光電廢棄物資源化製備奈米吸附材料及其應用於二氧化碳捕獲之研究

國立交通大學

環境工程研究所

博士論文

光電廢棄物資源化製備奈米吸附材料及

其應用於二氧化碳捕獲之研究

Optoelectronic industrial waste derived porous

adsorbents and their application for the capture of CO

2

greenhouse gas

研 究 生：林亮毅

指導教授：白曛綾

光電廢棄物資源化製備奈米吸附材料及

其應用於二氧化碳捕獲之研究

Optoelectronic industrial waste derived porous adsorbents and

their application for the capture of CO

2

greenhouse gas

研 究 生：林亮毅 Student：Liangyi Lin

指導教授：白曛綾 Advisor：Hsunling Bai

國立交通大學

環境工程研究所

博士論文

A Dissertation

Submitted to Institute of Environmental Engineering

College of Engineering

National Chiao Tung University

in partial Fulfillment of the Requirements

for the Degree of Doctor of Philosophy

in Environmental Engineering

June 2012

Hsinchu, Taiwan, Republic of China

光電廢棄物資源化製備奈米吸附材料及

其應用於二氧化碳捕獲之研究

研  究  生：林亮毅      指導教授：白曛綾 

國立交通大學環境工程研究所 

摘  要 

隨著京都議定書正式生效，同時二氧化碳捕獲及封存技術 (Carbon dioxide Capture and Storage, 簡稱 CCS) 也於 2005 年被聯合國之 IPCC 組織評估為可行方 式之一。其中利用固體吸附劑捕獲二氧化碳被視為是現行許多捕獲技術中最具有 潛力之一；在眾多吸附劑中，中孔洞二氧化矽因其有高比表面積、可調整之孔徑 大小與高抗熱性，目前逐漸被應用至二氧化碳控制。雖然中孔洞二氧化矽藉助於 奈米科技的創新與技術上之改良，使其材料製程與發展得以快速進化發展，不過 該類型材料價格昂貴且其製造過程亦較費時費能，產物取得較沸石與活性碳困 難，目前應用中孔洞氧化矽於二氧化碳控制之研究不若傳統沸石與活性碳廣泛。 此外為解決全球溫室效應問題所需削減之二氧化碳氣體排放量相當龐大，因此若 採用 CCS 技術將必須消耗大量之地球資源。另一方面，近年來隨著半導體與光 電產業的快速發展，大量含矽之廢棄粉末亦伴隨而生。此類型之廢棄物質輕且體 積龐大，需額外花費較多之成本委託廠商進行後續廢棄物處理。相對的，如能有 效利用廢棄物，將之加以資源化製成多孔材料，則不僅具有成本效益，且可解決 廢棄物處理與處置問題。 本研究旨在利用光電粉末廢棄物做為二氧化矽之前驅物，分別透過液相水 熱法以及氣相製程製備中孔洞二氧化矽，並將其應用做為吸附劑進行二氧化碳氣 體捕獲之研究。研究中亦探討中孔洞吸附劑孔洞特性對於二氧化碳吸附效能之影 響以及利用廢棄粉末製備吸附劑之經濟效益，以評估取代商業吸附劑之可行性。 研究結果指出以廢棄粉末做為矽源，透過離子型界面活性劑十六烷基三甲基溴化 銨(CTAB)作為模板並加入適量之氫氟酸與氫氧化銨可於常溫下製備出具有高比 表面積(788 m2g-1)、大孔徑(4.5 nm)以及大孔洞體積(1.1 cm3g-1)之中孔材料 MCM-41(DU)-F。為了進一步縮減吸附劑製備所需之成本，本研究亦嘗試利用非 離子型之三崁式界面活性劑F127 做為模板；相對於陽離子型界面活性劑(CTAB)

不僅在價格上較便宜外，在環境汙染程度上也相對較低。而研究成果顯示，界面 活性劑 F127 濃度於再生製備中孔洞材料 MS 上有顯著的影響。當 F127/Si 莫耳 比例為0.001 時，MS 材料為具有籠狀(cage-like)之中孔材料；而當當 F127/Si 莫 耳比例提升於0.0023 時，MS 材料則是轉變為具有囊泡狀(cellular foam)之材料， 而其孔徑與孔體積亦大幅提升。 另一方面，本研究亦開發出利用常溫鹼萃取法可將廢棄粉末分離為矽酸鹽 水溶液與沉澱物；沉澱物之成分經鑑定後主要為高純度之氟化鈉(>90%)。由於氟 化鈉是工業上常用之化學品，因此所回收之高純度氟化鈉可提供二次再利用的機 會；而經分離所得之矽酸鹽水溶液則可作為合成二氧化矽材料之前驅物。透過此 萃取法能夠將廢棄粉末轉變為兩種具有高度經濟價值的物質。而利用矽酸鹽經由 水熱法所製造之中孔材料 MCM-41(AF)其物化特性與利用純化學品所製備出 MCM-41(NaSi)之特性相似，顯示由 TFT-LCD 粉末廢棄物所製得之矽酸鹽的確為 一具有潛力的二氧化矽來源。本研究亦延伸以粉末廢棄物所製得之矽酸鹽之製備 與應用範疇，以無機鹽類做為模板透過連續式氣相製程製備出 MSP(AS)以及 MSS(HNO3)中孔材料。在價錢成本估算部分中，使用廢棄物粉末合成之中孔材 料MSS(HNO3)可相較於使用化學品合成 SBA-15 節省約 95 %的價錢，更為使用 化學品合成MCM-41 僅 2%的價錢。因此利用廢棄矽酸鹽為前驅物以一步氣膠合 成方式預期將可大幅減少化學材料成本以及製造時間，如此本研究所製得之奈米 材料即可大量製造，並應用於捕獲CO2溫室氣體上。 在二氧化碳吸附捕獲測試結果顯示，中孔洞材料其孔徑大小以及孔體積對 於二氧化碳捕獲效能有顯著的影響。經迴歸分析，可知孔洞體積為最影響吸附效 能之關鍵因子，其次為孔洞大小，比表面積之影響則相對較小。其中中孔材料 MSS(HNO3)在二氧化碳入流濃度 10%、吸附溫度 60oC 時吸附量可達到 122 mg-CO2/g-adsorbent ， 高 於 利 用 純 化 學 品 所 合 成 之 中 孔 洞 材 料 MCM-41 與 SBA-15。因此結果顯示利用廢棄物所合成之樣品 MSS(HNO3)具有價格便宜、高 二氧化碳吸附量以及快速之製備時間。本研究所製得之奈米材料因此可大量製 造，並應用於捕獲 CO2 溫室氣體上。綜合成本考量和後端二氧化碳應用，使用 此材料在未來二氧化碳捕捉的應用上具有前瞻性。 關鍵字：廢棄物資源化、光電廢棄粉末、中孔洞矽材料、氟化鈉、二氧化碳、氣 膠輔助製程   

Optoelectronic industrial waste derived porous adsorbents and

their application for the capture of CO

2

greenhouse gas

Student：Liangyi Lin Advisor：Hsunling Bai

Inatitute of Environmental Engineering

National Chiao Tung University

Abstract

The carbon dioxide (CO2) capture and storage (CCS) technologies have received

out-breaking concerns after the Kyoto Protocol came into force in 2005. Among capturing technologies, adsorption is regarded as one of the feasible approaches which can limit the CO2 emission. Mesoporous silica materials with high surface area, large

pore size and large pore volume are considered as good candidates for CO2 capture.

