Environmental Science and Pollution Research

https://doi.org/10.1007/s11356-019-04846-3

SEDIMENT RESEARCH AND MANAGEMENT

Theoretical framework for the estimation of H2S concentration

in biogas produced from complex sulfur-rich substrates
Iván Moreno-Andrade 1 & Gloria Moreno 1 & Guillermo Quijano 1

Received: 31 October 2018 / Accepted: 11 March 2019

# Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract
A theoretical framework was developed and validated for the estimation of H2S concentration in biogas produced from complex
sulfur-rich effluents. The modeling approach was based on easy-to-obtain data such as biological biogas potential (BBP),
chemical oxygen demand, and total sulfur content. Considering the few data required, the model fitted well the experimental
H2S concentrations obtained from BBP tests and continuous bioreactors reported in the literature. The model supported a
correlation coefficient (R2) of 0.989 over the experimental data, obtaining average and maximum errors of ~ 25 and ~ 35%,
respectively. The theoretical framework yielded good estimations for a wide range of experimental H2S concentrations (0.2 to
4.5% in biogas). This modeling approach is, therefore, a useful tool towards anticipating the H2S concentration in biogas
produced from sulfur-rich substrates and deciding whether the installation of a desulfurization technology is required or not.

Keywords Biogas . H2S concentration . Industrial effluents . Model validation . Sulfur-rich substrates

Nomenclature VGas Gas phase volume in the BBP bottle (L)

BBP Biological biogas potential (Lbiogas gCODremoved−1) VLiq Liquid phase volume in the BBP bottle (L)
COD Chemical oxygen demand (g L−1) VTot Total volume of the BBP bottle (L)
CW Cheese whey WV Wine vinasse
H Dimensionless Henry’s law constant (−)
H° Dimensionless Henry’s law constant at pH ≈ 7
and 298.15 K (−)
H2S% Total H2S concentration in the system (%)
H2 SGas
% H2S concentration in biogas (%) Introduction
H2 SLiq
% H2S concentration in liquid phase (%)
MW Sulfur molar weight (32.07 g mol−1) Biogas, mostly composed of ~ 60–65% CH4 and ~ 35–40%
OFMSW Organic fraction of municipal solid waste CO2, is a renewable fuel obtained from the anaerobic degra-
Scons Sulfur-to-COD consumption ratio (gSconsumed dation of organic matter (Noyola et al. 2006). Depending on
gCODremoved−1) the characteristics of the organic matter, the resulting biogas
T Temperature (K) may contain a sort of pollutants that must be removed prior its
T° Reference temperature (298.15 K) use in engines or turbines to produce heat and power (Dubois
and Thomas 2010). H2S is one of the most common biogas
pollutants, which causes corrosion to internal combustion de-
vices and industrial pipes by forming H2SO4 in the presence
Responsible editor: Philippe Garrigues of water (Maizonnasse et al. 2013; Pokorna and Zabranska
2015). In addition, H2S is toxic for humans, being lethal at a
* Guillermo Quijano concentration of 300 ppmv (0.03% v/v) (Aita et al. 2016).
GQuijanoG@iingen.unam.mx Among the most common devices used at industrial scale
1
for producing thermal and electrical power from biogas are
Laboratory for Research on Advanced Processes for Water
Treatment, Unidad Académica Juriquilla, Instituto de Ingeniería,
combined heat and power engines (Lo and Woon 2016),
Universidad Nacional Autónoma de México, Blvd. Juriquilla 3001, which can be operated with biogas containing no more than
Querétaro 76230, Mexico 500 ppmv of H2S (Aita et al. 2016). Therefore, biogas laden
 Environ Sci Pollut Res

