Valorization of Different Landrace and Commercial Sorghum (Sorghum bicolor (L.) Moench) Straw Varieties by Anaerobic Digestion

332 | GCB Bioenergy. 2023;15:332–345.wileyonlinelibrary.com/journal/gcbb Received: 10 October 2022 | Accepted: 18 November 2022 DOI: 10.1111/gcbb.13025 RESEARCH ARTICLE Valorization of different landrace and commercial sorghum (Sorghum bicolor (L.) Moench) straw varieties by anaerobic digestion Davidde la Lama- Calvente1 | María JoséFernández- Rodríguez2,3 | JacintoGandullo3 | IreneDesena2 | Clarade la Osa3 | Ana BelénFeria3 | AntoniaJiménez- Rodríguez2 | RafaelBorja1 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. GCB Bioenergy published by John Wiley & Sons Ltd. 1Spanish Scientific Research Council (CSIC) - Instituto de la Grasa (IG), Department of Food Biotechnology, Campus Universidad Pablo de Olavide, Sevilla, Spain 2Departamento de Sistemas Físicos, Químicos y Naturales, Universidad Pablo de Olavide, Sevilla, Spain 3Departmento de Biología Vegetal y Ecología, Facultad de Biología, Universidad de Sevilla, Sevilla, Spain Correspondence María José Fernández- Rodríguez, Departamento de Biología Vegetal y Ecología, Facultad de Biología, Universidad de Sevilla, Sevilla, Spain. Email: [email protected]; [email protected] Funding information regional government of Andalucía, Junta de Andalucía, Consejería de Transformación Económica, Industria, Conocimiento y Universidades, Grant/ Award Number: Project FEDER UPO- 1380782 Abstract To reduce the impact on the environment and enhance the sustainability of resources, it is necessary to promote and strengthen the use of landrace cultivars that advocate regenerative agriculture. In this study, the growth and development as well as the anaerobic digestion (AD) of six different landrace cultivars, two commercial hybrids cultivars and a public genotype of Sorghum bicolor have been evaluated. The landrace cultivars, in general, presented greater heights, biomass yields and compactness shoots as well as similar or an improvement in grain production compare to the commercial varieties. The AD of the different sorghum straws was performed in batch mode at mesophilic temperature (35°C). The landrace cultivar Zahina (ZH) obtained the highest final methane yield (413 ± 79 NL CH4 kg−1 VS, volatile solids) but the landrace cultivars Zahina gigante (ZHG) and Trigomillo (TG) were the ones that obtained the highest methane per biomass production (13.7 and 12.7 NL CH4 shoot unit−1, respectively). By contrast, the commercial varieties were the ones that obtained the lowest methane yields. Two mathematical models, firstorder kinetics and the Transference Function model, were used to fit the experimental data with the aim of describing and simulating the anaerobic biodegradation of these S. bicolor straw varieties and obtaining the kinetic constants. Both models allowed for adequately fitting the experimental results of methane production with time. In particular, the fastest biomethanization occurred using the commercial variety PR88Y20 (PR88) (specific rate constant k=0.148 ± 0.008 days−1), while the slowest one was obtained from Panizo (PAN) variety (k=0.064 ± 0.005 days−1). In addition, the highest values of the maximum methane production rate, Rm, were attained for the varieties ZH and PR88, which were 87.1% and 71.3% higher than that achieved for the PAN variety, which exhibited the lowest value. | 333 de la LAMACALVENTE et al. 1 | INTRODUCTION Sorghum is among the five most produced cereals in the world, and, although the production of maize, rice and barley are forecast to decline, the stocks of sorghum are expected to increase in the following years (FAO,2022). In the season 2021– 2022, the sorghum world production reached 62.659 million tons of which 1.015 million tonnes were produced in Europe (USDA,2022). Sorghum bicolor (L.) Moench is a C4 plant originally from tropical, subtropical and semiarid areas that has been recently introduced in warm regions of Europe due to new high tolerant varieties capable to adapt to a wide range of environmental and soil conditions produced by breeding techniques (Pasteris et al., 2021; Thomas et al.,2021). This crop is able to grow from clay soils to light sand, within a pH interval of 5.0– 8.5, although the most suitable conditions are soils with medium/high nutritional conditions (N ≥ 260, P ≥ 12 and K ≥ 120 kg ha−1), 0.6% of organic matter, 80 cm of depth, 1.4Mg m−3 density and a 50% waterholding capacity (Ahmad Dar et al.,2018). The low nutrient needs of this crop allow to reduce the use of fertilizers and it has been