The Combination of Nitrogen (N₂) Pyrolysis and Carbon Dioxide (CO₂) Activation for Regenerating Spent Activated Carbon

Abstract

In line with the principles of the circular economy, this study aimed to develop a pyrolysis-activation regeneration process capable of producing highly porous carbon materials from spent granular activated carbon (GAC), which was generated by a high-tech electronics manufacturing company in Taiwan. Thermogravimetric analysis (TGA) and other thermochemical analyses were first conducted to investigate the thermal decomposition behavior of the spent GAC. Subsequently, the thermal regeneration system was employed to perform the N₂ pyrolysis and CO₂ activation experiments under various process conditions (i.e., 800, 850, and 900 °C for holding 0, 30, and 60 min, respectively). Analytical instruments included a surface area and porosimeter for pore property analysis, scanning electron microscopy (SEM) for porous texture observation, and energy dispersive X-ray spectroscopy (EDS) for surface elemental distribution analysis. The results revealed that the pore properties of thermally regenerated GAC were significantly improved compared to the spent GAC, indicating the effective removal or decomposition of adsorbed organics and deposited substances under the process conditions. Additionally, thermal regeneration via physical activation with CO₂ led to enhanced pore properties compared to simple pyrolysis. The maximum BET surface area achieved exceeded 720 m²/g, which was greater than those of spent GAC (approximately 425 m²/g) and N₂-pyrolyzed GAC (approximately 570 m²/g) under the same regeneration conditions (i.e., 900 °C with a 30 min holding time).

1. Introduction

Activated carbons (ACs) are microporous carbon-based adsorbents characterized by their extensive pore properties (e.g., specific surface area, total pore volume), which enable significant adsorption capacities for removing contaminants from air and water streams. Commercially available AC is typically found in two forms: powdered AC (PAC) and granular AC (GAC) [1]. PAC is primarily used in wastewater and water treatment applications such as decolorization in the food processing industries and the removal of trace organics from drinking water, often generating sludge-like waste when PAC becomes exhausted in the liquid phase. In such cases, the regeneration of spent PAC becomes a challenge, and it is often disposed of without reuse. In contrast, GAC, either in granulated form or pelletized form, is widely applied in gas-phase adsorption, including organic vapor recovery, indoor air purification, gas separation, respiratory protection, and vent gas purification [2]. Given the high cost of AC and the ease of regeneration [3], spent GAC used in industrial applications can be regenerated or reactivated using various methods, which will be further addressed in the subsequent section.

Regenerating or reactivating AC offers both economic and environmental advantages, as it reduces the need for treating or disposing of spent AC. Various regeneration methods for exhausted or saturated AC have been reviewed in the literature, including thermal, chemical, biological, electrochemical, microwave, and oxidation techniques [4,5,6,7,8]. Although microwave thermal regeneration is considered a novel technique [9,10,11,12,13], conventional thermal regeneration (desorption) via pyrolysis under an inert gas (i.e., nitrogen) remains the most commonly applied method. However, one drawback of thermal regeneration is that it often fails to fully restore the original pore properties of the AC [14,15,16,17,18], thereby limiting its reuse in adsorption applications. To overcome this limitation, researchers have explored combinations of thermal and chemical oxidation methods. For instance, Álvarez et al. regenerated spent GAC at 850 °C using N₂ (pyrolysis alone) as well as N₂ and CO₂ combined (pyrolysis plus activation) [19]. According to their results, pyrolysis alone was not able to restore the pore properties due to residual charred products remaining on the carbon surface and pores; however, using an oxidant was capable of removing adsorbed organics completely. Cho et al. also regenerated spent GAC by combining N₂ pyrolysis at 500–800 °C with air activation at 600–800 °C for holding times between 0 and 60 min [20]. Although air is rarely used as an oxidant in thermal regeneration, their study suggested it may offer economic advantages due to lower energy consumption and shorter processing times. Toledo et al. evaluated regeneration using acid washing followed by thermal treatment (650–950 °C), demonstrating that acid washing prior to thermal regeneration was critical for reactivating spent GAC [21]. Nasir et al. assessed thermal regeneration of spent commercial activated carbon (SAC) by N₂ pyrolysis (Pyro-RAC) and CO₂ activation (CO₂-RAC) [22], indicating the order of Pyro-RAC > CO₂-RAC > SAC based on their specific area values. Fagbohun et al. used CO₂ activation and chemical activation with K₂CO₃/KCl, showing that combining K₂CO₃ and KCl under N₂ pyrolysis conditions yielded higher pore properties compared to CO₂ activation alone [23]. Márquez et al. showed that while N₂ pyrolysis at 900 °C could provide the highest surface area values [24], a simpler thermal regeneration process at (<350 °C) and 1 h using an oxidizing atmosphere (air) could effectively restore most of the pore properties of exhausted GAC. Yuan et al. regenerated spent GAC using N₂ pyrolysis (700–900 °C) followed by steam activation for 0.5–1.5 h [25], showing that the optimal conditions for pore property recovery of GAC were at 800 °C and 1 h.

