the performance and microbial community of anaerobic co- Biotechnol. 79, 673–677.
digestion of waste activated sludge and food waste. Chemosphere 105, 146–151. Passos, F., Ortega, V., Donoso-Bravo, A., 2017. Thermochemical pretreatment and
Han, S.K., Shin, H.S., 2002. Enhanced acidogenic fermentation of food waste in a con- anaerobic digestion of dairy cow manure: experimental and economic evaluation.
tinuous-flow reactor. Waste Manag. Res. 20, 110–118. Bioresour. Technol. 227, 239–246.
Hay, J.X.W., Wu, T.Y., Juan, J.C., Jahim, J.M., 2013. Biohydrogen production through Paudel, S.R., Banjara, S.P., Choi, O.K., Park, K.Y., Kim, Y.M., Lee, J.W., 2017.
photo fermentation or dark fermentation using waste as a substrate: overview, eco- Pretreatment of agricultural biomass for anaerobic digestion: current state and
nomics, and future prospects of hydrogen usage. Biofuel. Bioprod. Bior 7, 334–352. challenges. Bioresour. Technol. 245, 1194–1205.
He, M., Sun, Y., Zou, D., Yuan, H., Zhu, B., Li, X., Pang, Y., 2012. Influence of temperature Raganati, F., Procentese, A., Olivieri, G., Russo, M.E., Salatino, P., Marzocchella, A., 2014.
on hydrolysis acidification of food waste. Procedia Environ. Sci. 16, 85–94. MFA of Clostridium acetobutylicum pathway: the role of glucose and xylose on the
Hood, C.R., Randall, A.A., 2001. A biochemical hypothesis explaining the response of acid formation/uptake. Chem. Eng. Trans. 38, 337–342.
enhanced biological phosphorus removal biomass to organic substrates. Water Res. Rashed, E.M., Massoud, M., 2015. The effect of effective microorganisms (EM) on EBPR in
35, 2758–2766. modified contact stabilization system. HBRC J. 11, 384–392.
Izumi, K., Okishio, Y.K., Nagao, N., Niwa, C., Yamamoto, S., Toda, T., 2010. Effects of Raza, Z.A., Abid, S., Banat, I.M., 2018. Polyhydroxyalkanoates: characteristics, produc-
particle size on anaerobic digestion of food waste. Int. Biodeterior. Biodegrad. 64, tion, recent developments and applications. Int. Biodeterior. Biodegrad. 126, 45–56.
601–608. Rebecchi, S., Pinelli, D., Bertin, L., Zama, F., Fava, F., Frascari, D., 2016. Volatile fatty
Jankowska, E., Chwiałkowska, J., Stodolny, M., Oleskowicz-Popiel, P., 2015. Effect of pH acids recovery from the effluent of an acidogenic digestion process fed with grape
and retention time on volatile fatty acids production during mixed culture fermen- pomace by adsorption on ion exchange resins. Chem. Eng. J. 306, 629–639.
tation. Bioresour. Technol. 190, 274–280. Reddy, M.V., Mohan, S.V., 2012. Influence of aerobic and anoxic microenvironments on
Jiang, J., Zhang, Y., Li, K., Wang, Q., Gong, C., Li, M., 2013. Volatile fatty acids pro- polyhydroxyalkanoates (PHA) production from food waste and acidogenic effluents
duction from food waste: effects of pH, temperature, and organic loading rate. using aerobic consortia. Bioresour. Technol. 103, 313–321.
Bioresour. Technol. 143, 525–530. Reis, M., Albuquerque, M., Villano, M., Majone, M., 2011. Mixed culture processes for
Kim, D.H., Kim, M.S., 2013. Development of a novel three-stage fermentation system polyhydroxyalkanoate production from agro-industrial surplus/wastes as feedstocks.
converting food waste to hydrogen and methane. Bioresour. Technol. 127, 267–274. In: Fava, F., Agathos, S., Young, M. (Eds.), Comprehensive Biotechnology.
Kim, E., Shin, S.G., Jannat, M.A.H., Tongco, J.V., Hwang, S., 2017. Use of food waste- Amsterdam, pp. 669–683.
recycling wastewater as an alternative carbon source for denitrification process: a Ren, Y., Yu, M., Wu, C., Wang, Q., Gao, M., Huang, Q., Liu, Y., 2018. A comprehensive
full-scale study. Bioresour. Technol. 245, 1016–1021. review on food waste anaerobic digestion: research updates and tendencies.
