Development of lactic acid production from coffee grounds hydrolysate by fermentation with Lacticaseibacillus rhamnosus

Abstract

 

Spent coffee grounds (SCG) are commercial waste that are still rich in numerous valuable ingredients and can be further processed into useful products such as coffee oil, antioxidant extract, lactic acid, and lignin. The challenge and innovation is to develop the SCG processing technology, maximizing the use of raw material and minimizing the use of other resources within the sequential process. The presented research is focused on the aspect of biotechnological production of lactic acid from SCG by using the Lacticaseibacillus rhamnosus strain isolated from the environment. Thanks to the optimization of the processes of acid hydrolysis, neutralization, enzymatic hydrolysis of SCG, and fermentation, the obtained concentration of lactic acid was increased after 72 hr of culture from the initial 4.60 g/l to 48.6 g/l. In addition, the whole process has been improved, taking into account the dependence on other processes within the complete SCG biorefinery, economy, energy, and waste aspects. Costly enzymatic hydrolysis was completely eliminated, and it was proven that supplementation of SCG hydrolysate with expensive yeast extract can be replaced by cheap waste from the agri-food industry.

One-Sentence Summary

A process for efficient lactic acid production from spent coffee grounds using the Lacticaseibacillus rhamnosus strain was developed and optimized, including nutrient solution preparation, supplementation and fermentation.

Graphical Abstract

Graphical Abstract.

Introduction

Coffee has become one of the symbols of mass culture, and its taste and smell are appreciated by most of us. Cafés have become a permanent part of the landscape of European and American cities (Gavin, 2013; Tucker, 2017). The main producers of coffee (Our World in Data, 2024) are Brazil (3.13 million tons of beans in 2022), Vietnam (1.95 million tons), Indonesia (795 thousand tons), Colombia (665 thousand tons), and Ethiopia (496 thousand tons). On the other hand, the largest consumption per capita is in the Nordic countries and amounts to 12 kg/person in Finland, 9.9 kg/person in Norway, or 9.0 kg/person in Iceland (World Population Review, 2024). In Poland, annual coffee consumption is 2.2 kg/person. However, after drinking coffee, we are left with spent coffee grounds (SCG). It is estimated that 600–650 kg of SCG is obtained from 1 ton of green beans, and 2 tons of wet SCG are received from the production of 1 ton of instant coffee (Mussatto et al., 2011). Thus, we are dealing with a waste of enormous mass.

In recent years, there has been an increasing number of papers devoted to the reuse of SCG (Sampaio et al., 2013). The authors analyze the aspects of obtaining energy from SCG combustion (Limousy et al., 2013; Schmidt et al., 2020). This solution is currently commercialized due to the high calorific value of coffee grounds and the low content of ashes and metals (Mayson & Williams, 2021). Unfortunately, it is burdened with many drawbacks, such as high emissions of harmful gases, for example increased carbon monoxide (Nosek et al., 2020), or nitrogen oxides (Kang et al., 2017). Other uses of SCGs are also known, for example their use as a direct fertilizer or compost component (Kopeć et al., 2024; Hirooka et al., 2022; Kourmentza et al., 2018), for biogas production (Girotto et al., 2018) or using them for remediation (Kim et al., 2014; Vahabi et al., 2021) thanks to their porosity and ability to absorb compounds on their surface.

The long-term goal of our research team is to develop a technology for complete processing of SCG into other products that can be reintroduced to the market and therefore meet the criteria of the circular economy (Geisendorf & Pietrulla, 2018). SCG should be understood as a lignocellulosic raw material (Chen et al., 2017; Houfani et al., 2020; Wyman et al., 2004), the composition of which depends on the methods of growing, roasting, and brewing the beans (Kovalcik et al., 2018). Nevertheless, the overall analysis of the SCG allows to identify several of the most important components (Ballesteros et al., 2014; Murthy & Madhava, 2012; Mussatto et al., 2011; Obruca et al., 2015), that can be recovered or processed: cellulose 8.6–13.3% (m/m), hemicellulose 30.0–40.0% (m/m) (as: arabinose 1.7–3.6% [m/m], galactose 13.8–16.4% [m/m], and mannose 19.1–21.2% [m/m]), lignin 24.0–33.0% (m/m), oil 10.0–20.0% (m/m), polyphenols 1.5–2.5% (m/m) (i.e. caffeine 0.1% [m/m]).

Coffee oil extracted from SCG is a liquid substance with a characteristic dark brown color and an intense coffee aroma. Previous studies (Acevedo et al., 2013; Ribeiro et al., 2013, 2021; Vu et al., 2021) showed that the largest part of the lipid fraction is triacylglycerols (approx. 85%). In addition to them, there is also a much smaller amount of free fatty acids, monoacylglycerols, and diacylglycerols. SCG oil contains mainly linoleic, palmitic, and stearic acids, and quite significant amounts of arachidic and linolenic acid. Sterols are part of the lipid fraction (approx. 5.5% m/m). These compounds are able to reduce cholesterol levels in the body (Ribeiro et al., 2021) and possibly can be a healthy replacement for conventional fats used in the kitchen. In the cosmetics industry coffee oil may be added to creams since it has a positive effect on the skin's hydrolipid barrier (Ribeiro et al., 2021) and slows down its aging process (Kanlayavattanakul et al., 2021). SCG fatty acid methyl esters may also be used in the fuel industry as an additive to biodiesel (Al-Hamamre et al., 2012; Blinová et al., 2017; Phimsen et al., 2016).

