A Coffee-Based Bioadsorbent for CO₂ Capture from Flue Gas Using VSA: TG-Vacuum Tests

Abstract

In the energy sector and in other types of industries (cement, iron/steel, chemical and petrochemical), highly roasted coffee ground residue can be used as a source material for producing bioadsorbents suitable for CO₂ capture. In this study, a bioadsorbent was produced in a two-step process involving biowaste carbonization and biocarbon activation within a KOH solution. The physicochemical properties of the bioadsorbent were assessed using LECO, TG, SEM, BET and FT-IR methods. Investigating the CO₂, O₂ and N₂ equilibrium adsorption capacity using an IGA analyzer allowed us to calculate CO₂ selectivity factors. We assessed the influence of exhaust gas carbon dioxide concentration (16%, 30%, 81.5% and 100% vol.) and adsorption step temperature (25 °C, 50 °C and 75 °C) on the CO₂ adsorption capacity of the bioadsorbent. We also investigated its stability and regenerability in multi-step adsorption–desorption using a TG-Vacuum system, simulating the VSA process and applying different pressures in the regeneration step (30, 60 and 100 mbar_abs). The tests conducted assessed the possibility of using a produced bioadsorbent for capturing CO₂ using the VSA technique.

1. Introduction

Reducing greenhouse gas emissions, including carbon dioxide, alongside transitioning to a circular economy, is the primary action needed to slow down climate change. One of the solutions proposed for reducing carbon dioxide emissions is implementing CCS (Carbon Capture and Storage) and CCU (Carbon Capture and Utilization) technologies. Interest in these technologies stems from the need to reduce the enormous CO₂ waste streams, mainly originating from power plants and combined heat and power plants but also from other energy-intensive sectors such as the cement, iron/steel, chemical and petrochemical industries. The initial process in both CCS and CCU technologies is CO₂ capture. The exhaust gas streams from various industrial and energy sectors differ in terms of CO₂ content. For example, coal-fired power plants emit flue gases with a CO₂ content of 12–16% vol.; cement production produces 14–33% vol.; the iron/steel industry produces 20–27% vol.; the petrochemical industry produces varying levels, depending on the product—8% (ethylene), 12% (methanol), 18% (ammonia); refineries produce 3–13% vol.; and the chemical industry produces up to 100% [1,2]. These concentrations are important in the capture process, as carbon dioxide is easier and less expensive to separate from streams with a high concentration.

Carbon dioxide capture can be conducted in various ways. The most technologically developed method is based on liquid absorbents such as amine solutions, though that based on solid adsorbents is another worth taking into account. This method can be divided into the following techniques: PSA (Pressure Swing Adsorption), VSA (Vacuum Swing Adsorption) and TSA (Temperature Swing Adsorption). These techniques can also be combined, as exemplified by PTSA (Pressure Temperature Swing Adsorption), VPSA (Vacuum Pressure Swing Adsorption) and VTSA (Vacuum Temperature Swing Adsorption). The advantages of the adsorption method include its simple operation, absence of corrosion, low-cost adsorbents, high efficiency and low energy consumption. The VPSA method, based on gas separation through cyclic pressure changes, is proposed for capturing carbon dioxide from the energy sector and other industries [3,4]. In the case of the adsorption method, the type of adsorbent used plays a significant role in CO₂ separation efficiency and energy demand. The adsorbent proposed for use in VPSA installations should be characterized by a high carbon dioxide adsorption capacity, as well as selectivity. Moreover, the adsorbent should be fully regenerable without adsorption capacity loss and without any influence on its structure or physicochemical properties, since the adsorption process operates in multi-step cycles. The adsorbent’s stability is also of crucial importance, as this directly affects the material’s lifespan and therefore the need to replace it in adsorption installation. Another important factor influencing the durability of the adsorbent is its mechanical strength. Additionally, the adsorbent should exhibit tolerance to moisture and other contaminants found in flue gases, such as SOₓ and NOₓ [4,5].

Currently, the main types of adsorbents proposed for CO₂ capture installations are zeolites [6,7], activated carbons [5,8,9] and metal–organic frameworks (MOFs) [10]. Recent reports in the literature [11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31] indicate a growing interest in the use of adsorbents derived from various types of biowaste for CO₂ capture. The utilization of waste materials, including food waste, to produce adsorbents not only offers inherent environmental benefits but also constitutes a means of valorizing these waste streams.

Bioadsorbents can be produced from inexpensive and commonly available raw materials, such as biomass and industrial by-products from sectors including forestry, agriculture and the food industry. In particular, valorizing spent coffee grounds appears highly promising, since coffee is one of the most popular beverages globally and the market has steadily been expanding [23,24,25,26,27,28,29,30,31]. Global coffee production has increased from 151.3 mil. 60 kg bags in 2015/2016 [23] to 168.4 mil. 60 kg bags in 2019/2020 and 178.0 mil. 60 kg bags in 2023/2024 [32]. The residue generated during the brewing process (spent coffee grounds) constitutes about 45% w/w [33]. Therefore, recycling or upcycling this coffee ground waste can significantly reduce disposal costs and contribute to the development of a circular economy.

