The seeds of sweet sorghum of the cultivated variety (cv.) Botanichnyi were kindly provided by Prof. D. Rakhmetov, M.M. Gryshko National Botanical Garden at the National Academy of Sciences (NAS) of Ukraine [37]. The S. saccharatum plants were grown in a temperate continental climate zone, Shostka district of Sumy region of Ukraine, situated in the north-east part of the country (51°42′N, 33°12′E, 125 m a.s.l.), within a geographical zone named Polissia. The seeds were planted at the beginning of May, in a soddy podzolic soil, with 45 сm row spacing and 10 cm distance between the plants in a row. The crops were harvested in September after the kernels had matured (Fig. 1A).

Fig. (1). Sweet sorghum, cv. Botanichnyi: A – grown on the field; B – squeezed on the roller press.

The harvested sorghum leaves and peeled panicles were squeezed on the roller press producing 1.2 kg of juice out of 3 kg of the biomass. Then, sweet sorghum juice was filtered and concentrated under vacuum in a rotary evaporator at 60±2 °C. The concentrated juice (syrup) was stored at 4 °C. The methods of the chemical analyses are outlined in section 2.3.

Subsequently, the syrup was diluted with tap water to fractions containing 140 g/kg TS, 250 g/kg TS, 400 g/kg TS, autoclaved for 20 min at 121 °C, and used as a yeast growth medium, fermentation medium, and as a substrate for continuous fermentation accordingly. Other components of fermentation media (urea and orthophosphoric acid) were sterilized separately under the same conditions.

S. cerevisiae Meyen ex E.C. Hansen, Medd. Carlsberg Lab.: 29 (1883) Y-М-05 was obtained from the National Collection of Microorganism Strains and Plant Lines for Food and Agricultural Biotechnology of the Institute of Food Biotechnology and Genomics of the National Academy of Sciences of Ukraine. S. cerevisiae Y-M-05 strain was isolated from the surface of millet grain, it is suitable for yeast biomass generation purposes and ethanol production. The stock culture was maintained on 80 g/kg TS barley malt wort agar slants at 4 °C. Barley malt wort was purchased from beer manufacturer Obolon CJSC, Kyiv, Ukraine.

100 g/kg TS sterile barley malt wort was poured into agar slants with malt wort and S. cerevisiae Y-М-05 culture and stored for 3 h at 30 °C. Pre-cultured inoculum liquid was transferred from tubes to 500 ml flask with 300 ml sterile sweet sorghum 140 g/kg TS diluted syrup and cultivated aerobically for 24 h at 30±1 °C on the rotary shaker at 150 rpm. Then, the content of the flask with cultured broth and S. cerevisiae cells was transferred to 3 L Erlenmeyer flask with 1.5 L of 140 g/kg TS diluted syrup and cultivated aerobically at the same conditions as the previous cultivation.

Supplementing the sorghum juice medium with a nitrogen source can stimulate yeast cell growth and ethanol production [38]. For this reason, 0.33 g/kg or 0.64 g/kg urea was added to sweet sorghum diluted syrup. The control syrup had no urea added. For this experiment, 500 ml Erlenmeyer flasks with 300 ml of 140 g/kg TS sterile sorghum syrup were inoculated with 30 ml of seed culture and incubated at aerobic conditions for 12 h at 30±1 °C on the rotary shaker at 150 rpm.

Seed culture for bioethanol production was prepared in five 3 L Erlenmeyer flasks with 1.5 L of 140 g/kg TS sterile sorghum syrup and 300 ml inoculum, cultivated aerobically at 30±1 °C on the rotary shaker at 150 rpm. The selected amount of urea (0.33 g/kg) and 0.09 g/kg of 70% orthophosphoric acid were added to the syrup as a source of phosphorus to facilitate the biomass accumulation and to increase the ethanol output [38, 39]. When Brix of the fermentation broth decreased twofold at the exponential yeast growth phase, 15 ml of the inoculum was transferred to 1 L Erlenmeyer flask with 600 ml working volume comprising: sorghum 250 g/kg TS diluted syrup with 15 drops of Antifoam FD 20 PK for the anaerobic phase of alcoholic fermentation. Initial pH was maintained at 4.7 with 1 N sulfuric acid.