In the previous studies [26,27], spent bleaching earth (SBE) was thermally regenerated at 500–800 °C under pyrolysis gas nitrogen (N₂) and 800–900 °C under CO₂ activation. Compared to untreated SBE, the regenerated material exhibited enhanced pore properties due to new pore formation from the carbonization and activation of non-desorbed organics [28]. Although regeneration of spent GAC is a well-explored topic, this work uniquely focuses on using a combination of N₂ pyrolysis and CO₂ activation, making it different from most prior studies. In this context, proximate analysis, thermogravimetric analysis (TGA), and pore property measurements of a spent GAC from a local liquid crystal display (LCD) manufacturing plant were first determined to be used as baseline characteristics. The regeneration experiments were then performed under various conditions (i.e., 800, 850, and 900 °C with holding times of 0, 30, and 60 min, respectively), in compliance with Taiwan’s official regulation for spent GAC regeneration (i.e., ≥800 °C). The resulting pore properties (i.e., surface area, pore volume, and pore size distribution), along with textural and chemical characteristics of GAC samples (i.e., spent and regenerated GAC products), were analyzed and discussed in relation to the thermal regeneration conditions.

2. Materials and Methods

2.1. Spent GAC

The spent AC used in this study was obtained from a liquid crystal display (LCD) manufacturing plant (Tainan City, Taiwan). The material was in cylindrical form with a particle diameter of 4 mm. Proximate analysis of the as-received sample was performed in triplicate according to the American Society for Testing and Materials (ASTM) standards method. Prior to use in thermogravimetric analysis (TGA) and the pyrolysis/activation experiments, the spent GAC was pre-dried in an oven dryer at 105 °C for 12 h. To prevent the moisture sorption, the dried sample was stored in an air-circulation oven set at 60 °C.

2.2. Thermogravimetric and Elemental Analyses of Spent GAC

Thermogravimetric analysis (TGA) was used to determine the changes in weight of the test sample as a function of temperature. In this study, a TGA instrument (Model: TGA-51; Shimadzu Co., Tokyo, Japan) was used to analyze the thermal decomposition behavior of the dried sample (approximately 0.2 g) under an inert nitrogen atmosphere flowing at 50 cm³/min. The sample was heated from 25 to 1000 °C at a rate of 10 °C/min. Additionally, energy dispersive X-ray spectroscopy (EDS) conducted during scanning electron microscopy (SEM) was performed to obtain the elemental compositions on the surface of the spent GAC. Detailed information on the SEM-EDS instrumentation and its operating conditions is provided in Section 2.4. The SEM-EDS data were also used to compare the surface characteristics of the spent and regenerated GAC samples.

2.3. Thermal Regeneration Experiments

In this study, a vertical fixed-bed furnace was used to perform the thermal regeneration experiments with two different methods. The first method involved simple pyrolysis under N₂ flow (500 mL/min), where the samples were heated at a rate of 10 °C/min up to a specified temperature (i.e., 800, 850, and 900 °C) and held for varying holding times (i.e., 0, 30, and 60 min). Approximately 5 g of the dried GAC was used for each experiment. After completion, the furnace was cooled to below 150 °C before removing the regenerated sample. The regenerated GAC samples were labeled as KN-temperature-time. For example, KN-900-30 referred to GAC regenerated at 900 °C and a holding time of 30 min.

The second regeneration method was to reactivate the spent GAC samples by using a CO₂ atmosphere at the end of the first stage. In the pyrolysis-activation experiments, the first stage was to increase the system temperature to 500 °C at 10 °C/min. In the second stage, the gas atmosphere was switched to CO₂ flowing at 100 mL/min while stopping N₂ flow and further increasing the furnace to the target temperature. Samples regenerated under these conditions were labeled as KNC-temperature-time. For example, KNC-900-30 referred to a GAC sample thermally activated at 900 °C under CO₂ for 30 min.

2.4. Determinations of Textural and Chemical Characteristics of Regenerated AC Products

To assess changes in pore structure after thermal regeneration, nitrogen adsorption–desorption isotherms were measured using a high-performance system (Model: ASAP 2020 Plus; Micromeritics Co., Norcross, GA, USA) at −196 °C. Prior to the isotherm analysis, about 0.25 g of the test samples (i.e., spent and regenerated GAC products) were degassed under a vacuum (condition: ≤10 μmHg at 200 °C for about 10 h) to remove moisture and other impurities. The data on the Brunauer–Emmett–Teller (BET) surface area was calculated from the relative pressure (P/P₀) range of 0.05 to 0.30. The total pore volume was estimated on the assumption that the liquid nitrogen (liquid density at −196 °C = 0.8064 g/cm³) filled all pores at saturation (i.e., P/P₀ set at 0.995). Furthermore, micropore surface area and micropore volume were obtained by the Harkins and Jura thickness model (i.e., t-plot method) [29,30]. The pore size distribution, particularly in the mesoporous range, was calculated using the Barrett–Joyner–Halenda (BJH) method.

