Controlling pore size of activated carbon through self-activation process for removing contaminants of different molecular sizes

Journal of Colloid and Interface Science 518 (2018) 41–47

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

Regular Article

Controlling pore size of activated carbon through self-activation process for removing contaminants of different molecular sizes Yingji Wu a, Changlei Xia a,b,⇑, Liping Cai a,b, Sheldon Q. Shi a,⇑ a b

Department of Mechanical and Energy Engineering, University of North Texas, Denton, TX 76203, USA College of Materials Science and Engineering, Nanjing Forestry University, Nanjing, Jiangsu 210037, China

Internal pressure changes

-734 torr -306 torr (4 h), -201 torr (15 h) -81 torr (100 h) Self-activation

Iodine number (mg g -1 )

g r a p h i c a l a b s t r a c t 3000

R² = 0.9729

2000

1000

0 0.3

0.4

0.5

Large impurity

Small impurity

Tannin value (mg L -1 )

PVDFT-micropore (cm3 g-1 )

Controlling pore size by self-activation time

600 R² = 0.9417

400

200

0 0

0.5

1

1.5

PVDFT-mesopore (cm3 g -1 )

a r t i c l e

i n f o

Article history: Received 8 December 2017 Revised 31 January 2018 Accepted 5 February 2018 Available online 7 February 2018 Keywords: Self-activation Activated carbon Surface area Pore volume Chemical adsorption

a b s t r a c t Self-activation was employed for the manufacturing of activated carbon (AC) using kenaf core fibers, which is more environmentally friendly and cost-effective than the conventional physical/chemical activations. It makes the use of the gases emitted from the thermal treatment to activate the converted carbon itself. The mechanism was illustrated by the Fourier transform infrared spectroscopy and mass spectrometry analysis of the emitted gases, showing that CO2 served as an activating agent. The AC from self-activation presented high performance, for instance, the Brunauer-Emmett-Teller surface area was up to 2296 m2 g 1, Using the Density Functional Theory (DFT), the pore volume (PV) was determined to be 1.876 cm3 g 1. Linear relations of PVDFT-micropore/iodine number, and PVDFT-mesopore/tannin value were established, indicating a strong relationship between the pore structure of AC and its adsorbing preference. Adsorption results for copper (II) and rhodamine 6G also indicated that the pore size of AC should be designed based on the molecular size of the contaminants. Ó 2018 Elsevier Inc. All rights reserved.

1. Introduction Driven by the societal concerns, anthropogenic changes of the ecosystem have become an important part of the political agenda [1]. In this context, green process engineering is greatly attracting the public interests [2]. Traditionally, the manufacturing methods ⇑ Corresponding authors at: Department of Mechanical and Energy Engineering, University of North Texas, Denton, TX 76203, USA. E-mail addresses: [email protected] (C. Xia), [email protected] (S.Q. Shi). https://doi.org/10.1016/j.jcis.2018.02.017 0021-9797/Ó 2018 Elsevier Inc. All rights reserved.

of activated carbon (AC) can be sorted to two categories, i.e. physical and chemical activations [3]. The physical activation process uses gases presenting mild oxidation, for instance, carbon dioxide (CO2, a greenhouse gas), steam, air, etc. [4], to eliminate the volatile matters, followed by partial gasification with thermal treatment, at the expense of raising environmental issues [5–8]. Besides, chemical activation process introduces chemicals, e.g. ZnCl2, KOH, and H3PO4, into the activation processes [9–11]. The removal of those activating chemicals after the activation might introduce hazardous chemical wastes [9,12–14]. For instance, strong acids (e.g.