Carbon adsorbents can be obtained from spent coffee grounds via carbonization and chemical activation using agents such as zinc chloride, phosphoric acid, mixtures of zinc chloride and phosphoric acid, or potassium hydroxide, as well as through physical activation using CO₂, steam, or air. The resulting bioadsorbents are carbon-rich materials exhibiting hydrophobic properties, which can be further activated to develop high porosity, while still allowing for mild regeneration conditions. The selection of porous carbon precursors and the specific preparation method have a significant impact on the structure and porosity of the resulting carbon material. Carbon materials derived from spent coffee grounds find applications in various fields, including in reducing carbon dioxide emissions. Of particular interesting is the utilization of spent coffee grounds to produce adsorbents for CO₂ capture [24,25,26,27,28,29,30,31]. The appropriate processing of this biomass enables the production of biocarbon with tailored properties, such as a specific microstructure and surface area, optimized for efficient carbon dioxide adsorption. Data in the literature [24,25,26,27,28,29,30,31] indicate that bioadsorbents derived from spent coffee grounds exhibit a considerable CO₂ adsorption capacity. Plaza et al. [26,27] have already produced biocarbons from spent coffee grounds through both physical and chemical activation methods. Their preliminary studies demonstrated the significant potential of this waste material for CO₂ capture. The obtained carbon materials exhibited CO₂ adsorption capacities of up to 215.6 mg_CO2 g_A⁻¹ at 273 K and 132 mg_CO2 g_A⁻¹ at 298 K, which are higher than those of commercially activated carbons. KOH-activated samples showed improved textural development compared to those activated using CO₂. In the case of KOH activation, it was observed that increasing the activation temperature enhanced the development of micro- and mesopores at the expense of ultra-microporosity. Using a fixed-bed reactor under ambient pressure and a constant CO₂ concentration (30% vol.), Mukherjee et al. [19] investigated the CO₂ capture performance of adsorbents derived from spent coffee grounds at various activation temperatures (300–900 °C). The highest adsorption capacity, 123.2 mg_CO2 g_A⁻¹, was recorded at 300 °C. A comparable adsorption performance (117.5 mg_CO2 g_A⁻¹) at 35 °C and under pure CO₂ was reported by Liu S.H. and Huang Y.Y. [25], who studied the temperature-dependent CO₂ capture performance (35–50 °C), as well as the cyclic stability and selectivity of coffee-derived adsorbents.

Numerous studies [11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31] on the use of spent coffee ground-derived bioadsorbents present CO₂ adsorption capacity results at atmospheric pressure using the thermogravimetric method for pure carbon dioxide [25,27,30] or a gas mixture [26,27], as well as adsorption isotherms using the volumetric method [24,26,28,29,30]. There are few data on investigations of such materials in VSA systems in adsorption–desorption cycles [26], as well as few works presenting an analysis of the material parameters at each stage of their transformation [27], i.e., from biowaste, through biochar, to bioadsorbent.

Therefore, the present study of coffee-based bioadsorbent focuses on three aspects: (1) showing the changes in physicochemical properties at each stage of obtaining the product, starting from the raw material; (2) studies of the adsorbent selectivity and CO₂ adsorption capacity for different carbon dioxide concentrations in simulated gases, as well as variable temperatures showing their possible applications; and (3) multiple adsorption–desorption investigations to prove the bioadsorbent’s stability and confirm its suitability in CO₂ capture VSA installations.

2. Materials and Methods

2.1. Bioadsorbent Preparation Method

The method of preparing bioadsorbent from food waste is detailed in Figure 1.

The bioadsorbent (AHRoCG—Activated Highly Roasted Coffee Grounds) was produced in a two-step process. In step I (carbonization), highly roasted coffee ground post-brewing residue (HRoCG—Highly Roasted Coffee Grounds) was pre-dried at room temperature and then dried at 120 °C for 24 h to remove moisture. Next, the material was carbonized at 700 °C for 45 min in a nitrogen atmosphere (the high temperature contributed to the decomposition of organic matter and the nitrogen prevented oxidation). The final product was porous carbon structures characteristic of biocarbon (CHRoCG—Carbonized Highly Roasted Coffee Grounds). In stage II (modification), the chemical activation of the CHRoCG was carried out using KOH. For this purpose, solid KOH was mixed with the CHRoCG in a mass ratio of 2:1 and moisturized for 2 h. Next, the mixture was dried at 105 °C in an air dryer for 24 h. After this, it was heated in a muffle furnace at a rate of 5 °C min⁻¹ up to 700 °C in a nitrogen atmosphere. After 45 min of thermal activation at 700 °C, the mixture was cooled and washed with distilled water until pH = 7 to remove residual KOH. Finally the sample was dried at 105 °C in an air dryer for 12 h and denoted as AHRoCG.