proven to help with soil bioremediation processes enhancing the microbial communities (Carcedo et al.,2021; Pasteris et al.,2022). Due to the efficiency of nutrient uptake, the high growth rate and the low water needs, this crop has received the attention of the agroindustrial communities specially in those areas where water and other mineral nutrients are difficult to obtain, converting S. bicolor into a vital food for millions of people, although, in the global north its mainly used as animal feed (Taylor et al.,2006). In a present framework of high energy demand against the need to palliate climate change and global warming, biofuels are of great interest as an alternative to fossil fuels. Among the different biomasses studied and used for biofuels production, those with a high lignocellulosic fraction are the most effective options (Wang et al.,2019). Sorghum bicolor is a lignocellulosic biomass that contains both soluble and insoluble carbohydrates almost at the same ratio, which is one of the main reasons of why this crop is suitable as a substrate for biofuel production such as bioethanol and biogas, although the yield of these products is highly dependent on the cultivar, the year and the harvest period (Ahmad Dar et al.,2018). Sorghum bicolor grows fast (around 150 days), presents a high carbohydrate content (>50% d.m.), its grains are widely used as food and feed and the produced straw is suitable for animal feed or for further manipulation into biofuels; therefore, it is one the most efficient crops in terms of a circular economy system, which is why it is considered an Energy Crop for several authors (Gómez- Camacho et al.,2021). Among the different technologies related to biofuel production, anaerobic digestion (AD) is presented as a viable and wellknown biotechnology for converting organic substrates into biogas. In addition, the subsequent digestate fits the purposes of a circular economy model due to its recognized capacity as a biofertilizer (Bona et al.,2022; Rakascan et al.,2021). Furthermore, unlike other biofuel technologies, AD is feasible both at small scale and large scale, being of great use in developing countries and rural areas (Ahmad Dar et al.,2018). Thus, sweet sorghum straw and silage have been studied as feedstock for AD processes (Gómez- Camacho et al.,2021; Senghor et al.,2017). Among the bibliography, the reported methane yield of S. bicolor straw is ranging from 270 to 400 NL CH4 kg−1 volatile solids (VS), which supports the idea of a feasible process at larger scales (Ahmad Dar et al.,2018). Moreover, in a global warming scenario, where the CO2 concentration in the atmosphere is expected to keep raising, sorghum is presented as the less affected cereal crop by elevated CO2 concentrations (Senghor et al.,2017). However, methane yield from sorghum biomass is highly dependent on the cultivar itself and the cultivation area. Thomas et al.(2019) reported methane yields from 200 to 259 NL CH4 kg−1 total solids (TS) when assessing the biochemical methane potential (BMP) of four S. bicolor phenotypes cultivated at different locations and collected at the dough grain stage. While Wannasek et al.(2017) investigated the biomethane potential of the silage of five species of sorghum and reported no significant differences with an average methane yield of 338 NL CH4 kg−1, however, when the biomass yield of each specie was considered and the methane yield corrected per hectare, the differences were significant with methane yields ranging from 2300 to 6500 m3 CH4 ha−1. This research was carried out to assess the BMP at mesophilic temperature (35°C) of the residual straw after grain collection of 6 landrace cultivars of S. bicolor from different areas of the Iberian Peninsula, 2 commercial hybrids cultivars and a public genotype in controlled and optimum conditions. Two mathematical models, the firstorder and Transference Function (TF) model, were tested to predict the biomethane potential KEYWORDS anaerobic digestion, commercial varieties, kinetic parameters, landrace cultivars, regenerative agriculture, Sorghum bicolor 17571707, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/gcbb.13025 by Universidad De Sevilla, Wiley Online Library on [13/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 334 | de la LAMACALVENTE et al. of the sorghum straw varieties and to determine the kinetic parameters of the different anaerobic processes evaluated using the experimental results of methane production time. To get a better understanding of the feasibility of this technology, the methane yield has also been