However, the requirements of tedious processing time and expensive manufacture costs strongly limited their applications. Furthermore, the global emission quantity of CO2 is so huge that it may consume tremendous amount of resource materials to

capture the CO2 greenhouse gas. On the other hand, with the evolution of

semiconductor and optoelectronic industries, huge amounts of siliceous waste powder are significantly increased. Such waste powders are light-density with bulky volume and are thus difficult to be transported and disposed. Therefore, additional expenses on waste treatment and landfill disposal are needed. So if the captured sorbent can be obtained from product wastes, the cost-effectiveness of the CO2 capture technology

and the waste treatment and disposal problem will be resolved simultaneously. This study intends to reutilize the waste powder as an alternative resource for the production of mesoporous silica materials via either solution precipitation method or aerosol spray approach. The structural properties and cost-effectiveness of the recycled materials on CO2 adsorption performance was investigated as well. The

results showed that the waste powder can be directly converted in to mesoporous silica MCM-41(DU)-F with high surface area (788 m2g-1), large pore size (4.5 nm) and large pore volume (1.1 cm3g-1) with the assistance of ionic surfactant of CTAB, hydrofluoric acid as well as ammonium hydroxide. Through similar pathway, silica materials with hierarchically mesocellular structures can be facilely prepared by using single F127 surfactant. The concentrations of hydrofluoric acid and F127 were found to strongly affect the structural properties of the recycled materials.

On the other hand, a low-temperature alkali extraction was developed to effectively separate the silicate supernatant and the sediment of sodium fluoride (NaF) from the waste powder. The obtained sediment containing high purity of NaF (>90%), which provides further reuse possibility since NaF is widely applied in chemical industry. The supernatant is a valuable silicate source for synthesizing mesoporous silica material. In other words, the optoelectronic waste powder can be converted into two valuable resources, the supernatant as the silica precursor and the sediment of sodium fluoride. The mesoporous MCM-41 produced from the waste-derived silicate, namely MCM-41(AF), possessed high specific surface areas (1069 m2_{/g), narrow pore}

size distributions (3.0 nm) and large pore volumes (0.97 cm3/g), similar with those of the MCM-41(NaSi) fabricated using commercial silica precursors. This clearly suggests that the silicate supernatant from waste powder can be potential silica resource.

This study further extends the preparation of mesoporous materials using waste-derived silicate supernatant as precursors. It was demonstrated that mesoporous MSP(AS) and MSS(HNO3) materials can be synthesized by employing inorganic salts

as templating media. The cost analysis shows that the synthesized material of MSS(HNO3) is about five percent of the price of SBA-15 and two percent of the

MCM-41 made from commercial silica precursors.

Furthermore, the correlation between CO2 adsorption capacity and the pore

structure properties (pore size, pore volume and specific surface area) is studied. The result of the linear regression indicates that the CO adsorption capacity has the

strongest correlation with the total pore volume of the mesoporous materials (R2>0.9). The amine-impregnated MSS(HNO3) can achieve 122 mg/g adsorption capacity,

which is superior to that of the original MCM-41(115 mg/g) and SBA-15(117) made from commercial precursors under the same conditions. The MSS(HNO3) prepared

using optoelectronic industrial waste powder as the silica source via salt-templated aerosol spray approach exhibits several important advantages of simple and rapid synthesis, low manufacturing costs and superior CO2 adsorption performance.

Therefore, it could be considered as potential and competitive sorbents for CO2

capture from flue gas.

Keywords: resource recovery; optoelectronic industrial waste powder; mesoporous silica materials; sodium fluoride; carbon dioxide; aerosol assisted process

誌 謝

轉眼間博士班生涯到了尾聲，研究期間承蒙恩師 白曛綾教授對

於研究方向與細節的悉心指導，以及董瑞安教授、蔡春進教授、林錕

松教授、李壽南博士與張淑閔副教授於論文口試期間對於本文疏漏及

謬誤之處費心指正並提供寶貴之學術建議，在此表達萬分感謝。

回首這四年期間實驗室大夥的相處就像是一個大家庭；除了研究

上大家能夠互相討論給予協助，日常生活的關心照顧、嘻笑打鬧也讓

實驗室生活更加多采多姿。在此也感謝博士班錦德與承業學長對於研

究上的討論，以及劉凱、祈緯、瑋婷、建廷、侑霖、祐菖、詩婉、國

華、志成、崇瑋、佳錡、紘宇、智傑、世元、玫華、婉婷、太太、

Momo 與姥姥等學弟妹的生活相伴與協助，與你們一起度過的日子與

點點滴滴都會是未來美好的回憶，也在此致上最深的謝意。

最後要感謝我的家人在背後的支持關心，讓我能夠一路順利完成

學業。今日學位的完成，是一個階段的結束更是另一階段的開始，期

望自己未來能貢獻所學，取之於社會，還之於社會。

謹誌於 交大環工所

民國一百零一年六月

CONTENTS

CONTENTS ... IX LIST OF FIGURES ... XI LIST OF TABLES ... XVII

CHAPTER I INTRODUCTION ... 1

1.1 Background and Motivation ... 1

1.2 Objectives and Scope ... 2

CHAPTER II LITERATUTR REVIEW ... 3

2.1 Mesoporous silica materials ... 3

2.1.1 MCM-41 ... 3

2.2.1 SBA-15 ... 4

2.3.1 Mesostructured cellular foam (MCF) ... 4

2.2 Carbon capture and storage (CCS) ... 8

2.2.1 Amine-functionalized solid sorbent for CO2 removal ... 9

2.2.1.1 Microporous zeolite sorbents ... 9

2.2.1.2 Mesoporous silica sorbents ... 14

2.3 Effect of pore structure on CO2 adsorption performance ... 18

2.4 Siliceous solid wastes derived adsorbents for CO2 capture ... 22

CHAPTER III SILICA MATERIALS RECOVERED FROM OPTOELECTROMIC INDUSTRIAL WASTE POWDER: ITS EXTRACTION, MODIFICATION, CHARACTERIZATION AND APPLICATION ... 26

CHAPTER IV DIRECT CONVERSION OF WASTE POWDER INTO MESOPOROUS SILICA MATERIALS ... 42

CHAPTER V AEROSOL-ASSISTED SYNTHESIS OF MESOPOROUS SILICA PARTICLES VIA THE USE OF SODIUM METASILICATE PRECURSOR ... 73

CHAPTER VI AEROSOL PROCESSING OF MESOPOROUS SILICA PARTICLES

USING WASTE-DERIVED SILICATE ... 88

CHAPTER VII COMPARISON OF WASTE-BASED ADSORBENTS FOR THEIR CO2 CAPTURE PERFORMANCE ... 115

CHAPTER VIII CONCLUSIONS AND RECOMMENDATION ... 121

8.1 Conclusions ... 121

8.2 Recommendation for future work ... 122

REFERENCES ... 124

LIST OF FIGURES

Figure 2.1 A scheme for the formation mechanism of mesoporous MCM-41. ... 6 Figure 2.2 TEM image of the honeycomb structure of MCM-41 and a schematic

representation of the hexagonal shaped one-dimensional pores. ... 6 Figure 2. 3 Illustration of mesoporous SBA-15. ... 7 Figure 2. 4 Schematic cross section of the strut-like structure exhibited by MCFs. ... 7 Figure 2.5 Physisorption and chemisorption capacities of 15% CO2 on Y60(TEPA) at