42

Y. Wu et al. / Journal of Colloid and Interface Science 518 (2018) 41–47

HCl, HNO3, H2SO4) are normally used to eliminate the zinc compound formed during the ZnCl2 activation, which creates hazardous waste chemicals. Hence, both the conventional activation processes are accompanied by hazardous wastes, which needs the additional disposal and brings environmental bearings. As well-known, tremendous volume of gases will be released from biomass during the thermal decomposition processes, e.g. pyrolysis and gasification. It was reported that those gases from high-temperature decomposition of biomass mainly contained H2, CO2, CO, CH4, H2O, etc. [15,16], in which, CO2 and H2O have been greatly utilized as activating agents for producing AC [3,17]. Consequently, the CO2 and H2O emitted from the biomass thermal decomposition may serve as the agents during the activation process, so that no additional gas needs to be imported [18], which might revolutionarily solve the environmental issues during the activation processes. This activation process employing the emitted gases from thermal decomposition of biomass as activation agents was defined as self-activation process, which has been developed rapidly [15,18–20]. Recently, apart from the conventional activation processes, Wang et al. categorized self-activation as the third type of activation methods in a review article [21]. In addition to the advantages of the more environmental-friendly and cost-effective compared with the conventional activation approaches, the performance of ACs from self-activation process was very competitive with that of the traditional ACs. For instance, the surface areas (SAs) of ACs from physical and chemical activation processes were 1926 m2 g 1 (coconut shell-based) and 1720 m2 g 1 (cotton stalk-based), respectively [4], while the SAs from the self-activated ACs were 2738 m2 g 1 (pine-based), 2602 m2 g 1 (cellulose-based), and 2432 m2 g 1 (kenaf core-based) [15,18,20]. Kenaf, as a porous material and mainly composed of C, H and O, has been demonstrated to be an excellent feedstock for manufacturing ACs [18,22–24]. For instance, Soheil et al. prepared fiberlike AC from kenaf fiber using K2HPO4 activation for adsorbing phenolic contaminants [22]. Chowdhury et al. prepared AC from kenaf fiber for removing copper from waste water [23]. In this work, selfactivation was successfully applied to fabricate kenaf core fibers (KCFs)-based AC (KAC), and its mechanism was investigated. The performance of ACs from self-activation were evaluated by SAs, pore volumes (PVs), iodine numbers, tannin values, etc. It was found that the ACs’ pore structures owned a great effect on their performance of absorbing impurities with different molecular sizes. Commonly, the pore size can be sorted into three ranges in according with the definition from International Union of Pure and Applied Chemistry (IUPAC), i.e. micropore width less than 2 nm, macropre larger than 50 nm, and mesopore size at the range of 2 to 50 nm. The ACs with different pore structures could exhibit diverse adsorption preferences. Hence, the association between pore structure of AC and its adsorption capacity was investigated in this study.

2. Materials and methods 2.1. Self-activation process KCFs (3 – 30 meshes) from Biotech Mills Inc., Snow Hill, NC, USA, were selected for producing AC through the self-activation process. The dried KCFs own 49.0% of cellulose, 29.7% of hemicellulose, and 19.2% lignin. The results from proximate analysis and ultimate analysis of the dried KCFs, KAC-1, KAC-2, and KAC-3 are shown in Table 1. The chemical composition of KCFs from the proximate analysis ultimate analysis was similar to the reported values [25]. The fixed carbon (proximate analysis) and the carbon content (ultimate analysis) of the KACs were decreased with the activation

time increase, because of the increased amount of inorganic ash content (Table 1). The oxygen contents of the KACs presented very low values, compared with AC prepared by chemical activation [26]. The low oxygen of AC might benefit to its adsorption of specific chemicals, e.g. surfactant [27]. The detail processes of self-activation were described in our previous work [18]. Briefly, the self-activation process was conducted in the Sentro Tech Corp. STY–1600C box furnace that has vacuum capability. After placing 200 g KCFs (179 g dried mass and 21 g moisture) into the chamber in the air atmosphere, the furnace was pumped down to an approximate pressure of 734 torr (the absolute vacuum pressure is approximate 760 torr) and then all the valves were switched off. The furnace was kept in the state as a closed system (no mass transfer between the surrounding air and furnace) during the entire process. Then, the self-activation process was conducted by following three stages: (a) ramping (1 0 °C min 1) the furnace temperature to 1000 °C; (b) dwelling at 1000 °C for 4, 15, and 100 h, respectively; and (c) cooling down ( 10 °C min 1). Finally, the as-produced KACs were obtained and denoted as KAC-1 (4h), KAC-2 (15 h), and KAC-3 (100 h), respectively. In addition to the three groups of KACs from self-activation process, a commercial granular AC (8  16 meshes) from Calgon Carbon Corp. (CAC) was used as a comparison in this study. Prior to the tests of iodine number determination, the tannin value test, copper (II) and rhodamine 6G adsorptions, all four groups of ACs were grinded into the particle sizes of not less than 90% at 325 mesh, 95% at 200 mesh, and 99% at 100 mesh. This particle size requirement of AC is defined in the American Water Works Association (AWWA) B600–10 standard, which is a standard designed for powdered AC used in drinking water purification. 2.2. Surface area and pore volume The Micromeritics Instrument Corp. 3Flex 3500 surface area and pore size analyzer was employed for determining the SAs and PVs of the ACs. The N2 gas adsorption at 77 K was used to analyze the gas adsorption capabilities of the ACs. Prior to the tests, the Micromeritics Instrument Corp. VacPrep 061 degasser was employed for vacuum-degassing the ACs at 350 °C for 4 d. After being transferred to the analyzing ports, the ACs were in situ vacuum-degassed by the Edwards EXT75DX 63CF turbo pump at 350 °C for 20 h. The SAs of ACs were calculated from the isothermal plots through the instrumental software (3Flex Version 1.02) by means of the Brunauer-Emmett-Teller (BET) model. The PVs and pore size distribution of the ACs were investigated depending on the Density Functional Theory (DFT). 2.3. Fourier transform infrared spectroscopy The Fourier transform infrared spectroscopy (FT-IR) was employed to examine the concentration changes of CO2 and CO for the self-activation processes with different dwelling times. A Thermo-Fisher Nicolet 6700 FT-IR analyzer with a 20-mL Pike flow cell was employed. Three gas samples were collected by 1-L Tedlar bags containing the emitted gases after the self-activation processes at 1000 °C for 4 h, 15 h, and 30 h, respectively. 2.4. Iodine number determination The iodine number is a parameter to represent the adsorbing capacity of AC, which is an important characteristic for the chemical adsorption evaluation of AC. The iodine numbers of the powdered ACs were established using the sodium thiosulfate (Na2S2O3) volumetric method in accordance with the ASTM D4607 standard [28]. The iodine number can be used as an approx-