Among the antioxidants contained in SCG, the most numerous groups include phenolic acids and flavonoids (Anesini et al., 2012; Zuorro & Lavecchia, 2012). These compounds can be used in the food, pharmaceutical and cosmetic industries. In addition, they can be helpful in the fight against cancer and liver disorders (Sindhi et al., 2013), and in the prevention of skin aging processes (Zuorro & Lavecchia, 2012). Some studies also indicate their pro-regenerative effect on the nervous system and their impact on brain function, among other things, which may allow them to prevent the development of neurodegenerative diseases such as Alzheimer's disease (Pettinato et al., 2019). The most important phenol present in SCG is chlorogenic acid, which has antioxidant, antibacterial, and anti-inflammatory properties. It directly affects human metabolism by regulating insulin secretion and also supporting the breakdown of fats in the body (Naveed et al., 2018). Another group of compounds responsible for the reduction of free radicals are melanoidins. They can perform preservative and health-promoting functions in food (Iriondo-DeHond et al., 2020). The stimulating properties of caffeine are known (Ősz et al., 2022). However, this compound can be also used in the fight against cancer, for example leukemia, by acting on cell apoptosis and autophagy (Saiki et al., 2011).

SCGs are a rich source of polysaccharides, from which, by selecting the appropriate conditions for the decomposition of natural polymers such as cellulose and hemicellulose, it is possible to obtain a mixture of simple sugars, which are a source of carbon for lactic acid bacteria (LAB). In the literature, there are reports of the production of lactic acid on renewable waste, such as corn or rice straw and molasses, wood hydrolysate, or even cardboard waste (Bernardo et al., 2016; Chen et al., 2017; Cui et al., 2011; Qi & Yao, 2007; Vidra et al., 2017; Wang et al., 2015; Wee et al., 2004; Yáñez et al., 2005). Few articles report on lactic acid production using SCG (Hudečkova et al., 2018; Koo et al., 2019; Kovalcik et al., 2018), however, there are no reports of this type of solution being implemented on a scale larger than a laboratory scale. Lactic acid (LA) itself has a wide range of applications. It can be used primarily as a raw material for the synthesis of a biodegradable polymer—PLA (polylactide) (Swetha et al., 2023), as an alternative to the currently used petroleum-based polymers. By the way, the addition of lignin in such a composite can improve its properties (Spiridon et al., 2015). LA is also used in the food or cosmetics industries as a pH regulator or preservative (Alsaheb et al., 2015). In the pharmaceutical industry, lactic acid is used as an electrolyte in various parenteral/intravenous solutions (Ojo & Smidt, 2023).

The lignin remaining after the recovery of the oil, antioxidant extraction, and hydrolysis of cellulose and hemicellulose to soluble simple sugars, constitutes a large mass share in SCG. Chemically, it is a polymer formed by the enzymatic polymerization of phenolic precursors, mainly coniferyl alcohol, sinapinyl alcohol, and p-coumaryl alcohol (Lebo et al., 2014). The elemental composition and structural formula of lignin are difficult to determine because they depend on the type of plant, the method of isolation or the methods of purification (Santos et al., 2012). The global production of lignin is estimated at 100 million tons per year (Bajwa et al., 2019). The most commonly used on an industrial scale are lignosulfonates and Kraft lignin. However, both of these forms generate harmful by-products and contain sulfur, which translates into their limited use (Aro & Fatehi, 2017; Duval et al., 2013). The research of Carvalho and coworkers shows that lignin from coffee extraction is suitable for further processing (Carvalho et al., 2018). Isolated lignin can be used in many industries, for example in medicine, as an ingredient in wound dressings or pharmaceuticals (Yu & Kim, 2020), for electrochemical energy materials (Culebras et al., 2019; Mashhadimoslem et al., 2021) or for 3D printing (Li et al., 2021). Moreover, the use of lignin as a component in composites seems promising (Shakoor Shar et al., 2023; Souza de Miranda et al., 2015; Spiridon et al., 2015, 2011; Yang et al., 2019).

All SCG components can be extracted and valorized in a biorefinery where the biomass is sustainably processed so that it can be reused and profitable. This concept allows for the implementation of closed-loop technology in accordance with EU directives (European Parliament, 2024). Processing coffee waste on an industrial scale may have an impact on the cost of coffee production, due to the fact that coffee is treated not as a consumer good, but as an intermediate product for obtaining energy or producing useful substances. Various concepts of a biorefinery processing coffee waste are described in the literature (Kondamudi et al., 2008; Rajesh Banu et al., 2020; Vardon et al., 2013). As a rule, they are based on the extraction of coffee oil and further processing of residues and, for example, the extraction of renewable fuel sources with a high energy value. Some reports include the extraction of antioxidants (Cerino-Córdova et al., 2020). The more processing steps, the more difficult the process becomes and the more factors need to be taken into account (Caetano et al., 2017).

The concept of the biorefinery proposed by our team includes a developed sequence of processes which make it possible to obtain coffee oil, antioxidants, lignin, and lactic acid (as a product of the SCG hydrolysate fermentation) from SCG. The general diagram is shown in Fig. 1. The entire project is being carried out by the Polish company EcoBean in collaboration with Warsaw University of Technology and is currently at the stage of completed conceptual design (covered by patent application number P.447416). This article is focused on developing and optimizing the process of L-lactic acid production from SCG in the context of the total SCG processing.

Fig. 1.

A simplified schematic diagram showing products sourced from SCG in EcoBean's biorefinery concept.