30-70 ℃. ... 12 Figure 2.6 CO2 Reaction Pathways with Primary Amines. ... 12

Figure 2.7 Morphology of the amine composites prepared based on a) quartz, b) KA zeolite, c) NaA zeolite, d) NaY zeolite, e) as-synthesized MCM-41, and f) calcined MCM-41 loaded with 50 wt% TEPA. ... 13 Figure 2.8 Schematic diagram of PEI loaded in the mesoporous molecular sieve of

MCM-41. (A) MCM-41 support; (B) low PEI loading; (C) high PEI loading; (D) extremely high PEI loading. ... 17 Figure 2.9 Schematic diagram illustrating the influence of the template occluded in a

channel of MCM-41 on the distribution of TEPA . ... 17 Figure 3.1 (a) EDS spectrum and (b) FTIR spectrum of the optoelectronic waste powder. ... 29 Figure 3.2 XRD pattern of the optoelectronic waste powder. ... 29 Figure 3.3 (a) TGA analysis and (b) DTG profile of the optoelectronic waste powder.

... 31 Figure 3.4 (a) EDS spectrum and (b) XRD pattern of the sediment after silica extraction... 33 Figure 3.5 (a) TGA analysis and (b) DTG profile of the sediment. ... 33 Figure 3.6 (a) EDS spectrum of MCM-41(PWP) sample and (b) XRD patterns of the

MCM-41(PWP) and MCM-41(NaSi) samples. ... 37 Figure 3.7 N2 adsorption-desorption isotherms of the optoelectronic waste powder,

calcined MCM-41(NaSi) and MCM-41(PWP) samples. ... 38 Figure 3.8 BJH pore diameter distributions of calcined MCM-41(NaSi) and

MCM-41(PWP) samples. ... 38 Figure 3.9 TEM images of MCM-41(NaSi) and MCM-41(PWP) samples. ... 40 Figure 4. 1 CO2 adsorption via pack column system………. 44

Figure 4.2 Schematic procedures for (a) conventional alkaline fusion process and (b) direct utilization process for the recovery of mesoporous silica from the optoelectronic waste powder. ... 48 Figure 4.3 (a) Low angle XRD patterns of MCM-41(DU), MCM-41(DU)-F as well as

MCM-41(AF) and (b) wide angle XRD patterns of MCM-41(DU) and MCM-41(DU)-F samples... 50 Figure 4.4 N2 adsorption-desorption isotherms of raw optoelectronic waste powder,

MCM-41(DU), MCM-41(DU)-F and MCM-41(AF) samples. ... 52 Figure 4.5 BJH pore size distribution of MCM-41(DU), MCM-41(DU)-F and

Figure 4.6 (a) EDS spectrum of MCM-41(DU)-F; (b) TEM image of MCM-41(DU); (c) TEM image of MCM-41(DU)-F. ... 54 Figure 4.7 29Si NMR spectrum of MCM-41(DU) and MCM-41(DU)-F samples. ... 57 Figure 4.8 SEM images of (a) MCM-41(AF) and (b) MCM-41(DU)-F samples. ... 57 Figure 4.9 CO2 breakthrough curves for TEPA-functionalized adsorbents of

MCM-41(NaSi), MCM-41(AF), MCM-41(DU) and MCM-41(DU)-F samples. ... 60 Figure 4.10 (a) Relationship between the pore diameter of the mesoporous substrates

and CO2 uptake and (b) correlation of total pore volume of the

mesoporous substrates and CO2 uptake at 60 C. ... 60

Figure 4.11 BJH pore size distribution of TEPA-MCM-41(NaSi), TEPA-MCM-41(AF) and TEPA-MCM-41(DU)-F samples. ... 61 Figure 4.12 Schematic cartoon of the structure of the recycled MS (0.0023). ... 68 Figure 4.13 Representative TEM image of the recycled MS (0.0023). ... 68 Figure 4.14 N2 adsorption-desorption isotherms of mesoporous silica materials

synthesized with various nF127/nSi ratios. ... 69

Figure 4.15 Pore size distribution of mesoporous silica materials synthesized with various nF127/nSi ratios. ... 69

Figure 4.16 N2 adsorption-desorption isotherms of mesoporous silica materials

synthesized with various nHF/nSi ratios. ... 70

Figure 4.17 Pore size distribution of mesoporous silica materials synthesized with various nHF/nSi ratios. ... 70

Figure 4.18 Possible mechanism of recycling of mesostructured cellular foam from optoelectronic industrial waste powder. ... 71 Figure 4.19 TEM images of the recycled mesoporous silica samples synthesized by using different amounts of F127. a) MS(0.001), b) MS(0.01) ... 72 Figure 5.1 Experimental setup for the generation of mesoporous silica spherical

particles (MSPs) through aerosol-assisted evaporation induced self-assembly (EISA) process. ... 76 Figure 5.2 XRD patterns of calcined MSPs prepared from Na2SiO3 (denoted as

MSP(NaSi)) and TEOS (denoted as MSP(TEOS)), respectively. ... 78 Figure 5.3 Nitrogen adsorption-desorption isotherms of calcined MSP(NaSi) and

MSP(TEOS), respectively. ... 81 Figure 5.4 BJH pore diameter distribution of calcined MSP(NaSi) and MSP(TEOS),

respectively. ... 82 Figure 5.5 SEM images of calcined MSPs. (a1) MSP(TEOS) with scale bar of 10 m,

(a2) MSP(TEOS) with scale bar of 1 m (b1) MSP(NaSi)-18 with scale bar of 10 m, and (b2) MSP(NaSi)-18 with scale bar of 500 nm. ... 85 Figure 5.6 TEM images of calcined MSPs. (a) MSP(TEOS), (b) MSP(NaSi)-10, (c)

MSP(NaSi)-18 and (d) MSP(NaSi)-22. (The scale bar is 20 nm) ... 86 Figure 6.1 Experimental setup for the generation of mesoporous silica spherical

particles (MSPs) through aerosol spray process. ... 91 Figure 6.2 TGA system for the CO2 adsorption. ... 91

washed MSP (AS) samples. ... 94

Figure 6.4 (a) TGA and (b) DTG analyses of the washed MSP (AS). ... 94

Figure 6.5 (a) N2 adsorption-desorption isotherms and (b) BJH pore diameter distributions of the washed MSP and MSP (AS) samples. ... 97

Figure 6.6 Possible pathway for the formation of MSP and MSP (AS) samples through aerosol spray process. ... 99

Figure 6.7 SEM images (a) low-magnification, (b) high-magnification of the washed MSP (AS). (c) TEM image and (d) EDS spectrum of the washed MSP (AS). ... 101