The particles sizes of the biowaste subjected to the carbonization process, the biochar used in the activation process and the final product—the bioadsorbent—are presented in Figure 2.

Most HRoCG particles (Figure 2a) are in the range of 0.75–1.25 mm (the largest do not exceed 1.75 mm), CHRoCG (Figure 2b) are 0.5–1.0 mm (the maximum 1.75 mm), and, after activation, the bioadsorbent particles are (Figure 2c) 0.25–0.5 mm (the maximum ca. 1.25 mm). This confirms that the bioadsorbent production process reduces the size of the material at each step; however, from the point of view of application in fix beds, pelletization is required.

2.2. Material Characterization Methods

The physicochemical properties of the biowaste, biocarbon and bioadsorbent were characterized by a wide range of analyses. The chemical composition, including the carbon, hydrogen, nitrogen and sulfur contents in the sample, was ascertained using a CHNS analyzer (LECO TruSpec CHNS, LECO Corporation, St. Joseph, MI, USA). The thermal properties of the samples were characterized via thermogravimetric analysis (TGA/DSC1, Mettler Toledo, Greifensee, Switzerland). Each sample was heated at 10 °C min⁻¹ from 25 to 800 °C under a nitrogen flow (50 cm³ min⁻¹). The samples’ morphological features were investigated by scanning electron microscopy (S-3400N, Hitachi, Tokyo, Japan). Their porosity was determined via N₂ adsorption–desorption isotherm performed at −196 °C using a sorptometer (ASAP 2010, Micromeritics, Norcross, GA, USA). The specific surface area was calculated via the BET method from the linear part of the BET plot according to the IUPAC (International Union of Pure and Applied Chemistry) recommendations using the adsorption isotherm (relative pressure (p/p_o) = 0.05–0.23). The total pore volume (V_p) was determined based on the maximum amount of N₂ adsorbed within the pores of the samples at p/p_o of 0.99. The pore size distribution was calculated using the DFT method. The micropore volume (W_o) was determined based on N₂ adsorption isotherms using the Dubin–Raduszkiewicz equation and assuming an adsorbed phase density of 0.808 cm³ g⁻¹ and cross-sectional area of 0.162 nm² [34]. The average pore width (L_o) was calculated using the Stoeckli–Ballerini equation [35]. Prior to adsorption, the samples were degassed at 300 °C overnight in a high vacuum. The changes in spectral functional groups during the transformation of biowaste into biochar—and then into bioadsorbent—were shown using FT-IR analysis in the spectral range of 4000–400 cm⁻¹ (Nicolet iS10, ThermoFischer Scientific, Waltham, MA, USA).

2.3. Adsorption Capacity Investigations

2.3.1. Isotherms of CO₂, O₂, and N₂

The adsorption equilibriums of three pure gases (99.999%), CO₂, O₂ and N₂, on AHRoCG were ascertained using a gravimetric analyzer (IGA, Hiden Isochema, Warrington, UK). Before testing, the sample was regenerated by degassing for 24 h at 150 °C. Next, it was cooled to 25 °C and the adsorption process was started. Six equilibrium points were determined by stepwise pressure changes in the range of 0–1 bar. The measurement of a given isotherm point ended when the fluctuation of the sample mass reached 99.5% of the predicted asymptotic value. Based on the determined adsorption equilibrium points, the adsorption isotherms of the individual gases were plotted.

2.3.2. Thermogravimetry CO₂ Adsorption Tests Using Different Gas Mixtures

The CO₂ adsorption capacity of the bioadsorbent was examined using thermogravimetric analysis (TGA/SDTA 851^e, Mettler Toledo, Greifensee, Switzerland). The investigation was conducted according to the isothermal adsorption test. Before the adsorption process, the sample of AHROcG was regenerated by (1) heating from 30 to 120 °C in a N₂ atmosphere at a heating rate of 20 °C min⁻¹, (2) baking at this temperature for 30 min in a N₂ atmosphere (until a constant sample mass was achieved) and (3) cooling from 120 to 25 °C in a N₂ atmosphere with a cooling rate of −5 °C min⁻¹. Then, (4) the CO₂ adsorption process was conducted isothermally at a temperature of 25 °C/50 °C/75 °C for 60 min under atmospheric pressure and a flow rate of 100 cm³ min⁻¹, using four different gas compositions (expressed in % vol.): (1) 16.0% CO₂; 80.5% N₂; 3.5% O₂ (characteristic for flue gases from conventional coal-fired plants); (2) 30% CO₂; 60% N₂; 10% O₂ (in the case of exhaust gas from cement plants); (3) 81.5% CO₂; 14.5% N₂; 4.0% O₂ (a concentration of flue gas from coal oxy-combustion or a concentration of CO₂ in the enriched gas before the second stage of adsorption purification); and (4) 100% CO₂ (as a reference, for which the adsorption capacity is as reported in the literature).