SCG are obtained from commercial customers such as cafes and instant coffee producers, using a specially designed application available on Android and iOS. Employees of a café collect coffee grounds into the dedicated SCG collection boxes, which are waterproof and approved for contact with food, and their design prevents the load from spilling out during transport. After filling the box, an employee of the coffee grounds collection point, using the application, generates a waste transfer document, which is compliant with the mandatory in Poland “Database on products, packaging and waste management” and a bill of lading, thanks to which the courier company will be able to collect the shipment. The package is ultimately sent to the laboratories of the Warsaw University of Technology, where implementation studies of the SCG processing technology into further products are carried out. Once the planned pilot biorefinery has been built, a similar logistical solution will be implemented, which will ensure a raw material supply of 3 tons/day. Further plans include building coffee ground biorefineries in Poland and Europe.

List of abbreviations

SCG—spent coffee grounds; SCGw—wet spent coffee grounds, obtained directly from the café, humidity approx. 60% m/m; SCGc—cleaned spent coffee grounds, that is SCGw dried to approx. 5% m/m moisture content and sieved through a sieve with a mesh size of 600–650 µm; SCGx—spent coffee grounds after hexane extraction, that is SCGc that was subjected to the hexane extraction process and then dried; SCGe—spent coffee grounds after ethanol extraction, that is SCGx which was subjected to the process of extraction with an aqueous solution of ethanol and then dried; SCGh—spent coffee grounds after acid hydrolysis, that is SCGe that was subjected to the process of acid hydrolysis and then dried; AA—acetic acid; LAB—lactic acid bacteria; LA—lactic acid; L-LA—L-enantiomer of lactic acid; D-LA—D-enantiomer of lactic acid; WP—whey permeate; YE—yeast extract.

Materials and Methods

Materials

Coffee grounds

SCG were collected from a chain of cafes in Warsaw. No mold was found on them. Before further processing, all batches of SCG were dried at 80°C for 24 ± 2 hr to a water content of less than 5% m/m.

Lactic acid bacteria strain

The strain used in the research comes from the collection of WUT strains of the Chair of Drug and Cosmetics Biotechnology of the Warsaw University of Technology. It belongs to the genus Lacticaseibacillus rhamnosus. It was isolated from fermented beet juice native to Poland, Mazovia. The strain is stored in a 15% glycerol solution at −80°C in a 2 ml cryogenic tube.

Chemical reagents

H₂SO₄ min. 95% p.a., NaOH (microgranules) p.a., ethanol 96% p.a. from POCH, Poland, HCl 35–38%, p.a. and CaCO₃ solid pure, from Chempur, Poland, enzyme mixture FLASHZYME Plus 200 from AB Enzymes, Germany, Line-EtOH from Linegal, Poland, WP from Mlekovita, Poland, yeast extract and Milli-Q® from Merck Millipore, USA, anti-foaming agent ROKAmer G3500 and ROKAmer 2600 from PCC exol, Poland, calcium oxide CaO from VWR International, USA, L-Lactic Acid Kit, D-Lactic Acid Kit, Acetic Acid Kit from Megazyme, L-Histidine solid 99% pure, PanReac AppliChem from ITW Reagents, Tween 80 Reagent Grade from VWR CHEMICALS, Magnesium Citrate Anhydrous 95% pure, Tri-ammonium Citrate 95% pure, Iron(III) Citrate 5 hydrate 95% pure, Tricalcium Citrate 4 Hydrate 98% pure, from WARCHEM Sp. z o. o. Poland, D-(+)-Biotin 98% pure from POL-AURA, Poland.

Microbial media

The following media were used for bacterial cultures: liquid and solid (with 1.5% agar) MRS Broths, Merck; Nutrient Agar + Ampicillin, Biocorp, France; PCA medium, Biomerieux, France; Sabouraud medium, Merck Millipore, USA; self-prepared liquid media from SCG hydrolysates (SCG medium) without or with supplementation.

Analytical Methods

Analysis of simple sugars (arabinose, galactose, mannose, and glucose)

Ten microliter of the medium sample was injected onto the column (Repromer Ca, 9 µm, 300 × 8 mm) thermostated at 80°C, and the sugar content was analyzed using a UHPLC chromatograph (Nexera X3, Shimadzu) in combination with an RI detector that was thermostated at 35°C. 9.0 mM H₂SO₄ was used as the mobile phase.

Measurements were made in an isocratic flow rate of 1.2 ml/min. The collected samples were filtered through syringe filters into the vials, neutralized with an aqueous solution of 5 M NaOH to a pH value in the range of 3.0–3.5 and placed in an autosampler. The pH value was measured using an indicator strip (Lach-ner). The dilution factor was taken into account when reporting the result.

The chromatograms obtained after the separation were analyzed in LabSolution (version 5.111) program.

The score is the arithmetic mean of at least two repetitions.

Analysis of organic acids (L-lactic acid, D-lactic acid, and acetic acid)

D-lactic acid, L-lactic acid, and acetic acid concentrations in culture medium were assayed in 96-well plates on SPECTROstar Nano (BMG LABTECH GmbH) plate reader using Megazyme enzymatic tests and manufacturer protocols.

The score is the arithmetic mean of at least three repetitions along with the standard deviation.

Microbiological control

In order to monitor the growth of bacteria in the cultures, samples of the culture were plated on Petri dishes. The dishes were incubated at 37°C for 48 ± 2 hr. After this time, the colonies that grew were counted and their morphology was assessed. Biofilm homogeneity was evaluated to confirm that microorganisms of one species were present in the culture on PCA and MRS medium + 1.5% agar. Observations to determine whether the bioreactor has been contaminated with fungi were carried out using SAB medium with ampicillin or with chloramphenicol.

pH measurement

pH measurements were performed with a pH meter equipped with a SenTix 41 electrode at 22–25°C.

Moisture measurement

The moisture measurement was carried out on a calibrated MA.210.X2.A moisture analyzer (RADWAG).

