Production, Activation and CO₂ Uptake Capacity of a Carbonaceous Microporous Material from Palm Oil Residues

Abstract

While Malaysia produces about half of the world’s palm oil and is the largest producer and exporter worldwide, oil palm industries generate large amounts of lignocellulosic biomass waste as a sub-product with no economic market value other than feedstock for energy valorisation. With the aim to increase the sustainability of the sector, in this work we prepare new materials for CO₂ capture from palm oil residues (empty fruit bunches and kernels). The biochar is obtained through the carbonisation of the residues and is physically and chemically activated to produce porous materials. The resulting microporous samples have similar properties to other commercial activated carbons, with BET surfaces in the 320–880 m²/g range and pore volumes of 0.1–0.3 cm³·g⁻¹. The CO₂ uptake at room temperature for physically activated biochar (AC) was 2.4–3.6 mmolCO₂/gAC, whereas the average CO₂ uptake for chemically activated biochar was 3.36–3.80 mmolCO₂/gAC. The amount of CO₂ adsorbed decreased at the highest temperature, as expected due to the exothermic nature of adsorption. These findings confirm the high potential of palm oil tree residues as sustainable materials for CO₂ capture.

1. Introduction

Over 76 million tons of palm oil were produced worldwide in 2020 [1], of which 40% were solid wastes. As a result, the palm oil industry generates one of the most abundant biomass residues with a generation of circa 30 million tons. Malaysia dedicates 45.2% of its total land plantation to palm oil and is the world’s largest producer and exporter, contributing to about half of the total production [2,3]. Oil palm industries produce, together with oil palm as the main product, a great quantity of lignocellulosic biomass, considered as waste, including oil palm trunks (OPT), oil palm fronds (OPF), empty fruit bunches (EFB), palm oil fibres (POF), palm kernel shells (PKS) and palm oil mill effluent palm (POME). These large-scale industrial activities generate large amounts of waste with low or no economic value other than, in some cases, energy generation. Such disposal emergency has been recently aggravated after the Department of Environment in Malaysia banned open burning [4].

In addition to waste generation, the palm oil industry is responsible for a high amount of Greenhouse Gases (GHGs) that are generated at various stages (transportation of raw material, combustion of fuel in boilers or main oil production steps) [5]. Carbon, Capture and Storage (CCS) continues to develop as a central strategy to mitigate GHGs emissions. The main existing CCS technologies include adsorption using activated carbons (ACs), zeolites, and new materials such as metal−organic frameworks (MOFs), polymers, and metal oxides, designed to improve CO₂ adsorption efficiencies [6]. ACs are promising adsorbents for carbon capture due to their (i) abundant availability; (ii) low-cost and sustainability; (iii) high surface area and (iv) high adsorptive and high selectivity characteristics tailored by the number and size of pores generated during the activation process. The adsorptive capacity of ACs is induced when the structure of the raw biomass is modified by dehydration and carbonisation, followed by a chemical or physical activation step [7]. These preparation steps define the morphology of the ACs (i.e., surface area, micropore volume and size) which ultimately determines CO₂ uptake [8]. Khandaker et al. [8] stated that the control of micropores had greater importance for adsorbing high CO₂ compared to surface area and total pore volume and Deng et al. [9] reported that a pore size of 0.33–0.63 nm played an important role in maximizing the CO₂ uptake. It is also noteworthy, in any case, that temperature and pressure also affect gas adsorption to a great extent.

Production of activated carbon (AC) from agricultural byproducts has gained increased interest in recent years because of their abundance, high carbon content and cheap availability. Many agricultural residues have been tested as AC precursors, such as wheat, corn straw, olive stones, bagasse, birch wood, sunflower shell, almond shells, peach stones, straw, or rice husks, among others [10]. The production of ACs usually involves a first carbonisation step followed by an activation method: physical activation, with carbon dioxide, steam, or air at elevated temperatures [11], and chemical activation with chemical activating agents, such as dehydrating agents and oxidants. The most common chemical agents are ZnCl₂, KOH, H₃PO₄ or K₂CO₃ [12]. The selection of the activation method will result in different physicochemical properties of the activated carbon, such as surface area, micropore and mesopore distribution, total pore volume or surface functionality, which will define their adequacy and efficiency as adsorbent materials.

