MINISTRY OF EDUCATION AND TRAINING MINISTRY OF NATIONAL DEFENSE ACADEMY OF MILITARY SCIENCE AND TECHNOLOGY HOANG KIM HUE STUDY OF THE ADSORPTION OF 2,4-D AND 2,4,5-T FROM AQUEOUS SOLUTION ONTO CARBON NANOTUBES (CNTs) Specialization: Theoretical and Physical chemistry Code: 44 01 19 SUMMARY OF PhD THESIS IN CHEMISTRY HA NOI - 2019 The dissertation completed at: Academy of Military Science and Technology Academic supervisors: Dr Lam Vinh Anh Dr To Van Thiep Reviewer 1: Prof.Dr Ta Ngoc Don Hanoi University of Science and Technology Reviewer 2: Assoc.Prof.Dr Vu Anh Tuan Vietnam Academy of Science and Technology Reviewer 3: Assoc.Prof.Dr Nguyễn Manh Tuong Academy of Military Science and Technology The dissertation will be defended in front of Doctor Examining Committee held at Academy of Military Science and Technology at … … …, … … … …, 2019 The thesis can be found at: - The Library of Academy of Military Science and Technology - Vietnam National Library THE SCIENTIFIC PUBLICATIONS Hoang Kim Hue, Lam Vinh Anh, To Van Thiep, Pham Trung Kien (2016), Some initial results of the purification of carbon nanotubes (CNTs), synthesized by the chemical vapor deposition (CVD) method, Journal of catalysis and adsorption, 5(3), pp 52 - 62 Hoang Kim Hue, Lam Vinh Anh, To Van Thiep, Phung Thi Lan (2017), Study of the purification of carbon nanotube materials to be applied in the adsorption of 2,4-D herbicide in the aqueous solution, Journal of catalysis and adsorption, 6(2), pp 100 - 106 Hoang Kim Hue, Lâm Vinh Anh, To Van Thiep, Nguyen Hoang Dung (2017), Study of the activation of carbon nanotube materials to be applied in the adsorption of 2,4-dichlorophenoxyacetic acid, Journal of ScienceReseach and Military Technology, 52, pp 186 - 193 Hoang Kim Hue, Lam Vinh Anh, To Van Thiep (2018), Study of the adsorption of 2,4-dichlorophenoxyacetic acid from the aqueous solution onto carbon nanotubes, Vietnam Journal of Chemistry, 56(2), pp 191 - 197 Hoang Kim Hue, Lam Vinh Anh, Dinh Bao Trong (2018), Study of the adsorption of 2,4-dichlorophenoxyacetic acid from the aqueous solution onto activated carbon, Vietnam Journal of Chemistry, 56(2), pp 208 - 214 Hoang Kim Hue, Lam Vinh Anh, Le Minh Cam (2018), Comparative Study of 2,4-Dichlorophenoxyacetic Acid Adsorption onto Alkali-activated Carbon Nanotubes and Activated Carbon, Eleventh internation conference on the Remediation of chlorinated and Recalcitrant compounds, California, US, pp 32 Hoang Kim Hue, Lam Vinh Anh (2018), Study of the adsorption of 2,4,5T herbicide in the aqueous phase on activated carbon nanotubes, Journal of catalysis and adsorption, 7(1), pp 78 - 85 INTRODUCTION Due to Vietnam war’s consequences, Bien Hoa, Phu Cat and Da Nang air bases were severely polluted by 2,4-dichlorophenoxyacetic acid (2,4-D), 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) herbicides The estimated volume of contaminated soils and sediments is approximately 700000 m3 To date, a volume of nearly 90000 m3 at Da Nang air base has been treated by In-pile thermal desorption with the support of Vietnamese and American governments And a volume of 225000 m3 of contaminated soils and sediments has been treated by the isolation landfilling at Phu Cat, Bien Hoa and Da Nang air bases A huge volume of remaining contaminated soils, sediments and water at Bien Hoa air base requires a suitable treatment technology Nevertheless, all existing treatment technologies in Vietnam produce the solution of herbicides as a by-product, then a further treatment by adsorbents is required Currently, there is a fast development in the study and application of adsorbents in the environmental treatment Scientists continue to invent new materials with better adsorption capability In some recent decades, carbon nanotubes (CNTs) have drawn attention in study CNTs have uniformly porous structure, porous force, hydrophobicity and ability to form π - π interaction with molecules of 2,4-D, 2,4,5-T and Dioxin In addition, CNTs also possesses thermal stability, then, they could be recycled Therefore, CNTs are forecasted to be a promising adsorption material for the treatment of the solution of 2,4-D, 2,4,5-T and Dioxin The dissertation title: “study of the adsorption of 2,4-D and 2,4,5-T from the aqueous solution onto carbon nanotubes (CNTs)” The objective of the dissertation: To set up a purification and activation procedure from raw CNTs, synthesized in Vietnam to produce purified and activated CNTs to be used in the adsorption of 2,4-D and 2,4,5-T herbicides To study the adsorption characteristics of 2,4-D and 2,4,5-T herbicides on purified CNTs and activated CNTs in the aqueous solution Main content of the dissertation: Developing a purification procedure from raw CNTs in Vietnam Selecting activation condition for purified CNTs Studying factors, affecting the adsorption capability of 2,4-D on purified CNTs and activated CNTs Studying and setting up the adsorption isotherm, kinetics and thermodynamic parameters and activation energy of the adsorption of 2,4-D on purified CNTs and activated CNTs Studying and setting up the adsorption isotherm, kinetics and thermodynamic parameters and activation energy of the adsorption of 2,4,5-T on activated CNTs Academic and practical contributions of the dissertation: Studied and developed the purification procedure from raw CNTs in Vietnam and the activation methods from purified CNTs to increase purity, specific surface area, porous volume and 2,4-D adsorption capability of CNTs Studied and defined adsorption