The general diagram of conducting the sorghum diluted syrup fermentation in flasks is shown in Fig. (2). Control fermentations were performed at atmospheric pressure, while two test treatments were conducted after 18th h of fermentation to allow ethanol to accumulate in the cultured milieu. The treatment (1) consisted of 10 min periods at 20±5 mmHg reduced pressure and the treatment (2) was delivered at a constant 80±5 mmHg pressure throughout the fermentation period.

All flasks were placed on heated magnetic stirrers ММ3M (Mukachiv Factory Complete Laboratories JSC, Ukraine). The flasks under atmospheric pressure were closed with rubber stoppers containing two holes each for the thermometer and hydraulic seal.

To carry out the part of the experiment under reduced pressure (Fig. 2), the flasks were connected through a Vyurtz nozzle with a thermometer and Liebig condenser (500 mm length). On the other side, the condenser was joined to a receiving flask via an allonge for vacuum extraction. The round-bottom receiving flask was immersed in a container with ice. The vacuum pressure was maintained using a water jet pump generating a vacuum of maximum 10-20 mmHg, pumping speed of 0.15 L / sec. The water jet pump was connected via a vacuum meter and a tee to the allonge. The tee was used to control the vacuum and to stabilize the system for a long time under the relatively constant vacuum. The vacuum level was measured using a laboratory mercury vacuum meter. Yeast cells were dyed with methylene blue and counted with a Neubauer haemocytometer [40], using a JENAVAL optical microscope (Carl Zeiss, Germany).

Fig. (2). General scheme of sorghum diluted syrup fermentation (all experiments were performed in triplicate): 1 – a flask at atmospheric pressure; 2 – a flask at atmospheric pressure with repeated (10 min) connection to a vacuum system for boiling the medium and maintaining boiling (20±5 mmHg); 3 – a flask at constant reduced pressure without boiling (80±5 mmHg); 4 – Liebig condenser; 5 – tee; 6 – vacuum meter; 7 – water jet pump; 8 – thermometer; 9 – hydraulic seal; 10 – allonge for collecting fractions under vacuum; 11 – receiving flask; 12 – reservoir with ice; 13 – Vyurtz nozzle; 14 – heated magnetic stirrer.

To scale up the process, a 5 L fermenter was used, (Fig. 3). The inoculum preparation for fermentation in the bioreactor was the same as for reduced pressure experiment in flasks. The adaptation of inoculum was performed in the fermenter for 2 h at 30±1 °C. Yeast seeds (20 g/L) were placed into a 2 L displacement filled with 140 g/kg TS sorghum syrup supplemented with 0.33 g/kg urea and 0.09 g/kg of 70% orthophosphoric acid. Yeast adaptation and growth took place in aerobic conditions by bubbling filtered air through at 3 L/L, with a compressor. For yeast growth at atmospheric conditions, a hydraulic seal was used.

Bioethanol production was performed in a 5 L fermenter with 250 g/kg TS sorghum syrup, pH 4.7 was adjusted with 1 N sulfuric acid, in anaerobic conditions at 30±1 °C. The vacuum extraction was conducted at 20±5 mmHg for the separate 1.25 L fractions of the medium in 2 L Erlenmeyer flasks for 8 min. Following the extraction, the content of the flask was returned to the bioreactor for further fermentation.

The parameters of sweet sorghum syrup, including moisture (water and TS content), ash, pH, titratable acidity were determined according to the AOAC methods [41]. To analyze the water content, the samples were dried at 105 ºC until constant weight [42]. The ash content was established by incineration at 600 ºC until constant weight [42]. The pH value was checked with pH meter millivoltmeter pH-150-M model (Gomel Plant of Measuring Devices, the Republic of Belarus) standardized against standard buffer solutions.

The content of reducing sugars was analyzed by Lane and Eynon method (ISO 5377:1981) by titration the sample to a Fehling's solution with methylene blue as an internal indicator. The total sugars (reducing sugars + sucrose) content was analyzed by Lane and Eynon method after the inversion of sucrose with hydrochloric acid at 69±1 °C. Alternatively, digestible carbohydrate content was tested by the biological fermentation method (GOST 31683-2012).