Several activated carbons obtained from agricultural waste have been postulated for CO₂ adsorption in pure and mixed streams, [13,14] and more specifically, palm oil residues have been used in gas and water treatment. Promraksa et al., for example, pyrolyzed PKS, EFB and POF obtaining a biochar with 0.46 mmol/g capacity for CO₂ adsorption [15]. Ibrahim et al. carbonised oil palm biomass, which was further functionalized with nitric acid, facilitating the adsorption of various pollutants from aqueous solutions, such as dyes and heavy metals [16]. Lawal et al. produced activated biochar from palm oil biomass that was then used for water treatment in the palm oil effluents, making the wastewater suitable for reuse in palm oil mills and safe for discharge into the aquatic environment [17]. The same authors also predicted uptake yields and selectivity at varying conditions through molecular dynamic simulations [18].

In this work, we report the preparation and activation of carbons from palm oil residues (Section 3) that have been then used as CO₂ adsorbents for an equimolar mixture of CO₂/N₂. To do so, we have designed and assembled an experimental set-up to obtain breakthrough curves of CO₂ uptake in carbon-based materials (Section 3). In Section 3, we discuss our results and compare them with other adsorbent materials in terms of suitability and efficiency. The outcomes highlight that our strategy can be a sustainable alternative to the disposal and open-burning of palm oil residues, simultaneously giving added value to residues, and ultimately promoting circular economies.

2. Materials and Methods

2.1. Materials

Palm oil residues (empty fruit bunches and kernels, Figure 1a) were obtained from Paloh Oil Palm Mill, Kelantan, Malaysia. Samples were first washed in distilled water several times and then dried in an oven at 100 °C for one day to remove moisture. After drying, samples were manually crushed in a mortar and sieved to obtain particles with diameters between 1.5 mm and 3.5 mm. The resulting feedstock was stored for further use and analyses (Figure 1b). A 0.5M sodium hydroxide solution was prepared from solid granules (Acrōs Organics^®, Geelm Belgium) and used for the chemical activation step (see Section 2.3).

2.2. Preparation of Porous Carbons

Figure 2 shows the bench scale experimental set-up employed for carbonisation. Samples were prepared in a purpose-designed quartz reactor sat within a Nabertherm tubular furnace (RT 50-250/11). The quartz reactor was connected to a flow-controlled N₂ stream (in the range of 0–250 mL·min⁻¹), controlled through a gas flowmeter (Omega^®, Aberdeen, UK). The exit gas went through a Drechsel glass bottle filled with cold water, where the produced bio-oil was trapped. The part of the vessel that remained outside the furnace was covered with an insulating material to avoid heat loss.

Samples were initially weighed (20 ± 0.1 g) and heated in an electric tubular furnace at around 100 °C for 30 min to remove moisture and were further heated at a rate of 10 °C·min⁻¹ to the carbonisation temperature, 650 °C, which was held for an hour. The set-up was then allowed to cool down to room temperature to avoid thermal shock and finally, the carbonised sample was removed. The carbon-based material was weighed again to evaluate the initial burn-off percentage ($B$, %) promoted by carbonisation and calculated as [19]:

$$B\left(\%\right)=\frac{m_0-m_f}{m_0}100 ,$$

(1)

with $m_{0 }$ and $m_f$ the initial and final mass of the samples, respectively.

2.3. Carbon Activation Procedure

Activated carbons were prepared via either physical or chemical activation in the same electric furnace used for carbonisation, see Figure 3. An additional inner vessel, namely the activation vessel, was inserted into the external vessel’s top hole. The activation vessel’s top hole was closed using a rubber stopper, and the outlet gas was directed to a bottle with cold water before being exited into the atmosphere.

Physical activation was carried out using CO₂ as an activating agent using a flow equal to 12 mL·min⁻¹. The system was heated at a rate of 10 °C·min⁻¹ from 100 °C to 900 °C, and the temperature was then maintained for the duration of different times (390–1130 min) to yield a series of activated carbon with various degrees of burn-off. Sodium hydroxide (NaOH 0.5M) was used as the chemical activating agent at different wet impregnation ratios (1:1, 1:2 and 1:3) with respect to the carbonous substrate. The impregnated samples were heated at 700 °C, held for an hour and then left overnight at room temperature while stirring, to remove the residual NaOH [20,21]. Samples were then cleaned with distilled water, filtered, dried in an oven at 100 °C overnight and weighed.