with H2S concentrations higher than 500 ppmv must be puri- Theoretical framework
fied before its use for heat and electrical power generation.
There are many residual streams that have been reported as The total H2S concentration produced (H2S%) mainly de-
good candidates for producing biogas at industrial scale, in- pends on the amount of sulfur contained in the substrate.
cluding wastewaters from pulp and paper mills, sugar/ethanol Hence, in the present work, it was considered that the H2S
production from molasses, edible oil industry refineries, concentration in biogas is a function of the total mass of
cheese industry, and wine production (Lens et al. 1998; sulfur consumed per gram of COD removed (Scons) and the
Demirer et al. 2000; Barrera et al. 2013), as well as solid biochemical biogas potential (BBP) expressed as normal-
wastes such as the organic fraction of municipal solid waste ized liters of biogas produced per gram of COD removed as
(OFMSW) and cattle manure (Neves et al. 2008; Campuzano shown in Eq. 1:
and González-Martínez 2016; Díaz et al. 2016). A common 0 1
S cons
feature of these liquid and solid residual streams is their rela-  22:4
B C
tively high content of sulfur. The above-mentioned wastewa- H 2 S % ¼ @ MW A  100% ð1Þ
BBP
ters might contain sulfate concentrations between 0.5 and
50 g L−1 (Lens et al. 1998; Barrera et al. 2013), while the solid
wastes might contain sulfur concentrations between 0.1 and The factor 22.4 stands for the molar volume of the ideal gas
0.3% of the total solids in dry basis (mainly in the form of at normal pressure and temperature conditions (normalized L
proteins) (Campuzano and González-Martínez 2016). of H2S per mol), while MW is the sulfur molar weight
Therefore, the biogas produced from these streams will be (32.07 g mol−1). The central assumption of Eq. 1 is that all
laden with a proportional high concentration of H2S. the sulfur consumed is fully converted into H2S during the
Nowadays, there is a sort of physical-chemical and anaerobic digestion process, estimating therefore, the maxi-
biotechnological technologies for H2S removal from bio- mum H2S concentration that could be reached (in % vol/vol
gas (process known as desulfurization or biogas sweeten- units, which can be easily converted to ppmv by applying a
ing). Adsorption processes and chemical scrubbing in factor of 10,000). The total H2S produced in the system esti-
alkaline/amines solutions rank among the most common mated by Eq. 1 is then partitioned between the gas (H2 SGas
% )
physical-chemical treatments for biogas desulfurization −
and liquid phases (H2 SLiq 2−
% , dissolved as S or HS ) according
(Fortuny et al. 2008; Belmabkhout et al. 2009; Hao to Eq. 2:
et al. 2016). On the other hand, the commercially avail-
Liq
able THIOPAQ, SHELL-PAQUES, SULPHUS, and SOX H 2 S % V tot ¼ H 2 S Gas
% V Gas þ H 2 S % V Liq ð2Þ
technologies are reference biological-based technologies
for biogas desulfurization where H2S is oxidized to ele- H2 SLiq
% can be expressed in terms of the gas phase concen-
mental sulfur or sulfate by chemolithoautotrophic bacteria tration by using the Henry’s law constant:
(Quijano et al. 2018). In this regard, the H2S concentra-
tion in biogas is a key parameter to be considered for the H 2 S Gas
%
H 2 S Liq
% ¼ ð3Þ
selection of the desulfurization technology. A precise es- H
timation of the H2S concentration is therefore of para-
mount relevance towards deciding whether the installation Therefore, Eq. 2 can be written as follows:
of a desulfurization technology is required or not (based 0 1
on the maximum H 2 S allowable concentration of B V Tot C
% ¼ H 2S%@
H 2 S Gas
V Liq A
ð4Þ
500 ppmv), and if required, selecting the most suitable
V Gas þ
desulfurization technology (Aita et al. 2016). H
In this study, a modeling approach for the estimation of the
At this point, it is important to stress that the Henry’s law
H2S concentration in biogas produced from liquid and solid
constant is affected by both the pH value and temperature. H
residual streams was proposed and validated. The model esti-
values in literature are commonly given at reference condi-
mations were compared with H2S concentrations experimen-
tions of pH ≈ 7 and 298.15 K (Sander 1999). In this regard,
tally determined in BBP tests and also with data reported in
BBP tests are conducted under neutral pH values, NaHCO3
the literature. The theoretical framework considers easy-to-
being provided to avoid drastic pH changes during the exper-
obtain data such as biological biogas potential, COD, and total
iment (Angelidaki et al. 2009). However, to consider the im-
sulfur consumption. The capability of the model to yield pre-
pact of temperature (37 °C in BBP tests), the following equa-
cise estimations of H2S concentrations in biogas was evaluat-
tion can be used to correct the H value:
ed experimentally using cheesy whey, wine vinasses, and or- h  
ganic fraction of municipal solid waste as model liquid and − ddlnH  T −T °
1 1