43

Y. Wu et al. / Journal of Colloid and Interface Science 518 (2018) 41–47 Table 1 The results from proximate analysis and ultimate analysis of dried KCFs, KAC-1, KAC-2, and KAC-3. Sample

KCFs KAC-1 KAC-2 KAC-3

Proximate analysis (%)

Ultimate analysis (%)

Volatile matter

Fixed carbon

Ash

C

H

O

N

S

82.85 1.23 0.45 0.12

15.41 90.87 79.63 51.61

1.74 7.90 19.92 48.27

48.00 93.25 89.12 62.38

6.37 0.44 0.52 0.51

42.10 0.16 0.11 0.14

0.35 0.11 0.10 0.17

0.07 0.01 0.01 0.02

imation for micro-porosity of AC with acceptable accuracy, defined as the iodine adsorbed by 1 g of AC when the iodine concentration in the filtrate is 0.02 mol L 1. Briefly, the powdered ACs with three different amounts were individually added into 250-mL flasks with 10 mL 5 wt% hydrochloric acid (HCl) aqueous solution. Each mixture was heated to the boiling point and held at that temperature for 30 s, then cooled to room temperature. Sequentially, 100 mL of the 0.1 mol L 1 iodine solution was pipetted into each flask, which was immediately stopped and shaken vigorously for 30 s. The mixture was then quickly filtered through a sheet of folded filter paper (Whatman No. 2V), and the filtrate was collected into a flask. The iodine concentration in the filtrate was determined by Na2S2O3 titration using 2 mL of the starch solution (1 g L 1) as an indicator. The iodine amount adsorbed per gram AC (iodine/AC) was plotted against the iodine concentration in the filtrate, using the logarithmic axes. A three-point isotherm plot was drawn for each AC, from which, the iodine/AC value was defined as the iodine number when the iodine concentration in the filtrate was 0.02 mol L 1. 2.5. Tannin value test The tannin value is defined as the concentration of powdered AC in milligrams per liter (mg L 1) required to reduce the standard tannic acid (MP Biomedicals Corp., USA) concentration from 20 mg L 1 to 2 mg L 1 [29]. The chemical formula of tannic acid is presented in Fig. S1a. The tests for tannin values of the ACs were carried out in terms of the AWWA B600–10 standard. Briefly, 80 mg, 160 mg, 240 mg, and 320 mg of each powdered KAC was individually placed in 1-L beakers and each beaker was filled with 800 mL of 20 mg L 1 tannic acid test solution. The mixtures were stirred at 100 rpm for 1 h and vacuum-filtered (0.8 lm Millipore paper) to collect the filtrates. The concentrations of the filtrates were determined by UV absorbance at 275 nm by a UV–Vis spectrophotometer (UV-2600, Shimadzu Corp., Japan). After plotting the results on double logarithmic axes to establish the relationship between the AC dosage and the total remaining tannic acid in the filtrate, a linear fitting method was applied. The applied AC concentration (mg L 1) to obtain the filtrate with 2 mg L 1 of tannic acid was defined as the tannin value. 2.6. Copper (II) adsorption The copper bromide (CuBr2) (Sigma-Aldrich Corp., USA) was used as targeting material for the Cu (II) adsorption tests. For each test, 50 mg of powdered AC was added into 200 mL CuBr2 aqueous solution (200 mg L 1). The mixture was stirred for 24 h, and then filtered to clarify the solution by a 0.2 mm syringe filter. The residual concentration of CuBr2 aqueous solution was determined by the UV absorbance at 272 nm in according with the Beer’s Law. 2.7. Rhodamine 6G adsorption The rhodamine 6G (Fig. S1b) was purchased from Sigma-Aldrich Corp., USA. The removal of rhodamine 6G from water was carried out in accordance with the procedure from the literature [30,31].