2.3.3. Cyclic Adsorption–Desorption Test

The several adsorption–desorption cycles were assessed using a TG-Vacuum system consisting of a (1) thermogravimetry analyzer TGA/SDTA 851^e (Mettler Toledo, Greifensee, Switzerland), a (2) pressure control system (RVC 300, Pfeiffer, Aßlar, Germany), and a series of valves and a vacuum pump (Duo 5 M, Pfeiffer, Aßlar, Germany). In this way, we simulated the VSA CO₂ capture process in order to estimate the regenerability and stability of the bioadsorbent. This system (shown in Figure 3) can be used as a fast laboratory method for assessing the ability of adsorbents to work in multi-stage adsorption–desorption cycles and thus determining their suitability for CO₂ separation in VSA units.

Before each cyclic adsorption–desorption test, the bioadsorbent sample was prepared in four steps: (rI) heating from 30 to 120 °C at a rate of 20 °C min⁻¹ in a N₂ atmosphere and at a gas flow rate of 100 cm³ min⁻¹, (rII) baking for 30 min at 120 °C in a N₂ atmosphere and at a gas flow rate of 100 cm³ min⁻¹ in order to degas (remove any moisture and volatile substances), (rIII) cooling to 25 °C at a rate of −5 °C min⁻¹ in a N₂ atmosphere and at a gas flow rate of 100 cm³ min⁻¹, and (rIV) vacuum degassing at 25 °C for 15 min with no gas flow under the same pressure as applied during the cyclic desorption process, i.e., 30, 60 or 100 mbar_abs. This step was conducted to remove the N₂ from the measurement cell of the thermogravimetric analyzer, as well as the adsorbent pores. Then, after sample regeneration, a four-cycle adsorption–desorption test (ads I–IV/des I–IV) was carried out. The adsorption step was conducted in a CO₂ atmosphere (100% vol.) for 15 min and at a flow rate of 100 cm³ min⁻¹. This step was characterized by a gradual increase in pressure (from the set pressure in the desorption step to atmospheric pressure) in the measurement cell of the thermogravimetric analyzer in which the sample was placed. After CO₂ sorption, a desorption process took place. The adsorbent was regenerated at three different vacuum levels, i.e., 30, 60 or 100 mbar_abs, and maintained for 15 min without N₂ flow. During cycles, the temperature was constant at 25 °C (the optimal temperature for CO₂ sorption–desorption on physical sorbents). Based on the mass change, the following indicators were determined: CO₂ adsorption capacity during the adsorption step (mass increase) and regeneration capability in the desorption step (mass decrease). The values were recorded in mg_CO2 g_A⁻¹.

3. Results and Discussion

3.1. Physicochemical Properties of Bioadsorbent

The chemical composition of the HRoCG, as well as that of the obtained CHRoCG and AHRoCG, in the two-stage treatment process is shown in Figure 4. The initial biowaste had a carbon content of 44.8% w/w, which increased up to 66.7% w/w in the biochar and 68.9% w/w in the final bioadsorbent. This confirms the high-carbon content of the obtained material as a result of the carbonization of the HRoCG and following the activation of the CRoCG by KOH. The AHRoCG is characterized by higher nitrogen and lower hydrogen and oxygen contents compared to the HRoCG and CHRoCG. During the investigation, no sulfur was found in all three samples and, as such, these values are not included in the figure.

The three investigated materials’ thermal decomposition (TG curve) profiles and derivatives (DTG curve) during heating in an inert atmosphere (N₂) at a rate of 10 °C min⁻¹ are shown in Figure 5.

In the case of HRoCG, a twofold weight decrease can be observed. The first mass decrease (in the temperature range of 25–150 °C) results from the desorption of physically adsorbed water and amounts to about 3.2%. The second mass loss (220–550 °C) amounts to about 66.3% and is the result of the thermochemical biopolymer fraction conversion process. This includes cellulose and hemicellulose decomposition (the maximum peak observed on the DTG curve at 300 °C) and probably lignin degradation (at about 400 °C) [19]. In the case of CHReCG and AHReCG, significant dehydration-related mass loss (25–150 °C) can be observed; this was 7.5% in the first case, and 27.5% in the second.

The structure of the initial HRoCG, as well as that of the CHRoCG and AHRoCG obtained as a result of its modification, is shown through SEM photography in Figure 6.