Figure 6.8 Comparison on CO2 uptakes of TEPA loaded on MCM-41, SBA-15, MSP (AS) and NaY zeolite samples. ... 104

Figure 6.9 N2 adsorption-desorption isotherms (a) parent MCM-41, SBA-15, MSP (AS) and NaY zeolite and (b) TEPA loaded on MCM-41, SBA-15, MSP (AS) and NaY zeolite samples. ... 104

Figure 6.10 CO2 adsorption performance of TEPA-MSP (AS) with 10% and pure CO2 feed gas at 60℃. ... 105

Figure 6.11 XRD patterns of the as-synthesized MSS samples. ... 110

Figure 6.12 N2 adsorption-desorption isotherms of the washed MSS samples. ... 111

Figure 6.13 Pore size distribution of the washed MSS samples. ... 111

Figure 6.14 SEM images of the washed (a) MSS (H2SO4), (b) MSS (HNO3) and MSS (HCl) samples; TEM images of the washed (d) MSS (H2SO4), (e) MSS (HNO3) and (f) MSS (HCl) samples ... 112

Figure 6.15 Comparison on CO2 uptakes of TEPA loaded on MCM-41, SBA-15, MSS

(H2SO4), MSS (HNO3) and MSS (HCl) samples... 113

Figure 7.1 Correlation of the surface area of the mesoporous substrates and the CO2

uptake. ... 117 Figure 7.2 Correlation of the pore diameter of the mesoporous substrates and the CO2

uptake. ... 117 Figure 7.3 Correlation of total pore volume of the mesoporous substrates and the CO2

uptake. ... 118

LIST OF TABLES

 

Table 2. 1 Amine-functionalized mesoporous siliceous sorbents for CO2 capture ... 16

Table 3.1 Elemental analysis of optoelectronic waste powder and sediment analyzed by the SEM-EDS and ICP-MS analyses ... 35

Table 3.2 Physical properties of optoelectronic waste powder and mesoporous adsorbents ... .41

Table 4.1 Structural properties of MCM-41(NaSi) silica, waste powder and waste-derived mesoporous siliceous materials ... 62

Table 4.2 Physical parameters of the optoelectronic waste powder and the recycled mesoporous silica samples prepared at different nF127/nSi ratio. ... 73

Table 4.3 Physical parameters of the recycled mesoporous silica samples prepared at different nHF/nSi ratio. ... 73

Table 5.1 Physical properties of the mesoporous adsorbents ... 87

Table 6.1 Elemental analysis of optoelectronic waste powder and silicate supernatant analyzed by the SEM-EDS and ICP-MS analysis ... 96

Table 6.2 Structural parameters of the materials ... 96

Table 6.3 Structural parameters of mesoporous and microporous adsorbents ... 114

Table 7.1 TEPA-related adsorbents for CO2 capture in this work. ... 119

CHAPTER I INTRODUCTION

1.1 Background and Motivation

Worldwide issues including energy crisis, global warming and sustainable development are receiving great attention. The carbon dioxide (CO2) capture and

storage (CCS) technologies have attracted out-breaking concerns after the Kyoto Protocol came into force in 2005 [1]. In this regard, various capture technologies, including absorption, adsorption, cryogenics, membrane separation, etc., have been widely investigated [2–4]. Among them, the design of a full-scale adsorption process might be feasible and the development of a promising material that would efficiently adsorb CO2 with high adsorption capacity and low energy penalty for the regeneration

process will undoubtedly enhance the competitiveness of adsorptive separation in a flue gas application. However, the global emission quantity of CO2 is so huge that it

may consume tremendous amount of resource materials to capture the CO2

greenhouse gas.

Ever since the rises of flat panel industry, especially thin film transistor-liquid crystal display (TFT-LCD), demands on large scale panel display increase rapidly during recent years [5]. However, huge amounts of waste products have been created in the forms of waste solvents, sludge and solid waste, etc. Silicon (Si)-containing waste powder is a common waste product of chemical vapor deposition (CVD) process in TFT-LCD and semiconductor plants. The waste powders have problems of treatment and disposal due to their light-density and bulky volume. Besides, the small particle sizes of such waste powders would induce harmful effects on human beings if not properly treated. Currently, these waste powders have been disposed of without any profit due to the shortage of storage sites and have resulted in severe environmental issues.

Several research studies have attempted to reuse solid wastes for manufacturing mesoporous silica materials. This includes coal fly ash [6] and rice husk ash [7] which are by-products of the coal-fired power plant and agricultural activities. However, most recycling routes are energy-consumed and time-consuming, which would further increase the manufacturing costs. Unlike the coal fly ash or rice husk ash which

contains lots of complicated metal oxides [8,9], the primary components of optoelectronic waste powder might only contain Si-, N- and F- species. So far, the identification of the waste powder from CVD processes of either semiconductor or optoelectronic industry as well as the possibility of waste powder recovery for further environmental applications have not been reported yet.

1.2 Objectives and Scope

The research described in this study was motivated to develop cost-effective CO2

adsorbents from waste materials through convenient synthetic procedures. This work intends to utilize the waste powder from TFT-LCD plants as the alternative precursor of the mesoporous siliceous adsorbent for CO2 capture. So if the adsorbent can be

obtained from wastes product, the cost-effectiveness of the CO2 capture technology

and the waste treatment and disposal problem will be resolved simultaneously. The objectives of this study are listed:

1. To identify the chemical composition of the waste powder from CVD processes of optoelectronic industry as well as the possibility of waste powder recovery for further environmental applications.

2. To fabricate cost-effective mesoporous materials by employing waste powder as silica precursor via either solution precipitation method or aerosol spray process as supports of adsorbents for CO2 capture.

3. To investigate the effect of textural properties of the mesoporous sorbents on the CO2 adsorption performance.

CHAPTER II LITERATUTR REVIEW

2.1 Mesoporous silica materials

Porous materials have been intensively applied to the processes of adsorption, separation and catalysis, etc [10–14]. According to the IUPAC definition, porous materials are divided into three classes; microporous (pore size < 2nm), mesoporous (2-50nm), and macroporous (>50nm) materials [15]. In recent years, microporous zeolites have attracted strong attention due to their high specific area, well-define and crystalline pore structure for environmental protection applications[16–18]. However, zeolites have mass transfer limitations when it is being applied for the adsorption of large molecules. Thus attempts have been made to work on enlargement of the pore sizes into mesopores, which can allow large molecules to penetrate into the pore channels.

Mesoporous materials were first discovered by the Mobil Company in 1992. The characteristics of these materials include: (1) large specific surface area (~1000 m2/g) and pore volume (~1 cm3/g), (2) adjustable size and morphology, (3) uniformly distributed pore size, (4) highly ordered mesostructures and (5) controllable chemical composition and unique surface chemistry via functionalization. Typically, surfactant micelles are the structuring-directing template for the mesoporous materials. So far, many families of mesoporous materials have been developed, including M41S [19,20] , SBA [21,22], HMS [23], MCF [24–26], and KIT [27,28], etc. These materials were synthesized by using different surfactants, co-surfactants, synthesis conditions and methods.