For each test, 50 mg powdered AC was mixed with 50 mL rhodamine 6G aqueous solution (400 mM). After the 24 h stirring, the mixture was filtered by a syringe filter (0.22 mm pore size) to remove the solid particles. The residual concentration of rhodamine 6G aqueous solution was determined by the Vis absorbance at 528 nm in according with the Beer’s Law. 2.8. ESEM observation The Quanta 200 environmental scanning electron microscope (ESEM) was employed for the morphology study of the KAC. The morphology of KAC-2 was observed by ESEM. The operating voltage was set to be 20 kV, and the magnifications of the images were 100 and 500. 3. Results and discussion 3.1. Pore size control through the self-activation process In the self-activation process, the dwelling time affected the internal pressure, and simultaneously controlled the pore size of produced AC which might adsorb different-size impurities (Fig. 1a). The box furnace was initially vacuumed to 734 torr before the temperature ramping, however, the pressure was raised to be 81 torr after 100 h self-activation process (Fig. 1b). The difference of vacuum pressures between the initial and final in the furnace could confirm there was a significant amount of gases generated during the activation process. Intermediately, the final pressures in the furnace after self-activation for 4 h and 15 h were 306 torr and 201 torr, respectively (Fig. 1a). It was indicated that more and more amount of gases would be generated by increasing the time of self-activation process, which might be contributed by the CO2 and H2O activation processes. For instance, 1 mol of CO2/H2O would become 2 mol of gases after the reactions with carbon during the activation processes. The FT-IR and mass spectrometry (MS) were used for the examination of the emitted gases during the self-activation process, which could reveal the mechanism of self-activation. The MS results showed that the gases from self-activation of KCFs included H2, CO2, CO, H2O, CH4, etc. (Fig. S2), confirming the presentence of the activating agents of CO2 and H2O. In the FT-IR spectra (Fig. 2), the adsorptions of CO2 and CO are obviously discovered. The peaks at 2360 and 2341 cm 1 belong to CO2, and those at 2171 and 2117 cm 1 are attributed to CO [16]. The detail FT-IR absorbance of CO2 and CO, and their ratios are summarized in Table S1. As the activation time increased, the FT-IR absorbance of CO was increased, whereas that of CO2 was decreased (Fig. 2). The value of CO/CO2 was dramatically increased from 0.34 to 0.64, and then to 1.38 by extending the activating time from 4 h to 15 h, and then to 100 h (Table S1). It was suggested that CO2 reacted with carbon to generate more CO, which presented as an activation process [15,18]. This reaction also reflected as the increase of internal pressure with the increase of self-activation time. It was concluded that the emitted gases during the self-activation process can serve as activating agents. Furthermore, the FT-IR results revealed that CO2 was gradually turned into CO, which would be benefited by

44

Y. Wu et al. / Journal of Colloid and Interface Science 518 (2018) 41–47

(a) Internal pressure changes

-734 torr -306 torr (4 h), -201 torr (15 h) -81 torr (100 h)

Self-activation

Controlling pore size by self-activation time Large impurity

Small impurity

(b) 1200

400

800

0

Temperature (°C) Pressure (torr) 400

-400

0 0

20

40

60

80

100

120

Pressure (torr)

Temperature (°C)

Reaction time

-800 140

Time (h) Fig. 1. Illustration of self-activation process to control pore size of produced AC for adsorbing different-size impurities (a), and the internal temperature and pressure changes during the process (b).

0.8 CO2

Abs. (a.u.)