normalized as per shoot biomass unit in an attempt to include biomass production as a vital variable for preliminary energy sustainability analysis. Other factors such as biomass and grain production have been considered in the final conclusions. The novelty of this study relayed in the assessment of landrace and noncommercial cultivars of S. bicolor cultivated under controlled environmental and soil conditions. To the best of our knowledge, the methane potential of these endogenous species has not been evaluated before, furthermore, although other published studies investigated the differences between S. bicolor varieties, in most of those cases the environmental and soil conditions were different for each compared cultivar, thus, significant differences in methane production cannot be exclusively linked to the specific S. bicolor cultivar itself. Moreover, there is a general need to acquire relative data linking methane yield, biomass and grain production of the assessed crops, that is, in some cases sorghum crops are destined to grain production; thus, the valorization of the straw could be a positive process. 2 | MATERIALS AND METHODS 2.1 | Analytical methods The different S. bicolor (L.) Moench cultivars and the anaerobic sludge were analysed before the startup of the BMP assays. The subsequent digested effluents that resulted at the end of each assay were also analysed. Standard method 2540B & 2540 E (APHA,2017) was used as reference for the determination of TS, VS and mineral solids. Total chemical oxygen demand (CODtotal) was carried out as described by Raposo et al.(2008). Soluble chemical oxygen demand (CODsol) was performed by the closed digestion and colorimetric standard method 5220D (APHA,2017). pH and total alkalinity (TA) were carried out using a pH meter (Crison 20 basic), and TA was performed by titration to pH4.3 and following the standard method 2320B (APHA,2017). Elemental analysis for the determination of C, N and H content in the different substrates was carried out by a LECO CHNS- 932 Elemental Analyzer (Leco Corporation). Prior to the analysis, a representative sample of each S. bicolor cultivar was brought to dryness. Analyses of soluble parameters were performed after sample centrifugation (Eppendorf, 9000×g, 10min) and filtration (Albet, 47 mm glass fibre filter). 2.2 | Sorghum biomass 2.2.1 | Plant materials and growth conditions Nine S. bicolor varieties were used in this study. PR87G57 (PR87) and PR88Y20 (PR88) are two commercial hybrid cultivars from Pioneer Hibreed and PR898012 (P89) correspond to a public genotype. Six landrace cultivars from different Iberian regions available in the Plant Genetic Resource Centre (CRFINIA) belonging to the Spanish Government were selected (Figure1): Cañadú (CÑ), Milho painzo (MP), Panizo (PAN), Trigomillo (TG), Zahina (ZH) y Zahina gigante (ZHG). FIGURE 1 Plant material. Geographical origin of landrace cultivars selected along Iberian Peninsula. TG (Trigomillo), CÑ (Cañadú), MP (milho Painzo), ZH (Zahina), ZHG (Zahina Gigante), PAN (Panizo). 17571707, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/gcbb.13025 by Universidad De Sevilla, Wiley Online Library on [13/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 335 de la LAMACALVENTE et al. Sorghum bicolor plants were grown in a greenhouse (CITIUS, University of Seville, Seville, Spain) with a 14/10 photoperiod at 25°C and 60% of humidity. The experiments were carried out with 16 replicates per treatment and designed as a randomized controlled trial. The seeds were germinated in the soil in 96- cell seed trays (75 ml per cell). After 2 weeks, seedlings were transferred into 5- L pots. The soil was a standard substrate made of black peat enriched with perlite and coconut fibre (Blumenerde), supplied by Gramoflor substrate. The plants were watered three times a week with tap water. The plants were harvested and analysed when the seeds were completely mature, 4– 5months after sowing. 2.2.2 | Growth and productivity parameters Panicles were dried in an oven at 38°C for a week. Once dried, the seeds were threshed by hand and seed yield and weight were quantified. Shoot height was determined at the moment of harvesting, excluding the panicles in this measure. Shoot tissues were also dried in an oven at 65°C for a week and weighed to quantify shoot dry weight. Finally, the sorghum straw was ground in a mortar, homogenizing the size of the fragment of between 1 and 2 mm before the AD process. 