ð1=T Þ
solid substrates. H ¼ H °e ð5Þ
Environ Sci Pollut Res

where H° and T° stand for the Henry’s law constant and tem-
perature under reference conditions, respectively. The term
dlnH
d ð1=T Þ is a constant related to the enthalpy of solution that
describes the temperature dependence, whose value for H2S
is 2300 K (Sander 1999). Equations 1, 2, 3, 4, and 5 constitute
the proposed theoretical framework that allows for estimating
the H2S concentration in biogas at any temperature under
neutrophilic conditions. A first approach for estimating H2S
concentration in biogas was early proposed by Peu et al.
(2012). These authors developed a specific equation focused
on substrates with a high content of solids such as cover crops,
animal wastes, or vegetable residues. In the present study, a
general framework that can be applied to solid and liquid
substrates is proposed.
methane gas quantification (triplicate experiments for each
effluent). The raw biogas composition (CH4, CO2, and H2S)
was also characterized by gas chromatography to obtain the
BBP. The tests were performed in 500-mL glass bottles
(370 mL of liquid working volume) at 37°C according to
Angelidaki et al. (2009). An agitation rate of 144 rpm was
provided for 1 min every 3 min by means of a rotating shaft.
A ratio of 0.5 gCODsubstrate/gTVSinoculum was used to avoid
biological activity limitation (Holliger et al. 2016; Figueroa-
González et al. 2018). Controls without substrate (to evaluate
the endogenous production of the inoculum) and controls with
glucose (to evaluate the activity of the inoculum) were also
performed in triplicate.
Analytical methods

Materials and methods
Inoculum and substrates characterization
Granular anaerobic sludge from an UASB reactor (San Juan
del Río, Mexico) was used as inoculum in the BBP tests. The
sludge was characterized by total volatile solids and density of
0.1 g g−1 and 1.0 g mL−1, respectively. The substrates inves-
tigated were cheese whey (CW), red wine vinasses (WV), and
organic fraction of municipal solid waste (OFMSW, restaurant
waste). In the case of OFMSW, the bones and inert material
(paper and plastic) were discarded during OFMSW collection
and only the fermentable matter was preserved. After selecting
the waste, the OFMSW was crushed and homogenized in a
blender. Finally, the waste was frozen (− 20 °C) until it was
used in the BBP tests. The substrates were characterized in
terms of total sulfur content and COD. The total sulfur content
in the sludge used as inoculum was also determined. Table 1
presents the characteristics of the diluted substrates (Table 1).
COD of the substrates, total volatile solids, and density of the
inoculum were determined according to Standard Methods
(APHA 2005). CH4 and CO2 gas concentrations were mea-
sured using a SRI-8610C GC-TCD (SRI Instruments, USA)
equipped with a Carboxen 1010 Plot column (30 m ×
0.53 mm × 30 μm). The temperatures of the injector, column,
and detector were 200, 100, and 230 °C, respectively. N2 was
used as the carrier gas at 20 mL min−1. H2S gas concentration
was determined in a SRI 8610-C gas chromatograph (SRI
Instruments, USA) equipped with a flame ionization detector
and a Restek MXT-1 column (60 m × 0.53 mm × 5 μm). Total
sulfur content in the substrates and the sludge was determined
by inductively coupled plasma mass spectrometry (iCAPq,
Thermo, USA) according to Boulyga et al. (2007).
Results and discussion
Biochemical biogas production

Biochemical biogas potential
The BBP tests were performed in an Automatic Methane
Potential Test System (AMPTS II, Bioprocess Control,
Sweden). This experimental system allowed testing of the
three effluents in the same experimental run with automatic
The three substrates supported a similar biogas production with
BBP values ranging from 0.48 to 0.55 nL gCODremoved−1, the
standard deviation being in all cases lower than 5.5%. It was ob-
served that the highest BBP value was achieved with wine vinasse,
while the lowest BBP value was recorded with cheese whey
(Fig. 1a). These results are in agreement with previous studies

Table 1 Characteristics of the

diluted substrates used in the BBP Substrate Preparation for the BBP test COD (g L−1) Total sulfur content
tests (mg S L−1)

WV Raw wine vinasse was diluted by a factor 25.0 ± 2.4 1.74 ± 0.3
of 8.4 in distilled water.
CW Raw cheese whey was diluted by a factor 25.0 ± 0.8 9.63 ± 1.2
of 2.3 in distilled water.
OFMSW 150 g of OFMSW was diluted in 1.5 L 15.5 ± 2.9 2.69 ± 0.4
of distilled water.