Research Methods

Oil extraction

After drying, the SCGc were sieved on sieves and then subjected to the process of extracting coffee oil with hexane. The extraction lasted 30 min at 69°C. After the entire process, the suspension was filtered and the grounds were dried to obtain SCGx.

Antioxidant extraction

Dry SCGx were subjected to an antioxidant extraction process using an aqueous ethanol solution. The extraction lasted 2 hr at 30°C. After the entire process, the suspension was filtered and the grounds were dried to obtain SCGe.

Acid hydrolysis

Dried coffee grounds, regardless of the previous treatment, were hydrolyzed in 2.5% (m/V) aqueous solution of sulfuric acid (POCH). The ratio of coffee grounds to sulfuric acid, as well as the time and temperature of hydrolysis were optimized. Most of the fermentation processes described in this study were carried out on a medium obtained in one of the three hydrolysis variants, that is,

SCG: acid solution in a ratio of 1:9 (m/V) in an autoclave (121°C for approx. 30 min)—variant A; SCG: acid solution in a ratio of 1:6 (m/V), in an autoclave (121°C for approx. 30 min)—variant B; SCG: acid solution in a ratio of 1:6 (m/V) at 100°C for 3 hr at atmospheric pressure—variant C.

After the hydrolysis had been completed, and the entire suspension filtered, the obtained SCGh sediment was retained for further processing, while the filtrate, after neutralization and enzymatic hydrolysis, was used to prepare the bacterial medium.

Filtrate neutralization

Solid calcium carbonate was added to the filtrate to obtain the pH of the solution in the range of 5.0–5.5/25°C. To prevent foam formation, ROKAMer G3500 antifoam was added in the amount of 0.25% m/V in relation to the volume of the entire solution. The resulting calcium sulfate precipitate was filtered and dried.

Enzymatic hydrolysis

A blend of cellulolytic and hemicellulolytic enzymes, FLASHZYME Plus 200, was added to the neutralized hydrolysate in the amount of 6% m/V relative to the volume of the total solution. The mixture was then incubated in a shaker (IKA) for 24 ± 1 or 72 ± 2 hr at 50°C.

Preparation of bacterial medium

Filtered and autoclaved SCG hydrolysate, without supplementation or properly supplemented, was used as a medium for LAB culture. Yeast extract (Merck), WP (Mlekovita), glucose, mannose, galactose, arabinose, histidine, biotin, Tween 80, triammonium citrates, calcium, magnesium, and iron were used as supplements in optimizing process. Selected supplements were added to the medium prior to the autoclave sterilization, with the exception of histidine and biotin, which were filtered using a 0.22 µm syringe filter and added to the sterile medium.

Inoculum preparation

The frozen strain was plated reductively in a Petri dish with solid medium MRS + 1.5% m/m agar. The dish was then incubated for 72 ± 2 hr at 37°C and a single colony was transferred to an Erlenmeyer flask with 100 ml of sterile liquid MRS medium and incubated for 24 ± 1 hr at 37°C.

Fermentation in flasks

SCG medium, without or with suitable supplementation, was adjusted to a pH of 7.2 ± 0.2/25°C, poured into Erlenmeyer flasks and autoclaved. Then, an inoculum was added in the amount of 1% V/V in relation to the volume of the hydrolysate solution. Fermentation was carried out, depending on the variant of the experiment, in a volume of 60, 80, 150, and 200 ml.

At 0 hr and every 24 hr during the culture, 2.5 ml of samples were withdrawn. The samples were centrifuged for 5 min at 6000 rpm and supernatants were stored at −20°C till further analysis.

The results are the arithmetic mean of at least two biological repetitions.

Fermentation in a bioreactor

SCG medium with suitable supplements, was poured into the bioreactor vessel (Sartorius 2 l or Biotehniskais centrs 4 l) and autoclaved. Fermentations were conducted in bioreactors at constant pH 6.0, 6.5, or 7.0. with mixing (25 rpm). Properly supplemented SCG medium was sterilized in a culture vessel and inoculated with 1% (V/V) Lacticaseibacillus rhamnosus overnight culture. Four milliliter of samples were withdrawn at 0 hr and every 24 hr during the culture. The samples were centrifuged for 5 min at 6000 rpm and supernatants were stored at −20°C till further analysis.

Statistical calculations

Flask cultures were conducted in at least two biological repetitions, whereas bioreactor experiments were performed in a single approach. Determination of L- and D-lactic acid concentrations using the enzymatic method was performed in at least three repetitions for a specific sample dilution. After rejecting gross errors, the mean and standard deviation were calculated using an Excel formula to determine the spread of individual results around the mean.

where: x = consecutive result, = average result, n = number of measurements.

The obtained results of the concentrations of individual acids were recorded in the form ± SD [g/l].

Results

Biotechnological production of lactic acid is one of the unit processes of the SCG total biorefining technology currently under development. Therefore, the stages preceding the fermentation process may significantly affect the obtained LA concentration.

Based on the research carried out, a general scheme for obtaining lactic acid from SCG was adopted:

where: SCGe stands for spent coffee grounds after hexane and ethanol extraction, AH = acid hydrolysis, N = neutralization, EH = enzymatic hydrolysis, S = hydrolysate supplementation, LAB = lactic acid bacteria, F = fermentation, LA = lactic acid.

Each of the presented stages was optimized in terms of the obtained concentration of lactic acid and the economics of the process. The Lacticaseibacillus rhamnosus strain used in this study produces predominantly L-enantiomer of lactic acid. The ratio of L-LA to D-LA varies depending on the variant of the experiments carried out, but the median for the results discussed in this paper is 95:5.