The resulting activated carbon samples were labelled as POR-AC, where A and S refer to the type of activation (P, physical and C, chemical) and settings (burn-off percentage for physical activation and C:NaOH ratio for chemical activation). For example, POR-C11 will be a palm oil residue activated chemically with a ratio C:NaOH equal to 1:1.

The surface area and porosity of all activated carbons were calculated from the N₂ adsorption–desorption isotherms measured at −196 °C with a Micrometrics TriStar 3000 analyser. The specific surface area (S_BET) was estimated by using the Brunauer–Emmet–Teller (BET) equation. The total pore volume was calculated using Gurvich’s law. The Horvath–Kawazoe (HK) method was used to calculate the micropore size distribution. Following the IUPAC classification, pore sizes of 2 and 50 nm represent the boundary of micropores–mesopores and mesopores–macropores, respectively.

2.4. CO₂ Uptake

All the CO₂/N₂ adsorption experiments were carried out in a home-made system at the School of Engineering (University of Aberdeen, Scotland), shown schematically in Figure 4. The set-up consisted of the following elements: CO₂ (UN1013, Linde^®, Aberdeen, UK) and N₂ (UN1066-OFN of Linde^®) pressure bottles; CO₂ and N₂ manual flowmeters (Omega^®); pressure transducer (Omega^®) connected to one computer; brass adsorption cell; digital flowmeter (Agilent Technologies^®, Cheadle, UK) and CO₂ analyzer (G110 Geotech^®, Aberdeen, UK).

A gas adsorption cell was designed and built to carry out both adsorption and permeability tests, with one or two outputs, respectively (Figure 5a). The gas went through the cell from the bottom side and passed through the sample holder (Figure 5b) where the carbonaceous material under test was placed. The bottom side had a second outlet for the collection of retentate when necessary (and was closed and sealed for adsorption tests to avoid leaks). The cell was covered with rock wool to avoid heat dissipation (Figure 5b).

Activated carbons obtained as described previously were pelletised using ashes from palm oil as the binding agent, in a mass ratio of 3:2. Corresponding amounts of ACs and ashes (around 0.25 g of ACs and 0.17 g of ashes) were ground in a mortar and two droplets of distilled water were added to create a paste that was placed inside a pellet die set (Figure 6a). The resulting pellets (diameter of 16 mm, Figure 6b) were then obtained using a mould by pressuring at 100 bar for 1 min.

The feed gas for the CO₂ uptake tests was an equimolar mixture of nitrogen and carbon dioxide at a total flow rate of 100 mL·min⁻¹. All the CO₂ breakthrough curves were obtained at atmospheric pressure and varying temperatures (25 °C, 125 °C and 225 °C). An electrical heater (Omega^®) and a Taisuo^® temperature controller (STP-321, Nigbo, China) with a thermocouple type J were introduced in the system for tests at higher temperatures.

3. Results and Discussion

3.1. Production and Characterisation of Activated Carbons

Initially, 20 g of the pre-treated palm oil residues were carbonised following the procedure described in Section 2.1. An average of 70% (weight %) of the initial mass was removed during carbonisation in all samples. The remaining 30% (wt.) of the carbon-based materials was physically or chemically activated, resulting in the samples described in the following paragraphs.

3.1.1. Physical Activation

Physical activation of the carbonised samples (~2.5 g) was carried out using CO₂. Different burn-off percentages were reached after multiple activation cycles (

Table 1. Physically activated samples.

Sample	Activation Time (min)	Burn-Off (%)
POR-P20	390	23.13
POR-P40	1130	39.57

), with the total time shown as the sum of the duration of each individual cycle. As shown in the table, the longer the activation time, the further development of S_BET and V_TOTAL is found. Hence, ACs with higher burn-off showed increased CO₂ uptakes compared with their analogous half-burn-off values (20% vs. 40%).

3.1.2. Chemical Activation

Chemical activation of carbonised samples (~1.5 g) was carried out using NaOH, as described in Section 2. The resulting burn-off percentages are gathered in

Table 2. Chemically activated samples.

Sample	C:NaOH Mass Ratio	Burn-off (-)
POR-C11	1:1	4.34
POR-C12	1:2	4.90
POR-C13	1:3	3.39

.

The average pore diameter, surface area and total pore volume of all the ACs were obtained from the isotherms, as described above, and the main results are shown in

Table 3. BET results for all the produced ACs (correlation coefficient = 0.99) ¹.