Fig. (3). General view of the fermentation reactor for bioethanol production: 1 – the working chamber of the bioreactor; 2 – hydraulic seal; 3 – thermostatic bath with circuit system; 4 – fermenter control unit with an air compressor; 5 – electric drive of the mixer; 6 – electric drive of mechanical defoamer; 7 – air filter; 8 – thermoregulatory unit; 9 – power supply of the thermoregulatory unit.

Calcium content in sweet sorghum syrup was determined by EDTA complexometric titration method [43]. The nitrate content was evaluated by the titrimetric method after the reduction to ammonia under alkaline conditions with Devarda’s alloy as the reducing agent [44]. The nitrogen content was defined by Kjelhdal’s method after titration of distilled ammonia with K₂SO₄ and CuO used as catalysts for acid mineralization at 1.4±0.14 kW heating electrical power in UDC-7 mineralizer (Spectro Lab LTD, Ukraine) [45]. The total phosphorus content was tested in the ash [46] by the colorimetric Brigg’s method [47].

The fermentation process was monitored via analysis of the alcohol content in the distilled samples and measuring the Brix of the medium and broth using saccharometer AC-3 (Shatlyhin & Co LTD, Ukraine). Following the extraction of volatile substances and water vapor, the Brix value of the distillery sludge was measured after adjusting it to the volume of the culture liquid sample. Undigested sugars in mature culture liquid were determined by the colorimetric micro method with resorcinol [48] using the Concentration Photoelectric Photometer model KFK-3-01 (Zagorsk Optical-Mechanical Plant, Russia).

The bioethanol yield (Y_P/S) was calculated [49] as the actual ethanol produced and expressed as the weight to weight ratio, g ethanol per g total sugar utilized (g/g). The bioethanol productivity (Q_p, g L^-1 h^-1) and the conversion efficiency of bioethanol yield (E_y, %) were calculated by the following equation [49]:

where P – the actual bioethanol concentration produced (g L^-1), t – the fermentation time (h) and 0.54 is the maximum theoretical bioethanol yield of sucrose consumption.

All measurements have been done in triplicates and the data presented are the mean values of three independent experiments expressed as mean ± errors. The data have been statistically analyzed by t-test using Origin 2016, OriginLab Corporation, USA. Differences between means at 5% (p < 0.05) level have been considered significant.

The sorghum biomass yield at the time of harvest in the ripen seeds vegetation phase reached 92 t/ha, the stem juice yield was 36.8 t/ha, 14 °Brix, 99.5 g/kg of fermented carbohydrates. As was previously reported by Widianto et al. [50], sweet-stalk sorghum juice does not contain substances inhibiting yeast growth, which is testified with the lag phase duration similar to the standard medium. The main fermentable sugars in sorghum juice are sucrose, glucose, and fructose [29, 50, 51]. Fructose is mostly utilized as the last option [50]. In the tested sample of sweet sorghum syrup, sucrose accounted for approximately 68% of fermented sugars and the reducing sugar share amounted to 32% of the total fermented sugars output (Table 1), which matches the literature (65-68% sucrose, 17-22% glucose, 12-15% fructose) [52]. The differences in sugar content and variations in their distribution in the sorghum juice might be attributable to the impact of environmental factors, managing practices, soil properties [52], and to a large extent, to different sorghum varieties [17]. A linear relationship was established between the total sugar content in the sorghum juice and Brix data, which can be convenient to estimate the potential ethanol output of the raw source by the field analytical test [17].

Appropriate nutritional conditions such as sufficient nitrogen levels in accessible form affect the adaptation of S. cerevisiae cells to higher ethanol concentrations [53]. Davila-Gomez et al. showed [11] that concentration of soluble nitrogen-containing compounds in sorghum juice depends on the variety, but is generally below the level, required for structural and enzymatic protein synthesis in S. cerevisiae, hence nitrogen supplementation should be necessary for quality sustainable fermentation. The most efficient nitrogen sources for bioethanol production are ammonium sulfate and urea [20]. Yue et al. showed [54] that adding urea to sweet sorghum helped in the growth of maximum ethanol concentration in the broth as well as increased the number of yeast cells. In contrast, ammonium sulfate did not show positive effects. Previous studies have shown that urea, added as a nitrogen source, protects the yeast from osmotic stress [38, 55], and acts as a molecular filter binding to proteins and preventing their destruction during alcohol production [30, 55].