characteristics of 2,4-D and 2,4,5-T on purified CNTs and activated CNTs Structure of the dissertation: Introduction; Chapter 1: Overview; Chapter 2: Study objects and methods; Chapter 3: Results and discussions; Conclusion CHAPTER OVERVIEW 1.1 Introduction of 2,4-D and 2,4,5-T herbicides 1.1.1 Use history and toxicity 2,4-D and 2,4,5-T were used in agriculture since 40th decade of the previous century Their compounds were studied and mixed into herbicides to spray in the south from 1961 to 1971 during the Vietnam war Currently, 2,4-D and 2,4,5-T have been prohibited in many countries due to its severe toxicity to eyes, nerve system, endocrine, immune system and its possibility to cause blood cancer Especially, 2,4,5-Trichlorophenol molecules as a by-product of the decomposition of 2,4,5-T could combine together to form Dioxin if existing long time in the environment 1.1.2 Structure, chemical and physical properties of 2,4-D and 2,4,5-T - 2,4-D and 2,4,5-T are among acids with phenoxy group, and there is conjugated π electron system of benzene ring in their molecules - pKa,2,4-D = 2.73; pKa,2,4,5-T = 2.88 - logKOW,2,4-D = 2.81; logKOW,2,4,5-T = 4.00 - In aqueous environment, 2,4-D and 2,4,5-T dissolve into acidic radicals 1.1.3 Source and current situation of 2,4-D and 2,4,5-T pollution in Vietnam - Herbicides used in the agriculture - Herbicides used by US army during the Vietnam war 1.1.4 Some treatment methods for the source of herbicides, used in Vietnam war 1.1.5 Current situation of research on the adsorption of 2,4-D and 2,4,5-T onto carbon materials in the aqueous environment 1.2 Carbon nanotube materials and their adsorption characteristics to organic compounds 1.2.1 Overview of carbon nanotubes 1.2.2 Structure of CNTs 1.2.2.1 Crystalline structure Carbon nanotubes (CNTs) are made by rolling up of sheet of graphene into a hollow cylinders with two ends covered by fullurene hemispheres The diameter of the carbon nanotube is about a few nm, with a length of some μm to several centimeters The covalent carbon atoms together make up the tightened six-sided rings Each carbon atom has four electrons in the outer layer that make up the three bonds σ that have sp2 and π orbital hybrids However, according to number of walls in the tube, carbon nanotues are divided into single walled nanotubes (SWCNTs) and multi walled nanotubes (MWCNTs) Due to the interaction force π-π, carbon nanotues tend to aggregate into bundles, the distance between tubes in a bundle is approximately 0.34 nm In addition, carbon nanotubes could have hetero-elemental defects, expanding or narrowing the six-sided rings 1.2.2.2 Pore structure The surface area of the carbon nanotubes depends on the tube diameter, the number of walls, open ends or closed ends, the aggregation into bundles or seperation, functional groups on the surface of tubes or metal impurities in the tube core The tube core of SWCNTs has a diameter of less than nm while, the diameter of the tube core of MWCNTs is normally from to 15 nm 1.2.3 Surface chemistry of CNTs 1.2.4 Adsorption characteristics of organic compounds on CNTs 1.2.4.1 Non-uniform adsorption 1.2.4.2 Multi-mechanisms of interaction Hydrophobic interaction, electrostatic interaction, π-π interaction, hydrogen bonding interaction 1.2.4.3 Factors affecting the adsorption - Affected by the properties of CNTs: surface area, capillary volume, tube diameter, oxygen functional groups, aggregation into bundles - Affected by organic compounds: molecular geometry, functional groups - Affected by environmental conditions: temperature, pH, ionic force 1.2.5 Method of preparing CNTs 1.2.5.1 Method of synthesis Carbon nanotubes can be synthesized by arc, laser and chemical vapor deposition (CVD) methods Among them, the CVD method is widely used because of the high synthesis yield, but CNTs synthesized by this method contain many impurities such as amorphous carbon, aromatic organic compounds and metals 1.2.5.2 Purification methods - Chemical methods using techniques such as: oxidation in the gas phase, oxidation in liquid phase, electro-oxidation and treatment with HCl - Physical methods using techniques such as filtering, centrifugation and tempering - Integration method: combining physical and chemical methods 1.2.5.3 Activation method Chemical methods is more effective in increasing the surface area and in narrowing capillary size distribution of CNTs than physical methods 1.3 Theory of adsorption applied in the dissertation 1.3.1 Concept and classification of adsorption 1.3.2 Equation of adsorption isotherm The forms of the Langmuir (1.12) and Freundlich (1.15) can be expressed respectively as: KL Ce (1.12) (1.15) qe = qm qe = KF Cne + KL Ce Whereas qmax is the maximum adsorption capacity (mg/g), and K L is the Langmuir adsorption equilibrium constant (L/mg), which is related to the free energy of adsorption KF (mg1-n.g-1.Ln) represents the adsorption capacity when adsorbate equilibrium concentration equals to 1, and n represents the degree of adsorption dependence at equilibrium concentration KT is the equilibrium binding constant, BT is related to heat of adsorption, T is the absolute temperature (K) Whereas, Ce (mg/L) is the concentration of the solute at the equilibrium, qe: the the adsorbed amount at the equilibrium (mg/g), K L: Langmuir constant, KF (l/g) and: Freundlich constants 1.3.3 Equation of adsorption kinetics - Two equations of pseudo first and second order apparent kinetics can be expressed as (1.17) and (1.18), respectively: t 1 = + t (1.18) v0 = k2 q2e (1.19) ln(qe - qt ) = lnqt - k1 t (1.17) q t v0 q e Whereas, k1 (minutes-1) and k2 (g/mg.minutes): the rate constant of first and second pseudo apparent kinetics, qt: the adsorbed amount at the moment t (mg/g), v0: initial adsorption rate - Weber - Morris diffusion kinetics model is expressed as: (1.21) qt = kd t1⁄2 + L Where kd (mg/g min1/2) is the intra-particle diffusion rate constant (mg/g.minutes0.5) If L is equal to 0: the intra-particle diffusion controls the adsorption rate If L is different to 0: both film diffusion and intra-particle diffusion affect the adsorption 1.3.4 Thermodynamic and kinetic conditions CHARPTER STUDY OBJECTS AND METHODS 2.1 Study objects - 2,4-D and 2,4,5-T herbicides in the aqueous environment - CNTs synthesized in Vietnam (CNT-TH) 2.2 Chemicals and equipment 2.2.1 Hóa chất Reference substances of 2,4-D and 2,4,5-T have the purity of 99.9 % Toluene, ACN, HCl, HF, HNO3, CH3COOH, KOH, NaOH and CaCl2 have analytical purity grade, N2 is made in Vietnam with the purity of 99.999 %, CNT-TH is synthesized in Vietnam by CVD method with tube diameter of 10 to 30 nm and SBET: 170 - 200 m2/g CNT-TQ is MWCNTs of China with tube diameter of 10 to 20 nm and tube length of - 15 µm and the purity of over 97 %, Shirasagi activated carbon - Z1 (AC) is made in Japan, used in the study of the integration technology 2.2.2 Equipment Agilent HP-1100: high performance liquid chromatography (HPLC), Agilent GC-6890: gas chromatograph and Agilent MS 5975: signal detector, AAS - 300 - USA: atomic absorption spectrometer, Hanna HI 2211: pH meter, Mettler Toledo AB204: analytical Balance, Soxhlet extractor, Ultrasound machine, Shaking machine with temperature control, Vacuum dryer; SRJX Tube Furnace - 2,5 - 13 2.3 Study methods 2.3.1 Set up the purification procedure for CNT-TH The purification process is based on techniques that include: Soxhlet extraction, oxidation in liquid phase with HNO3, oxidation in air, treatment with HCl and HF, tempered at 900 ˚C in N2 gas The efficiency (HTC) of the purification process and Fe removal efficiency (HFe) are calculated according to equations (2.1) and (2.2) as follows: mFe,s ms ( ) H % = ∙100 (2.2) ( ) HTC % = ∙100 (2.1) Fe mFe,t mt Whereas, mt: the sample weight (g) before the purification, ms: the sample weight after the purification (g), mFe,t: the amount of Fe in the sample before the purification, mFe,s: the amount of Fe in the sample after the purificaiton 2.3.2 Investigation of activation conditions for CNT-TC The mixture of CNT-TC and KOH at the ratio of a/1 was mechanically ground, then heated at a temperature of x ˚C in N2 gas atmosphere with flow rate of z mL/min Rinsed the product with HCl and distilled water to neutral medium, dried and stored in a desiccator 2.3.3 Investigation of the adsorption process 2.2.3.1 Preparation of the adsorption solution 2.3.3.2 Adsorption conditions The adsorption process was studied by batch method The investigated concentrations of was in the range from 52.2 to 205.7 mg/L with 2,4-D and from 53.0 to 200.0 mg/L with 2,4,5-T - Adsorption isotherm was studied at: volume of solution (V): 50 mL, 2,4D adsorbent amount (m2,4-D): 50 mg, 2,4,5-T adsorbent amount (m2,4,5-T): 25 mg, pH = 6, temperature: 30 ˚C, shaking speed: 150 rpm, sampling time: 24 hours - Adsorption kinetics was studied at: V: 100 mL, m2,4-D: 100 mg, m2,4,5-T: 50 mg, pH = 6, temperature: 30 ˚C, shaking speed: 150 rpm, sampling time: 1, 2, 5, 8, 10, 12, 15, 20, 30, 40, 60, 90 and 120 minutes - Condition investigation: temperature: 10, 20, 30 and 40 ˚C, pH: - 9, ionic force (concentration of CaCl2) at 0, 0.005, 0.01, 0.1, 0.5 and mol/L Sample volume, filtered by ultrafiltration membrane: 0,5 mL 2.3.3.3 Determination of the adsorption capacity of the material The adsorption amount at the time (qt,mg/L) and at the equilibrium (qe,mg/L) are calculated as follows: C0 - Ct C0 - Ce (2.3) (2.4) qt = ∙V qe = ∙V m m Adsorption efficiency (HHp): C0 - Ce HHP = ∙100 (2.5) C0 Whereas, C0, Ct, Ce (mg/L) are concentrations of 2,4-D or 2,4,5-T in the solutions at the initial time, moment t and at the equilibrium, respectively 2.2.3.4 Set up adsorption isotherm Following the linear regression of experimental data 2.2.3.5 Set up adsorption kinetics Following the linear regression of experimental data 2.2.3.6 Evaluation of the fitting of the isothermal model 10 (a) 100 95 90 85 (b) 1700 SBET (m2/g) Hiệu suất hấp phụ (%) 3.1.3.2 Purity of CNT-TC The TGA results showed that CNT-TC lost about 99.91 % by weight due to the burning of CNTs at 500 - 700 °C The residue obtained after 900 °C is about 0.09 %, being consistent with the amout of Fe of 0.08 %, determined by AAS In addition, there were almost no signals of Al, Si and Fe in EDX diagram of CNT-TC There were only signals Hình 3.20: Giản đồ EDX CNT-TC for 97.70 % of C and 2.30 % of O Therefore, the purity of CNT-TC is about 97.61 % and CNT-TC has the higher quality grade than CNT-TQ 3.2 Study of material chemistry 3.2.1 Activation conditions of CNT-TC 1200 700 200 80 CNT-TC CNT-HK CNT-HNa CNT-TC CNT-HK CNT-HNa AC AC 100 98 96 94 92 90 600 (a) SBET (m2/g) Hiệu suất hấp phụ (%) Figure 3.21: Adsorption capacity of 2,4-D (a) and surface area (b) of CNT-TC, CNT-HK, CNT-HNa and AC (C0 = 52.2 mg/L) (b) 500 400 300 200 8 Ratio of KOH/CNT-TC Ratio of KOH/CNT-TC Figure 3.22: Effect of KOH ratio, used to activate CNT-TC on the adsorption of 2,4-D (a) and the surface (b) of HKi (C0 = 52.2 mg/L) 98 96 94 92 99 Adsorption efficiency (%) Adsorption efficiency (%) Adsorption efficiency (%) 100 98 97 500 600 700 800 900 1000 Temperature (˚C) 98 96 94 96 90 100 0.5 1.5 2.5 Time (hours) 500 1000 1500 Gas blowing rate (mL/min) Figure 3.23 Figure 3.24 Figure 3.25 Figure 3.23, 3.24, 3.25: Effect of temperature, time and N2 blowing rate on the adsorption of 2,4-D on HKi (C0 = 52.2 mg/L) 11 Appropriate conditions to activate CNT-TC included: KOH as activating agent, ratio of KOH/CNT-TC: 5/1, temperature: 800 ˚C, activation time: hour and N2 blowing rate: 500 mL/min 3.2.2 Physico chemical characteristics of CNT-HKi 0.12 Absorbance(Abs) dV/dD (cm3/g.nm) Figure 3.27: TEM image of CNT-HK5 0.09 0.06 0.03 0.05 0.04 0.03 CNT-TC CNT-HK5 0.02 0.01 D (nm) CNT-TC CNT-HK5 3500 CNT-HK3 CNT-HK7 Wavenumber Relative intensity Figure 3.29: Capillary size distribution of CNT-TC, CNT-HKi 1.2 0.8 0.6 0.4 0.2 1000 2500 1500 500 (cm-1) Figure 3.30: IR Spectra of CNT-TC and CNT-HK5 CNT-TC CNT-HK5 1200 1400 1600 1800 Frequency (cm-1) Figure 3.32: Raman spectra of CNT-TC and CNT-HK5 Table 3.8: pHPZC of CNT-TC and CNT-HKi Material CNT-TC CNT-HK3 CNT-HK5 CNT-HK7 pHPZC 8.45 7.40 7.10 6.80 According to the XRD results, the activation did not destroy the crystalline structure of MWCNTs, but increased the defect (Figure 3.27) The specific surface area was significantly increased from 267 to 540 m2/g and the volume of the capillary in the diameter range from 3.2 to 4.2 nm was increased to 2.5 times (Figure 3.29) Activated CNTs did not have any different functional groups from CNT-TC (Figure 3.30), but the pHPZC value decreased with the increase in the KOH/CNT-TC ratio (Table 3.8) In addition, Raman analysis showed that the density per unit area of sp2 carbon decreased after activation (Figure 3.32) 12 Adsorption efficiency (%) Adsorption efficiency (%) qe(mg/g) qe(mg/g) qe (mg/g) 3.3 Study of adsorption kinetics of 2,4-D and 2,4,5-T on carbon nanotubes 3.3.1 Some factors affecting the adsorption capacity of 2,4-D on CNT-TC and CNT-HKi CNT-TC CNT-HK3 3.3.1.1 Effect of initial concentration of 2,4-D CNT-HK5 CNT-HK7 3.3.1.2 Effect of temperature 150 The adsorption capacity of 2,4-D of 120 90 CNT-TC and CNT-HKi decreased with the 60 increase of temperature (Figure 3.34) This is a 30 50 75 100 125 150 175 200 sign of exothermic adsorption and physical Cₒ (mg/l) nature Figure 3.33: Effect of initial 3.3.1.3 Effect of pH concentration of 2,4-D The adsorption capacity of 2,4-D on 48 (a) CNT-TC and CNT-HKi decreased as the pH of 46 the solution approached pHPZC, then negligibly 44 decreased if the pH of the solution exceeds the 42 40 this pH value (Figure 3.35) This could be 10 20 30 40 50 explained by the electrostatic interactions T (˚C) Indeed, at pKa, 2,4-D= 2.73 < pHdd < pHPZC, 52 (b) 2,4-D existed as anion form in the solution while the surface of adsorbents was positively 51.5 51 charged, therefore, there was electrostatic interaction between them The electrostatic 50.5 10 20 30 40 50 attraction force decreased as the pH T (˚C) approached pHPZC and the electrostatic repulsion force occured at pHdd > pHPZC as the Figure 3.34: Effect of temperature surface of the material is negatively charged 100 (a) Thus, the adsorption capability decreased as 95 90 the electrostatic attraction decreased and the 85 negligibly changed when the repulsion force 80 between them appeared 75 10 3.3.1.4 Effect of ionic force pH The 2,4-D adsorption capacity of CNT-TC 100 (b) 99 and CNT-HKi increased as the concentration 98 of CaCl2 in the solution increased Because the 97 presence of salts in the solution induced 96 95 pressing on the diffusion layer on the material, 10 then favored the electrostatic attraction and pH thus favored the adsorption Figure 3.35: Effect of pH 13 3.3.2 Investigation of adsorption isotherm models of 2,4-D on CNT-TC and CNT-HKi 3.3.2.1 Set up Langmuir adsorption isotherm model Langmuir isotherm parameters were found to describe the adsorption equilibrium of 2,4-D on CNT-TC/CNT-HKi and displayed in the table 3.9 Table 3.9: Langmuir isotherm parameters of the adsorption of 2,4-D on CNT-TC and CNT-HKi T qm qmdt KL ARE Material RL R2 (˚C) (mg/g) (µg/m ) (L/mg) (%) 10 84.03 313.56 0.1475 0.1149 0.9980 5.87 20 84.03 313.56 0.0982 0.1631 0.9955 6.69 CNT-TC 30 83.33 310.95 0.0837 0.1861 0.9944 4.52 40 79.37 296.14 0.0754 0.2024 0.9947 4.49 10 140.85 304.86 0.5000 0.0369 0.9979 12.18 20 138.89 300.63 0.3258 0.0555 0.9964 14.92 CNT-HK3 30 135.14 292.50 0.2426 0.0731 0.9948 9.56 40 128.87 281.10 0.1778 0.0972 0.9944 11.33 10 156.25 289.35 0.4476 0.0410 0.9968 13.85 20 151.52 280.58 0.3568 0.0509 0.9964 13.87 CNT-HK5 30 147.06 272.33 0.2528 0.0704 0.9944 12.83 40 142.86 264.55 0.1813 0.0955 0.9929 11.25 10 158.73 287.56 0.4286 0.0427 0.9940 16.49 20 153.85 278.71 0.3283 0.0551 0.9942 16.10 CNT-HK7 30 149.25 270.39 0.2659 0.0671 0.9937 13.80 40 147.06 266.41 0.1915 0.0908 0.9908 12.95 In table 3.9, the dependence of Ce/qe on Ce was displayed as linear lines with the correlation coefficient R2 larger than 0.9908 Thus, parameters of Langmuir models were highly reliable On the other hand, the value R L is between and 1, indicating that the adsorption is favorable in the investigated concentrion range However, the value of ARE is quite large with lowest value from 4.49 to 16.49 % Therefore, Langmuir model could be used to describe the adsorption isotherm equilibrium of 2,4-D on CNT-TC and CNT-HKi The value of qm (mg/g) characterized the saturated monolayer per unit weight of the material showing that the material with high specific surface area and