2.1.1 MCM-41

The first-ordered mesoporous material, namely M41S family, was firstly synthesized by Mobile Corporation [29]. These materials were prepared from ionic surfactants, such as quaternary ammonium ions. A liquid crystal templating mechanism was proposed to elucidate the formation of the inorganic-organic composites, which is based on electrostatic interactions between the positively

charged surfactants and the negatively charged silicate species in solution as shown in

Figure 2.1 [13]. MCM-41 is the most widely studied M41S material. It is often used as a model to compare with other materials or to study fundamental aspects in sorption, catalysis, etc [30–32]. It consists of an amorphous silicate framework with hexagonal pores. MCM-41 has high surface areas of up to 1200 m2/g and large pore volumes. The pores are very uniform causing narrow pore size distributions [33,34]. The pores are unidirectional and arranged in a honeycomb structure over micrometer length scales as shown in Figure 2.2 [35].

2.2.1 SBA-15

In 1998, a new family of highly ordered mesoporous silica materials has been synthesized in an acid medium by the use of commercially available non-ionic triblock copolymers (EOnPOmEOn) with large polyethyleneoxide (EO)n and

polypropyleneoxide (PO)m blocks [15]. Different materials with a diversity of

periodic arrangements have been prepared and denoted as SBA materials. A wide variety of SBA materials has been reported in the literature, such as SBA-15 (2D hexagonal) and SBA-16 (cubic cage-structured) [36–39]. SBA-15 immediately attracted a lot of attention because of its desirable features and is now the most intensely studied SBA structure.

SBA-15 is a combined micro- and mesoporous material with hexagonally ordered tunable uniform mesopores (4-14 nm) [21,40]. The size of the micropores was found to depend on the synthesis conditions and can vary between 0.5 and 3 nm in size [21,41,42]. It consists of thick microporous silica pore walls (3-6 nm) responsible for the high hydrothermal stability of SBA-15 compared to other mesoporous materials with thin pore walls like MCM-41, MCM-48 and HMS [43–45].

2.3.1 Mesostructured cellular foam (MCF)

Among various mesoporous materials, a new family of mesostructured cellular foam (MCF) material, which consists of uniform spherical cells interconnected by windows with ultra-large mesocellular pores and three-dimensional porous structure

(Figure 2.4) has received increasing attentions recently. MCF material is originally made by addition of a swelling agent such as 1,3,5-trimethylbenzene (TMB) to the synthesis of SBA-15 [26,46,47]. The swelling agent enlarges of the micelle, resulting in a sponge-like foam with three-dimensional structure with large uniform spherical cells (15-50 nm), accessible via large windows (5-20 nm). Therefore, MCF is a very open structure with large uniform pore diameters and large pore volumes. It has thick pore walls resulting in a high hydrothermal stability. The addition of ammonium fluoride can selectively enlarge the windows by 50-80% [48–50].

Figure 2.1 A scheme for the formation mechanism of mesoporous MCM-41 [13].

Figure 2.2 TEM image of the honeycomb structure of MCM-41 and a schematic representation of the hexagonal shaped one-dimensional pores [35].

Figure 2. 3 Illustration of mesoporous SBA-15 [35].  

Figure 2. 4 Schematic cross section of the strut-like structure exhibited by MCFs [47].  

2.2 Carbon dioxide capture and storage (CCS)

The capture of CO2 greenhouse gas from industrial and utility plants has become a

world issue since the enforcement of Kyoto Protocol [51]. For controlling the emissions of CO2, several significant efforts have been devoted to develop efficient

and low-cost techniques, including the improvement of the efficiency of energy utilization, increasing the use of low-carbon energy sources and CO2 capture and

sequestration. Up to date, Carbon dioxide Capture and Sequestration (CCS) has been identified as the most effective and feasible technique limiting the emissions of CO2

[51].

Numerous capture processes including adsorption, absorption and membrane separation are proposed to separate and recover CO2 emitted to the atmosphere [52]. It

has been conducted the chemically absorptive removal of CO2 using ammonia (NH3)

solvent process to replace conventional monoethanolamine (MEA) process. It was found that the potential for removing CO2 via ammonia scrubbing may be very

promising. Although these liquid-phase systems are effective, there are still many drawbacks existing in such systems, such as high energy consumption, limitation of amine concentration in aqueous solutions and corrosion by sulfur oxides [53,54].

Development of a dry-based sorbent in CO2 post-combustion capture is always

desirable when liquid sorbents used in industries are still facing many constraints, such as corrosion, foaming, low removal rate with large-size equipments requirement, etc. Solid sorbents are normally easier for handling and causing fewer issues during the operation. In addition, solid sorbents are expected to be more cost-effective materials over liquid solvents because of its low energy requirement for regeneration.

In this regard, possible dry adsorbents including silica-gel [55], activated carbons [16], carbon nanotubes [56], and zeolite materials [57]have been widely investigated on CO2 adsorption in terms of adsorption capacity, regeneration ability and tolerance

2.2.1 Amine-functionalized solid sorbent for CO2 removal

2.2.1.1 Microporous zeolite sorbents

Zeolites are the most widely studied adsorbents for CO2 capture so far [58].

Zeolites are typically employed at ambient temperature and elevated pressure (above 2 bar). Siriwardane et al. [59] reported that the CO2 adsorption capacity of zeolite 13X,

zeolite 4A and activated carbon was about 160, 135 and 110 mg-CO2/g-adsorbent,

respectively, under 25℃ and 1 atm CO2 partial pressure. However, their adsorption

capacity has been confirmed to be significantly influenced by the increasing temperature and the presence of moisture in the gas mixture. Moreover, since all the gases are physically adsorbed on these adsorbents, the separation of various gases is difficult to be achieved. For practical applications, many separation processes were carried out under relatively higher temperatures, i.e., 50℃ to 80℃, which is a typical temperature of the exit of flue gas. Therefore, adsorbents with high adsorption capacity and high CO2 selectivity, which can be operated at relatively higher

temperatures, are desired.

In order to enhance the sorption properties of zeolite materials for low pressure applications such as those relevant in flue streams, there have been several reports on development of amine-functionalized zeolite materials as alternative sorbents for CO2

capture. Chatti et al. [60] performed the MEA-modified zeolite 13X with enhanced adsorption capacity of 23 mg-CO2/g-adsorbent, which is 1.4 times higher than the

bare 13X at 75℃. Unlike the adsorption process driven by physisorption at lower temperature, the chemical interaction between impregnated amine and CO2 gas may

play an important role in the CO2 adsorption performance at higher temperatures.

Similar observations were found in the report of Su et al. [57], who prepared tetraethylenepentamine  (TEPA)-modified zeolite Y60 for CO2 sorption at various

temperatures. A comparison of physisorption and chemisorption for CO2 sorption

from 30℃ to 70℃ was investigated and shown in Figure 2.5. It was found that physisorption mainly determines the CO2 sorption capacity in raw Y60 material,

whereas chemisorption of CO2 plays a predominant role instead of physisorption in

TEPA-functionalized Y60 sorbent and it increased with increasing temperatures. Furthermore, moisture in the stream was found to be beneficial in the enhancement of

CO2 sorption of TEPA-functionalized Y60 and the sorption capacity could achieve

4.27 mmol g-1 in the presence of 15% CO2 at 70℃ as well as 7% water vapor in the

stream. However, the sorption capacity would decrease with further increasing water contents due to the strongly competitive adsorption of water vapor with CO2 gas. A

possible CO2 reaction scheme with amine solid sorbent is represented in Figure 2.6. It

is showed that amine could react with CO2 molecules to form intermediates of

carbamate or bicarbonate under mild temperatures (40~100℃). Besides, this process is reversible; therefore, the CO2 can be easily desorbed by heating treatment and the

adsorbent can be regenerated.