The effect of urea addition on sugar utilization is shown in Fig. (4). Adding urea significantly impacted the degradation of medium components starting from the 5th h of yeast cultivation. Control values started to drop slowly from 5th to 7^th h which can be attributed to the adaptation of yeast cells to urea supplementation. When Brix of the supplemented fermentation broth decreased two-fold, the addition of 0.33 g/kg and 0.64 g/kg urea caused practically no significant differences in the fermentation efficiency in sweet sorghum syrup medium.

о – the control means and test values are not significantly different (p < 0.05);

+ – the control means and test values are significantly different (p < 0.05);

* – the control means are significantly different compared to test values (p < 0.05), and test values are not significantly different (p < 0.05).

The data in Fig. (4) shows that adding 0.33 g urea to 1 kg of diluted 140 g/kg TS syrup (0.15 g/kg nitrogen) provided the best result while further increase of urea supplementation did not lead to a significant reduction in dissolved solids in the medium, which is consistent with prior research [39]. However, in the case of very high gravity (VHG) due to adding extra sugar to the sorghum juice, or when the concentrated juice contains over 270 g/L TS [53], much greater supplementation of urea is required [38, 56].

Parameters of fermentation processes under reduced pressure depend on the physical properties of the ethanol-water mixture and the biochemical properties of the yeast fermentation process [57].

In this experiment, under the lowest pressure conditions, the ethanol concentration in the broth at the end of the fermentation reached maximum 5.4% vol. while this parameter under atmospheric pressure produced 11% vol. ethanol (Fig. 5B). Higher alcohol content was obtained in glucose-yeast-peptone fermentation by S. cerevisiae: 6.33% vol. ethanol in a flask under 47 mmHg pressure and 14.75% vol. under atmospheric pressure conditions [57]. Although 10-12% of ethanol concentration strongly depresses the specific ethanol generation rate and the specific growth rate of the yeast cells [27-29], significant metabolite changes can be observed at 5% vol [58].

The lower concentration of ethanol in flasks under vacuum conditions resulted in substantial bioethanol productivity compared to control under atmospheric pressure: 22% higher under constant reduced pressure and 55% higher under periodic reduced pressure (Table 2). Periodic reduced pressure allowed the fermentation medium to boil periodically which resulted in significantly higher (33%) ethanol extraction. The Brix parameter was also the highest in periodic reduced pressure mode, (Fig. 5A, C) for the reason of boiling-off the water vapor and a higher concentration of cultured broth. At the same time, the residual undigested sugars in mature cultured broth were 65% higher at atmospheric pressure compared to periodic pressure treatment (Table 2).

Fig. (4). Time course of Brix in sweet sorghum syrup medium with addition of nitrogen source at 0.33 g/kg and 0.64 g/kg urea (test values) and no urea addition (control).

The process of bioethanol production can be prolonged and enhanced by applying the reduced pressure fermentation approach and with the addition of more feeding, since the inhibiting concentrations of ethanol in cultured broth are accumulated slower than under atmospheric fermentation. We performed two additional feedings in fermentations with a reduced pressure approach (Fig. 5A), which resulted in the increase of Brix (Fig. 5A, C). After the second feeding, ethanol concentration decreased due to dilution with a new portion of medium (Fig. 5B). In the case of fermentation under atmospheric pressure, only one additional feeding was applied (Fig. 5A).

Fig. (5). Time course of S. saccharatum cv. Botanichnyi syrup fermentation under reduced and atmospheric pressure: (A) – Brix of cultured broth; (B) – ethanol concentrations in the cultured broth; (C) – Brix of cultured broth after distillation.

According to Samnuknit and Boontawan [59], in the case of extractive fermentation under reduced pressure, feeding supplementation can be repeated up to 9 times with the same inoculum of yeast. However, after that, yeast cell viability weakened presumably due to the accumulation of approximately 70 g/L of lactic acid. We believe that the accumulation of non-volatile metabolites can be further limited by cultural liquid output as was also suggested by literature data [60].