high capillary size distribution in the range of 3.2 - 4.2 nm high will have higher 2,4-D adsorption capacity However, comparing the adsorption amount per unit surface area of the material (qmdt, µg/m2) showed the opposite side Thus, activation reduced 14 the adsorption capacity per unit area of the surface of the material On the other hand, according to the Raman results, activation reduced the number of sp2 hybrid carbon per unit area of the surface, while sp2 hybrid carbon atom CNTs has π orbital that can interact π-π with benzene ring in the 2,4-D molecule Thus, it can be assumed that the decrease in π-π interaction reduced the adsorption capacity per unit surface area of the material after activation 3.3.2.2 Freundlich isotherm model Table 3.10: Freundlich isotherm parameters of the adsorption of 2,4-D on CNT-TC and CNT-HKi Material T (˚C) KF (L/g) n 1/n R2 ARE (%) 10 35.823 5.7703 0.1733 0.9826 1.80 20 30.229 4.9628 0.2015 0.9748 2.71 CNT-TC 30 27.259 4.5935 0.2177 0.9900 1.33 40 24.179 4.3197 0.2315 0.9736 2.66 10 65.543 5.0582 0.1977 0.9563 6.16 20 59.033 4.8591 0.2058 0.9891 3.09 CNT-HK3 30 52.342 4.5208 0.2212 0.9818 3.17 40 45.966 4.2845 0.2334 0.9934 1.83 10 66.853 4.5086 0.2218 0.9766 4.74 20 61.529 4.3821 0.2282 0.9803 3.06 CNT-HK5 30 53.678 4.0733 0.2455 0.9861 3.42 40 46.829 3.8565 0.2593 0.9907 2.39 10 67.518 4.5025 0.2221 0.9818 4.66 20 61.886 4.3802 0.2283 0.9922 2.75 CNT-HK7 30 55.857 4.1425 0.2414 0.9894 3.01 40 49.043 3.9063 0.2560 0.9926 2.32 Freundlich isotherm parameters of the adsorption of 2,4-D on CNT-TC and CNT-HKi in the table 3.10 indicated that the relationship between lnCe and lnqe is linear with the correlation coefficient R2 larger than 0.9563 The value of ARE is from 1.33 to 6.16 %, smaller than Langmuir model Therefore, Freundlich model is more fitting than Langmuir model The fitting of Freundlich model indicated that the surface of the material is not uniform As mentioned in above discussions, the adsorption of 2,4-D on CNT-TC and CNT-HKi was dominated by the π-π interaction and electrostatic force 15 3.3.3 Determination of thermodynamic parameters of the adsorption of 2,4-D on CNT-TC and CNT-HKi Thermodynamic parameters of the adsorption of 2,4-D on CNT-TC and CNT-HKi were determined and presented in the table 3.13 Table 3.13: Thermodynamic parameters of the adsorption of 2,4-D on CNT-TC and CNT-HKi ΔG0T (kJ/mol) ΔH0298 ΔS0298 Material 283 K 293 K 303 K 313 K (kJ/mol) (J/mol.K) CNT-TC -13.387 -11.007 -11.508 -9.541 -49.144 -127.250 CNT-HK3 -16.749 -16.148 -14.804 -13.862 -45.069 -99.593 CNT-HK5 -15.802 -15.411 -14.223 -13.203 -41.184 -89.010 CNT-HK7 -15.887 -15.445 -14.654 -13.609 -37.382 -75.447 Table 3.13 indicated that the adsorption enthalpy change (∆H0298 ), entropy change (ΔS0298 ), Gibb free engery change (∆G0T ) at 283, 293, 303 and 313 K were all negative Therefore, the adsorption was spontaneous, exothermic and the order of the system was increased The adsorption heat was in the range from 20 to 80 kJ/mol, characterizing the electrostatic interaction and physical adsorption With results from the investigation of pH effect in the item 3.3.1.3, it was concluded that the electrostatic interaction had significant impact on the adsorption capacity On the other hand, experiments were done at pH of 6, smaller pHPZC of CNT-TC and CNT-HKi, then the electrostatic interaction was the attaction force It is notable that the 2,4-D adsorption heat on the study materials decreased with the reduction of the pHPZC value It was possible that the narrow gap between pHPZC value of the adsorbents and pH value of the solution led to the decrease in the electrostatic attraction force, reducing the exothermic heat of the system 3.3.4 Study of the adsorption of 2,4,5-T on CNT-HK5 and its comparison with 2,4-D 3.3.4.1 Comparison of hydropobicity of 2,4-D and 2,4,5-T Bipolar momentums of the bond between C-OCH2COOH and C-Cl were named as: µ1 and µ2 μ2,4-D = √μ21 + μ22 - μ1 μ2 (3.15) μ2,4,5-T = √μ21 + μ22 - 2μ1 μ2 (3.16) From the expressions (3.15) and (3.16), it was observed that: μ2,4,5-T < 𝜇2,4−𝐷 indicated that 2,4,5-T was less disolved in water than 2,4-D 16 Therefore, 2,4,5-T was more hydrophobic than 2,4-D This result was also consistent with the value KOW, 2,4,5-T larger than KOW, 2,4-D 3.3.4.2 Set up the adsorption isotherm of 2,4,5-T on CNT-HK5 Isothermal parameters of Langmuir and Freundlich models established to decribe the adsorption of 2,4,5-T on CNT-HK5 were presented on the table 3.14 Table 3.14: Isothermal parameters of Langmuir and Freundlich models for the adsorption of 2,4,5-T on CNT-HK5 Parameters of isothermal models T (˚C) Langmuir qm (mg/g) KL (L/mg) RL R2 ARE (%) 10 200.000 0.4505 0.0402 - 0.0110 0.9977 14.15 20 196.078 0.4080 0.0442 - 0.0121 0.9982 12.95 30 196.078 0.3091 0.0575 - 0.0159 0.9967 13.41 40 192.307 0.2905 0.0610 - 0.0169 0.9980 10.49 Freundlich KF (L/g) n 1/n R2 ARE (%) 10 117.378 8.4962 0.1177 0.9752 3.08 20 113.137 8.2169 0.1217 0.9828 2.50 30 107.254 7.7640 0.1288 0.9886 1.91 40 101.291 7.2993 0.1370 0.9905 1.83 3.3.4.3 Determination of thermodynamic parameters of the adsorption of 2,4,5-T on CNT-HK5 Table 3.15: Thermodynamic parameters of the adsorption of 2,4,5-T on CNT-HK5 T (˚C) ΔG0T (kJ/mol) ΔH0298 (kJ/mol) ΔS0298 (J/mol.K) 283 - 25.225 293 - 25.137 - 43.685 - 64.443 303 - 24.211 313 - 23.350 3.3.4.4 Comparison of the adsorption capacity of 2,4,5-T and 2,4-D on CNT-HK5 Two typical parameters of the adsorption capacity of the system is qm and KF These values of the adsorption of 