Kim, H., Kim, J., Shin, S.G., Hwang, S., Lee, C., 2016. Continuous fermentation of food Bioresour. Technol. 247, 1069–1076.
waste leachate for the production of volatile fatty acids and potential as a deni- Riemenschneider, W., 2000. Carboxylic acids, aliphatic. Ullmann's Encycl. Ind. Chem.
trification carbon source. Bioresour. Technol. 207, 440–445. https://doi.org/10.1002/14356007.a05_235.
Koller, M., Maršálek, L., De Sousa Dias, M.M., Braunegg, G., 2017. Producing microbial Shahriari, H., Warith, M., Hamoda, M., Kennedy, K., 2013. Evaluation of single vs. staged
polyhydroxyalkanoate (PHA) biopolyesters in a sustainable manner. N. Biotech. 37, mesophilic anaerobic digestion of kitchen waste with and without microwave pre-
24–38. treatment. J. Environ. Manag. 125, 74–84.
Komemoto, K., Lim, Y.G., Nagao, N., Onoue, Y., Niwa, C., Toda, T., 2009. Effect of Shen, D., Yin, J., Yu, X., Wang, M., Long, Y., Shentu, J., Chen, T., 2017. Acidogenic
temperature on VFA's and biogas production in anaerobic solubilization of food fermentation characteristics of different types of protein-rich substrates in food waste
waste. Waste Manag. 29, 2950–2955. to produce volatile fatty acids. Bioresour. Technol. 227, 125–132.
Lee, M., Hidaka, T., Tsuno, H., 2008. Effect of temperature on performance and microbial Shen, F., Yuan, H., Pang, Y., Chen, S., Zhu, B., Zou, D., Liu, Y., Ma, J., Yu, L., Li, X., 2013.
diversity in hyperthermophilic digester system fed with kitchen garbage. Bioresour. Performances of anaerobic co-digestion of fruit & vegetable waste (FVW) and food
Technol. 99, 6852–6860. waste (FW): single-phase vs. two-phase. Bioresour. Technol. 144, 80–85.
Li, Y., Jin, Y., Li, J., Li, H., Yu, Z., Nie, Y., 2017. Effects of thermal pretreatment on Shin, H.S., Youn, Y.H., Kim, S.H., 2004. Hydrogen production from food waste in anae-
degradation kinetics of organics during kitchen waste anaerobic digestion. Energy robic mesophilic and thermophilic acidogenesis. Int. J. Hydrogen Energy 29,
118, 377–386. 1355–1363.
Lim, J.W., Chiam, J.A., Wang, J.Y., 2014. Microbial community structure reveals how Singhania, R.R., Patel, A.K., Christophe, G., Fontanille, P., Larroche, C., 2013. Biological
microaeration improves fermentation during anaerobic co-digestion of brown water upgrading of volatile fatty acids, key intermediates for the valorization of biowaste
and food waste. Bioresour. Technol. 171, 132–138. through dark anaerobic fermentation. Bioresour. Technol. 145, 166–174.
Lim, S.J., Kim, B.J., Jeong, C.M., Choi, J.D., Ahn, Y.H., Chang, H.N., 2008. Anaerobic Sosulski, F.W., Imafidon, G.I., 1990. Amino acid composition and nitrogen-to-protein
organic acid production of food waste in once-a-day feeding and drawing-off bior- conversion factors for animal and plant foods. J. Agric. Food Chem. 38, 1351–1356.
eactor. Bioresour. Technol. 99, 7866–7874. Stenmarck, A., Jensen, C., Quested, T., Moates, G., Buksti, M., Cseh, B., Juul, S., Parry, A.,
Liu, C.G., Xue, C., Lin, Y.H., Bai, F.W., 2013. Redox potential control and applications in Politano, A., Redlingshofer, B., Scherhaufer, S., 2016. Estimates of European Food
microaerobic and anaerobic fermentations. Biotechnol. Adv. 31, 257–265. Waste Levels. Reducing Food Waste through Social Innovation. Fusion, Stockholm.