The Effect of SCG Processing on LA Production and Initial Nutrient Solution Supplementation

The technological line of SCG biorefining and their full valorization has been arranged in such a way that the sequence of processes ensures the maximization of the efficiency of obtaining a given product, but emphasis has also been placed on the best possible efficiency of production in the next technological node.

Recently we have demonstrated the need for supplementation of SCG hydrolysate in order to obtain better LA yield (Wysocki et al., 2022). By investigating the effect of different concentrations of yeast extract on LA production, this additive was identified as an essential basic supplement to the SCG medium which at 1% (m/V) concentration increased the production of L-LA from 4.60 g/l to 8.45 g/l after 72 hr of culture, bringing this value closer to the result obtained for the culture on MRS medium (11.21 g/l).

Hudečkova and coworkers found that in the process of sugar hydrolysis the best ratio of SCG mass to sulfuric acid solution volume is 1:9 (Hudečkova et al., 2018). This was our starting point for further optimization. As a result of the conducted research (not shown), the ratio of sulfuric acid to water used for hydrolysis was reduced to 1:6 (Materials and Methods, variant B).

In order to check if coffee oil and antioxidant extractions from SCG influence the process of lactate fermentation, the same amounts of SCGc, SCGx, and SCGe were subjected to acid hydrolysis at 1:6 SCG: sulfuric acid ratio and the efficiency of lactic acid production on the obtained media supplemented with 1% YE was checked. The results are presented in Fig. 2.

Fig. 2.

The effect of SCG hexane and ethanol extractions on lactic acid production.

Cultures were carried out in flasks in 60 ml of medium and at a temperature of 37°C for 72 hr.

It was shown that the extraction of SCG with hexane and ethanol solution, preceding hydrolysis and preparation of the medium, promotes the production of lactic acid, since after 72 hr of LAB culture on SCGe hydrolysate, the production of LA increased by nearly 40% compared to SCGc hydrolysate. The removal of the oil fraction also significantly improved L-LA enantiomer excess (for details see Table S1 in Supplementary Data).

Optimization of the SCG Hydrolysis and Neutralization Process

When scaling a technological process, it is necessary to take care of the efficiency, but also the energy consumption of each unit process. Acid hydrolysis in an autoclave was effective, but on a technical scale it is impractical—it would translate into high investment and operating costs. Accordingly, acid hydrolysis was carried out at three different temperatures and times at atmospheric pressure (for details see Table S2 in Supplementary Data). Dry SCGe were subjected to acid hydrolysis in a ratio of 1:6 (mass of SCGe: volume of sulfuric acid solution) and the resulting hydrolysate was neutralized with solid CaCO₃, but the enzymatic hydrolysis process was not performed. After filtering, 1% YE was added to the obtained medium. Lacticaseibacillus rhamnosus was cultured for 72 hr at 37°C in a volume of 200 ml (Fig. 3).

Fig. 3.

The effect of SCGe acid hydrolysis temperature and time on LA production.

The obtained results allowed us to conclude that the determining parameter in the case of SCG acid hydrolysis is the temperature of the process. For each hydrolysis time, more than fourfold higher LA concentrations were achieved in the culture on hydrolysate obtained by boiling than in the lowest temperature tested. An additional advantage of the method used is the fact that a higher concentration of lactic acid was obtained in the boiled hydrolysate than in the autoclaved hydrolysate. In addition, the hydrolysis of coffee grounds at 100°C yielded 97.5–97.9% LA in the form of L-LA enantiomer (depending on the time of hydrolysis), whereas using other hydrolysis methods the best result was 96.5% (see Table S2 in Supplementary Data).

The process of neutralization of acidic hydrolysate has a double meaning in the technology being developed. First, the removing of sulphate ions is unfavorable from the point of view of membrane downstream processes, and second, obtaining the optimal pH for the culture of microorganisms. Regardless of the mineral acid used for SCG hydrolysis, the main environmental problem would be the release of anions into the wastewater, which would subsequently generate pollutants. Therefore, the selection of sulfuric acid for hydrolysis and choosing the appropriate substance for the precipitation of sulfate ions was the basis of the technology. CaCO₃ was chosen as a neutralizing agent due to its low price and safety of use and storage. Table 1 shows the effect of different methods of calcium carbonate application on the concentration of L-LA obtained after culture on a medium prepared with the use of a given neutralization variant. Equal amounts of dry SCGe were used for each variant, which were hydrolyzed according to variant A. Each time, the neutralization compound was added, the pH was measured continuously to 5.0–5.5. The values obtained were then corrected if necessary, using 1 M hydrochloric acid. Enzymatic hydrolysis was performed after neutralization. The nutrient solution was supplemented with 1% YE and 3% WP. The culture was carried out for 72 hr at 37°C.

Table 1.

Average Values of L-LA Concentration Obtained in Culture Media Neutralized by Different Methods, Results After 72 hr of Culture in a Volume of 150 ml in Flasks

Neutralization variant	L-LA [g/l]
CaCO3 in powder form manual mixing	11.6 ± 0.82
CaCO3 in powder form mixing with a mechanical stirrer	12.2 ± 0.51
CaCO3 + anti-foaming I	12.4 ± 0.26
CaCO3 + anti-foaming II	12.4 ± 0.05
CaO	7.70 ± 0.18
CaCO3 aqueous solution	11.3 ± 0.23

It has been shown that the use of CaCO₃, regardless of the form of addition, does not affect the production of lactic acid (Table 1). The effect of two selected antifoaming agents (both from PCC Exol) was also checked, since the carbon dioxide released during the reaction caused intense foaming of the suspension. The use of an anti-foaming agent solved this problem, and could also slightly contribute to increasing LA production efficiency. We have also attempted to use calcium oxide as a neutralizing agent. In this case, no foam was formed, however, due to the two-stage chemical reaction involving first slow formation of calcium hydroxyoxide and then the more rapid formation of Ca(OH)₂, obtaining a stable pH value was extremely difficult. Due to these problems, too much calcium oxide was added, and the medium pH was then adjusted to 5.0–5.5 with a larger amount of 1 M hydrochloric acid. This was most likely not favorable for growing of bacteria and translated into lower lactic acid production (Table 1). Therefore, CaO was rejected as a neutralizing reagent. Due to easy application and the simple system which can be implemented on an industrial scale, we decided to implement CaCO₃ in powder form.