2,4-D and 2,4,5-T on CNT-HK5 at 10, 20, 30 and 40 ̊C were displayed in the table 3.16 17 Table 3.16: Comparison of qm and KF values of the adsorption of 2,4-D and 2,4,5-T on CNT-HK5 at different temperatures qm (mg/g) KF (L/g) Temperature˚C 2,4-D 2,4,5-T 2,4-D 2,4,5-T 10 156.250 200.000 66.853 117.378 20 151.515 196.078 61.529 113.137 30 147.059 196.078 53.678 107.254 40 142.857 192.307 46.829 101.291 Table 3.16 indicated that at all investigated temperatures, the maximum adsoprtion capacity of 2,4,5-T on CNT-HK5 was always higher that than value of 2,4-D on CNT-HK5 and KF value of 2,4,5-T on CNT-HK5 was always higher than that value of 2,4-D on CNT-HK5 On the other hand, the adsorption heat of 2,4,5-T on CNT-HK5 was also higher than that value of 2,4-D on CNT-HK5 While, 2,4-D and 2,4,5-T have similar molecular structure, the hydrophobicity of 2,4,5-T is higher than that of 2,4-D Therefore, it could be concluded that the hydrophobic interaction force played a role in the adsorpion of 2,4-D and 2,4,5-T on CNT-HK5 3.3.5 Summarization of study results of adsorption thermodynamics The results of the study can be summarized as follows: The adsorption of 2,4-D and 2,4,5-T on CNT-TC and CNT-HKi showed signs of physical adsorption, occurring on the heterogeneous surface of the material in terms of energy The adsorption equilibrium was best described in the Freundlich model There were several types of adsorption forces that were active simultaneously in the adsorption of 2,4-D and 2,4,5-T on CNT-TC and CNT-HKi such as electrostatic forces, hydrophobic interaction forces and π-π interaction forces, where electrostatic forces had the primary influence on the adsorption The adsorption of 2,4-D and 2,4,5-T on CNT-TC and CNT-HKi was spontaneous, exothermic and order-increased The heat generated in the adsorption was from 37.382 to 49.144 kJ/mol 3.4 Kinetics study of the adsorption of 2,4-D and 2,4,5-T on CNTs 3.4.1 Effect of contact time on the adsorption capacity of 2,4-D on CNT-TC and CNT-HKi In all cases of the adsorption of 2,4-D on CNT-TC and CNT-HKi, increasing the contact time from to minutes led to the significant increase in the adsorption capacity at all different concentrations, then, the increase was slow and the equilibrium was reached 18 3.4.2 Establishing the adsorption kinetics of 2,4-D on CNT-TC and CNT-HKi 3.4.2.1 Động học khuếch tán Weber - Morris 45 35 10 y = 0.6087x + 48.114 R² = 0.8998 46 43 y = 6.2948x + 35.779 R² = 0.8608 12 46 43 40 qt (mg/g) qt (mg/g) 49 10 12 52 (c) y = 0.0594x + 50.666 R² = 0.9239 y = 0.6363x + 48.191 GĐ1 R² = 0.9214 GĐ2 y = 5.6994x + 37.081 GĐ3 R² = 0.9151 t1/2(min1/2) t1/2(min1/2) 52 GĐ1 GĐ2 GĐ3 40 30 (b) y = 0.0814x + 50.396 R² = 0.9114 49 qt (mg/g) y = 0.0954x + 43.07 R² = 0.7858 y = 1.6004x + 36.591 GĐ1 R² = 0.9303 GĐ2 GĐ3 y = 6.5579x + 25.396 R² = 0.9391 40 qt (mg/g) 52 (a) y = 0.0609x + 50.685 R² = 0.8621 49 y = 0.3875x + 49.316 R² = 0.8494 46 GĐ1 GĐ2 GĐ3 y = 3.49x + 42.545 R² = 0.8645 43 (d) 40 t1/2(min1/2) 10 12 10 12 t1/2(min1/2) Figure 3.48: Weber - Morris diffusion kinetics of the adsorption of 2,4-D on CNT-TC (a), CNT-HK3 (b), CNT-HK5 (c) and CNT-HK7 (d) Based on Weber - Morris model, the adsorption of 2,4-D on CNT-TC could be separated into three steps: 1-5 minutes, 5-20 minutes and 20-120 minutes The value of Li is different from 0, indicating both film diffusion and intra-particle diffusion governed the adsorption rate 3.4.2.2 Pseudo first order adsorption kinetics The correlation value R2 of linear lines, displaying the relationship between t and ln(qe-qt) was low, therefore, the pseudo fist order adsorption model was not fit to decribe the whole adsorption process of 2,4-D on CNT-TC and CNT-HKi 3.4.2.3 Pseudo second order adsorption kinetics The correlation value R2 of linear lines, displaying the relationship between t/qt and t, and parameters of pseudo second order kinetics model of the adsorption of 2,4-D on CNT-TC and CNT-HKi were displayed in the table 3.18 In the table 3.18, the correlation value R2 in all cases are greater than 0.9996 The adsorption capacity at the equilibrium, calculated from the pseudo second order model (qe,tt) was similar to that empirical value (qe,tn) Therefore, the adsorption of 2,4-D on CNT-TC and CNT-HKi was well decribed by the pseudo second order kinetics model In the table 3.18, the initial concentration of 2,4-D had different impact on the initial adsorption rate (v₀), depending on the adsorption material 19 Table 3.18: Parameters of the pseudo second order kinetics equation of the adsorption of 2,4-D on CNT-TC and CNT-HKi Parameters Con qe,tn qe,tt k2 v0 R2 (mg/L) (mg/g) (mg/g) (g/mg.min) (mg/g.min) CNT-TC C01 43.973 44.248 0.0460 90.090 C02 53.331 53.476 0.1093 312.500 C03 60.512 60.976 0.0708 262.158 C04 71.266 71.429 0.0332 169.492 0.9999 C05 72.885 72.993 0.0196 104.167 0.9998 C06 81.237 80.645 0.0116 75.758 0.9996 CNT-HK3 C01 51.240 51.282 0.0809 212.766 C02 70.849 70.922 0.0497 250.000 C03 91.123 90.909 0.0484 400.000 C04 104.602 105.263 0.0475 526.316 C05 112.284 112.360 0.0720 909.091 C06 128.285 128.205 0.0608 1000.00 CNT-HK5 C01 51.279 51.282 0.0951 250.000 C02 72.517 72.464 0.0501 263.158 C03 93.400 93.458 0.0477 416.667 C04 112.342 113.636 0.0430 555.556 C05 118.186 117.647 0.0723 1000.00 C06 141.327 140.845 0.0630 1250.00 CNT-HK7 C01 52.292 51.282 0.1311 344.828 C02 72.596 72.464 0.0732 384.615 C03 94.845 95.238 0.0735 666.667 C04 115.747 116.279 0.0561 769.231 C05 121.309 121.951 0.0672 1000.00 C06 148.170 149.254 0.0561 1250.00 With CNT-TC, v0 