2.3 | Anaerobic inoculum BMP assays were set up using inoculum as an anaerobic sludge collected from a fullscale upflow sludge blanket reactor located in Seville, Spain. The anaerobic sludge was acclimated to brewery wastewaters and prior to use was degassed at 35 ± 2°C for 48 h. To determine the methanogenic activity, microcrystalline cellulose (MCC) (Avicel® PH- 101, Fluka) was used as a positive control. The main physicochemical parameters of the selected inoculum were as follows: TS, 47.3 ± 0.6 g kg−1; VS, 34.4 ± 0.6g kg−1; pH, 7.14 and TA, 3334 mg CaCO3 L−1. 2.4 | Experimental procedure The BMP assays were set up following the recommendations described by Holliger et al.(2016). Each assay, including blanks and positive control, was carried out in triplicate. Blanks were prepared following the same indications described below but with no substrate, and, as a positive control, MCC was used. In total, 250 ml reactors were filled with inoculum, substrate, 20% macro and micronutrient solution (Raposo et al.,2006) and distilled water up to a working volume of 240 ml. The relation between the substrate and the inoculum was stablished at a ratio of 0.5 based on the VS. Temperature was controlled within the mesophilic range (35 ± 1°C) by a water bath and the reactors were flushed with nitrogen gas at the beginning of the experiment so the anaerobic conditions were maintained. Methane was measured by allowing the biogas to pass through a 2N NaOH solution. This method assumes that the liquid displacement is due to the produced methane while the generated CO2 is trapped in the alkaline solution, as is widely accepted and reported in the literature (Casallas- Ojeda et al., 2022; Fernández- Rodríguez et al.,2021). The experiment was finished when the accumulated volume of methane was less than 1% for three consecutive days, this period was about 31 days. The accumulated methane yield was normalized to standard temperature and pressure conditions (273.15 K and 101.325 kPa) and the endogenous production from the inoculum subtracted. 2.5 | Data analysis 2.5.1 | Biodegradability The theoretical methane potential of each substrate was calculated as proposed by Nielfa et al. (2015), by both CODtotal (Equation1) and elemental composition analysis (Equation3). Biodegradability was then calculated by comparing the theoretical values with the experimental methane yield. where BMPthCOD is the theoretical methane yield (NLCH4 kg−1 VSadded); nCH4 the methane moles determined by Equation(2) (mol); R the ideal gas constant (0.082 atm L mol−1K−1); T the standard temperature (273 K); p the standard pressure (1atm); VS the volatile solids of the substrate (kg) where CODtotal is the chemical oxygen demand of the substrate (g O2 kg−1 TS); 64 the stoichiometric molecular weight of methane combustion (g mol−1). (1) BMP thCOD = n CH4 RT pVS added (2) n CH4= CODtotal 64 17571707, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/gcbb.13025 by Universidad De Sevilla, Wiley Online Library on [13/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 336 | de la LAMACALVENTE et al. where BMPthAtC is the theoretical methane yield ( NLCH4 kg−1 VSadded); n, a, b and c are the stoichiometric parameters of C, H, O and N, respectively, based on the atomic composition of the substrates. 2.5.2 | Kinetic models Firstorder kinetic model The kinetics behaviour and the process performance of the AD of the nine sorghum straw varieties studied were analysed by the following firstorder kinetic model: where G is the cumulative specific methane production (NL CH4 kg−1 VSadded), Gmax the ultimate methane production (NL CH4 kg−1 VSadded), k the specific rate constant (days−1) and t the digestion time (days). This kinetic model is normally applied to assess the kinetics of the batch AD processes for different types of biodegradable substrates (Li et al.,2012; Scarcelli et al.,2020). This model assumes that the microbial cell mass does not affect the methane production, which is treated as onlyproportional to the amount of substrate (Wang et al.,2017). Transference function model In addition, to the firstorder model, a second mathematical model, for example, the TF model was assessed as suitable to fit the experimental data of the abovementioned BMP results (Equation5): where B (NL CH4 kg−1 VSadded) is the cumulative specific methane production, Bmax (NL CH4 kg−1 VSadded) is the ultimate methane production, Rmax the maximum methane production rate (NL CH4 (kg VSadded·d)−1), t (d) is the digestion time and λ (d) is the lag time. This model assumes that the kinetics parameters of methane production are proportional to the bacterial growth rate inside the batch reactors (Wang et al.,2017), and it has been applied successfully by several authors (Donoso- Bravo et al., 2010; Fernández- Rodríguez et al.,2019; Li et al.,2012). The goodness- of- fit and the accuracy of the results for both models have been assessed by the determination coefficients (R2) and the standard errors of estimates (S.E.E.). The kinetic parameters for each experiment and mathematical adjustment were determined numerically from the experimental data obtained by nonlinear regression using the software Sigma- Plot (version 11). 