Optimization of the Fermentation Conditions

The next step was to optimize the fermentation process itself. The results obtained on a small scale (Figs. 2 and 3) had to be verified on a larger scale in a bioreactor. Table 2 collects the data on the optimization of this stage. Due to the high nutritional requirements of the LAB (Hayek & Ibrahim, 2013), it was decided to add another supplement—WP, which has been tested by other authors for effectiveness in promoting of LAB growth (Amrane, 2005). The concentration of WP was optimized (not shown), resulting in an increase of the produced LA at 3% (m/V) concentration of this supplement. Taking into account that WP is a waste itself, its addition to the medium causes a further increase in the valorization of waste in the discussed technology. In addition, it was investigated whether the use of SCG from other café would affect the efficiency of the process. Regardless of the source, all coffee grounds were processed in the same way, that is they were dried (SCGc was obtained), oil was extracted (SCGx was obtained), and antioxidants were extracted (SCGe was obtained). Each time, dry SCGe were hydrolyzed with sulfuric acid, neutralized and, depending on the variant of the experiment, enzymatic hydrolysis was performed. Appropriate supplements were added to the hydrolysate obtained in this way and culture was carried out for 48 hr at a given temperature.

Table 2.

The Influence of Various Parameters of the Fermentation Process on the Production of L-LA, D-LA, and AA, Cultures Were Carried Out For 48 hr in a Volume of (Depending on the Variant) From 2500 ml to 3500 ml

Acid hydrolysis variant	Enzymatic hydrolysis	Breeding supplementation	Initial pH of the culture	Breeding temperature [°C]	L-LA [g/l]	D-LA [g/l]	AA [g/l]
A	✔	-	6.0	37	8.83 ± 1.6	0.431 ± 0.01	0.479 ± 0.02
A	✔	1% YE	6.0	37	16.2 ± 0.86	0.740 ± 0.06	0.740 ± 0.04
A	✔	1% YE + 3% WP	6.0	37	29.8 ± 1.1	1.67 ± 0.01	4.42 ± 0.57
A	✔	1% YE + 3% WP	6.0	42	31.0 ± 0.26	1.78 ± 0.19	nd
A	✔	1% YE + 3% WP (SCGw from other café)	6.0	42	32.9 ± 0.78	1.74 ± 0.02	nd
A	✔	1% YE + 3% WP	7.0	37	34.0 ± 0.56	1.67 ± 0.16	1.27 ± 0.03
C	✔	1% YE + 3% WP	6.5	37	46.3 ± 2.3	2.22 ± 0.17	1.63 ± 0.17
C	✘	1% YE + 3% WP	6.5	37	45.0 ± 4.7	2.10 ± 0.18	1.52 ± 0.04
C	✘	1% YE + 3% WP + 0.25% GFS	6.5	37	44.3 ± 3.1	2.50 ± 0.23	2.08 ± 0.44

Note. YE = yeast extract; WP = whey permeate; GFS = glucose-fructose syrup; nd = no data.

Nutrient solution supplementation was confirmed in flask cultures to be essential for bacterial growth, however, the sharp increase in LA production was achieved in bioreactor cultures, where constant pH, temperature, and mild mixing were maintained. The increase in lactic acid production in the WP supplemented medium is due to provision of additional nutrients, as, according to manufacturer's internal technological card, 100 g of dry WP contains 78 g of sugars and 12 g of proteins. An initial pH equal to 6.5 resulted in an increase in LA production of almost 27% relative to pH 7.0 and over 35% relative to pH 6.0. The comparison of LA production on medium prepared using acid and enzymatic hydrolysis with LA production on medium obtained using only acid hydrolysis shows that enzymatic hydrolysis does not significantly increase the concentration of lactic acid obtained. In summary, the best results have been achieved with the medium obtained by acidic hydrolysis of SCGe slurry at 100°C for 3 hr, supplemented with 1% yeast extract and 3% WP, and by conducting the culture at 37°C, at pH = 6.5.

During the development of the technology for the total biorefining of SCG, we have considered it appropriate to separate, after acid hydrolysis, the lignin-containing solids from the filtrate containing the sugars, instead of carrying out neutralization of the whole mixture for subsequent enzymatic hydrolysis and filtration at the end. The latter approach would result in contamination of the lignin with calcium sulphate precipitate. The enzymatic hydrolysis did not improve the production of lactic acid, nor did the addition of well-absorbed (Carvalheiro et al., 2011) glucose-fructose syrup (Table 2). Therefore, considering the prices of enzyme preparations, time factor, and the quality of the material for lignin purification, the elimination of enzymatic hydrolysis would significantly contribute to the simplicity and profitability of the technology.

In addition, the samples were tested for the content of acetic acid, which is often produced by LAB during fermentation (Lee et al., 2021), to confirm that L-LA is the major metabolite in the resulting post-culture fluid. In all the variants studied, the concentration of produced acetic acid remained approximately 4% in relation to the total concentration of lactic acid. However, the lowest amount (2.67%) was determined for the sample, where the pH was maintained at 6.0, and the highest (14.1%) for the pH of 7.0, while maintaining the same supplementation and other process conditions.