increased from 90.090 to 312.500 (mg/g.min) as the initial concentration of 2,4-D increased from 52.2 to 75.833 (mg/L) However, if C0 was continued to increase to 205.672 (mg/L), then v0 decreased to 75.758 (mg/g.min) Because, when the initial concentration of 2,4-D was low, the number of adsorption sites on the surface of CNT-TC was sufficient for the adsorption However, as the number of adsorption sites 20 was unchanged, continuing to increase the initial concentration of 2,4-D led to the adsorption competition between 2,4-D molecules on adsorption sites, then, reduced v0 With CNT-HK3, CNT-HK5 and CNT-HK7, v0 increased from 212.766 to 1000.00 (mg/g.min), from 250.000 to 1250.00 (mg/g.min) and from 344.828 to 1250.00 (mg/g.min), respectively as the initial concentration of 2,4-D increased in the investigated range Therefore, v0 of CNT-HKi was much higher than that value of CNT-TC It could be explained by the fact that the capillary volume in the diameter range of 3.2-4.2 nm of CNT-HKi was 2.5 times higher than that value of CNT-TC In addition, the aggregation of bundles of CNT-HKi created more capillaries with open ends at middle of tubes, then, the capillary force of CNT-HKi was higher than CNT-TC This was also the reason why the value of v0 of CNT-HKi was higher than CNT-TC 3.4.3 Effect of temperature and activation energy on the adsorption of 2,4-D on CNT-TC and CNT-HKi Table 3.19: Pseudo second order kinetics parameters of the adsorption of 2,4-D on CNT-TC and CNT-HKi at different temperatures (C03 = 102.026 mg/L) T qe,tn qe,tt k2 v0 Material R2 (˚C) (mg/g) (mg/g) (g/mg.phút) (mg/g.phút) 10 65.489 65.790 0.0340 147.059 20 64.802 64.935 0.0516 217.391 0.9996 CNT-TC 30 60.512 60.976 0.0708 262.158 40 56.761 57.471 0.1044 344.828 0.9999 10 94.641 95.238 0.0230 208.333 20 92.882 93.458 0.0358 312.500 CNT-HK3 30 91.123 90.909 0.0484 400.000 40 88.399 88.496 0.0751 588.235 10 97.192 98.039 0.0196 188.679 20 95.934 96.154 0.0309 285.714 CNT-HK5 30 93.400 93.458 0.0477 416.667 40 91.598 91.743 0.0743 625.000 10 97.835 98.039 0.0274 263.158 20 96.613 97.087 0.0442 416.670 CNT-HK7 30 94.845 95.238 0.0735 666.667 40 92.243 92.593 0.1060 909.091 Parameters of the pseudo second order kinetics of the adsorption of 2,4-D on CNT-TC and CNT-HKi at the initial concentration C03 of 102.026 mg/L at 10, 20, 30 and 40 ˚C were presented in the table 3.19 21 From the table 3.19, the value k2 increased with an increase of temperature If the dependence of k2 on temperature was assumed to follow the Arrhenius equation, then the apparent activation energy of the adsorption (Ehp) was calculated and presented in the table 3.20 Table 3.20: Activation energy of the adsorption of 2,4-D on CNT-TC and CNT-HKi Material CNT-TC CNT-HK3 CNT-HK5 CNT-HK7 Ehp (kcal/mol) 6.370 6.784 7.785 8.052 In the table 3.20, Ehp of the adsorption systems was always smaller than 8.052 kcal/mol, consistent with published studies, stated that the activation energy of the physical adsorption was less than kcal/mol or 40 kj/mol Therefore, it could be assumed that the adsorption of 2,4-D on CNT-TC and CNT-HKi had physical nature and this was consistent with the results of adsorption thermodynamics 3.4.4 Kinetics study of the adsorption of 2,4,5-T on CNT-HK5 and its comparison with 2,4-D 3.4.4.1 Kinetics study of the adsorption of 2,4,5-T on CNT-HK5 The adsorption of 2,4,5-T on CNT-HK5 could be considered as three stages when describing by Weber - Morris and the pseudo second order kinetics model was also fit to describe the whole adsorption process like the adsorption of 2,4-D on CNT-HK5 3.4.4.2 Effect of the initial concentration on the adsorption kinetics Parameters of apparent adsorption kinetics of the adsorption of 2,4,5-T on CNT-HK5 at different initial concentration of 2,4,5-T were displayed in the table 3.22 Table 3.22: Parameters of apparent adsorption kinetics at different initial concentration of 2,4,5-T Parameters Con qe,tn qe,tt k2 v0 R2 (mg/L) (mg/g) (mg/g) (g/mg.min) (mg/g.phút) C01 104.095 104.167 0.0174 188.679 C02 142.202 142.857 0.0094 192.308 C03 164.869 163.934 0.0133 357.143 C04 177.122 178.571 0.0112 357.143 0.9999 C05 178.817 178.571 0.0224 714.286 0.9999 C06 195.454 196.078 0.0217 833.333 0.9999 22 vₒ (mg/g.min) 1500 In the figure 3.59, at all investigated 1200 concentrations, the initial adsorptionr ate 900 600 of 2,4,5-T on CNT-HK5 was slower than 300 that value of 2,4-D on CNT-HK5 C₀₁ C₀₂ C₀₃ C₀₄ C₀₅ C₀₆ although the hydrophobic interaction Adsorbate concentration 2,4-D/CNT-HK5 2,4,5-T/CNT-HK5 with adsorption sites was stronger with 2,4,5-T However, each molecule of Figure 3.59: Initial adsorption 2,4,5-T had one more chlorine atom at the rate of 2,4-D and 2,4,5-T on CNT-HK5 at different fifth position, compared to the molecule concentrations of 2,4-D, then, the bulky state of 2,4,5-T caused its difficulty in diffusion into the core, channels and hollow space of bundles of tubes in CNT-HK5 3.4.4.3 Effect of temperature and activation energy Kinetics parameters of the adsorption of 2,4,5-T on CNT-HK5 at the initial concentration C03 of 101.717 mg/L were presented in the table 3.23 Table 3.23: Pseudo second order kinetics parameters of the adsorption of 2,4,5-T on CNT-HK5 at different temperature (C03 = 101,717 mg/L) T qe,tn qe,tt k2 v0 R2 (˚C) (mg/g) (mg/g) (g/mg.min) (mg/g.min) 10 172.073 175.439 0.0044 135.135 20 168.764 169.492 0.0083 238.095 30 164.869 163.934 0.0133 357.143 40 161.308 161.290 0.0275 714.286 The activation energy of the the adsorption of 2,4,5-T