a
0.60
BBP (nLBiogas /gCODremoved)
0.50
0.40
0.30
0.20
0.10
0.00
0 10 20 30
Time (h)
b of pH on the Henry’s law constant (and consequently on the
CW 60
OFMSW 61
WV 67
0% 20% 40%
Biogas composion
Fig. 1 a Biochemical biogas production curves obtained with wine
vinasses (WV), cheese whey (CW), and organic fraction of municipal
solid waste (OFMSW); b composition of the biogas produced from the
substrates, CH4 (dark bars) and CO2 (clear bars)
reporting BBP values ranging from 0.3 to 0.6 nLBiogas CODfed−1
for red wine vinasses, cheese whey, and organic fraction of munic-
ipal solid waste (Demirer et al. 2000; Lalov et al. 2001; Rodríguez-
Pimentel et al. 2015). The composition of the biogas obtained from
each substrate in terms of CH4 and CO2 is shown in Fig. 1b. In all
cases, at least 60% of the gas was methane. The biogas obtained
with wine vinasses contained the highest methane concentration
(67%). The H2S concentration in all cases was ≤ 0.21%, and thus,
for simplicity, it was neglected from the overall composition
depicted in Fig. 1b. It has been reported that sulfur-reducing bac-
teria outcompete with methanogens for hydrogen and acetate,
resulting in a lower methane production during the anaerobic pro-
cess (Janke et al. 2016). However, in the present BBP tests with
initial sulfur concentrations of up to ∼3 mg S gCODfed−1 (consid-
ering the additional sulfur from the inoculum, see next section) the
biogas obtained had enough quality to be used in energy produc-
tion devices such as combined heat and power (CHP) engines/
turbines considering the CH4 content.
Experimental H2S concentrations vs model
estimations
Table 2 presents the biogas production, the experimental Scons
values, and initial/final pH values recorded in the BBP tests.
Environ Sci Pollut Res
The amount of sulfur coming from the sludge (total sulfur
content of 290 ± 47 mg S L−1) was considered in the calcula-
tion of Scons. BBP tests were provided with 50 mL of sludge,
which accounted for additional 14.5 mg S in all tests. The
granular sludge used as inoculum was obtained from an an-
aerobic reactor treating wastewaters from food and paper in-
WV CW OFMSW dustries, which explains its high sulfur content. It is important
to note that NaHCO3 was provided to avoid drastic pH chang-
es during the anaerobic digestion process. The differences
40 50 60 70 observed between the initial and final pH values in the BBP
tests were ≤ 3% regardless of the substrate studied. The effect
partitioning characteristics of H2S) was therefore considered
40
as negligible. Once the effect of pH was discarded, the impact
of temperature on the Henry’s law constant was determined by
means of Eq. 5. According to Sander (1999), the dimensionless
39 H value under reference conditions is 0.41. Considering the
working temperature of 310 K in Eq. 5, the corrected dimension-
less H value was 1.09. The experimental values of Scons and BBP,
33 as well as the corrected H value, were entered in Eqs. 1 and 4 for
the estimation of the H2S concentration in biogas.
Experimental H2S concentrations in biogas ranged from
60% 80% 100%
1718 ± 272 to 2133 ± 150 ppmv. As shown in Fig. 2a, the model
estimations were in very good agreement with the experimental
H2S data. In the worst case, the model overestimated ~ 27% the
H2S concentration obtained with CW, which was attributed to the
potential formation of other sulfur compounds different from
H2S. In this context, the theoretical framework herein presented
assumes that all the sulfur consumed is converted into H2S, and
therefore it estimates the maximum H2S concentration that can
be achieved. The experimental data obtained with WV, OFMSW,
and CW showed that this assumption is reasonably good for such
complex substrates. In addition, it must be highlighted that the
presence of sulfur in the inoculum led to an H2S production that
exceeded the limits for operating CHP devices regardless of the
substrate tested (H2S concentrations up to ~ 4 times above the
limit of 500 ppmv). Such high H2S concentrations would not be
anticipated if the model only considers the sulfur contained in the
substrates. Due to the high content of sulfur in the inoculum, the
model output indicates that the biogas generated during the
startup of the process must be discarded (e.g., burned in a flare)
and not used in CHP devices to avoid corrosion issues. Once the
sulfur from the inoculum is depleted, then the sulfur contained in
the substrates will produce an H2S concentration low enough to
be used in CHP devices. Therefore, all sulfur sources must be
taken into account for Scons calculation. This is particularly
important when performing BBP tests since good model
estimations can only be retrieved when the total sulfur in the
system is considered. As mentioned above, a potential sulfur
contribution from the inoculum would be temporary in
continuous digestion processes, making necessary the monitor-
ing of H2S concentration until the generated biogas fulfills the
requirements for use.
Environ Sci Pollut Res

Table 2 Sulfur/COD removal performances and pH changes recorded in the BBP tests

Substrate Initial COD Final COD COD Scons Initial pH Final pH pH BBP
(g L−1) (g L−1) removal (gSconsumed gCODremoved−1) difference (nL gCODremoved−1)
(%) (%)

CW 6.34 ± 0.17 0.94 ± 0.03 85 0.0027 7.27 ± 0.03 7.46 ± 0.02 2.6 0.48
WV 6.53 ± 0.15 0.15 ± 0.01 98 0.0024 7.30 ± 0.03 7.40 ± 0.09 1.3 0.52
OFMSW 5.92 ± 0.28 0.23 ± 0.04 96 0.0031 7.31 ± 0.05 7.54 ± 0.02 3.0 0.55