It is generally accepted that fermentation conditions affect yeast cell viability and reproduction capacity. As demonstrated by Xie et al. [61], fermentation under reduced pressure can inhibit yeast reproduction although the ethanol productivity gets enhanced. However, our data show that pressure does not particularly affect pH: while the initial pH was 4.7, the final pH after fermentation was 4.7±0.1 and 4.8±0.1 under reduced and atmospheric pressure, respectively.

In the case of ethanol fermentation under vacuum, yeast cells suffer oxygen deficiency stress, intracellular and extracellular pressure difference, and accumulation of non-volatile toxic metabolites [62]. Prior to comprehensive analyses of proteomic [63] and lipidomic [64], adaptive responses of S. cerevisiae to repeated vacuum fermentation explain the higher viability of cells compared to the same parameter under constant reduced pressure conditions.

As presented in Table 2, the highest bioethanol output, efficiency and productivity of the process were received with periodically reduced pressure fermentation at the level of vacuum close to the boiling point of the cultured broth. At the same time, this mode did not significantly affect the final yeast cell concentration that allowed a more complete utilization of the sorghum syrup sugars (Table 2). The maximal ethanol output was received when its concentration was below 5.5% vol (Fig. 5B).

The fermentation process of the yeast occurs at the temperature 30-35 °C. Depending on the ethanol-water concentration, the mixture boils within the range from 78.3°C up to 100°C [57]. When the fermentation process takes place under reduced pressure, the boiling point of this mixture also decreases. The main feature of volatile components separation from the alcohol mash is the possibility of boiling the cultured liquid under reduced pressure and the temperature of fermentation 30-40 °C. The advantage of fermentation under reduced pressure is the possibility to minimize the inhibition of the producer by volatile products of metabolism. Nevertheless, a thorough technical and financial analysis of the reduced pressure fermentation technology is required before a decision can be made on its practical industrial application.

These scaling up experiments were performed in a continuous stirred 5 L tank reactor (CSTR) (Fig. 3). It allowed to increase the volume more than 8 times (from 600 ml working volume in the flask), automatic control of the temperature, aeration and defoaming as well as stirring throughout the fermentation. Periodic vacuum extraction of a quarter of the cultured broth at 20±5 mmHg was conducted as shown in Fig. (6). This treatment helped to maintain the ethanol concentration in the broth at 5.2% vol. prior to distillation and enabled feeding with 160 ml portions of 400 g/kg TS sorghum syrup as shown in Fig. (7).

The parameters of yeast survival rate, undigested sugars of cultured liquid, and bioethanol yield in the bioreactor were similar to those under periodic vacuum extraction in the flasks even though the medium was subjected to the reduced pressure only partially. This performance has been achieved due to a quality mass transfer in the bioreactor. The distillate obtained in the cold trap had an alcohol concentration of about 32% vol. and needed less energy for further dehydration than ethanol produced in the bioreactor. Such a positive feature of the bioethanol fermentation under reduced pressure was also reported by Nguyen et al. [57].

Fig. (6). General scheme of bioethanol production process from sorghum juice or syrup in a fermenter with periodic vacuum extraction of the cultured broth fraction.

Fig. (7). Time course of Brix and bioethanol concentration in sweet sorghum diluted syrup fermentation in 5 L bioreactor with periodical feeding and vacuum extraction of the cultured broth samples.

When creating a vacuum for the entire volume of the fermenter, its production rises in price because of the need for thick walls. To overcome such restrictions, fermentation under the atmospheric pressure, and vacuum extraction of broth portions in a separate fermentation tank (flash fermentation) was carried out and proposed for further implementation. The technology layout illustrated in Fig. (6) is suitable for implementation on bioethanol plants. It has the potential to increase the productivity due to avoiding inhibition of yeasts by high alcohol concentrations, the gradual feeding of additional sugar source and a prolonged use of the yeast biomass in the active growth phase via continuous extraction of volatile metabolites. VHG approach with high initial concentrations of sugars can also be performed with ethanol extraction at least from a part of the cultured liquid [60].