2.5.3 | Statistical analysis All the analyses and results were at least carried out in triplicates and values are given by means (standard deviation, SD). For single comparisons, a twotale Student's ttest was carried out, while for multiple comparisons, the oneway analysis of variance (ANOVA) and the Duncan's multiple range test was the selected method. For the purposes of data discussion, a p < 0.05 value was accepted as statistically significant. 3 | RESULTS 3.1 | Plant biometry and biomass Both the landrace cultivars and the public genotype cultivar were significantly taller plants than the commercial hybrid cultivars (Figure2a). The highest cultivars were TG, PAN, MP and ZÑ (275 ± 6; 242 ± 9; 205 ± 3 and 202 ± 3 cm, respectively). Instead, the commercial hybrid cultivars presented a significantly lower height (PR87: 60 ± 3 cm and PR88: 46 ± 4 cm). The public genotype (P89) presented a height of 146 ± 5 cm. No significant differences were found between the heights obtained for the cultivars ZHG (156 ± 8cm), P89 (146 ± 5 cm) and ZH (164 ± 4 cm; Figure2a). Regarding the biomass obtained (Figure2b), the cultivars with a dry weight shoot significantly higher than the other ones studied were ZHG (53 ± 3g) and TG (40 ± 2 g). The commercial hybrid cultivars reached a significantly lower biomass than all the other cultivars studied in this work (PR87, 15 ± 1g; PR88, 9.5 ± 0.8 g). No significant differences were found between the public genotypes P89 and the landrace cultivars MP and ZH (23 ± 2; 23 ± 1 and 26 ± 1 g, respectively) but these data were significantly lower than those obtained for CÑ, PAN and ZH (28.8 ± 0.9; 30 ± 2 and 26 ± 1 g, respectively) (Figure2b). The cultivar that presented the highest compactness was ZHG (0.36 ± 0.04 g cm−1), showing very tall stems and a high biomass (Figure2c). No significant differences were found between the compactness of the 6 landrace cultivars studied (Figure2c). ZH (0.159 ± 0.005 g cm−1) did not present statically significant differences either with the public phenotype cultivar (P89, 0.157 ± 0.008 g cm−1) or with the commercial hybrid cultivar PR88 (0.21 ± 0.01 g cm−1). However, ZH did present significant compactness differences with the (3) BMP thAtC = 22.4(n∕2+a∕8−b∕4−3c∕8) 12n+a+16b+14c (4) G =G max ⋅ [ 1−exp (−k⋅t) ] (5) B =B max × ( 1−exp [ − ( R max (t−𝜆) ) ∕B max]) 17571707, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/gcbb.13025 by Universidad De Sevilla, Wiley Online Library on [13/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 337 de la LAMACALVENTE et al. commercial hybrid cultivar PR87 (0.25 ± 0.02 g cm−1). The PR87 commercial hybrid cultivar presented a greater compactness than that presented by ZH landrace cultivar (Figure2c). The variety with the highest seed weight was the public phenotype (Figure2d: p89, 4.36 ± 0.1 g 100 seed −1), followed by the landrace cultivar ZH (3.9 ± 0.1 g 100 seed −1) and for the commercial hybrid PR87 (2.8 ± 0.2 g 100 seed −1). Very similar seed weights were obtained from the other varieties (Figure2d). 3.2 | Use of Sorghum bicolor (L.) Moench for biogas recovery 3.2.1 | Sorghum bicolor (L.) Moench cultivars' main characteristics The main characteristics of the nine different varieties of S. bicolor studied are presented in Table1. The VS content of all varieties was higher than 809 ± 11 g kg−1. Highlighting the volatile solids content of CÑ (862 ± 5 g kg−1), there being no significant differences between the VS content of the landrance cultivars PAN and TG (859 ± 6 and 852 ± 5 g kg−1), respectively. The S. bicolor straw of the nine different varieties studied presented a high VS/TS ratio (>88%) reaching values of 94% for CÑ variety and 92% for PAN, ZHG and TG varieties. The CODtotal values obtained for the different S. bicolor straws studied were higher than 980 ± 21 g O2 kg−1 (Table1). The nine varieties studied allowed to establish three groups in relation to their C/N contents. The first group includes the cultivars with the lowest C/N ratio (PR87, 77 ± 14; PR88, 53 ± 4) the public genotype (P89) and the landrace cultivar MP which presented C/N ratio values of 61 ± 2 and 72 ± 1, respectively. Another group would be for the TG, CÑ and ZH cultivars with C/N ratio values of 107 ± 7, 150 ± 80 and 281 ± 78, respectively. Finally, ZHG and PAN cultivar showed C/N ratios higher than 300. In FIGURE 2 Plant height (a), dry weight shoot (b), compactness (c) and seed weight (d) of the two commercial hybrid cultivars (PR87 and PR88), the public genotype (P89) and the six landrace cultivars (CÑ, MP, PAN, TG, ZH and ZHG) of Sorghum bicolor. Values represent means ± standard deviation. Different letters show significant differences according to Ducan's multiple range test, p < 0.05. 