The influence of support on CO2 capturing performance was studied by Fisher et al. [55], who demonstrated a comparative study on TEPA-modified silica, alumina and zeolite beta, respectively, for CO2 capture. The results showed that TEPA-zeolite beta

possessed superior sorption capacity of 2.08 mmol g-1 in the presence of 10% CO2 at

30℃ than those of TEPA-SiO2 and TEPA-Al2O3 sorbents. It was suggested that

zeolite beta with relatively higher specific surface area could lead to better dispersion of TEPA on support, which would result in higher sorption performance. Besides, Yue et al. [61–63] employed a series of TEPA-modified porous materials as adsorbent for CO2 removal and their morphologies are represented in Figure 2.7. Obviously, the

amine-containing quartz, KA zeolite and NaA resulted in gel-like composite. It was due to the fact that the lack of pores of quartz and small pore openings of KA zeolite (0.4 nm) and NaA zeolite (0.3 nm), the TEPA could only be loaded on the external surface. In contrary, the amine-supported NaY zeolite, as-synthesized MCM-41 and calcined MCM-41 were in the form of powder instead of gels. This was ascribed to the larger pore diameter and pore volume of the support, which allowed the amines penetrated insides the channels freely.

From the above studies, it is clear that zeolite materials provides advantages of high specific area and well-defined porous structures, which may be considered as candidates of supports for loading amines for efficient removal of CO2. There have

been studies showing that TEPA-functionalized zeolites can achieve the CO2 bench

mark of 2 mmol g-1; however, since TEPA is thermally unstable, these TEPA-containing sorbent composites would suffer from continuous decay of CO2

To develop thermally stable amine-supported sorbents, numerous reports on various types of amines such as TEPA [64], ethylenediamine (EDA) [65], polyethyleneimine (PEI) [66] and 3-aminopropyltriethoxysilane (APTES) [56], etc., have been investigated in terms of their molecular structure, adsorption capability and thermal stability on CO2 adsorption. The results showed that most amines like

EDA-supported zeolites could not contribute capacity of 2 mmol g-1; despite they are more thermally stable than those of TEPA-containing sorbents during cyclic tests. Thus, it is necessary to prepare such amine-modified sorbents with higher loadings to achieve the bench mark value for commercialization (2 mmol g-1). However, zeolite supports with small pore diameters and small pore volumes may not provide sufficient spaces to accommodate high amounts of amines as shown previously in Yue et al. [62]. Consequently, it is necessary to work on the pore expansion from micropores to larger pores for the accommodation of higher amine loadings.

Figure 2.5 Physisorption and chemisorption capacities of 15% CO2 on Y60(TEPA) at 30-70 ℃ [57].

Figure 2.7 Morphology of the amine composites prepared based on a) quartz, b) KA zeolite, c) NaA zeolite, d) NaY zeolite, e) as-synthesized MCM-41, and f) calcined MCM-41 loaded with 50 wt% TEPA [62].

2.2.1.2 Mesoporous silica sorbents

Recent advances in the development of efficient amine-functionalized adsorbents revealed that amine molecules can be introduced and stabilized insides the pore channels of the mesoporous silica materials, leading to high adsorption performance [68]. A new concept called “molecular basket” was firstly proposed by Xu and colleagues [69], who impregnated PEI onto mesoporous silica MCM-41 and it has high specific surface area (1486 cm2g-1), large pore diameter (2.8 nm) and large pore volume (1cm3g-1) which can serve as a good sorbent for CO2 adsorption. It was

demonstrated that there is a synergetic effect of MCM-41 on the PEI for the adsorption of CO2. The 75 wt% PEI-supported MCM-41 could achieve capacity of

3.02 mmol g-1 in the presence of pure CO2 at 75℃, which is much higher than

unmodified MCM-41(0.2 mmol g-1_{) and pure PEI (2.47 mmol g}-1_{) under the same}

conditions. This clearly reveals that the addition of MCM-41 as support for loading PEI can significantly enhance the sorption performance and this could be due to the fact that high surface area and uniform mesoporous channel of MCM-41 increase the dispersion of PEI.

The synergetic effect between MCM-41 and the PEI was further studied by altering the PEI loadings on MCM-41 [70] and the schematic illustrations are presented in Figure 2.8. As the bare MCM-41 was employed alone for CO2 adsorption

(Figure 2.8A), physical adsorption driven by capillary condensation was almost negligible (0.14 mmol g-1_{). As the PEI loading increases, higher sorption capacity was}

obtained (Figure 2.8B), and the highest synergetic effect adsorption gain was obtained when MCM-41 was loaded with 50wt% PEI as shown in Figure 2.8C. With PEI loading higher than 50wt%, parts of PEI were coated on the external surface of the MCM-41, which resulted in the blockage of pores and active adsorption sites (Figure 2.8D), and the sorption capacity tended to decreased.

Later on, various types of mesoporous silica materials including MCM-48 [71], SBA-15 [72], SBA-16 [73], HMS [23], KIT-6 [28], etc were utilized as supports of adsorbents in terms of CO2 adsorption. Huang and Yang [74] employed the

amine-modified MCM-48 for acid gas removal from natural gas. It was demonstrated that APTES-modified MCM-48 shows superior selective adsorption of CO2 from the

novel TEPA-impregnated silica monolith materials, which exhibited 5.80 mmol g-1 sorption capacity in the presence of pure CO2 at 75 ℃.Very recently, mesocellular

silica foam (MCF) with large pore volume (1.82cm3g-1) were found to be efficient CO2 sorbents, which can achieve 3.45 mmol g-1 sorption capacity in the presence of

15% CO2 at 75 ℃ [76]. Until now, there have been numerous works on the

development of novel amine-functionalized mesoporous silica sorbents for the capture of CO2 as summarized in Table 2.1 in terms of their starting precursors, amine types,

operational conditions as well as CO2 adsorption capacities.

Besides having good CO2 capture ability, cost is another important issue to be

considered in the development of any potential adsorbent. Specifically, the most enormous barrier for CCS application is its relatively high costs for the capture and separation of CO2 from flue gas. In terms of the costs of CCS, the capture of CO2

represents over 80% of the total costs associated with the CO2 capture, transport and

storage cycle [77]. As seen from Table 2.1, most amine-supported mesoporous materials can exhibit superior performance over the benchmark value for commercialization (2 mmol g-1). However, the utilization of expensive precursors including the preferred silica sources and organic additives would lead to the high manufacturing costs of the adsorbents. This would prevent cost-effective and large-scale production of the mesoporous materials to replace industrially manufactured zeolite materials [34,78]. The use of cheaper starting precursors in the manufacturing process would be a great contribution to industrial applications, especially for the capture of abundant CO2 greenhouse gas where massive quantities

of adsorbents are required. However, so far, information on manufacturing cost of adsorbents is rarely discussed.