The surface morphology and elemental composition of the spent and regenerated GAC samples were examined using scanning electron microscopy (Model: S-3000N; Hitachi Co., Tokyo, Japan) and energy-dispersive X-ray spectroscopy (Model: X-stream-2; Oxford Instruments, Abingdon, UK). During SEM analysis, a 15 kV acceleration potential was applied. Prior to imaging, the samples were gold-coated using an ion sputter coater (Model: E1010; Hitachi Co., Tokyo, Japan) to provide a conductive surface. SEM-EDS results were used to compare the structural and chemical changes between the spent and regenerated samples.

3. Results

3.1. Thermochemical Characteristics of Spent GAC

As shown in

Table 1. Proximate analysis of spent GAC (denoted as KN).

Property a	Value
Moisture (wt%)	9.99 ± 0.50
Ash (wt%)	13.58 ± 0.66
Volatile matter (wt%)	13.34 ± 0.95
Fixed carbon b (wt%)	73.08

^a Mean ± standard deviation for three determinations of an as-received spent GAC sample. ^b By difference.

, the proximate analysis revealed that the spent GAC contained significant amounts of ash and volatile matter, indicating that the exhausted adsorbent was contaminated by the deposited particles and adsorbed organics from the vent and process streams. TGA was performed to further explore the thermal desorption and decomposition behaviors of the spent GAC, as illustrated in Figure 1. Three distinct regions were observed in the derivative thermogravimetry (DTG) curve. In the temperature range of 100–300 °C, the observed weight loss is likely attributed to the thermal desorption of moisture, light gases (e.g., CO₂) and low-molecular-weight organics [31]. At the higher temperature range, 300–500 °C, it is likely due to the thermal desorption or decomposition of heavier organics adsorbed onto the sample. The thermal decomposition that occurred above 700 °C is a significant observation, which may be due to the breakdown of the limited inorganics with moderate melting points and adsorbed organics with high decomposition features [16] or the char forming of residual/trapped organics [32].

Based on these results, a process temperature above 800 °C was selected, implying the production of highly carbonized char with relatively low yields. Based on the TGA/DTG results and considering energy consumption, the thermal regeneration of spent GAC was conducted at 800, 850, and 900 °C. The holding time varied between 0, 30, and 60 min after reaching the target temperature.

3.2. Pore Properties of Spent GAC and Its Regenerated Products

The pore properties, such as the BET surface area, total pore volume, micropore surface area, and micropore volume of the spent GAC and regenerated products, are summarized in

Table 2. Pore properties of spent GAC (denoted as KN) and regenerated GAC products.

Spent GAC (KN) and Regenerated GAC Product a	SBET b (m2/g)	Smicro c (m2/g)	Vt d (cm3/g)	Vmicro c (cm3/g)
KN	425.17	324.04	0.184	0.129
KN-800-00	605.38	385.99	0.272	0.156
KNC-800-00	538.00	426.68	0.231	0.169
KN-800-30	516.43	422.74	0.220	0.170
KNC-800-30	578.22	405.79	0.254	0.163
KN-800-60	523.99	413.95	0.224	0.165
KNC-800-60	647.15	470.20	0.283	0.188
KN-850-00	531.67	444.32	0.223	0.175
KNC-850-00	555.93	400.13	0.249	0.161
KN-850-30	519.14	397.93	0.222	0.158
KNC-850-30	541.13	399.76	0.239	0.160
KN-850-60	521.42	420.76	0.222	0.168
KNC-850-60	634.97	433.99	0.278	0.174
KN-900-00	558.38	451.83	0.237	0.179
KNC-900-00	569.21	409.04	0.260	0.164
KN-900-30	569.32	411.53	0.249	0.165
KNC-900-30	723.23	457.25	0.327	0.185
KN-900-60	496.08	387.65	0.212	0.154
KNC-900-60	699.88	385.89	0.316	0.155

^a Sample codes were denoted as the regenerated GAC produced at 800, 850, and 900 °C for holding 0, 30, and 60 min. ^b BET surface area (S_BET) was calculated from a relative pressure range of 0.05–0.30 (15 points). ^c Micropore surface area (S_micro) and micropore volume (V_micro) were estimated by the t-plot method. ^d Total pore volume (V_t) was obtained at a saturated relative pressure (about 0.995).

. The nitrogen adsorption–desorption isotherms (at −196 °C) and the corresponding mesopore size distributions of the regenerated samples are presented in Figure 2 and Figure 3.