0.6

4h 15 h 100 h

CO

CO2

CO

0.4

3.3. Relations between pore structure and capacity of absorbing different-size impurity

0.2

0.0 2500

was observed that the cell walls of the carbon became thinner after the self-activation process, which could dramatically increase the porosity of the carbon. The yields of KAC-1, KAC-2, and KAC-3 were determined to be 12.7%, 4.7%, and 2.0%, respectively. It was similar with the reported yields of cellulose-based ACs [15], i.e. 1.6–16.7%. To investigate the self-activation process, the effects of dwelling time on the SA and PV are evidenced from the N2 adsorption measurements (Fig. 4). In terms of the isothermal plots (Fig. 4a), the SABET of CAC, KAC-1, KAC-2, and KAC-3 were calculated to be 836, 901, 2266, and 2296 m2 g 1, respectively (Fig. 4). The highest SABET is highly competitive with those of commercially available ACs (Table S2). It was observed that CAC and KAC-1 presented lower SABET than KAC-2 and KAC-3. By increasing the activating time, the SABET will enlarge and reach a maximum, presenting as SABET dramatically increase from KAC-1 to KAC-2 but slight change from KAC-2 to KAC-3. This phenomenon was reported and explained by a pore expansion-combination activation model developed in our previous studies [18,20]. Briefly, the pore expansion dominated in the early stage of the self-activation process, resulting in the increase in SABET. However, the pore combination increased as the self-activation process continued, which limited the further increase of SABET. Apart from the investigation of SABET, the PVs of ACs were examined, which were considered as important parameters of AC absorbing capacities. The DFT pore size distributions of the ACs are shown in Fig. 4b. The PVs of the ACs at micropore, mesopore and macropore ranges are presented in Fig. 4. In general, KAC-3 owned the highest PVDFT-total (in addition of PVDFT-micropore, PVDFT3 1 , followed mesopore, and PVDFT-macropore in Fig. 4b) of 1.876 cm g 3 1 3 1 by KAC-2 of 1.253 cm g , KAC-1 of 0.502 cm g , and CAC of 0.362 cm3 g 1. The PVDFT-macropore of the ACs presented smaller values compared with those of PVDFT-micropore and PVDFT-mesopore, which was consistent with other reports [15,20]. The reason could be the limitation of N2 adsorption method on analyzing macropore volumes, since the highest pore width detected was approximate 100–300 nm for the ACs (Fig. 4b). The PVDFT-mesopore had the same trend as PVDFT-total of the ACs, however, PVDFT-micropore exhibited differently, i.e. PVDFT-micropore of CAC (0.349 cm3 g 1) and KAC-1 (0.321 cm3 g 1) were close and those of KAC-2 (0.543 cm3 g 1) and KAC-3 (0.513 cm3 g 1) were also similar. These changes of pore volumes with the increase in activation time could be demonstrated and explained by the pore expansion-combination activation model [18]. The micropore volume would present a small reduction because of the pore expansion from micropore (2 nm) to mesopore (2–50 nm) range. However, the mesopore would not expand to macropore (50 nm) range in at least 100 h so that a dramatical increase was observed.

2400

2300

2200

2100

Wavenumber (cm-1)

2000

1900

Fig. 2. FT-IR absorbance of the emitted gases (CO2 and CO) from self-activation processes at 1000 °C for 4, 15, and 100 h, respectively.

either burning directly as syngas or furtherly converting into valuable products, e.g. methanol [32]. 3.2. Performance of self-activated carbon Based on the activation model shown in our previous studies [18,20], the porosity will significantly increase during the selfactivation process. Fig. 3 shows the ESEM images of the KAC-2. It

The iodine number was employed to determine the adsorption capacity of ACs for small-molecule adsorbates, which is a fundamental parameter used to characterize micropore [14]. The minimum iodine number requirement of ACs for drinking-water cleaning is 500 mg g 1 based on the AWWA B600–10 standard. Table 2 shows that the iodine numbers of CAC (1053 mg g 1) and KAC-1 (952 mg g 1) were much lower than those of KAC-2 (2040 mg g 1) and KAC-3 (2120 mg g 1), and each pair owned similar iodine numbers. The results were highly consistent with the DFT (PVDFT-micorpore) results of the ACs (Fig. 4b). Furthermore, the iodine number and PVDFT-micropore presented a good linear relationship (R2 = 0.9729, in Fig. 5a). It was confirmed that the iodine molecules preferred to be absorbed by the micropore. Therefore, the iodine number can represent the capacity for absorbing small-molecule impurities of the ACs [33].

45

Y. Wu et al. / Journal of Colloid and Interface Science 518 (2018) 41–47

Fig. 3. ESEM observations of KAC-2 at the magnifications of 100 (a) and 500 (b).

CAC

SABET (m2 g-1): 836

Quantity adsorbed (cm3 g-1 STP)

1000 0

KAC-1 901

1000 0

KAC-2

2266

1000 0

KAC-3 2296

1000

(b) 5 Differential pore volume (cm3 g-1)

(a)

PVDFT-micropore:

PVDFT-mesopore:

PVDFT-macropore:

(cm3 g-1)

0.349

0.012

0.001

CAC

0 5

0.321

0.169

0.012

KAC-1 0 5

0.543

0.682

KAC-2

0 5

0.028

0.513

1.337

0.027

KAC-3 0

0 0.0

0.2

0.4

0.6

0.8

1.0

Relative pressure (P/P0)

1

10

100

Pore width (nm)

Fig. 4. Isothermal plots with relevant surface areas (a), and DFT pore size distributions and pore volumes (b) of CAC, KAC-1, KAC-2, and KAC-3, respectively.

Table 2 Iodine numbers and tannin values of CAC, KAC-1, KAC-2, and KAC-3, respectively. Sample

Iodine number (mg g 1)

Tannin value (mg L 1)

CAC KAC-1 KAC-2 KAC-3

1053 952 2040 2120

633 469 266 103

CAC. The results were fully consistent with the PVDFT-mesopore values of ACs (Fig. 4b), i.e. higher PVDFT-mesopore of the AC had lower tannin value. It was found that the tannin value and PVDFT-mesopore exhibited a great linear relationship (R2 = 0.9417, in Fig. 5b), indicating that the adsorption capacity of AC is strongly associated with its PVDFT-mesopore value.