Table 2.1 Amine-functionalized mesoporous siliceous sorbents for CO2 capture

Support Si source Amine

type Temp(C) Ce (mg/g) CO2 conc.(%) Ref. MCM-41 Colloid silica+CTAB PEI 75 111 >99 [66] MCM-41 Fumed silica+CTAB PEI 75 133 >99 [69] MCM-41 Fumed silica+CTAB DEA 25 55 >99 [79] PE-MCM-41 Fumed silica+CTAB+T MB DEA 25 104 >99 [79] As-syn MCM-41 Silica aerosol+CTAB TEPA 75 237 >99 [62] Al-MCM-41-100 Fumed silica+CTAB PEI 75 127 >99 [70] MCM-48 RHA+CTAB TREN 25 70 >99 [71] MCM-48 Colloid silica+ CTAB PEI 75 119 >99 [66]

KIT-6 TEOS+F127 TEPA 60 129 10 [27]

KIT-6 TEOS+F127 PEI 75 135 >99 [66]

SBA-15 TEOS+P123 EDA 22 86 >99 [65]

SBA-15 TEOS+P123 PEI 75 127 >99 [66]

SBA-15 RHA+P123 TREN 25 80 >99 [72]

SBA-15 TEOS+P123 PEI 75 105.2 15 [80]

As-syn SBA-15 TEOS+P123 TEPA 75 173 >99 [61]

SBA-16 TEOS+F127 PEI 75 129 >99 [66]

SBA-16 TMOS+F127 AEAPS 60 32 15 [73]

HMS TEOS+DDA PEI 75 128 >99 [23]

As-syn MSU-1 Sodium silicate TEPA 75 170 10 [81]

MCF TEOS+P123+ TMB PIE 105 151 15 [82] MCF TEOS+P123+ TMB PEI 75 152 15 [76] MCF Sodium silicate+P123 PEI 75 255 80 [83] MC400/10 TEOS+CTAB TEPA 75 347.6 10 [64]

Figure 2.8 Schematic diagram of PEI loaded in the mesoporous molecular sieve of MCM-41. (A) MCM-41 support; (B) low PEI loading; (C) high PEI loading; (D) extremely high PEI loading [70].

Figure 2.9 Schematic diagram illustrating the influence of the template occluded in a channel of MCM-41 on the distribution of TEPA[62].

2.3 Effect of pore structure on CO

2

adsorption performance

As described previously, mesoporous materials provides several advantages of high specific surface area (>1000 m2g-1), large pore size (2-50 nm) and large pore volume (~1 cm3g-1) in the enhancement of CO2 adsorption performance. Xu et al.

[69,70,84] clearly suggested the synergic effect between the impregnated amines and mesoporous supports. In other words, the structural properties of the supports might strongly influence the behaviors of the impregnated amines and the CO2 adsorption

performance. Fisher et al. [55] conducted the CO2 adsorption tests by employing three

different supports of alumina, silica and beta zeolite as adsorbents. They claimed that when the TEPA was loaded on the materials with high surface area, more CO2 affinity

sites were exposed to the adsorbate and thus the adsorption capacity increased. This is in line with the work reported by Xu et al. [69], who synthesized PEI-impregnated MCM-41 for the adsorptive removal of CO2.

Zelenák et al. [85] studied the effect of textural properties of the mesoporous supports including MCM-41, SBA-12 as well as SBA-15 on CO2 capture. They

claimed that the sorption capacity was in the subsequence of pore size of support and follow the order of SBA-15＞ SBA-12＞ MCM-41. As SBA-15 with larger pore size (7.1 nm) could promote more amines into the channel easier and avoid the blockage of accessible adsorption sites for CO2 gas, higher capacity can be obtained.

Their results are in agreement with the works of Son et al. [86], who employed a series of PEI-functionalized mesoporous silica of MCM-41, MCM-48, SBA-15, SBA-16, and KIT-6 with different textural properties to evaluate their CO2 sorption

performance. The CO2 capacities were found to be in the following order: MCM-41

(2.8 nm, 1D) < MCM-48 (3.1 nm, 3D) < SBA-15 (5.5 nm, 1D) ≈ SBA-16 (4.1 nm, 3D) < KIT-6 (6.5 nm, 3D).

Moreover, PEI-impregnated SBA-15 materials with different pore diameters and pore volumes were developed by Yan and his groups [80]. It was reported that 50wt% PEI-SBA-15 can perform 2.39 mmol g-1 sorption capacity in the presence of 15% CO2

at 75 ℃. They intended to address the effect of textural properties of the SBA-15 on CO2 sorption performance. It was demonstrated that the total pore volume of the raw

SBA-15 support plays an important role instead of pore diameter on CO2 sorption. In

the PEI-impregnated mesocellular silica foams largely influenced the adsorption performance. Besides, similar observations were also found in a recent report by Qi et al. [64], who prepared the PEI-impregnated mesoporous hollow spheres as CO2

adsorbents. The results revealed that larger particle size and higher interior void volume of the mesoporous hollow spheres all improved the CO2 capacity of the

sorbents.

Table 2.2 summarizes the studies on the CO2 adsorbents in terms of their

adsorption capacities, test conditions and structural properties. Based on the above observations, it could conclude that the adsorption performance could be affected by the pore structure of the substrates; however, most research works were carried out by using different porous supports and lack of systematical and quantitative information on the correlations between the adsorption performance and the structural properties of the supports. Consequently, we intend to address the relationship between the adsorption performance and the structural properties of the supports by using the same mesoporous substrate with different surface area, pore size as well as pore volume as supports. Subsequently, comparison on various substrates on CO2 adsorption was also

Table 2.2 Studies on the effect of pore structure on CO2 capture. Materials a_S BET(m2/g) b_d BJH (nm) c_V p(cm3/g)

Condition Capacity Note:

(key factors) Reference

Al2O3, SiO2, Beta zeolite a₆₈₀ b_- c_- 10%,CO2 30 ℃ 91

High surface area:

Beta zeolite with high surface area can lead to better dispersion of amines, which can

enhance the performance.

[55] MCM-41 (Amine: 50wt. %PEI) a₁₄₈₀ b_2.75 c_1.0 >99%,CO2 75 ℃ 133

High surface area and uniform pore structure:

MCM-41 with high surface area can lead to better dispersion of amines, which can

enhance the performance.

[69] KIT-6 (Amine: 50wt. %PEI) a₈₉₅ b_6.0 c_1.22 >99%,CO2 75 ℃ 135

Large pore size:

KIT-6>SBA-16=SBA-15>

MCM-48>MCM-41. Pore size is the main parameter determining the performance.

MCF(Amine: 50wt. %PEI) a₅₃₂ b_11.3 c_1.82 15 %,CO2 75 ℃ 177.8

Large window size: & large pore volume:

Both Window size and pore volume resulted

in increasing CO2 adsorption [76] SBA-15 (Amine: 50wt. %PEI) a_802.9 b_8.6 c_1.14 15 %,CO2 75 ℃ 105.2

Large pore volume:

Pore volume plays an predominant role

instead of pore size on CO2 adsorption

[80] MC400/10 (Amine: 83wt. %TEPA) a₇₂₅ b_3.1 c_0.73 10 %,CO2 75 ℃ 245 mg/g

Mesoporous hollow particle with larger

particle, larger pore volume as well as thinner wall thickness prefer to have

higher adsorption performance.