The key findings of the results presented in Table 2 and Figure 2 and Figure 3 are summarized in the following points:

• The regenerated samples exhibited significantly improved pore properties compared to the spent GAC (e.g., BET surface area of 425.17 m²/g and total pore volume of 0.184 cm³/g). The variations in pore property trends were not consistent with pyrolysis temperature and holding time. Interestingly, the maximum pore properties (BET surface area of 605.38 m²/g and total pore volume of 0.272 cm³/g) were obtained under the minimal process conditions (800 °C and 0 min holding time).
• The CO₂-activated products generally displayed even higher pore properties than both the spent GAC and the pyrolysis-regenerated products except in the cases of KN-800-00 and KNC-800-00. It can be attributed to further activation by the interaction of the CO₂ gas with carbon (AC or char) due to the following chemical reaction [1]:

  C_(solid) + CO_2(gas) → 2CO_(gas)
  Above 800 °C, CO₂ gas reacted with the carbon matrix, thus opening the pores, which resulted in higher pore properties. The maximal pore properties (BET surface area 723.23 m²/g, and total pore volume 0.327 cm³/g) of the CO₂-activated GAC products were produced at 900 °C and a 30 min holding time. These results show that, under optimized pyrolysis-activation conditions, approximately 70% of the pore properties can be achieved, making these regenerated GACs highly promising for water treatment applications.
• As shown in Figure 2, the nitrogen adsorption–desorption isotherms of spent GAC (KN) and the regenerated products (KN-900-30 and KNC-900-30) displayed both Type I (major) and Type VI (minor) patterns, typically indicating the presence of both micropores and mesopores [29]. Type I curves exhibited steep uptake at low relative pressures, corresponding to high micropore adsorption potential. On the other hand, the Type VI isotherms featured slight hysteresis loops, starting around a relative pressure of approximately 0.30. Figure 3 further confirms the presence of mesopores with the recorded peak at approximately 4.0 nm, while an upward trend below 2.0 nm on the left of the curve indicates a microporous structure. For more precise micropore distribution analysis, models such as Dubinin–Astakhov (DA) and Horvath–Kawazoe (HK) are recommended [29,30].

3.3. Textural and Chemical Characteristics of Spent GAC and Its Regenerated Products

SEM and EDS analyses were employed to examine the surface morphology and elemental composition of the spent and regenerated GACs. Figure 4 shows the SEM images of the spent GAC and two of the regenerated samples (KN-900-00 and KNC-900-00) at 1000× and 3000× magnifications. It can be seen that fine particles were loosely deposited or packed on the GAC surface. Although the samples exhibited similar porous textures, the CO₂-activated product (KNC-900-00) appeared more porous, as seen in Figure 4c, which corresponds with its higher measured pore properties compared to those of spent GAC and pyrolysis-regenerated GAC (i.e., KN-900-00).

Energy dispersive X-ray spectroscopy (EDS) was used in this work in order to acquire the spectra for the purpose of determining the changes in the elemental compositions of spent GAC and its regenerated products [33]. The analysis showed the carbon content of spent GAC to be as high as 61.20 wt% (Figure 5a). The spent GAC also contained significant amounts of other non-carbon elements like oxygen (O, 30.14 wt%), iron (Fe, 3.03 wt%), silicon (Si, 3.28 wt%), and aluminum (Al, 2.34 wt%). In comparison, analysis showed the typical regenerated GAC products (KN-900-00 and KNC-900-00) (Figure 5b,c), to have higher carbon and lower oxygen content. These results could be attributed to the desorption and the activation (charring) of adsorbed constituents, thus releasing some oxygen-containing gases (e.g., H₂O, CO, CO₂) during the thermal regeneration process.

4. Conclusions

The regeneration of spent or exhausted activated carbon (AC) generally refers to the desorption and/or decomposition of adsorbed contaminants without damaging the original carbon structure, thereby restoring its adsorption capacity for reuse. In this study, spent cylindrical GAC from an electronics manufacturing facility was thermally regenerated using two approaches: simple pyrolysis and CO₂-enhanced pyrolysis-activation. Based on the pore properties of spent GAC and its regenerated products, gasification of a carbon matrix carried out above 800 °C using CO₂ was able to open and create more pores, thereby improving the pore properties as compared to those of spent GAC and its N₂-pyrolyzed regenerated products. The maximal pore properties (BET surface area of 723.23 m²/g and total pore volume of 0.327 cm³/g) of the CO₂-activated GAC products were produced at 900 °C and a 30 min holding time. Approximately 70% of the pore properties can be restored based on the investigated pyrolysis-activation regeneration conditions. SEM images confirmed enhanced surface porosity, while EDS analyses showed an increase in carbon content and a reduction in surface oxygen after regeneration.