Introduction

The global coffee production was approximately 10 million tons in 2020, which has been steadily increasing yearly [1]. Additionally, it is known that 65% of coffee produced becomes spent coffee grounds (SCGs) after extracting the beverage [2, 3], which becomes mostly garbage [4, 5]. Based on the coffee made in 2020, SCGs are generated in approximately 6.6 million tons. Most SCGs are incinerated, thrown away, landfilled, and pollute the environment. Because efforts are being made to minimize petroleum consumption and greenhouse gas emissions following global climate change [6], research on using discarded SCGs as resources should be conducted.

The primary components of SCGs are approximately 50–60% carbohydrates, 6.7–13.7% protein, and 10–15% lipid [4, 7]. Most of the carbohydrates in SCGs exist in cellulose and hemicellulose [8]. Thus, sugars in SCGs can be extracted by hydrolysis and used for bioenergy and biochemical production, including diesel, ethanol, butanol, and polyhydroxyalkanoates (PHA) [9,10,11]. Nevertheless, there is no research on hydrogenase-based hydrogen production in dark fermentation using spent coffee grounds (SCGH).

The world is looking for sustainable energy to replace fossil-based resources and reduce global warming, and hydrogen is a representative resource. It has a high energy storage capacity of 121 MJ/kg and is an environmentally friendly energy source because it is oxidized and produces only water as a byproduct [12, 13]. Currently, most of the hydrogen is yielded by reforming fossil-based fuels. In this situation, hydrogen can be produced eco-friendly and carbon-neutrally through a biological process using SCGs. Fermentative bacteria can produce biohydrogen via dark and photo fermentation. Dark fermentation has the advantages of using different organic substrates and having a fast hydrogen production rate [14]. Nevertheless, biohydrogen still has a problem of low production produced. Theoretically, from 1-mol glucose, 4-mol hydrogen can be created when 2-mol acetate is produced, and 2-mol hydrogen can be made when 1-mol butyrate is produced [14]. However, it is difficult to achieve a hydrogen yield of more than 2 mol-H2/mol-consumed sugar in the actual reactor. This is because it also produces other metabolites such as lactate. Therefore, various approaches have been attempted to increase hydrogen productivity in dark fermentation. Recently, research using conductive material has been actively conducted. It has been shown that the physicochemical properties of the conductive material can not only increase performance of hydrogen production, but also enhance the electron transfer mechanism, and induce upregulation of related genes. Magnetite (Fe3O4) is a relatively inexpensive and readily available iron oxide and significantly increases the hydrogen production and yield because of its conductivity. Based on a study, adding magnetite increases sugar and lactate consumption, leading to higher hydrogen production and outcome [15].

This study aims to verify the applicability of SCGs in dark fermentative hydrogen production and to assess the impact of magnetite supplementation. Consequently, a series of experiments were performed as follows. i) Optimization of SCG hydrolysis conditions ii) Comparison of hydrogen production between synthetic substrates and SCGH iii) Evaluation of the impact of increasing hydrogen production by the magnetite supplementation in SCGH.

Materials and Methods

Strain and Hydrogen Production Using SCGH

Clostridium butyricum DSM10702 was used as hydrogen-producing bacteria. All batch tests were performed in serum bottles containing 100 mL of defined reinforced clostridial medium (RCM) in triplicate. Furthermore, it was sealed under anaerobic conditions using 99.9% N2 gas in an anaerobic chamber (Coy Laboratory Products, MI, USA). Defined RCM was composed as follows: peptone 10 g/L, beef extract 10 g/L, L-cysteine hydrochloride 0.5 g/L, yeast extract 3 g/L, sodium acetate 3 g/L, sodium chloride 5 g/L, sodium bicarbonate 5 g/L, and antifoam 100 μL/L.

Two experiments were performed to verify the applicability of SCGH as a substrate and to enhance biohydrogen productivity. First, the hydrogen production was conducted using SCGH media and simulated with mannose or galactose, known as SCGH's main sugar, to compare productivities. The media were defined as RCM containing 10 g/L of total sugar, and the pH was changed to 5.5 using HCl. The incubation conditions were 37 °C, 60 rpm, and 31 h. Second, 10 g/L of magnetite was supplemented with SCGH media to increase the hydrogen production. SCGH was diluted at a sugar concentration of 7.5 g/L, and the pH was adjusted to 5.5. The serum bottles were incubated at 37 °C and 60 rpm for 32 h.

Diluted-Acid Hydrolysis of SCGs

This experiment obtained SCGs from a local cafe in South Korea. SCGs were dried at 105 °C for 24 h. Dried SCGs were hydrolyzed at a solid/liquid ratio (S/L ratio) of 10–20% (w/w) and H2SO4 of 0.5–1.0% (w/w) to optimize hydrolysis conditions. Under each state, SCGs were hydrolyzed at 130 °C for 1 h with a working volume of 100 mL. Each sample was centrifuged at 3500 rpm for 40 min to separate the hydrolysate from the remaining solids. The hydrolysate analysis of sugar and organic matter was performed using high-performance liquid chromatography. Subsequently, SCGH was mass-produced under the optimal conditions derived above, sterilized at 121 °C for 15 min to remove the possibility of contamination, and refrigerated for future experiments.

