1. Introduction
US EPA has reported that metal water pollution is a severe worldwide problem due to the rapid development of industry and technology [
1]. Some specific metal compounds are toxic, nonbiodegradable, persistent, bioaccumulated, and cause cancer, so treating metal pollutants in aqueous solutions is a significant issue [
2,
3]. Among many metals, copper is commonly found in industrial wastewater [
4]. It has been widely used in semiconductors, electronic product manufacturing, and electroplating. Copper poisoning can cause nausea, diarrhea, liver, and kidney failure due to long-term exposure to copper through contaminated food and water sources [
5]. For these reasons, many techniques have been developed to effectively remove metals from wastewater to meet the discharge standard of water pollutants, such as ion exchange [
6], electrochemical treatment [
7], chemical precipitation [
8], reverse osmosis [
9], and adsorption [
5,
10,
11].
In these methods, adsorption by activated carbon (AC) is considered as a prospective approach to eliminate copper from wastewater, because of its easy implementation, high efficiency, low initial cost, and reusability [
5,
12]. It has widely been used in the removal of various contaminants from wastewater due to its high porosity parameter, great surface area, surface functional groups, and insensitivity to toxic environments [
13,
14,
15]. On account of its many advantages, there have been many biowastes used for the fabrication of activated carbon materials, such as sawdust [
16], paulownia wood [
17], rice straw [
18], and waste wood-based panels [
19]. These biowastes also contain lignin and celluloses as commercial activated carbon precursors that can be a considered choice and low-cost sources to produce activated carbon [
3].
In addition, with effluent standards becoming stricter in order to control pollution impacts, the modification of carbon materials has attracted extensive attention in recent years. The adsorption capacity of carbon materials is significantly relative to their pore structure and surface chemistry [
13]. The surface modifications of a material can enhance the adsorption capacity and removal efficiency by increasing the active functional groups with a high affinity for metals on an adsorbent surface [
20]. For example, HNO
3 oxidized biochar showed remarkably higher adsorption performance than the unmodified carbon via introducing functional groups [
10]. Zuo et al. functionalized biochar with hydrogen peroxide, which caused several oxygen-containing groups on the material surface, enhancing the affinity for copper [
14]. Among the many surface modifiers, fixing iminodiacetic acid (IDA) chelating functional group into materials and its use as an adsorbent can be helpful for wastewater restoration [
15]. Katarzyna et al. modified graphene oxide with IDA for effectivity preconcentration and removal of Cu(II) from water samples (108.4 mg/g) [
21]. Razak et al. grafted kenaf fiber with IDA and increased the adsorption capacity by about 200% for Cu(II) removal [
22]. Both oxygen and nitrogen on IDA can provide electron pairs to form coordination with transition metals. This endows the adsorbent with the ability to effectively attract copper to remove metals from the water [
23].
Therefore, activated carbon derived from waste wood-based panels has been oxidized with HNO3 to enhance the number of oxygen-containing functional groups on the carbon material in this study. Afterwards, we grafted IDA to enhance the adsorption ability, affinity, and removal efficiency. A batch experiment was conducted to examine the influence of the solution pH, reaction time, initial concentration, and temperature on adsorption. The data from the batch sorption were verified with equilibrium and kinetic adsorption models to identify the adsorption mechanisms from the adsorption procedure. The revitalization experiment was carried out to make the adsorbent more economical. Lastly, the column system was also used to study AC, OAC, and IDA-OAC.
2. Materials and Methods
2.1. Chemicals and Materials
The H3PO4 (≥85.0%), nitric acid (≥65.0%), hydrochloric acid (≥37%), iminodiacetic acid (IDA) (≥98%), epichlorohydrin (EPI) (>99%), NaHCO3 (≥99.7%), Na2CO3 (≥99.8%), NaOH (≥98%), NaCl (99.9%), NaNO3 (>99%), and Cu(NO3)2·3H2O (99%) were purchased from Uni-Onward Corp. Chemicals used in this research were all of analytical quality. N2 (99.5%) was purchased from Yun Shan Gas Co. Waste wood-based panels (WWP) were collected from the local recycling center. They were cut into small pieces (1~4 mm) and then washed several times with DI water. They were then dried in the oven at 333 K for 24 h and stored in a scintillation vial for subsequent usage.