3.4. Chemical adsorptions Tannin value test is another performance-based evaluation test described in the AWWA B600–10 standard. The tannin value is an index of the ability of absorbing high-molecular-weight impurities, such as organic compounds in the water by decayed vegetation. Tannin value is defined as the amount of AC requires to reduce the standard concentration of the tannic acid from 20 mg L 1 to 2 mg L 1 (90% removed from 20 mg L 1 standard tannic acid aqueous solution). Table 2 shows the tannin values of CAC, KAC-1, KAC2, and KAC-3 were 633, 469, 266, and 103 mg L 1, respectively. From the definition of tannin value, it was learned that the lower tannin value means the smaller amount of AC is needed for removing 90% tannic acid from 20 mg L 1 tannic acid aqueous solution. Hence, KAC-3 exhibited the highest quality of removing highmolecular-weight impurities, followed by KAC-2, KAC-1, and

Copper is a necessary micro-essential trace element for all the biological creatures including humans, animals, plants, and microorganisms [34]. However, if the concentration of copper exceeds the safety limit, it will result in various injurious health issues. For instance, the excess copper may cause problems with the liver, bone, central nervous system, immune system impairment and even pernicious anemia [35,36]. Thus, copper ion adsorption is important for drinking water. Fig. 6 shows that the residual copper (II) after being treated with CAC, KAC-1, KAC-2, and KAC-3 can be 5.62, 7.23, 5.09, and 5.09 mg L 1, respectively. Among them, KAC-2 and KAC-3 showed better performance for copper (II) adsorption, which matched the results of PVDFT-micropore and iodine number of the ACs (Fig. 4b and Table 2). As a small-molecule material, copper ion prefers to be absorbed by the micropore of AC.

46

Y. Wu et al. / Journal of Colloid and Interface Science 518 (2018) 41–47

(a) 4

3000

Randimine 6G (µM)

Iodine number (mg g–1)

(a)

R² = 0.9729

2000

1000

0 0.3

0.35

0.4

0.45

0.5

Tannin value (mg L–1)

3

2

1

0.55

0

PVDFT-micropore (cm3 g–1)

(b)

3.26

600

CAC

0.28

0.28

0.27

KAC-1

KAC-2

KAC-3

(b) R² = 0.9417

400 200 0 0

0.5

1

1.5

PVDFT-mesopore (cm3 g–1) Fig. 5. Relationships between (a) iodine number and PVDFT-micropore; (b) tannin value and PVDFT-mesopore.

10

CuBr2 (mg L–1)

5.62

5.09

5.09

KAC-2

KAC-3

4 2 0

KAC-1

KAC-2

KAC-3

CAC

Fig. 7. (a) Rhodamine 6G adsorptions (initial rhodamine 6G concentration: 400 mM) using CAC, KAC-1, KAC-2, and KAC-3, respectively. (b) A comparison of the pure water, and rhodamine 6G solutions after using CAC, KAC-1, KAC-2, and KAC-3, respectively.

7.23

8 6

Water

CAC

KAC-1

Fig. 6. Copper (II) adsorptions (initial CuBr2 concentration: 200 mg L KAC-1, KAC-2, and KAC-3, respectively.

1

) using CAC,

Since rhodamine 6G is a common organic dye with a molecular size larger than 2 nm, the mesopore structure of AC may play a critical role in the rhodamine 6G adsorption. Fig. 7 shows the rhodamine 6G aqueous solution after being treated with ACs for 24 h. It was observed easily that the solution being treated by CAC still presented red color, while the others (i.e. KAC-1, KAC-2, KAC-3) were illustrated to be as clear as water (Fig. 7b). The UV–vis analysis showed that, after the treatments of KAC-1, KAC-2, and KAC-3, 0.27 – 0.28 mM residual rhodamine 6G was found, while after being treated with CAC, 3.26 mM residual was detected (Fig. 7a). The reason probably was that the relatively low mesopore structure in CAC (Fig. 4b) can be used to absorb the rhodamine 6G molecules. 4. Conclusions High-performance ACs were successfully manufactured through self-activation process. The mechanism was investigated

using FT-IR and MS, indicating that the emitted gases functioned as activating agents. As the third type of activation methods, the self-activation was considered to be more environmental-friendly and cost-effective compared with the conventional activation approaches [21]. In addition, the performance of ACs from selfactivation was actually competitive with that of the traditional ACs. The SABET of KACs was up to 2296 m2 g 1, which was higher than those of most commercial ACs [37,38]. The self-activated ACs showed excellent performances on water cleaning. Compared to the commercial CAC, the self-activated ACs owned comparable or better adsorbing capacities. In addition, it was found that the adsorbing preference was strongly affected by the ACs’ pore structures. For instance, the AC micropore preferred to absorb smallmolecule impurities, and mesopore was good at absorbing largemolecule impurities. The quantitative study showed that PVDFTmicropore and PVDFT-mesopore presented good linear relations with the iodine number and tannin value, respectively. The future work is recommended to focus on the utilization of the emitted gases from the biomass activation process by taking the advantages of self-activation. For instance, the conversion of CO to valuable products, e.g. methanol, because of the great amount of CO produced during the self-activation process. Acknowledgements The research was partially discussed on the doctoral dissertation of Dr. Changlei Xia at University of North Texas. The authors are grateful for the support of the USDA NIFA Foundation Program Award (2017-67021-26138). We thank the Laboratory of Imaging Mass Spectrometry (LIMS) at University of North Texas on the MS and FT-IR analyses.