on CNT-HK5 was determined to be 10.493 kcal/mol 3.4.5 Summarizing the study of the adsorption kinetics Based on the study results of the adsorption of 2,4-D on CNT-TC, CNTHKi and the comparison with between the adsorption of 2,4-D and 2,4,5-T on CNT-HK5, the following conclusions were drawn: The adsorption included three stages: film diffusion, intra-particle diffusion and adsorption Among them, the intra-particle diffusion was the slowest and the most difficult The adsorption was well described by the pseudo second order kinetics model The initial adsorption rate increased with an increase of the specific area and the capillary volume in the capillary diameter range of 3.2-4.2 nm With each adsorption system, the adsorption rate increased with the initial concentration of 2,4-D, 2,4,5-T and temperature 23 CONCLUSION From the dissertation, the following conclusions were made: Successfully set up a purification procedure for CNTs, synthesized in Vietnam The purification procedure was an integration of chemical and physical techniques With this procedure, impurites such as organic compounds, amorphous carbon, Al, Si and Fe were eliminated The residual obtained after heating at 900 ˚C was about 0.09 % by weight and the percentage of Fe was about 0.08 % The purification process incresaed the purity of the material from 75.43 to 97.61 %, the specific surface area from 170 to 267 m2/g and the capillary volume from 0.897 to 1.426 cm3/g At the same time, the adsorption capacity of 2,4-D on the material increased from 64.39 to 83.24 % Conditions for the activation of purified CNTs by KOH were proposed The specific area increased to 540 m2/g and the capillary volume in the diameter range of 3.2-4.2 nm increased 2.5 times The activation process increased the defect, reduced the density of sp2 hybrid carbon on an unit of area of activated CNTs, compared to purified CNTs The value of pHPZC of activated CNTs decreased with an increase of the weight ratio of KOH/purified CNTs The effect of the 2,4-D initial concentration, temperature, pH and ionic force on the adsorption capacity of 2,4-D on purified and activated CNTs were investigated The adsorption equilibrium was described by both Langmuir and Freundlich isotherm models A comparative study of the adsorption capacity of 2,4-D 2,4,5-T as two compounds with calculated bipolar momentum to define the hydrophobicity on activated CNTs was also carried out Results indicated that the 2,4-D adsorption capacity on activated CNTs was higher than that of purified CNTs with the increase of specific area and the capillary volume in the diameter range of 3.2-4.2 nm The adsorption equilibrium of 2,4-D and 2,4,5-T on materials was well described by Freundlich model with the heterogeneous surface, active different forces for the adsorption of 2,4-D and 2,4,5-T such as electrostatic force, hydrophobic interaction and π - π interaction The adsorption of 2,4-D on purified CNTs, 2,4-D and 2,4,5-T activated CNTs was described by three kinetics models: Weber-Morris, pseudo first 24 order kinetics, pseudo second order kinetics Results indicated that the adsorption included three stages: film diffusion, intra-particle diffusion and adsorption The whole adsorption process was well fit with the pseudo second order kinetics The initial adsorption rate of activated CNTs was greater than that of purified CNTs and this value increased with an increase of the surface area and the capillary volume in the diameter range of 3.2-4.2 nm The activation energy of the the adsorption was determined to be in the range of 6.370-10.493 kcal/mol The values of thermodynamic parameters ΔH0298 , ΔS0298 and ΔG0T of the adsorption of 2,4-D 2,4,5-T on purified and activated CNTs at 283, 293, 303 and 313 K were all negative The adsorption heat was determined in the range of 37.382-49.144 kJ/mol New Contributions: The purification and activation procedures for domestic raw CNTs to increase the initial adsorption rate and capacity of 2,4-D onto CNTs were studied and developed The heterogeneity of the surface of purified and activated CNTs was determined The adsorption of 2,4-D and 2,4,5-T was governed by the porosity of CNTs due to the activation process Thermodynamics and kinetics of the adsorption of 2,4-D and 2,4,5-T from the aqueous environment onto purified CNTs and activated CNTs were systematically and thoroughly studied The fitting of Freundlich model to the adsorption isotherm was proven and the adsorption kinetics was well described the pseudo second order equation Further research directions: Research on the recycle of carbon nanotube material for use in the treatment technology Research on synthesis and activation methods to increase the specific surface area of carbon nanotube material Research on the treatment of Agent Orange/Dioxin herbicides by heterogeneous catalysis of transition metals, carried on activated carbon nanotube material ... volume and 2,4- D adsorption capability of CNTs Studied and defined adsorption characteristics of 2,4- D and 2,4, 5 -T on purified CNTs and activated CNTs Structure of the dissertation: Introduction;... forecasted to be a promising adsorption material for the treatment of the solution of 2,4- D, 2,4, 5 -T and Dioxin The dissertation title: “study of the adsorption of 2,4- D and 2,4, 5 -T from the aqueous... purified CNTs and activated CNTs Studying and setting up the adsorption isotherm, kinetics and thermodynamic parameters and activation energy of the adsorption of 2,4, 5 -T on activated CNTs Academic