An SEM image of the HRoCG (Figure 6a) shows a compact structure with a small number of visible primary pores, resulting from its lignocellulosic composition. The CHRoCG obtained through carbonization (Figure 6b) exhibits porosity (the presence of macropores), as formed by the removal of volatile components and the degradation of the lignin structure. Significant porosity developed after the biocarbon modification stage using KOH solution. The SEM image of AHRoCG (Figure 6c) reveals the presence of numerous fine micropores forming a highly regular network, which confirms the substantial microporosity of the bioadsorbent and, consequently, its potential for CO₂ capture.

Table 1. Porous structure properties of biowastes, biocarbon, and bioadsorbents.

Samples	SBET	Vp	W0	L0
	m2 g−1	cm3 g−1	cm3 g−1	nm
HRoCG	0.21	0.00006	n/a	1.54
CHRoCG	18	0.008	0.004	1.36
AHRoCG	1580	0.84	0.5	0.96

summarizes the textural parameters and specific surface area of raw biowaste, biocarbon and bioadsorbent.

As presented in Table 1, the specific surface area of the HRoCG (0.21 m²/g) increased to 18 m²/g for the CHRoCG. Although carbonization effectively removes some volatile compounds and initiates pore development [15], only chemical activation using KOH (the most frequently studied method for activated carbon synthesis due to its ability to form a large quantity of micropores inside activated carbon [13]) leads to a significant increase in the specific surface area (up to 1580 m²/g) of the AHRoCG. During this process, micropores are created and expanded, and blocking organic fractions are removed, which significantly increases the availability of the sorption surface of the material [15]. The modification process not only enhances the surface area, but also leads to increased total pore (0.84 cm³/g) and micropore volumes (0.5 cm³/g), and a decrease in the average pore diameter of the AHRoCG to 0.96 nm.

Figure 7 presents the N₂ adsorption–desorption isotherms for the AHRoCG (Figure 7a), as well as for the HRoCG and CHRoCG obtained through its modification (Figure 7b).

The AHRoCG (Figure 6a) exhibits a Type I isotherm according to the IUPAC, indicating that most of the N₂ is adsorbed at low pressures without the presence of a hysteresis loop. A Type I isotherm suggests that the adsorbent is predominantly microporous [34]. In the case of the CHRoCG (Figure 7b), the presence of a Type IV isotherm indicates the existence of micropores as well as mesopores [36]. A noticeable H2-type hysteresis loop is also observed, which is characteristic of pore structures with narrow necks and wider bodies, causing diffusion limitations. At low relative pressures, adsorbate layer formation on the adsorbent surface occurs. The low pore volume also supports the mesoporous nature of this material. The raw HRoCG (Figure 7b) exhibits minimal N₂ adsorption at high vapor pressures, which suggests a homogeneous structure with a negligible specific surface area and no porosity. The pore volume distributions by pore diameter of AHRoCG, CHRoCG and HRoCG are presented in Figure 8a,b.

As shown in Figure 8a, the majority of the total pore volume in AHRoCG is attributed to pores with a diameter of 0.86 nm, which classifies them as micropores (diameter < 2 nm). This confirms the dominant contribution of micropores in the structure of the resulting bioadsorbent. In turn, the CHRoCG (Figure 8b) is characterized by a total pore volume at a diameter of 1.36 nm (diameter < 2 nm), a small contribution at a diameter of 2.52 nm (diameter > 2 nm) and a negligible volume above 14 nm. This classifies the CHRoCG material as both microporous and mesoporous. In contrast, the raw HRoCG exhibits no porosity. In summary, the physicochemical properties, particularly the porous structure parameters of the obtained AHRoCG, indicate its significant potential for CO₂ capture.

FT-IR analyses conducted (Figure 9) show spectrum changes in biowaste after the carbonization and activation processes.

The 850, 1050 and 1350 cm⁻¹ bands correspond to the C-H, C-O-C and C-H₂ bonds, which indicate the presence of cellulose and hemicellulose aromatics and disappear after thermal treatment. The 1600 cm⁻¹ band is characteristic of lignin occurring in biomass, whereas in the case of biochar and bioadsorbent it indicates the appearance of a graphite-like structure. The bands in the 1700 cm⁻¹ range represent the C=O stretching of carbonyl groups. The 2900 cm⁻¹ band corresponds to the vibrations of C-H bonds in the aliphatic-chain structure and is not observed after the carbonization and activation processes. The last one at 3400 cm⁻¹ is specific for -OH bonds of chemisorbed water. These FTIR spectrum results are similar to those reported by Ramos et al. [8].

3.2. CO₂ Capture

3.2.1. Adsorption Isotherms of CO₂, O₂ and N₂

The adsorption isotherms (Figure 10) show the relationship between the pure gas pressure in the bulk phase (in bar) and its concentration in the adsorbed phase (in mg g⁻¹_A).