Y. Wu et al. / Journal of Colloid and Interface Science 518 (2018) 41–47

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.jcis.2018.02.017. References [1] H.J. Schellnhuber, S. Rahmstorf, R. Winkelmann, COMMENTARY: why the right climate target was agreed in Paris, Nat. Clim. Change 6 (7) (2016) 649–653. [2] N. Brun, P. Hesemann, D. Esposito, Expanding the biomass derived chemical space, Chem. Sci. 8 (7) (2017) 4724–4738. [3] D. Mohan, C.U. Pittman, Activated carbons and low cost adsorbents for remediation of tri-and hexavalent chromium from water, J. Hazard. Mater. 137 (2) (2006) 762–811. [4] M.A. Yahya, Z. Al-Qodah, C.Z. Ngah, Agricultural bio-waste materials as potential sustainable precursors used for activated carbon production: a review, Renew. Sustain. Energy Rev. 46 (2015) 218–235. [5] O. Ioannidou, A. Zabaniotou, Agricultural residues as precursors for activated carbon production—a review, Renew. Sustain. Energy Rev. 11 (9) (2007) 1966– 2005. [6] A.B. Rao, E.S. Rubin, A technical, economic, and environmental assessment of amine-based CO2 capture technology for power plant greenhouse gas control, Environ. Sci. Technol. 36 (20) (2002) 4467–4475. [7] X. Zhang, X. Ma, Y. Wu, Y. Li, Enhancement of farmland greenhouse gas emissions from leakage of stored CO2: simulation of leaked CO2 from CCS, Sci. Total Environ. 518 (2015) 78–85. [8] S.H. Knox, C. Sturtevant, J.H. Matthes, L. Koteen, J. Verfaillie, D. Baldocchi, Agricultural peatland restoration: effects of land-use change on greenhouse gas (CO2 and CH4) fluxes in the Sacramento-San Joaquin Delta, Global Change Biol. 21 (2) (2015) 750–765. [9] A.C. Lua, T. Yang, Effect of activation temperature on the textural and chemical properties of potassium hydroxide activated carbon prepared from pistachionut shell, J. Colloid Interface Sci. 274 (2) (2004) 594–601. [10] W. Tsai, C. Chang, M. Lin, S. Chien, H. Sun, M. Hsieh, Adsorption of acid dye onto activated carbons prepared from agricultural waste bagasse by ZnCl2 activation, Chemosphere 45 (1) (2001) 51–58. [11] Y. Guo, K. Yu, Z. Wang, H. Xu, Effects of activation conditions on preparation of porous carbon from rice husk, Carbon 41 (8) (2003) 1645–1648. [12] S.Q. Shi, W. Che, K. Liang, C. Xia, D. Zhang, Phase transitions of carbonencapsulated iron oxide nanoparticles during the carbonization of cellulose at various pyrolysis temperatures, J. Anal. Appl. Pyrol. 115 (2015) 1–6. [13] H. Dolas, O. Sahin, C. Saka, H. Demir, A new method on producing high surface area activated carbon: the effect of salt on the surface area and the pore size distribution of activated carbon prepared from pistachio shell, Chem. Eng. J. 166 (1) (2011) 191–197. [14] C. Saka, BET, TG–DTG, FT-IR, SEM, iodine number analysis and preparation of activated carbon from acorn shell by chemical activation with ZnCl2, J. Anal. Appl. Pyrol. 95 (2012) 21–24. [15] C. Bommier, R. Xu, W. Wang, X. Wang, D. Wen, J. Lu, X. Ji, Self-activation of cellulose: a new preparation methodology for activated carbon electrodes in electrochemical capacitors, Nano Energy 13 (2015) 709–717. [16] H. Yang, R. Yan, H. Chen, D.H. Lee, C. Zheng, Characteristics of hemicellulose, cellulose and lignin pyrolysis, Fuel 86 (12) (2007) 1781–1788. [17] T. Zhang, W.P. Walawender, L. Fan, M. Fan, D. Daugaard, R. Brown, Preparation of activated carbon from forest and agricultural residues through CO2 activation, Chem. Eng. J. 105 (1) (2004) 53–59. [18] C. Xia, S.Q. Shi, Self-activation for activated carbon from biomass: theory and parameters, Green Chem. 18 (7) (2016) 2063–2071.