Figure 2b shows the experimental H2S concentration in the
BBP controls performed. Concentrations of 2652 ± 418 and 31
± 9 ppmv were observed in the glucose and endogenous con-
trols, respectively. These results indicated that the organic mat-
ter contained only in the inoculum was not able to support the
total conversion of sulfur into H2S. On the contrary, the glucose
control showed that the sulfur reduction was triggered by the
organic matter contained in the substrate. These results are not
surprising since the endogenous control lacked of readily avail-
able organic matter and produced a limited amount of biogas. It
is well known that autotrophic routes of sulfur reduction are
linked to H2 or CH4 availability (Flores-Alsina et al. 2016; Hao
et al. 2014). Hence, in the glucose control (and in the tests with
CW, OFMSW, and CW), sulfur can be reduced via heterotro-
phic processes, while the autotrophic pathways are also
a 4,000
enhanced due to the higher biogas production and the higher
availability of H2 and CH4 relative the endogenous control.
Therefore, the sulfur contained in the inoculum must be includ-
ed in the calculation of Scons to avoid overestimations of the
H2S production potential of the inoculum.
Model estimations vs data available in the literature
The presence of H2S has been reported as a major drawback
for the use of biogas in the generation of renewable energies
(Barrera et al. 2013). Thus, several authors studied the H2S
production in complex sulfur-rich effluents. Table 3 summa-
rizes the BBP, Scons, and experimental H2S concentrations in
biogas produced from seaweed/pig slurry, ethanol distillation
vinasse, wastewater from rubber latex industry, and sulfate-
rich wastewater. All of these studies were performed in con-

3,500
Model esmaon Experimental data
tinuous bioreactors. As shown in Fig. 3a, the model estima-
tions were in all cases in the same magnitude order compared
H2S in biogas (ppmv)

3,000
with the experimental H2S data. In the worst case, the model
estimation was ~ 35% higher than that experimentally deter-
2,500
mined. However, this maximum overestimation was observed
2,000 for seaweed mixed with pig slurry, whose experimental error
1,500 in H2S concentration accounted for ~ 22%. For the other ef-
fluents reported in the literature, the model showed
1,000
overestimated H2S concentrations in the range of 18 to 30%
500 relative experimental data. It is worth noting that in all cases,
0 the model estimations were higher than the corresponding
WV OFMSW CW
experimental data, which strongly suggested that other sulfur
b
3,500
compounds different from H2S were also accumulated.
3,000 However, considering the minimum knowledge of the
H2S in biogas (ppmv)

substrates/inoculum required by the present theoretical frame-

2,500
work, a maximum error of 35% can be considered as a very
2,000 good approximation for anticipating the H2S concentration in
biogas obtained from complex sulfur-rich substrates.
1,500
The H2S concentration in biogas as a function of the Scons
1,000 value is shown in Fig. 3b. It was observed that the resulting H2S
500
concentration increased as the Scons value increased. The model
yielded larger overestimations as the Scons value increased.
0 Since the model estimates the maximum H2S that can be pro-
Glucose control Endogenous control
duced (assuming that all sulfur consumed is fully converted into
Fig. 2 a Experimental H2S concentrations in biogas from WV, OFMSW,
and CW (white bars) and model estimations (dark bars); b experimental
H2S), these results suggested that the higher the Scons value, the
H2S concentration in the biogas produced in the glucose and endogenous higher the accumulation of other sulfur compounds different
controls from H2S. In this context, several key sulfur compounds can
 Environ Sci Pollut Res

Table 3 Characteristics of the effluents reported in the literature and the H2S concentration experimentally determined in biogas

Substrate BBP Scons H2S in biogas Reference

(LBiogas gCODremoved−1) (gSconsumed gCODremoved−1) (ppmv)

Mixture of seaweed and pig slurry 0.574 0.049 45,000 ± 10,000 Peu et al. 2011
Ethanol distillation vinasse 0.321 0.015 28,470 ± 1423 Barrera et al. 2014
Wastewater from rubber latex industry 0.529* 0.012 11,422 ± 996 Chaiprapat et al. 2015
Sulfate-rich wastewater 0.397** 0.014 17,082 ± 2847 Liu et al. 2015