17571707, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/gcbb.13025 by Universidad De Sevilla, Wiley Online Library on [13/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 338 | de la LAMACALVENTE et al. previous studies, differences in the C/N ratio content have been observed, being highly dependent on the degree of drought of the biomass under study. Drought generally causes an imbalance between carbon and nitrogen, which is reflected in an increase in the carbon:nitrogen (C:N) ratio in mature leaves (Chen et al.,2015). 3.2.2 | AD and biodegradability The final methane yield for the nine varieties studied after 31 days of testing ranged between 413 ± 79 NL CH4 kg−1 VS and 287 ± 5 NL CH4 kg−1 VS (Figure3). The landrace cultivar ZH, TG and CÑ obtained the highest final methane yield (413 ± 79, 399 ± 18 and 387 ± 34 NL CH4 kg−1 VS, respectively). Instead, the lowest biogas productions were obtained with the commercial hybrid cultivar PR88, PR87 and with the public genotype P89 (287 ± 5, 316 ± 35 and 318 ± 22 NL CH4 kg−1 VS, respectively). Total COD- based biodegradability (Equation2) ranged between 60% and 90% for genotypes PAN and ZH, respectively. Regarding the elemental analysis based on biodegradability (Equation3), it ranged between 63% and 91% for landrace cultivars PAN and ZH, respectively. 3.3 | Estimation of the model parameters by kinetic modelling 3.3.1 | Firstorder kinetic model Table 3 shows the kinetic parameters obtained when the firstorder model was applied to the methane production experimental results. The low values of the SEEs and the high R2 values confirmed the suitability of the model to describe the experimental results. The highest specific rate constant obtained in the present research was found for the commercial hybrid cultivar PR88 (0.148 ± 0.008 days−1). This value was 32.1% and 49.4% higher than that obtained for the landrace cultivars ZH and ZHG respectively, which presented the highest values among the different traditional varieties tested in the present research (0.112 ± 0.006 and 0.099 ± 0.004 days−1, respectively). On the other hand, the landrace cultivars PAN (0.064 ± 0.005 days−1), CÑ (0.084 ± 0.004 days−1) and TG (0.067 ± 0.003 days−1) showed the lowest k values although these last two cultivars showed high values of the ultimate methane yield, Gmax (409 ± 9 and 435 ± 12 NL CH4 kg−1 VS, respectively). 3.3.2 | TF model Table 4 summarizes the parameters obtained for the nine sorghum straw varieties tested in this study when the TF model was applied. The low S.E.E.s and the high R2 confirmed the acceptable fit of the experimental results of methane productiontime to the suggested model. As can be seen in Table 4, the highest Rmax values were found for the landrace cultivar ZH (45 ± 2 NL CH4 (kg VS·d)−1) and the commercial hybrid cultivar PR88 (Pioneer Hi- Breed) (41 ± 2 NL CH4 (kg VS·d)−1) as also occurred with the k constant from the firstorder kinetic model. These Rmax values were 87.1% and 71.3% higher, respectively, than the lowest value, which was found for the landrace cultivar PAN (24 (1) NL CH4 (kg VS·d)−1). It is also relevant to stand out that the S. bicolor landrace cultivar ZH simultaneously also showed high Rmax and ultimate methane yield, Bmax (402 ± 9 NL CH4 kg−1 VS) values, which revealed that this landrace cultivar generates one of the highest methane yields with one of the highest rates. 3.4 | Methane production, biomass yield and methane yield per shoot Table2 comparatively shows the values of methane yield, biomass straw yield and methane yield per shoot. Comparing the biogas production and the total biomass of the plants obtained (Table2), the landrace TABLE 1 Main characteristics of the nine Sorghum bicolor straw varieties tested Parameters PR87 PR88 P89 CÑ MP PAN TG ZH ZHG ST (g kg−1) 922 ± 4 917 ± 5 924 ± 2 917 ± 4 916 ± 1 934 ± 3 927 ± 2 916 ± 5 918 ± 4 SV (g kg−1) 838 ± 5 809 ± 11 817 ± 2 862 ± 5 832 ± 4 859 ± 6 852 ± 5 829 ± 8 841 ± 5 SV/ST 0.91 0.88 0.88 0.94 0.91 0.92 0.92 0.90 0.92 CODtotal (g O2 kg−1) 1026 ± 42 1017 ± 65 980 ± 21 981 ± 32 1062 ± 32 1092 ± 24 1079 ± 27 1018 ± 26 1702 ± 124 C/N 77 ± 14 53 ± 4 61 ± 2 150 ± 80 72 ± 1 >500 107 ± 7 281 ± 78 304 ± 95 Note: The two commercial hybrids cultivars (PR87 and PR88), the public genotype (P89) and the six landrace cultivars (CÑ, MP, PAN, TG, ZH and ZHG) of Sorghum bicolor. Values represent the mean ± standard deviation. 