Further Optimization of Nutrient Solution Supplementation

Numerous studies on LAB's nutritional requirements (Waller, 1970 Pham et al., 2000; Tharmaraj & Shah, 2003; Landete et al., 2006; Haokok et al., 2023;) show that LAB need not only simple sugars to grow, but also proteins and organic salts, such as citrates. The collected results showing the effects of chosen supplements are presented in Table 3. Dry SCGe was used in the study, which was subjected to acid hydrolysis in variant A, neutralization, and enzymatic hydrolysis. The medium, containing 1% YE, was supplemented with other additives listed in Table 3. The culture was carried out for 72 hr at a temperature of 37°C in flasks in a volume of 80 ml.

Table 3.

The Effect of Additional Supplementation of Medium Obtained From SCGe, Results After 72 hr in a Volume of 80 ml

Additional supplementation	L-LA [g/l]	D-LA [g/l]	L-LA enantiomeric excess [%]
None	13.0 ± 2.3	0.560 ± 0.02	95.5
1.0 g/l histidine	13.2 ± 0.40	0.593 ± 0.01	95.4
20 mg/l biotin	10.1 ± 0.63	0.464 ± 0.03	94.9
1.0 ml/l Tween 80	9.67 ± 0.17	0.487 ± 0.01	95.3
20 mg/l biotin + 1.0 ml/l Tween 80	10.5 ± 0.78	0.493 ± 0.05	95,3
2 g/l tri-ammonium citrate	10.2 ± 0.13	0.668 ± 0.06	94.9
2 g/l magnesium citrate	13.6 ± 0.32	0.703 ± 0.07	95.2
0.08 g/l iron citrate	12.1 ± 0.49	0.581 ± 0.04	94.3
0.20 g/l calcium citrate	9.78 ± 0.11	0.560 ± 0.08	95.4
2 g/l tri-ammonium citrate + 2 g/l magnesium citrate + 0.08 g/l iron citrate + 0.20 g/l calcium citrate	13.2 ± 0.39	0.607 ± 0.03	95.1
2 g/l tri-ammonium citrate + 2 g/l magnesium citrate + 0.08 g/l iron citrate + 0.20 g/l calcium citrate + 3% whey permeate	13.4 ± 0.97	0.658 ± 0.01	95.5

The addition of histidine and magnesium citrate, as well as combined citrates slightly improved lactic acid production, while biotin, Tween 80 and the other citrate salts used separately had a rather negative effect on the LA concentration obtained. Given that the increase in LA production is insignificant, it is not economically justified to use the supplements studied (Table 3).

The beneficial effects of WP on lactic acid production (Table 2) have inspired attempts to replace expensive yeast extracts with low-cost supplements that are waste from various industries. Other research teams have already attempted to use in LA production of wheat bran (Li et al., 2010), soybean meal (Chi & Cho, 2016), yeast flakes (Luana et al., 2014), or dried distillery corn broth (Iram et al., 2020). In addition, the use of another production waste, that is molasses, which can also be used to produce lactic acid was checked (Ni et al., 2017). Each time, dry SCGe was used, which was hydrolyzed according to variant C and neutralized. No enzymatic hydrolysis was used. The filtered hydrolysate was supplemented with a selected additive. Culture was carried out for 48 hr in a bioreactor at 37°C and at pH of 6.5. The results are summarized in Table 4.

Table 4.

Effect of Selected Nutrient Supplements on the Production of Lactic Acid After 48 hr of Culture in a Volume of 2000 ml

Supplement (% m/V)	L-LA [g/l]	D-LA [g/l]	L-LA enantiomeric excess [%]
1% yeast extract	31.7 ± 0.92	0.890 ± 0.03	97.2
1% yeast extract + 3% whey permeate	45.9 ± 2.2	2.70 ± 0.21	94.1
1% wheat bran	29.8 ± 0.28	1.80 ± 0.06	94.0
1% soybean meal	25.8 ± 0.08	1.54 ± 0.09	94.0
1% dried distillery corn broth	12.6 ± 0.85	1.20 ± 0.06	90.5
1% yeast flakes	23.2 ± 3.0	1.43 ± 0.04	93.8
1% molasses	4.85 ± 1.1	0.407 ± 0.02	91.6
1% molasses + 1% yeast extract	25.4 ± 5.5	1.98 ± 0.05	92.2

The results confirm the possibility of replacing the expensive yeast extract with other cheaper supplement. The best results were achieved for cultivation on hydrolysate with the addition of wheat bran, which was as effective as yeast extract, nevertheless, soybean meal, and yeast flakes gave only slightly lower results. Moreover, it was found that the addition of waste from the sugar industry, such as molasses, cannot be used as a supplement in LAB cultivation, because as an independent additive it even inhibits the production of LA, and in the variant with the combination of YE and molasses, a lower concentration of lactic acid was obtained than in the variant with the addition of only YE by as much as 20%. With the use of industrial-derived supplements, the ratio of L-lactic acid to D-lactic acid changes. For the variant with the addition of YE only, the highest content of the L enantiomer in the produced lactic acid was obtained, amounting to as much as 97.2%. The use of WP, wheat bran, soybean meal, and yeast flakes resulted in a slightly lower enantiomeric excess of LA, that is around 94%. Of the low-cost raw materials used as nutrient supplements, dried distiller's corn broth resulted in the production of increased amounts of D-lactic acid (9.5%) relative to L-lactic acid (90.5%).