2.2. Preparation of Adsorbents
2.2.1. Fabrication of Activated Carbon from WWP
Activated carbon was fabricated via chemical activation. Specifically, WWP was impregnated with H3PO4 (85.0% w/w) at 298 K and placed in an oven at 383 K for 24 h. The impregnation ratio was 3:1 (H3PO4/WWP mass ratio). After that, a weighed amount of the dried mixture was put in the porcelain boat placed into a furnace, and heated up to 823 K for 2 h under continuous N2 flow (150 mL/min STP) at a constant heating rate (10 K/min). The activated carbon was cooled to room temperature and washed with distilled water until constant pH. This resulting sample was dried at 333 K for 24 h and sieved to a uniform size fraction of 0.595 mm–0.149 mm.
2.2.2. Activated Carbon Oxidation
To increase the bond with the metal and the subsequent grafting site, we oxidized the activated carbon to increase oxygen-containing groups (such as hydroxyl, carboxyl) on the surface with HNO3. For this, the 1 g of dried AC was impregnated in 10 mL of 7 N nitric acid solution for three hours at 353 K. After treatment, the residual material was filtered with filter paper and washed with distilled water until constant pH, and dried at 333 K for constant weight.
2.2.3. Synthesis of IDA-OAC
For fixing iminodiacetic acid on OAC, epichlorohydrin was reacted with primary hydroxyl on the OAC through electrophilic substitution. Afterward, nucleophilic substitution occurred between the amino groups on IDA and the epoxy rings on the OAC. Specifically, 1.0 g of oxidized activated carbon (OAC) was dispersed in aliquots of ethanol (20 mL), 1 M of NaOH (10 mL), and epichlorohydrin (10 mL). The mixture was agitated for 3 h at 313 K. After that, it was cooled to room temperature and collected by filtration, and washed with DI water dried at 333 K for 24 h, and abbreviated as EPI-OAC. Next, EPI-OAC (1.0 g), IDA (4.0 g), 100 mL of distilled water were mixed and stirred at 333 K for 10 h. The system was controlled to maintain the pH of the mixture higher than 10 during the reaction by adding NaOH solution. Finally, it was dried at 333 K for 24 h after the mixture solution was collected, and washed with diluted acetic acid and water, and abbreviated as IDA-OAC.
2.3. Characterization
The concentrations of Cu(II) ions were measured by ICP-OES (Varian, Palo Alto, CA, USA, Vista-MPX). The characteristics of the AC, OAC and IDA-OAC adsorbents were determined as follows: FTIR spectra for assessing the functional group’s presence were recorded within a range from 4000 to 500 cm
−1 using potassium bromide pallet in a Perkin Elmer, Spectrum one spectrometer. The changes of element content of C, N, H, and S were analyzed by an elemental analyzer (Unicube, Elementar Analysensysteme GmbH, Hanau, Germany). The surface morphology was determined with the Scanning Electron Microscopy (SEM) (EFE-SEM; SU-5000, HITACHI, Chiyoda ku, Japan). The BET surface area (Brunauer–Emmett–Teller) and pore structure were analyzed by adsorption and desorption with nitrogen at 77 K (Micromeritics ASAP 2020 specific surface analyzer). Before the gas adsorption measurement, the samples were degassed at 105 °C for 8 h. The Boehm titration method was performed to determine the acidic surface functional groups on the materials [
24]. The pH drift method was used to determine the adsorbent’s pH point of zero charges (pH
PZC) [
25].
2.4. Batch Sorption of Copper
In the batch adsorption process, Cu(NO
3)
2·3H
2O were used to prepare Cu(II) ions. A series of sorption was studied by adding 0.02 g of AC, OAC, IDA-OAC into 20 mL of a specific concentration of Cu(II) solution and agitated at a rate of 110 rpm in a shaker incubator. The temperature and the time factors were adjusted in the experiment. The influence of pH on adsorption experiments was processed at 298 K, and the HNO
3 or NaOH (0.01–1.0 M) solutions were used to adjust the pH value from 2 to 6. The effect of cations on the removal of Cu(II) was determined by adding 1000 to 5000 mg/L NaNO
3 with a Cu(II) concentration of 300 mg/L, and the mixture was shaken for 24 h at 298 K. For the determination of the impact of the different initial concentrations (5, 10, 30, 50, 80, 100, 150, and 200 mg/L), the solution was agitated for 24 h at a suitable pH value (pH = 5). The factor of contact time (4, 8, 16, 32, 60, 120, 180, 240, and 300 min) was determined at 298 K at optimum pH with a Cu(II) concentration of 200 mg/L. Further, the influence of temperature (298, 308 and 318 K) was analyzed. After each adsorption process, the filtrate was collected and analyzed by ICP-OES to measure the Cu(II) concentration. The adsorption capacity (
qe, mg/g) of the copper and removal efficiency (R%) were calculated using the following equation [
11]:
where
C0 is the initial concentration and
Ce is the equilibrium concentration of the copper solution (mg/L), and
V(L) and
W(g) are the volume of the Cu(II) solution and the weight of the adsorbent usage, respectively.