17571707, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/gcbb.13025 by Universidad De Sevilla, Wiley Online Library on [13/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 339 de la LAMACALVENTE et al. cultivars ZHG and TG were the ones that obtained the highest methane per biomass production (13.7 and 12.7 NL CH4 shoot unit−1, respectively). The landrace cultivars CÑ and ZH obtained values of 9.6 and 8.9 NL CH4 shoot unit−1. By contrast, the commercial varieties again were the ones that obtained the lowest methane per biomass production (PR87, 3.8 NL CH4 shoot unit−1 and PR88, 2.2 NL CH4 shoot unit−1). 4 | DISCUSSION Sorghum bicolor is known to be a lowdemanding crop and very well adapted to marginal environments. Sorghum bicolor is worldwide grown for grain as well as for livestock production. When sorghum crop is used for forage production, the main drawback is that must to be harvested and processed quickly to avoid nutritional losses (Tabacco et al.,2011). Exploring the genetic diversity of sorghum germplasm, it has been selected six landrace cultivars based on their longitudinal distribution along the Iberian Peninsula, focusing mainly in cultivars from Andalusian, in the south of Spain (Figure1). To characterize the six landrace cultivars of S. bicolor (CÑ, MP, PAN, TG, ZH, ZHG) and compare with the two commercial hybrid cultivars (PR87 and PR88) and the public one (P89), the height and biomass of each cultivar were studied, as well as the weight of the FIGURE 3 Variation in methane production versus time for the different substrates tested. The two commercial hybrids cultivars (PR87 and PR88), the public genotype (P89) and the six landrace cultivars (CÑ, MP, PAN, TG, ZH and ZHG) of Sorghum bicolor. Values represent means ± standard deviation. TABLE 2 Biodegradability, methane yield, biomass straw and methane yield per shoot of the nine Sorghum bicolor straw varieties tested PR87 PR88 P89 CÑ MP PAN TG ZH ZHG Biodegradability CODtotal (%) 66 64 74 80 75 60 80 90 74 Biodegradability C/N (%) 84 67 78 89 83 63 90 91 73 Methane yield (NL CH4 kg−1 SV) 316 ± 35 287 ± 5 318 ± 22 387 ± 34 344 ± 22 327 ± 42 399 ± 18 413 ± 41 315 ± 14 Biomass yield (g SV shoot unit−1) 12.5 7.7 18.8 24.8 18.9 26.4 33.8 21.6 44.8 Methane yield per shoot (NL CH4 shoot unit−1) 3.8 2.2 6.0 9.6 6.5 7.3 12.7 8.9 13.7 Note: The two commercial hybrids cultivars (PR87 and PR88), the public genotype (P89) and the six landrace cultivars (CÑ, MP, PAN, TG, ZH and ZHG) of Sorghum bicolor. Values represent the mean ± standard deviation. 17571707, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/gcbb.13025 by Universidad De Sevilla, Wiley Online Library on [13/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 340 | de la LAMACALVENTE et al. grains they produced. The height values of the cultivar obtained (Figure2a) agree with previous values reported by Mekasha et al.(2022) and are intermediate height values between those published for tall (<323 cm) and short type S. bicolor varieties (<182 cm) (Yosef et al.,2009). The previous studies have reported the possibility of increasing plant height by adding extra water to plants grown under drought conditions (Yosef et al.,2009). Regarding the biomass obtained (Figure2b), the cultivars with a dry weight shoot significantly higher than the other ones studied were ZHG and TG. The commercial hybrid cultivars reached a significantly lower biomass than all the other cultivars studied in this work. No significant differences were found between the public genotypes P89 and the landrace cultivars MP and ZH, respectively, but these data were significantly lower than those obtained for CÑ, PAN and ZH (Figure2b). The compactness is a useful parameter to evaluate the crop growth and development. ZHG was the cultivar that presented the highest compactness (Figure2c). The variety with the highest seed weight was the public phenotype (Figure2d) followed by the landrace cultivar ZH and for the commercial hybrid PR87. Since the 1960s, and later with the green revolution, the use of local varieties has been decreasing, favouring the use of commercial