Summing up the studies on the optimization of lactic acid production with the use of Lacticaseibacillus rhamnosus strain, the highest concentration of L-lactic acid was obtained in culture at pH = 6.5 on a 1% YE and 3% WP supplemented medium, obtained as a result of SCGe hydrolysis at 100°C for 3 hr, with the ratio of SCGe:2.5% (m/V) sulfuric acid = 1:6 (according to variant C), without enzymatic hydrolysis. The use of sugars during cultivation and the total production of LA enantiomers as a result of this modified process is shown in Fig. 4. The culture was carried out for 72 hr at 37°C in a bioreactor in a volume of 2000 ml. The obtained lactic acid concentrations were converted into the amount of LA obtained from 100 g of SCGe.

Fig. 4.

Production of lactic acid on coffee grounds medium and consumption of sugars during culture in a 2-litre bioreactor.

The results indicate that during the culture in the bioreactor, the total consumption of sugars and the maximum production of LA occurs in less than 48 hr. After 24 hr of culture, 35.7 g of lactic acid was obtained (efficiency at the level of 92.7%), and after 48 hr 37.2 g of lactic acid (efficiency of 96.6%), and after 72 hr the level of 38.5 g was reached. The product yield coefficient for the substrate (sugars) is then 1.10 g LA/g sugars, which is higher than the values obtained in the previously cited papers (Hudečkova et al., 2018; Koo et al., 2019).

Discussion

The chain of processes has been developed that generate new products from SCG, a waste from the HoReCa branch. The sequence of the individual unit processes is the essential know-how developed during the research carried out. Preceding the lactic acid production process with extraction of coffee oil and antioxidants, as well as carrying out acidic hydrolysate neutralization in a way that facilitates the preparation of highly purified lignin, proved to be beneficial and is a valuable novelty in the described technology. Hydrolysis is an intermediate step necessary for further processing of SCG. In the course of our research, it was possible to optimize these steps in terms of energy consumption, and thus also costs, and completely eliminate the use of expensive enzymes. It has been proven that supplementation of SCG hydrolysate with expensive yeast extract can be replaced by cheap waste from the agri-food industry. The lignin, remaining after the acid hydrolysis and filtration processes, which is currently being investigated for better purification, is also used, among others, as a filler in composites under development in EcoBean company. On the other hand, calcium sulphate sediment obtained in the neutralization process and bacterial biomass after the fermentation process can be a useful food additive (European Parliament, 2012, 2019).

In the course of this study, the single acid hydrolysis production variant was optimized, although our most recent research has shown that it is possible to obtain nearly twice as much LA from the same amount of SCG. After drying, SCGh were re-hydrolyzed under identical conditions and then used to prepare a medium that was supplemented with 1% YE. After 72 hr in the flask cultures, the following were obtained: 15.3 ± 0.51 g/l L-LA and 0.529 ± 0.05 g/l D-LA, which shows that it is possible to obtain an additional amount of lactic acid from the same portion of coffee grounds. Therefore, it is necessary to consider carrying out a sequence of two hydrolyses, which, however, would be a certain difficulty from the point of view of the whole technology. A thorough economic analysis and further research are necessary to improve the technology.

In the presented studies, the increase in LA between the 48 and 72 hr of cultivation was insignificant, therefore, from the point of view of energy consumption during the entire process, fermentation can be carried out for 48 hr or even in a shorter time. In the papers Hudečkova et al. (2018) and Kovalcik et al. (2018), it can be noted that the values of the obtained lactic acid are at the level of approx. 25 g/l after approx. 72 hr of culture. In the study of Koo et al. (2019), much higher final concentration of lactic acid was obtained, approx. 110 g/l, however, the culture lasted a very long time, that is 468 hr. Nevertheless, the process parameters we have developed can still be improved, so further research is being conducted.

So far, there are no reports on the implementation of the complete biorefinery of SCG on a scale larger than laboratory scale. There are also no reports on attempts to combine the production of lactic acid with the valorization of other components of coffee grounds, which is presented in this paper. In the course of the work to improve the efficiency of the LA production process on the SCG, other types of culture, such as continuous or semi-continuous cultures with concentrated medium dosing, or with constant product intake and biomass recirculation, should also be considered, taking into account the economics and safety of the process.

Scaling up biotechnological processes is a difficult and complex task (Bhatt et al., 2023; Tufvesson et al., 2010), therefore, it is important to test as many aspects of the technology under development as possible during the research and implementation phases. The described in this article fermentation process, carried out and optimized on laboratory scale, has been also conducted on a semi-technical scale in independent center (Institute of Agricultural and Food Biotechnology in Warsaw) where the fermentation process was carried out in a volume of 90 l on the hydrolysate supplemented with 1% YE and 3% WP, produced in the semi-technical hall of Faculty of Chemistry at Warsaw University of Technology. The results obtained after 48 hr of cultivation were as follows: 39.4 ± 0.60 g/l L-LA, 1.63 ± 0.01 g/l D-LA, and 1.64 ± 0.00 g/l AA. This shows that an approximately 40-fold increase in the scale does not significantly affect the efficiency of the process.

A quarter-technical scale process is currently underway where we are able to convert 10 kg of SCG per week into usable products. However, we are currently in the process of designing a plant capable of processing approximately 3 tons of SCG in 1 day. At the same time, we are working on scaling up the entire technology. Time and money of the investment come first here, so the most important aspects that are requested by a technology designer, are verified first. The conducted research made it possible to significantly facilitate the design process. The study took into account every aspect of technology and its impact on the efficiency of lactic acid production. The next planned stage of technology development is the work on the down-stream lactic acid purification process.

Funding

The research was supported by the UE framework program Horizon Europe (ID project: 27170) ‘Upcycling coffee waste into useful raw materials and green products’ and Ministry of Science and Higher Education program ‘Implementation PhD’.

Conflict of Interest

The authors declare no conflict of interest.