*S-sulfate was considered as the total S

**It was assumed that biogas contained 60% methane

be found under the reducing conditions prevailing in the anaer-
obic digestion processes, including thiosulfate, elemental sulfur,
mercaptans, dimethyl sulfide, dimethyl disulfide, dimethyl tri-
sulfide, ethyl methyl sulfide, and H2S (S2− or HS− in liquid
phase) (Fisher et al. 2018; Hao et al. 2014). In spite of the
variety of sulfur compounds that can be accumulated, the re-
sults herein obtained confirmed that H2S was the main sulfur
compound produced regardless of the Scons value tested since
the model supported a maximum overestimation of ~ 35%.
Evaluation of the theoretical framework performance
As shown in Fig. 4a, the model supported a correlation coef-
ficient (R2) of 0.989 over the experimental data. Therefore,
assuming that all sulfur consumed is converted into H2S is a
good approximation. The comparison of the experimental-vs-
model data plot with the perfect correlation line clearly
showed that model overestimation increases with the experi-
mental H2S concentration. However, considering the few data
required by the model, a maximum overestimation of ~ 35% is
quite reasonable for the wide range of experimental H2S

Fig. 4 a Experimental vs model data correlation plot showing the best

Fig. 3 a Experimental H2S concentrations in biogas obtained by several
authors from complex substrates (clear bars) and model estimations (dark
bars), b experimental (circles) and model (asterisks) H2S concentration in
biogas as a function of the Scons value (experimental data herein obtained
and data from Table 3 were used in this plot)
linear regression fit (continuous line) and the perfect correlation line
(dotted line), b model under/overestimations vs Scons value (circles) and
average error considering all the absolute error values (continuous line).
Experimental data herein determined and data from Table 3 were used in
all plots
Environ Sci Pollut Res
concentrations studied (0.2 to 4.5% H2S in biogas). It is also
important to stress that for the lower Scons values tested
(0.0011 and 0.0013 gSconsumed gCODremoved−1), the model
estimated H2S concentrations between 13 and 19% lower than
the experimental values. However, the standard deviation of
such experimental H2S values was 8–16% relative the mean
value. Hence, taking into account these deviations, the model
estimations were in good agreement with the experimental
data. An overview of the model under/overestimations as a
function of the S cons values is depicted in Fig. 4b.
Considering the absolute value of the under/overestimations,
an average error of ~ 25% was obtained for the present theo-
retical framework.
The results obtained showed that the theoretical approach
herein proposed constitutes a useful tool for anticipating the
H2S concentration in the biogas produced from complex
sulfur-rich substrates. The model requires a minimum knowl-
edge of the substrate and yields good estimations of H2S
concentration data from BBP tests and also from continuous
bioreactors. This information is of paramount relevance
towards deciding whether the installation of a desulfurization
technology is required or not considering the maximum H2S
concentration of 500 ppmv allowable for most CHP equipment
running on biogas. Moreover, knowledge on the magnitude
order of the H2S concentration in biogas allows for anticipating
the most suitable desulfurization technology that should be
installed.
Conclusions
To the best of our knowledge, this is the first report on the
successful application of a comprehensive theoretical frame-
work for estimating the H2S concentration in biogas produced
from liquid and solid substrates. The model yielded good es-
timations for the H2S concentrations determined experimen-
tally in the present study as well as for H2S concentrations
reported in the literature. From a process implementation point
of view, the modeling approach here presented constitutes a
useful tool for anticipating if a desulfurization process will be
required based on relatively simple parameters. The model
supported a correlation coefficient R2 of 0.989 and yielded
accurate estimations of the H2S concentration for liquid and