ones. Commercial hybrid cultivars are usually smaller size plants and good seed production, but landrace cultivars are usually better adapted to the climatic conditions of the area (Di Miceli et al.,2022). The recovery of local varieties would not only entail a wealth of genetic variety with greater ease of adaptation to both biotic and abiotic factors, but it would also provide more productive and highquality varieties (Di Miceli et al.,2022). In 2018, the European Parliament directive on waste (Directive (EU) 2018/851) was approved and the Spanish transposition was approved last April 2022. This law prohibits the stubble burning backing the recovery of the straw stubble as a fundamental basis. The methane production from the S. bicolor straw would not only avoid the process of burning it, but would also provide added value to the sowing of this cereal (Rakascan et al.,2021). All VS/TS ratio values obtained for the biomass of S. bicolor straw were higher than 88% (Table1), indicating the suitability as substrates for AD (Wang et al., 2016). CODtotal values confirmed the high organic matter content of the studied substrates. The most important characteristic for biogas production by AD is the organic fraction of the feed since it is the degradation of shortchain organic compounds the responsible to drive the evolution of the anaerobic process (Wang et al.,2016). The C/N ratio in all the S. bicolor straws studied was higher than the values established as suitable for the anaerobic process (20– 30), highlighting the lignocellulosic content of the substrates (Table1). The lignocellulosic biomass contains a high content of cellulose and hemicellulose which are ideal substrates for AD, but the low substrate nitrogen content and the complex lignocellulosic biomass framework can inhibit the anaerobic process (Paul & Dutta,2018). In this work, the feasibility of using the straw of different varieties of S. bicolor as the substrate for AD to obtain energy was studied. MCC used as a positive control achieved a final methane yield of 379 ± 16 NL CH4 kg −1 VS within the range required for validation. The obtained cellulose methane yield represented a 92% of the theoretical yield for this substrate and thus the high methanogenic activity of the sludge was confirmed (Holliger et al.,2016). Despite the high C/N values observed in the different varieties of S. bicolor, the methanogenesis stage did not show any sign of inhibition (Figure3). During AD batch assay, all S. bicolor varieties tested showed a slight delay in methane production, which lasted approximately 1 day, due to the lack of readily biodegradable material and the difficulty of hydrolytic bacteria in degrading the lignocellulosic material (Fernández- Rodríguez et al.,2022). Then, from day 1 an almost exponential methane production growth was observed, which lasted about 15 days. Finally, during the last 5 days, the daily methane production dropped achieving values of almost zero (Figure3). Głąb et al. (2019) obtained methane production values comparable with those reported in the present work. No differences in final methane yield using different fertilizers were found, instead the final methane yield was affected by the cellulose, hemicellulose and lignin concentration (Głąb et al.,2019). Values of final methane yield between 232 and 427 NmLCH4.gVS−1 were reported for 57 genotypes of different varieties of S. bicolor (Thomas et al., 2021). In this study, the greatest differences in methane production were related to stem structure. They showed that both biochemical and histological factors of the S. bicolor stem had a decisive influence on the final methane yield (Thomas et al.,2021). Another study about the AD of S. bicolor, carried out with ensiled one, reported values between 231.25 and 321.31 NL CH4 kg−1 VS (Pasteris et al.,2021). Pasteris et al.(2021) observed a great variability in the final methane yield depending on the S. bicolor cultivar and informed about the need of studying each landrace S. bicolor cultivars biomass to produce methane. Anaerobic biodegradability was determined based on the CODtotal concentration and on the elemental analysis (Nielfa et al.,2015) obtaining in both cases very similar values (Table2). The highest specific rate constant obtained by the firstorder kinetic model in the present research was found for the commercial hybrid cultivar PR88 (Table3). It is worth 17571707, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/gcbb.13025 by Universidad De Sevilla, Wiley Online Library on [13/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License