The CO₂ isotherm is characterized by nonlinearity and a sharp increase in carbon dioxide adsorption at low pressures. With increasing pressure, the capacity gradually decreases; however, it is still much greater than in O₂ and N₂ isotherms, which feature characteristic linearity and low adsorption of these gases.

In the case of a gas mixture, the parameter that determines the separation efficiency of one gas component with respect to another is the selectivity coefficient, which can be determined according to the following formula [7]:

$$S_{separated/existing in mixture}={\displaystyle \frac{x_{component 1}/x_{component 2}}{y_{component 1}/y_{component 2}}}$$

(1)

where $x_{component 1}$ is the molar fractions of separated components in the adsorbed phase, $x_{component 2}$ is the molar fractions of another existing components in the adsorbed phase, $y_{component 1}$ is the molar fractions of separated components in the bulk phase and $y_{component 2}$ is another existing components in the bulk phase.

This is important for the studies described in Section 3.2.2 and Section 3.2.3. The selectivity coefficients of CO₂/O₂ and CO₂/N₂ for three gas mixtures are listed in

Table 2. CO₂ selectivity coefficient of components.

Gas Mixture	${\mathit{S}}_{\mathbf{C}\mathbf{O}2/\mathbf{O}2}$	${\mathit{S}}_{\mathbf{C}\mathbf{O}2/\mathbf{N}2}$
16.0% CO2; 3.5% O2; 80.5% N2	12.7	22.3
30% CO2; 10% O2; 60% N2	7.2	42.7
81.5% CO2; 4.0% O2; 14.5% N2	27.1	491.9

.

As can be seen from Table 2, the selectivity coefficient depends on the partial pressures of CO₂, O₂, and N₂ in gas mixtures. An increase in the partial pressure of one gas, e.g., CO₂ (from 16.0 to 81.5%), at an almost constant partial pressure with the second gas (e.g., O₂) increases the selectivity by more than double (from 12.7 to 27.1); in the case of a simultaneous reduction in the partial pressure of the second gas (e.g., N₂ from 80.5 to 14.5%), a very significant increase—more than 22-fold (from 22.3 to 491.9)—is observed. Coffee-derived bioadsorbents presented in the literature show similar CO₂/N₂ selectivity coefficients. The coffee adsorbent obtained by Plaza M.G et al. [26] showed CO₂/N₂ selectivity ranging from 15 to 25 at 50 °C and 130 kPa for CO₂ concentrations from 9% to 31%. In turn, the coffee-based bioadsorbent CAC-2–800 obtained by Wang et al. [29] showed a CO₂/N₂ coefficient equal to 33.

3.2.2. Effect of Temperature and CO₂ Concentration on Carbon Dioxide Adsorption Capacity of Bioadsorbent

Figure 11, Figure 12, Figure 13 and Figure 14 present the carbon dioxide adsorption profiles of bioadsorbent at different feed gas CO₂ concentrations (simulating the exhaust gases from various types of industries (16 vol.% CO₂, 30 vol.% CO₂, 81.5 vol.% CO₂ and, for reference, 100 vol.% CO₂), and at three different adsorption temperatures (25 °C, 50 °C and 75 °C). The initial sample weight of about 6.2 mg was subjected to temperature regeneration (comprising three steps: heating, baking and cooling). Next, the carbon dioxide adsorption process was carried out within 60 min at atmospheric pressure. The mass change during the whole process is indicated with a solid black line and the temperature change with a dashed red line. To better illustrate the adsorption capacity of the applied adsorbent (mg_CO2 g_A⁻¹) in terms of unit mass, the main graph was supplemented with a nested graph (blue one).

The CO₂ adsorption profiles in Figure 11 show that the adsorption capacity of AHRoCG at 16% vol. CO₂ in the simulated gas mixture is 39.3 mg_CO2 g_A⁻¹ at 25 °C (Figure 11a). This decreases with rising temperature to 23.5 mg_CO2 g_A⁻¹ at 50 °C (Figure 11b) and 14.7 mg_CO2 g_A⁻¹ at 75 °C (Figure 11c). The adsorption curves shown in Figure 11a–c indicate that the sorption process proceeds similarly and relatively quickly at each tested temperature. Plaza M.G et al. [26] studied the CO₂ adsorption of coffee adsorbents and also showed fast adsorption kinetics, indicating their suitability as candidates for rapid swing adsorption cycles.