solid effluents with average and maximum errors of ~ 25 and
~ 35%, respectively. The modeling approach was able to esti-
mate adequately H2S concentrations obtained from both BBP
tests and continuous bioreactors.
Funding information The technical support of Jaime Pérez Trevilla and
Ángel Avizua Hernández is acknowledged. This study was financially
supported by DGAPA-UNAM through the PAPIIT project IA100719 and
by Fondo de Sustentabilidad Energética SENER– CONACYT (Mexico)
through the project 247006 Gaseous Biofuels Cluster.
References
Aita BC, Mayer FD, Muratt DT, Brondani M, Pujol SB, Denardi LB,
Hoffmann R, da Silveira DD (2016) Biofiltration of H2S-rich biogas
using Acidithiobacillus thiooxidans. Clean Techn Environ Policy
18:689–703. https://doi.org/10.1007/s10098-015-1043-5
Angelidaki I, Alves M, Bolzonella D, Borzacconi L, Campos JL, Guwy
AJ, Kalyuzhnyi S, Jenicek P, van Lier JB (2009) Defining the
biomethane potential (BMP) of solid organic wastes and energy
crops: a proposed protocol for batch assays. Water Sci Technol 59:
927–934. https://doi.org/10.2166/wst.2009.040
APHA (2005) Standard methods for examination of water and wastewa-
ter, 21st edn. APHA, AWWA, WPCR, New York
Barrera EL, Spanjers H, Dewulf J, Romero O, Rosa E (2013) The sulfur
chain in biogas production from sulfate-rich liquid substrates: a re-
view on dynamic modeling with vinasse as model substrate. J Chem
Technol Biotechnol 88:1405–1420. https://doi.org/10.1002/jctb.4071
Barrera EL, Spanjers H, Romero O, Rosa E, Dewulf J (2014)
Characterization of the sulfate reduction process in the anaerobic
digestion of a very high strength and sulfate rich vinasse. Chem
Eng J 248:383–393. https://doi.org/10.1016/j.cej.2014.03.057
Belmabkhout Y, De Weireld G, Sayari A (2009) Amine-bearing mesopo-
rous silica for CO2 and H2S removal from natural gas and biogas.
Langmuir 25:13275–13278. https://doi.org/10.1021/la903238y
Boulyga SF, Heilmann J, Prohaska T, Heumann KG (2007) Development
of an accurate, sensitive, and robust isotope dilution laser ablation
ICP-MS method for simultaneous multi-element analysis (chlorine,
sulfur, and heavy metals) in coal samples. Anal Bioanal Chem 389:
697–706. https://doi.org/10.1007/s00216-007-1485-7
Campuzano R, González-Martínez S (2016) Characteristics of the organ-
ic fraction of municipal solid waste and methane production: a re-
view. Waste Manag 54:3–12. https://doi.org/10.1016/J.WASMAN.
2016.05.016
Chaiprapat S, Wongchana S, Loykulnant S, Kongkaew C, Charnnok B
(2015) Evaluating sulfuric acid reduction, substitution, and recovery
to improve environmental performance and biogas productivity in
rubber latex industry. Process Saf Environ Prot 94:420–429. https://
doi.org/10.1016/j.psep.2014.10.002
Demirer GN, Duran M, Güven E, Ugurlu Ö, Tezel U, Ergüder TH (2000)
Anaerobic treatability and biogas production potential studies of
different agro-industrial wastewaters in Turkey. Biodegradation 11:
401–405. https://doi.org/10.1023/A:1011659705369
Díaz I, Figueroa-González I, Miguel JA, Bonilla-Morte L, Quijano G
(2016) Enhancing the biomethane potential of liquid dairy cow ma-
nure by addition of solid manure fractions. Biotechnol Lett 38:
2097–2102. https://doi.org/10.1007/s10529-016-2198-9
Dubois L, Thomas D (2010) Comparison of various alkaline solutions for
H2S/CO2-selective absorption applied to biogas purification. Chem
Eng Technol 33:1601–1609. https://doi.org/10.1002/ceat.
201000171
Figueroa-González I, Moreno G, Carrillo-Reyes J, Sánchez A, Quijano
G, Buitrón G (2018) From mesophilic to thermophilic conditions:
one-step temperature increase improves the methane production of a
granular sludge treating agroindustrial effluents. Biotechnol Lett 40:
569–575. https://doi.org/10.1007/s10529-017-2490-3
Fisher RM, Le-Minh N, Alvarez-Gaitan JP et al (2018) Emissions of
volatile sulfur compounds (VSCs) throughout wastewater biosolids
processing. Sci Total Environ 616–617:622–631. https://doi.org/10.
1016/j.scitotenv.2017.10.282
Flores-Alsina X, Solon K, Kazadi Mbamba C, Tait S, Gernaey KV,
Jeppsson U, Batstone DJ (2016) Modelling phosphorus (P), sulfur
(S) and iron (Fe) interactions for dynamic simulations of anaerobic
digestion processes. Water Res 95:370–382. https://doi.org/10.
1016/j.watres.2016.03.012