Liu, F., Tian, Y., Ding, Y., Li, Z., 2016. The use of fermentation liquid of wastewater https://www.ivl.se/webdav/files/Rapporter/C186.pdf, Accessed date: April 2018.
primary sedimentation sludge as supplemental carbon source for denitrification Tao, Y., He, Y., Wu, Y., Liu, F., Li, X., Zong, W., Zhou, Z., 2008. Characteristics of a new
based on enhanced anaerobic fermentation. Bioresour. Technol. 219, 6–13. photosynthetic bacterial strain for hydrogen production and its application in was-
López-Garzón, C.S., Straathof, A.J.J., 2014. Recovery of carboxylic acids produced by tewater treatment. Int. J. Hydrogen Energy 33, 963–973.
fermentation. Biotechnol. Adv. 32, 873–904. Tong, J., Chen, Y., 2007. Enhanced biological phosphorus removal driven by short-chain
Lucifero, N., 2016. Food Loss and Waste in the EU Law between sustainability of well- fatty acids produced from waste activated sludge alkaline fermentation. Environ. Sci.
being and the implications on food system and on environment. Agric. Agric. Sci. Technol. 41, 7126–7130.
Proc. 8, 282–289. Traverso, P., Pavan, P., Bolzonella, D., Innocenti, L., Cecchi, F., Mata-Alvarez, J., 2000.
Ma, J., Duong, T.H., Smits, M., Verstraete, W., Carballa, M., 2011. Enhanced Acidogenic fermentation of source separated mixtures of vegetables and fruits wastes

287
G. Strazzera et al. Journal of Environmental Management 226 (2018) 278–288

from supermarkets. Biodegradation 11, 407–414. volatile fatty acid accumulation in thermophilic anaerobic digestion of food waste.
Vavouraki, A.I., Angelis, E.M., Kornaros, M., 2013. Optimization of thermo-chemical Waste Biomass Valoriz 6, 1–12.
hydrolysis of kitchen wastes. Waste Manag. 33, 740–745. Young, V.R., Pellett, P.L., 1994. Plant proteins in relation to human protein and amino
Veluchamy, C., Raju, V.W., Kalamdhad, A.S., 2018. Electrohydrolysis pretreatment for acid nutrition. Am. J. Clin. Nutr. 59, 1203–1212.
enhanced methane production from lignocellulose waste pulp and paper mill sludge Yu, X., Yin, J., Wang, K., Shen, D., Long, Y., Chen, T., 2016. Enhancing food waste hy-
and its kinetics. Bioresour. Technol. 252, 52–58. drolysis and the production rate of volatile fatty acids by prefermentation and hy-
Wang, K., Yin, J., Shen, D., Li, N., 2014. Anaerobic digestion of food waste for volatile drothermal pretreatments. Energy Fuels 30, 4002–4008.
fatty acids (VFAs) production with different types of inoculum: effect of pH. Zhang, B., Zhang, L.L., Zhang, S.C., Shi, H.Z., Cai, W.M., 2005. The influence of pH on
Bioresour. Technol. 161, 395–401. hydrolysis and acidogenesis of kitchen wastes in two-phase anaerobic digestion.
Yasin, N.H.M., Mumtaz, T., Hassan, M.A., 2013. Food waste and food processing waste for Environ. Technol. 26, 329–340.
biohydrogen production: a review. J. Environ. Manag. 130, 375–385. Zhang, Y., Wang, X.C., Cheng, Z., Li, Y., Tang, J., 2016. Effect of fermentation liquid from
Yin, J., Yu, X., Wang, K., Shen, D., 2016a. Acidogenic fermentation of the main substrates food waste as a carbon source for enhancing denitrification in wastewater treatment.
of food waste to produce volatile fatty acids. Int. J. Hydrogen Energy 41, Chemosphere 144, 689–696.
21713–21720. Zhen, G., Lu, X., Kobayashi, T., Kumar, G., Xu, K., 2016. Anaerobic co-digestion on im-
Yin, J., Yu, X., Zhang, Y., Shen, D., Wang, M., Long, Y., Chen, T., 2016b. Enhancement of proving methane production from mixed microalgae (Scenedesmus sp., Chlorella sp.)
acidogenic fermentation for volatile fatty acid production from food waste: effect of and food waste: kinetic modeling and synergistic impact evaluation. Chem. Eng. J.
redox potential and inoculum. Bioresour. Technol. 216, 996–1003. 299, 332–341.
Yirong, C., Heaven, S., Banks, C.J., 2015. Effect of a trace element addition strategy on

288