Sample	Average Pore Diameter, dp (nm) *	BET Surface Area, SBET (m2·g−1) **	Total Pore Volume, VTOTAL (cm3·g−1) ***
POR-P20	0.67	653.5	0.22
POR-P40	0.68	881.5	0.30
POR-C11	0.80	319.0	0.11
POR-C12	0.77	356.0	0.22

¹ POR-C13 not available, * average error: 0.01 nm, ** average error: 32 m²·g⁻¹, *** average error: 0.031 cm³·g⁻¹.

. A representative example of N2-adsorption curves is reported in Appendix A.

Table 4. CO₂ uptake results of the physical and chemical activated samples tested at 25 °C.

Sample	CO2 Uptake (mmol/g)
POR-P20	2.41
POR-P40	3.58
POR-C11	3.36
POR-C12	3.75
POR-C13	3.81
Commercial AC [27]	3.84

gathers the average CO₂ uptake for both physically and chemically activated samples at 25 °C. The increasing activation time in the physically activated biochars (higher burn-off percentages) resulted in an increase in the adsorption capacity. This is due to the further developed textural properties (i.e., S_BET, V_Total, see Table 3) exhibited by the activated carbons with higher burn-off values. Even though chemical activation was carried out in a single heating step (with lower burn-off percentages), the corresponding ACs had higher adsorption capacity [6] than that showed by the physically activated ones. It is worth noting that the samples with higher NaOH impregnation ratios also show a slightly higher CO₂ uptake. This is consistent with the higher surface areas and lower average pore diameters promoted at higher NaOH percentages.

Moreover, especially if the subsequent activation process is carried out at high temperatures and for long periods of time, with greater activation times, the adsorption capacity of the activated carbon increases, obtaining good adsorption results even with samples carbonised at low temperatures [22,23]. It can be deduced then that the process that distinguishes the adsorption capacity of ACs is the activation step, with CO₂ uptake values independent of the carbonisation process. Remarkably, previous studies on palm empty fruit bunch showed similar CO₂ uptake values compared with this work [24,25,26]. Commercial AC is also included in the table for comparison.

3.2. Influence of Temperature on CO₂ Uptake

Table 5. CO₂ uptake results of the samples tested at 125 °C and 225 °C.

Sample	CO2 Uptake (mmol/g)
	125 °C	225 °C
POR-C12	1.20	0.32
POR-C13	0.78	0.10

gathers the CO₂ uptake associated with the breakthrough curves of the two samples with the best performance (chemically activated at ratio 1:2 and 1:3) at two temperatures (125 °C–225 °C). The amount of CO₂ adsorbed decreased at the highest temperature, as expected due to the exothermic nature of adsorption (favoured at low temperatures). Interestingly, at higher temperatures, samples with the lowest impregnation ratio performed better, as also observed in other works [28].

The average CO₂ uptake value obtained for physically activated carbons in this work was 3.1 mmolCO₂/gAC and the average CO₂ uptake value for chemically activated carbons was 3.6 mmolCO₂/gAC. Figure 7 shows a comparison of the CO₂ uptakes obtained in this study with other porous carbons reported in the literature. The capacities of CO₂ uptake of our materials are among the best performing in carbon-based adsorbents, studied under similar operating conditions. It is important to highlight that the key advantages of AC materials are their low cost, their insensitivity to moisture and the possibility of their production/synthesis from numerous carbon-based naturally existing or spent materials.

Finally, it is worth noting that, even though not carried out in this work, activated carbons can be completely regenerated (i.e., adsorption is reversible) due to the establishment of weak interactions between CO₂ molecules and the sorbent surface active sites [40]. This feature is highly desirable for potential industrial-scale applications and would help in promoting further circular economy strategies.

4. Conclusions

We have successfully activated biochars obtained from palm oil residues (empty fruit bunches and kernels), using physical and chemical processes. The resulting activated carbons are suitable for CO₂ adsorption, in equimolar mixtures with N₂, and show gas uptakes competitive to other existing carbonaceous materials. The utilisation of NaOH as a chemical activator increased the CO₂ adsorption, with respect to the physically activated materials, attributed to an increased selectivity towards CO₂ adsorption. The CO₂ adsorption capacity decreases with increasing temperature in accordance with the exothermicity of the process.

Our results suggest that the activated carbons obtained from oil palm residues can be used as efficient adsorbents for CO₂, with potential in carbon capture applications. Our experimental process can promote circular economy strategies by alleviating the accumulation of palm oil wastes. To close the cycle, the captured CO₂ could be re-used in the process of physical activation.