The CO₂ adsorption profiles presented in Figure 12 show that the adsorption capacity on the AHRoCG at 30 vol.% CO₂ is 65.8 mg_CO2 g_A⁻¹ at 25 °C (Figure 12a), 41.2 mg_CO2 g_A⁻¹ at 50 °C (Figure 12b) and 27.0 mg_CO2 g_A⁻¹ at 75 °C (Figure 12c), decreasing with increasing temperature. A similar situation occurs for the highest analyzed CO₂ concentrations in processing gases, i.e., for 81.5% vol. CO₂ the adsorption capacity is 113.7 mg_CO2 g_A⁻¹ at 25 °C (Figure 13a), 76.1 mg_CO2 g_A⁻¹ at 50 °C (Figure 13b) and 51.7 mg_CO2 g_A⁻¹ at 75 °C (Figure 13c), while for 100% vol. CO₂: 123.6 mg_CO2 g_A⁻¹ at 25 °C (Figure 14a), 84.0 mg_CO2 g_A⁻¹ at 50 °C (Figure 14b) and 57.4 mg_CO2 g_A⁻¹ at 75 °C (Figure 14c).

Figure 15 and Figure 16 summarize the CO₂ adsorption capacity of the bioadsorbent depending on the temperature during the adsorption process and the concentration of carbon dioxide in the feed gas.

In the case of all tested CO₂ concentrations in the simulated gases, the adsorption capacity of the bioadsorbent decreased with increasing process temperature, which is typical for physical adsorbents (Figure 15). This indicates that the most favorable temperature for the adsorption process is the lowest set, i.e., 25 °C.

Increases in temperature (from 25 °C to 50 °C and then to 75 °C) caused decreases in adsorption capacity of −40% and −63% w/w, respectively, in the case of 16% vol. CO₂ in the exhaust gas. The higher the concentration of CO₂ in the feed gas subjected to separation, the smaller the observed decrease in adsorption capacity: −37% and −59% in the case of 30% vol. CO₂, −33% and −55% in the case of 81.5% vol. CO₂ and −32% and −54% in the case of 100% vol. CO₂.

Similar conclusions were drawn by Mukherjee et al. [19], who studied the CO₂ capture of coffee adsorbents at different temperatures (from 30 °C to 90 °C) at a constant CO₂ concentration (30 vol.%) under ambient pressure in a fixed-bed reactor. The authors found that the CO₂ adsorption capacity decreased significantly with increasing temperature (from 30 °C to 90 °C), which is typical for physical adsorption. The coffee adsorbent they studied had the highest CO₂ adsorption capacity of 123.2 mg_CO2/g_A⁻¹ at 30 °C.

An increase in CO₂ concentration in the feed gas increased the adsorption capacity of bioadsorbents (Figure 16). This is due to the higher concentration gradient between the gas molecules and the adsorbent surface, which results in a higher mass transfer coefficient. However, the decisive influence on mass transfer is, ultimately, adsorbent pore diffusion, which limits its rate [37].

Increasing the CO₂ concentration from 16 to 30% vol. CO₂, then to 81.5% vol. CO₂, and finally to 100% vol. CO₂ increased the bioadsorbent adsorption capacity by 68% w/w, 190% w/w and 215% w/w at 25 °C. However, the higher the temperature in the adsorption step, the more and more noticeable the mass increase becomes (76%, 225%, 258% w/w at 50 °C and 84%, 252%, 291% w/w at 75 °C); however, the capacities achieved in relation to 25 °C were lower.

The AHRoCG already showed a significant adsorption capacity of 39.3 mg_CO2 g_A⁻¹ at 16% vol. CO₂ in the feed gas and a temperature 25 °C. This indicates that it can be applied for gases with low CO₂ concentrations such as flue gas from the power industry (10–16% vol. CO₂). However, the higher the CO₂ concentration in the feed gas, the greater capacity, which is seen at 30% vol. CO₂ (65.8 mg_CO2 g_A⁻¹ at 25 °C); this concentration is characteristic of the cement industry (20–30% vol. CO₂).

3.2.3. VSA Process Simulation Tests

CO₂ capture was simulated using the VSA technique and a TG-Vacuum system. After the adsorbent regeneration process (comprising four steps: (Ir) heating, (IIr) baking, (IIIr) cooling, (IVr) vacuum degassing), a sample with an initial weight of about 6 mg was subjected to four cyclic steps of carbon dioxide adsorption and desorption (ads I–IV/des I–IV). The results are presented in Figure 17, Figure 18 and Figure 19. In addition to the main graph presenting the mass changes in the sample during the whole experiment (solid black line), as well as the temperature changes (dashed red line), the nested graph (blue one) was also included, which shows the CO₂ adsorption capacity of the adsorbent for four adsorption–desorption cycles, normalized to the initial value. The adsorption steps (ads I–IV) were carried out by slowly compressing the thermogravimeter measurement cell using pure CO₂ from the set negative pressure (in the regeneration step) to ambient pressure (in the adsorption step). The ambient pressure was usually reached 4 min before the end of the adsorption step. The results were illustrated in the form of growing mass curves, reaching equilibrium before the end of the step. In turn, the desorption steps (des I–IV) were conducted by quickly depressurizing the thermogravimeter measurement cell to the set negative pressure, usually within about 3 min, and maintaining this at constant level until the end of the step with no purge gas flow. This process shows a rapid decrease in mass until the set vacuum pressure is reached, after which the mass decrease slows down significantly; this is related to the rate at which the component diffuses out of the adsorbent. In adsorbent use for industrial installations, it is not necessary to conduct full regeneration (which may take longer). In this way, these tests allow us to estimate adsorbent stability and provide information about its suitability for continuous operation in industrial installations.