Fortuny M, Baeza JA, Gamisans X, Casas C, Lafuente J, Deshusses MA,
Gabriel D (2008) Biological sweetening of energy gases mimics in
biotrickling filters. Chemosphere 71:10–17. https://doi.org/10.1016/
j.chemosphere.2007.10.072
Hao T w, yu XP, Mackey HR et al (2014) A review of biological sulfate
conversions in wastewater treatment. Water Res 65:1–21
Hao X, Hou G, Zheng P, Liu R, Liu C (2016) H2S in-situ removal from
biogas using a tubular zeolite/TiO2 photocatalytic reactor and the
improvement on methane production. Chem Eng J 294:105–110.
https://doi.org/10.1016/J.CEJ.2016.02.098
Holliger C, Alves M, Andrade D, Angelidaki I, Astals S, Baier U, Bougrier
C, Buffière P, Carballa M, de Wilde V, Ebertseder F, Fernández B,
Ficara E, Fotidis I, Frigon JC, de Laclos HF, Ghasimi DSM, Hack G,
Hartel M, Heerenklage J, Horvath IS, Jenicek P, Koch K, Krautwald J,
Lizasoain J, Liu J, Mosberger L, Nistor M, Oechsner H, Oliveira JV,
Paterson M, Pauss A, Pommier S, Porqueddu I, Raposo F, Ribeiro T,
Rüsch Pfund F, Strömberg S, Torrijos M, van Eekert M, van Lier J,
Wedwitschka H, Wierinck I (2016) Towards a standardization of
biomethane potential tests. Water Sci Technol 74:2515–2522.
https://doi.org/10.2166/wst.2016.336
Janke L, Leite AF, Batista K, Silva W, Nikolausz M, Nelles M, Stinner W
(2016) Enhancing biogas production from vinasse in sugarcane
biorefineries: effects of urea and trace elements supplementation
on process performance and stability. Bioresour Technol 217:10–
20. https://doi.org/10.1016/j.biortech.2016.01.110
Lalov IG, Krysteva MA, Phelouzat JL (2001) Improvement of biogas
production from vinasse via covalently immobilized methanogens.
Bioresour Technol 79:83–85. https://doi.org/10.1016/S0960-
8524(01)00045-1
Lens PNL, Visser A, Janssen AJH, Pol LWH, Lettinga G (1998)
Biotechnological treatment of sulfate-rich wastewaters. Crit Rev
Environ Sci Technol 28:41–88. https://doi.org/10.1080/
10643389891254160
Liu Y, Zhang Y, Ni BJ (2015) Zero valent iron simultaneously enhances
methane production and sulfate reduction in anaerobic granular
sludge reactors. Water Res 75:292–300. https://doi.org/10.1016/j.
watres.2015.02.056
Lo IMC, Woon KS (2016) Food waste collection and recycling for value-
added products: potential applications and challenges in Hong
Kong. Environ Sci Pollut Res 23:7081–7091. https://doi.org/10.
1007/s11356-015-4235-y
Environ Sci Pollut Res
Maizonnasse M, Plante J-S, Oh D, Laflamme CB (2013) Investigation of
the degradation of a low-cost untreated biogas engine using preheated
biogas with phase separation for electric power generation. Renew
Energy 55:501–513. https://doi.org/10.1016/j.renene.2013.01.006
Neves L, Gonçalo E, Oliveira R, Alves MM (2008) Influence of compo-
sition on the biomethanation potential of restaurant waste at
mesophilic temperatures. Waste Manag 28:965–972. https://doi.
org/10.1016/J.WASMAN.2007.03.031
Noyola A, Morgan-Sagastume JM, López-Hernández JE (2006)
Treatment of biogas produced in anaerobic reactors for domestic
wastewater: odor control and energy/resource recovery. Rev
Environ Sci Bio/Technology 5:93–114. https://doi.org/10.1007/
s11157-005-2754-6
Peu P, Sassi JF, Girault R, Picard S, Saint-Cast P, Béline F, Dabert P
(2011) Sulphur fate and anaerobic biodegradation potential during
co-digestion of seaweed biomass (Ulva sp.) with pig slurry.
Bioresour Technol 102:10794–10802. https://doi.org/10.1016/j.
biortech.2011.08.096
Peu P, Picard S, Diara A, Girault R, Béline F, Bridoux G, Dabert P (2012)
Prediction of hydrogen sulphide production during anaerobic diges-
tion of organic substrates. Bioresour Technol 121:419–424. https://
doi.org/10.1016/j.biortech.2012.06.112
Pokorna D, Zabranska J (2015) Sulfur-oxidizing bacteria in environmen-
tal technology. Biotechnol Adv 33:1246–1259. https://doi.org/10.
1016/j.biotechadv.2015.02.007
Quijano G, Figueroa-González I, Buitrón G (2018) Fully aerobic two-
step desulfurization process for purification of highly H2S-laden
biogas. J Chem Technol Biotechnol 93:3553–3561. https://doi.org/
10.1002/jctb.5732
Rodríguez-Pimentel RI, Rodríguez-Pérez S, Monroy-Hermosillo O,
Ramírez-Vives F (2015) Effect of organic loading rate on the per-
formance of two-stage anaerobic digestion of the organic fraction of
municipal solid waste (OFMSW). Water Sci Technol 72:384–390.
https://doi.org/10.2166/wst.2015.223
Sander R (1999) Compilation of Henry’s law constants for inorganic and
organic species of potential importance in environmental chemistry.
http://www.henrys-law.org/henry.pdf. Accessed 5 Feb 2019
Publisher’s note Springer Nature remains neutral with regard to juris-
dictional claims in published maps and institutional affiliations.