Metabolites Analysis

Biogas and metabolites were analyzed under the same conditions as previous studies [16]. Produced gas in the serum bottle was measured using a glass syringe at 37 ℃ and 1 atm. Subsequently, the gas was sampled using the 1 mL syringe (Hamilton Company, NV, USA) was injected into a gas chromatography (6890 N, Agilent Technologies, CA, USA) to evaluate the hydrogen content. High purity N2 gas (99.999%) was used as the carrier gas, and the thermal conductivity detector was heated to 250 ℃. Variations in the composition of metabolites in the culture were analyzed using high-performance liquid chromatography. For analysis, a 1 mL sample was obtained and centrifuged for 10 min at 4 °C and 13,000 rpm. After that, the supernatant was filtered using a PTFE membrane filter with 0.45 μm of pore size. The analysis conditions were 5 mmol H2SO4 as a mobile phase, 300-mm × 7.8-mm Aminex HPX-87H ion excursion column, 60 °C of column temperature, 55 °C of detector temperature, refractive index detector, 25 μL of injection volume, 0.6 mL/min of flow rate, and 55 min of running time.

Kinetic analysis

The observed hydrogen and biogas were converted to values at standard pressure and temperature before being applied to the modified Gompertz model below [17].

$$ H = P \times \exp \left\{ { - \exp \left[ {\left( {\left( {R_{{\text{m}}} \times e} \right)} \right)/P \times \left( {\lambda - t} \right) + 1} \right]} \right\}, $$

H is the estimated hydrogen production (mL). P is the hydrogen production potential (mL). Rm is the maximum hydrogen production rate (mL/h), λ is the lag phase (h), t is the time (h), and e is 2.7182… [17].

Results and Discussion

Acid Hydrolysis of SCGs

To obtain the optimal hydrolysis conditions, SCGs were hydrolyzed under conditions of S/L ratio 10–20% (w/w), H2SO4 0.5%, 1.0% (w/w), 130 °C, and 1 h. Figure 1A, B, and Table 1 indicate the sugar concentration after hydrolysis of SCGs; the total sugar concentration of 0.5% H2SO4 hydrolysate was 8.78, 8.60, and 8.96 g/L under the conditions of 10–20% S/L ratio. For 1.0% H2SO4 hydrolysate, the total sugar concentration was 26.77, 29.85, and 35.93 g/L with a 10–20% S/L ratio. Furthermore, the sugar recovery rate against SCGs carbohydrates content (62.7%) was 14.0%, 9.1%, and 7.1% at H2SO4 0.5% and S/L ratio 10–20%, respectively, and 42.7%, 31.6%, and 28.7% at H2SO4 1.0%, and S/L ratio 10–20%, respectively. Therefore, the highest sugar recovery rate was obtained under 1.0% H2SO4 and 10% of the S/L ratio. Additionally, most of the sugar in SCGH comprises mannose and galactose (MG). Figure 1C, D show the composition of short-chain fatty acid (SCFA) and furan derivatives (5-hydroxymethylfurfural (5-HMF) and furfural). Similar to the sugar production tendency, the concentrations of SCFA increased as the S/L ratio and the H2SO4 concentration increased. And 5-HMF and furfural, known as an inhibitor that affects microbial growth, were 0.27 g/L and 0.19 g/L, respectively, under H2SO4 1.0% and S/L ratio of 10%. Kim et al. and Haroun et al. reported that hydrogen production was inhibited at above 0.5 g/L of 5-HMF [15] and above 2.0 g/L of furfural [18]. Usually, previous studies have shown that 5-HMF concentrations high enough to affect the hydrogen production have been produced in various biomass hydrolysates such as rice straw (1.2 g/L) [19, 20] and red algae (2.4 g/L) [21]. Then, to remove this inhibitor, a post-treatment has been performed using powdered activated carbon or a physicochemical method. In comparison, a relatively minimal concentration of 5-HMF (0.27 g/L) was produced in SCGH under conditions of 1.0% H2SO4 and 10% S/L ratio. Therefore, it was possible that the post-treatment could be omitted. Subsequently, SCGH was derived at 1.0% H2SO4, and a 10% S/L ratio was used as a substrate for biohydrogen production.