2.5. Equilibrium Isotherms, Adsorption Kinetics, and Thermodynamic Studies
To better understand the adsorption reactions with the primary mechanism and to investigate how copper ions interacted with adsorbents, the adsorption results with different initial concentrations were fitted with Langmuir isotherm (Equation (3)) and Freundlich isotherm (Equation (4)) models [
15,
26].
where
qmax (mg/g) signifies the maximum adsorption capacity.
qe (mg/g) is the concentration of Cu(II) solution under equilibrium.
KL is a Langmuir constant related to the adsorption energy (L/mg).
Ce (mg/L) is the concentration of Cu(II) solution in equilibrium.
KF (mg/g) represents the Freundlich constants relative adsorption capacity of the adsorbent, and 1/
n is the Freundlich coefficient about heterogeneity.
Adsorption kinetics of Cu(II) on adsorbent used experimental data about the effect of contact time to fit separately by pseudo-first (Equation (5))- and second (Equation (6))-order kinetic models [
27].
where
qt (mg/g) is the quantities of Cu(II) adsorbed at the time, and
qe (mg/g) is the adsorbed amounts of Cu(II) under equilibrium, while
kf (1/min) and
ks (g/ (mg·min)) are the corresponding rate constants of pseudo-first- and second-order kinetic models, respectively.
The data from the effect of temperature on the adsorption process were used for determining three thermodynamic parameters: change in free energy (ΔG°), change in the enthalpy (ΔH°) and entropy variation (ΔS°). These parameters were calculated from the van’t Hoff equation [
28]:
where
R (8.314 J/(mol K)) is the universal gas constant,
Kd (L/mol) is the adsorption coefficient,
T is the absolute temperature in Kelvin. Using the slope and intercept of the plot of ln
Kd versus 1/
T, one can obtain the value of the ΔS° and ΔH°.
2.6. Reusability of IDA-OAC
To make the IDA-OAC adsorbent more economical and practical, regeneration and reusability experiments were carried out. In the regeneration experiment, the adsorbent which loaded copper ion was eluted (solid/liquid ratio of 1 g/L at 298 K for 24 h) with 0.1 M HCl, 0.1 M HNO3, and 0.1 M H2SO4, respectively. After regeneration, the adsorbent was collected, washed, and dried for the subsequent three adsorption/desorption experiments.
2.7. Column Sorption of Copper
The column-scale experiments were evaluated in continuous adsorption. About 0.2 g of the AC, OAC and IDA-OAC was wet-packed in a plexiglass column which was 4 mL in volume and 11.2 mm in diameter. The effect of initial Cu(II) concentration (50 mg/L) was analyzed. Before feeding the influents into the column, a peristaltic pump was pumped downward through the column with distilled water for about 3 h. The flow rate for fixed-bed experiments was 1 mL/min at pH 5 (optimum value from batch-scale studies) and 12 BV/h. The samples were collected from the exit with a fraction collector until a saturation state occurred. The following equations were used for data analysis of fixed-bed adsorption [
29].
where
qtot is the total mass of the adsorbent in the column,
Qv has been determined as the flow rate (mL/min),
Ci (mg/L) and
Ct (mg/L) are the inlet and outlet of Cu(II) concentration, the
mtot (mg) and R% present the total quantity of adsorbate flow through the column and the removal rate, respectively. The area under the breakthrough curve is marked as A and the plots of
Ct/Ci against bed volume are called the breakthrough curves.
Ct/Ci = 0.05 and 0.95 are signed as the breakthrough points (
pb) and exhaustion points (
pe), respectively.