47

[19] S.Q. Shi, C. Xia, Porositization process of carbon or carbonaceous materials, US Patent. 9/533,281, 2017. [20] C. Xia, C. Kang, M.D. Patel, L. Cai, B. Gwalani, R. Banerjee, S.Q. Shi, W. Choi, Pine wood extracted activated carbon through self-activation process for highperformance lithium-ion battery, Chemistryselect 1 (13) (2016) 4000–4007. [21] J. Wang, P. Nie, B. Ding, S. Dong, X. Hao, H. Dou, X. Zhang, Biomass derived carbon for energy storage devices, Journal of Materials Chemistry A 5 (6) (2017) 2411–2428. [22] S. Aber, A. Khataee, M. Sheydaei, Optimization of activated carbon fiber preparation from Kenaf using K2HPO4 as chemical activator for adsorption of phenolic compounds, Bioresour. Technol. 100 (24) (2009) 6586–6591. [23] Z.Z. Chowdhury, S.M. Zain, R.A. Khan, M.S. Islam, Preparation and characterizations of activated carbon from kenaf fiber for equilibrium adsorption studies of copper from wastewater, Korean J. Chem. Eng. 29 (9) (2012) 1187–1195. [24] C. Xia, S.Q. Shi, Self-activatin process to fabricate activated carbon from kenaf, Wood Fiber Sci. 48 (2016) 62–69. [25] M. Shamsuddin, N. Yusoff, M. Sulaiman, Synthesis and characterization of activated carbon produced from kenaf core fiber using H3PO4 activation, Procedia Chem. 19 (2016) 558–565. [26] A. Kumar, H.M. Jena, Preparation and characterization of high surface area activated carbon from Fox nut (Euryale ferox) shell by chemical activation with H3PO4, Results Phys. 6 (2016) 651–658. [27] P. Pendleton, S.H. Wu, A. Badalyan, Activated carbon oxygen content influence on water and surfactant adsorption, J. Colloid Interface Sci. 246 (2) (2002) 235–240. [28] A.A. Ceyhan, Ö. Sßahin, C. Saka, A. Yalçın, A novel thermal process for activated carbon production from the vetch biomass with air at low temperature by two-stage procedure, J. Anal. Appl. Pyrolysis 104 (2013) 170–175. [29] W.c. Ying, W. Zhang, Q.g. Chang, W.x. Jiang, G.h. Li, Improved methods for carbon adsorption studies for water and wastewater treatment, Environ. Prog. Sustain. Energy 25(2) (2006) 110–120. [30] G. Annadurai, R.-S. Juang, D.-J. Lee, Adsorption of rhodamine 6G from aqueous solutions on activated carbon, J. Environ. Sci. Health, Part A 36 (5) (2001) 715– 725. [31] D. Zhang, S.Q. Shi, C.U. Pittman, D. Jiang, W. Che, Z. Gai, J.Y. Howe, K.L. More, A. Antonyraj, Versatile and biomass synthesis of iron-based nanoparticles supported on carbon matrix with high iron content and tunable reactivity, J. Nanopart. Res. 14 (8) (2012) 1023. [32] T.N. Pham, S. Ono, K. Ohno, Ab initio molecular dynamics simulation study of successive hydrogenation reactions of carbon monoxide producing methanol, J. Chem. Phys. 144 (14) (2016). [33] C. Elliott, T. Colby, H. Iticks, Activated carbon obliterans alter aspiration of activated charcoal, Chest 96 (3) (1989) 672–674. [34] A. Ghaedi, M. Ghaedi, A. Vafaei, N. Iravani, M. Keshavarz, M. Rad, I. Tyagi, S. Agarwal, V.K. Gupta, Adsorption of copper (II) using modified activated carbon prepared from Pomegranate wood: optimization by bee algorithm and response surface methodology, J. Mol. Liq. 206 (2015) 195–206. [35] V.L. Goodman, G.J. Brewer, S.D. Merajver, Control of copper status for cancer therapy, Curr. Cancer Drug Targets 5 (7) (2005) 543–549. [36] B.R. Stern, Essentiality and toxicity in copper health risk assessment: overview, update and regulatory considerations, J. Toxicol. Environ. Health, A 73(2-3) (2010) 114–127. [37] P. Smith, T. Jiang, T. Tran, A. Mansour, High Energy Density Ultracapacitors, 2009. https://www1.eere.energy.gov/vehiclesandfuels/pdfs/merit_review_ 2009/energy_storage/esp_22_smith.pdf (Accessed January 18 2018). [38] T. Otowa, R. Tanibata, M. Itoh, Production and adsorption characteristics of MAXSORB: high-surface-area active carbon, Gas Sep. Purificat. 7 (4) (1993) 241–245.