Fig. 1
figure 1

Comparison of spent coffee grounds hydrolysate following S/L ratio and sulfuric acid concentration A sugar concentration under 0.5% H2SO4, B sugar concentration under 1.0% H2SO4, C by-products concentration under 0.5% H2SO4, D by-products concentration under 1.0% H2SO4

Full size image
Table 1 Concentration of sugars, short-chain fatty acid, and furan derivatives according to S/L ratio and sulfuric acid concentration
Full size table

Comparison of Hydrogen Production in SCGH and Synthetic Sugars Media

A test was performed to validate the availability of the SCGH yielded above experiment as a substrate for biohydrogen production. Cultivation under conditions comprising synthetic sugar of mannose or galactose, the main sugars of SCGH, was performed together to compare productivity. (Fig. 2 and Table 2) Also, a comparison of productivity with other biomass-derived biohydrogen was shown in Table 3. As depicted in Fig. 2B, lag times were 6.25 h, 5.57 h, and 5.36 h under the SCGH, Mannose, and Galactose media, respectively. SCGH media shows the longest lag time. However, no significant difference was observed between the synthetic media. Nevertheless, as depicted in Fig. 2C, the hydrogen production amount using SCGH was 125.83 mL. This demonstrated an increase of approximately 34.7% and 65.5% compared with that of Mannose and Galactose media, which yielded 93.39 mL and 76.02 mL of hydrogen production, respectively. Additionally, as depicted in Fig. 2D, which depicts VFA production and sugar consumption on the test, the amount of VFA production and sugar consumption in SCGH media were 3.28 g/L and 4.76 g/L, respectively. These mannose media were 2.24 g/L and 3.21 g/L and galactose media were 1.78 g/L and 3.68 g/L, respectively. It was discovered that in SCGH, VFA production increased by 46.4% and 84.3%, respectively, and sugar consumption increased by 48.3% and 29.3%, respectively, compared with those of mannose and galactose media.

Fig. 2
figure 2

Comparison of dark fermentation productivities based on SCGH, mannose, and galactose media A biogas profile, B hydrogen profile, C production amount of biogas and hydrogen, and hydrogen yield, D VFA production and sugar consumption

Full size image
Table 2 Comparison of dark fermentation productivities (production amounts of biogas, hydrogen, and VFA, sugar consumption, hydrogen yield, and lag time) in each substrate
Full size table
Table 3 Comparison of hydrogen productivity derived from the various biomass
Full size table

The favorable impact of formic and levulinic acid present at a specific quantity in SCGH may have contributed to the enhanced the hydrogen production, yield, sugar consumption, and VFA production expressed in SCGH media. Previous studies have reported that hydrogen production using C. butyricum has a hormesis effect that gradually increases hydrogen productivity when formic and levulinic acid are present in a medium below 1.5 g/L and 3 g/L, respectively [15]. The formic and levulinic acid concentrations in SCGH were both 0.08 g/L, which seems to have induced the hormesis effect. Through this, it was verified that the hydrogen production using SCGH could be performed effectively.

Effect of Magnetite Supplementation in SCGH

Magnetite supplementation was performed to increase the hydrogen production in dark fermentation using SCGH. Figure 3 and Table 4 depict the comparison of hydrogen productivity based on the magnetite supplementation. In Fig. 3B, lag times under the negative control and magnetite supplementations were 5.54 h and 4.99 h, respectively, and magnetite supplementation has shown that it can reduce lag time. As shown in Fig. 3C, which shows hydrogen yield and production, the experimental group increased hydrogen yield by 13.6% and hydrogen production by 35.7% compared with the negative control group. Figure 3D shows VFA production and sugar consumption in the test. In the experimental group, both increased compared to the negative control. Additionally, in the previous study, lactate was consumed with magnetite supplementation [22]. According to Kim et al., magnetite is a conductive iron oxide that is thought to act as an electron transfer carrier. Fe ions that dissociate from it are supposed to become parts of the enzyme hydrogenase, which increases the consumption of sugar, and raises hydrogen productivity. Also, magnetite may play a role in accelerating the mechanism by upregulating the genes expression related to the lactate-utilizing pathway [15, 22]. As the effect of magnetite identified under the synthetic conditions was applied equally to a hydrolysate, it was confirmed that the hydrogen productivity could increase hydrogen production using SCGH.

Fig. 3
figure 3

Comparison of dark fermentation productivities based on the presence of magnetite supplementation A biogas profile, B hydrogen profile, C production amount of biogas and hydrogen, and hydrogen yield, D VFA production and sugar consumption

Full size image
Table 4 Comparison of dark fermentation productivities (production amounts of biogas, hydrogen, and VFA, sugar consumption, hydrogen yield, and lag time) according to the presence of magnetite supplementation
Full size table

Conclusions

This study discovered that SCGs, which are discarded in large quantities yearly, can be used as a substrate for hydrogen production by Clostridium species. In the SCG hydrolysis, the maximum sugar recovery rate (42.7%) and 26.77 g/L of overall sugar concentration were obtained under an S/L ratio of 10% (w/w), H2SO4 1.0% (w/w), 130 °C, for 1 h. Additionally, it was confirmed that furan derivatives were created relatively less in SCGH, resulting in cost reduction by omitting the post-treatment process. When magnetite was added to SCGH, hydrogen productivity increased even further, surpassing that of synthetic substrates when using the ideal SCGH determined above. Thus, hydrogen production was 35.7% higher than the negative control group. Overall, findings confirm that SCGs can be used as a substrate for C. butyricum. Furthermore, suppose it is identified why hydrogen production using SCGH is higher than that of synthetic substrates. In that case, it is anticipated that the economic feasibility of dark fermentation can be enhanced in the future.