The tests conducted using the TG-Vacuum system, simulating the VSA process, confirmed the very good sorption–desorption performance of the investigated bioadsorbent (Figure 17, Figure 18 and Figure 19). The resulting profiles indicate that stable adsorption–desorption cycle was gradually reached, demonstrating decreasing differences between the initial and final mass of the adsorbent in a particular cycle (solid black line). However, the working CO₂ adsorption capacities (solid blue lines) remain at a similar level across successive cycles (starting from cycle 3). This demonstrates the bioadsorbent’s achieved stability. The working CO₂ adsorption capacity is equal ca. 121 mg_CO2 g_A⁻¹ at a desorption pressure of 30 mbar_abs, ca. 108 mg_CO2 g_A⁻¹ at 60 mbar_abs, and ca. 92 mg_CO2 g_A⁻¹ at 100 mbar_abs. As shown in Figure 17, Figure 18 and Figure 19, no reduction in the working CO₂ adsorption capacity was observed, which confirms the adsorbent’s good stability and potential for use in multicyclic processes. The possibility of using coffee-based adsorbents in VSA systems was investigated by Plaza M.G et al. [26]. They tested various VSA cycle configurations at 50 °C in a fixed-bed adsorption unit to evaluate the adsorbent’s performance under cyclic operation. The tested coffee-derived adsorbent showed no signs of deactivation during extended operation in the VSA system.

A preliminary evaluation of the bioadsorbent using the TG-Vacuum system to simulate the VSA process determined the stability of the obtained material during adsorption–desorption cycles (Figure 20). This feature is of key importance in practical applications, i.e., for filling adsorption columns in VSA installations.

Table 3. Comparison of CO₂ adsorption capacity of selected bioadsorbents obtained from coffee grounds.

Bioadsorbent	Porous Structure Parameters	CO2 Sorption	Literature
	SBET, m2 g−1 Vp, cm3 g−1 W0, cm3 g−1	a, mgCO2 gA−1 (25 °C, 1 bar)
CG 400 2-1; CG 400 4-1; CG 700 2-1; CG 700 4-1	1624–2785 0.66–1.36 0.59–0.79	123.6–194.5	[24]
NCLK1; NCLK2; NCLK3; NCLK4; NCLK3b; NCHA29; NCHA36; NCHA41	568–1082 0.24–0.51 0.22–0.41	92.4–132.0	[27]
CACs-1–800; CACs-2–800; CACs-3–800; CACs-4–800; CACs-2–700; CACs-2–900	654–1931 0.32–1.31 -	106.5–168.1	[29]
CG600-1; CG600-5; CG700-1; CG700-5; CG800-1; CG800-5	645–2337 0.26–1.15 -	151.8–199.8	[31]
AHRoCG	1580 0.84 0.5	123.6	this study

presents a comparison of the CO₂ sorption capacities of different coffee ground-derived bioadsorbents, as presented both in this work and in the literature.

In this work, the CO₂ adsorption capacity of the coffee ground-derived bioadsorbent (at 25 °C) was close to the values reported in the literature, ranging from 92.4 mg_CO2 g_A⁻¹ [27] to 199.8 mg_CO2 g_A⁻¹ [31]. The high adsorption capacity, combined with its good regenerability and stability in multi-stage adsorption–desorption processes, indicates these adsorbents’ significant potential for CO₂ capture in the power industry and other sectors.

4. Conclusions

CO₂ adsorption studies conducted on bioadsorbent derived from highly roasted coffee grounds showed that, due to the porous structure developed (large specific surface area and large pore volume, particularly including micropores), the adsorbent is suitable for carbon dioxide removal in the power industry and other sectors in VSA installations. The pore structure developed, particularly the existence of narrow micropores, confers high adsorption potential and is the basis for effective CO₂ adsorption. In addition to its high CO₂ adsorption capacity, the bioadsorbent demonstrated high selectivity in terms of CO₂, regenerability and stability. The CO₂ adsorption capacity of the bioadsorbent rose with increasing feed gas CO₂ concentration and decreased with increasing temperature. The use of the bioadsorbent in several subsequent adsorption–desorption cycles did not reduce its working adsorption capacity. Tests using the TG-Vacuum system confirmed that the use of a vacuum at the desorption step is beneficial due to the bioadsorbent’s rapid regeneration.