2.4 Siliceous solid wastes derived adsorbents for CO

2

capture

Current concerns of increasing generation of various waste residues have initiated extensive efforts toward the more efficient disposal of these waste materials. In particular, the development of processes for converting waste by-products into high value-added materials is highly desirable for global resource conservation and sustainable environment. Previously, intensive efforts have been made to promote the recycling of waste materials such as municipal sewage sludge [90], coal fly ash [91] and blast-furnace slag [92] into heterogeneous catalyst, adsorbent and catalyst support, and these all target a recycling-beneficial use of the residues to set-off part of the resources depletion problem. This would be expected to be economically beneficial in industrial manufacture of huge amount of adsorbents, especially for the environmental protection applications which require massive quantity of adsorbents or catalysts.

There have been numerous reports on the reutilization of siliceous wastes for the production of mesoporous silica materials as supports of adsorbents for CO2 removal

as listed in Table 2.3. Siliceous solid wastes including coal fly ash, rice husk ash and bottom ash, which are rich of silica and alumina, are confirmed to be potential resources for the production of silica-based sorbents [9,71,78]. However, the procedures in such recycling routes usually require complex preparation steps including high energy-consumed fusion or hydrothermal treatment for the pre-extraction of silica. Besides, the need of expensive organic additives such as CTAB and TMB would also significantly lead to the high manufacturing costs. Considering the mass production of adsorbents for potential commercialization, the synthesis of mesoporous adsorbents by avoiding costly surfactants seems highly desirable.

On the other hand, the development of flat panel display, especially thin film transistor-liquid crystal display (TFT-LCD), has been continuously increased and it is second only to the semiconductor industry over past decade [5]. According to statistics from Industrial Economics & Knowledge Center (IEK) of Taiwan’s Industrial Technology Research Institute (ITRI), , the production value of the global display panel industry reached $104.90 billion in 2007; the production value of large-area TFT-LCD constituted the majority of global LCD industry production value, making up almost 70% of total LCD production value. The production value of Taiwanese

large-area TFT-LCD was $32.73 billion, making Taiwan the world leader in TFT-LCD production and representing the first time that a single industry in Taiwan had exceeded 1 trillion in production value [93].

However, large quantity of sub-micrometer waste powders produced during the chemical vapor deposition (CVD) process has increased annually due to the growing markets. In a typical CVD process, gaseous reagents of silane (SiH4) and ammonia

(NH3) are introduced and silicon derived compounds such as silica and/or silicon

nitride (SiNX) thin films are formed on the substrates. The cleaning agent of nitrogen

trifluoride (NF3) is subsequently introduced to clean up the accumulative particles

formed on the wall of the reaction chamber. The waste powders are then collected by baghouses located in the exhaust of the CVD process. Such waste powders are light-density with bulky volume and are thus difficult to be transported and disposed of. Therefore, additional expenses on waste treatment and landfill disposal are needed. To date, these waste powders have been disposed of without any profit due to the shortage of storage sites and could result in environmental concerned issues, such as leaching of toxic compounds into the groundwater and harmful effects on human beings through direct inhalation. It is, therefore, essential to develop new technologies exploiting the nature of these waste powders so that they can be reutilized instead of being wasted.

If the adsorbent can be produced from the TFT-LCD waste powder, it could be expected that the cost-effectiveness of the CO2 capture technology and the waste

treatment and disposal problem will be resolved simultaneously. Therefore, we aim to develop convenient synthetic procedures for recycling of TFT-LCD waste powder into valuable mesoporous silica materials as supports of adsorbents for the capture of CO2

Table 2.3 Waste-based CO2 adsorbent for low temperatures. Raw

material Sample Manufacture process

a_S BET (m2/g) b_d BJH (nm) c_V p (cm3/g) Conditions Amine type CO2 capacity (mg/g) Reference Coal fly ash MCM-41

Alkali fusion extraction & hydrothermal

treatment (100 oC, 48h) Additives: CTAB a₆₁₀ b_6.29 c_1.03 10%,CO2 75 ℃ PEI 112 [94] SBA-15

Alkali fusion extraction (550oC, 1 h) &

hydrothermal treatment (100 oC, 72h) Additives: CTAB a₄₀₇ b_7.2 c_0.7 >99%,CO2 75 ℃ PEI 110 [95] Bottom fly ash MS

Alkali fusion extraction (600oC, 1 h) &

hydrothermal treatment (100 oC, 72h) Additives: P123 a₈₄₇ b_10.7 c_1.08 >99%,CO2 75 ℃ PEI 118 [96] MS-b

Alkali fusion extraction (600oC, 1 h) &

hydrothermal treatment (100 oC, 72h) Additives: 1-butanol a₇₈₅ b_24.4 c_1.33 >99%,CO2 75 ℃ PEI 153 [96] MS-m

Alkali fusion extraction (600oC, 1 h) &

hydrothermal treatment (100 oC, 72h) Additives: TMB a₉₈₆ b_37.4 c_2.40 >99%,CO2 75 ℃ PEI 178 [96]

SBA-15

Alkali fusion extraction (550oC, 1 h) &

hydrothermal treatment (100 oC, 72h) Additives: P123 a₆₄₅ b_10.1 c_1.47 >99%,CO2 75 ℃ PEI 169 [97] Rice husk ash MCM-41

Alkali extraction (70oC, 24 h) &

hydrothermal treatment (100 oC, 72h) Additives: CTAB a₁₁₀₁ b_0.96 c_3.54 >99%,CO2 25 ℃ TEPA 40 [98] MCM-41

Alkali extraction (70oC, 24 h) &

hydrothermal treatment (100 oC, 72h) Additives: CTAB a₁₀₉₉ b_0.96 c_3.51 15%,CO2 75 ℃ 3-CPA 57 [71] MCM-48

Alkali extraction (70oC, 24 h) &

hydrothermal treatment (100 oC, 48h) Additives: CTAB a₁₁₂₄ b_3.89 c_0.98 >99%,CO2 25 ℃ TEPA 30 [98] MCM-48

Alkali extraction (77oC) &

hydrothermal treatment (100 oC,48h)

Additives: CTAB, PLE

a₁₀₂₄ b_2.58 c_4.02 1%,CO2 25 ℃ APTS 28 [78] SBA-15

Alkali extraction (70oC, 24 h) &

hydrothermal treatment (40 oC, 24h) Additives: P123 a₇₁₂ b_5.82 c_0.68 >99%,CO2 25 ℃ TEPA 30 [98]

CHAPTER III SILICA MATERIALS RECOVERED FROM

OPTOELECTRONIC INDUSTRIAL WASTE POWDER: ITS

EXTRACTION, MODIFICATION, CHARACTERIZATION AND

APPLICATION

3.1 Motivation

Unlike the coal fly ash or rice husk ash which contains lots of complicated metal oxides [8,9], the primary components of optoelectronic waste powder might only contain Si-, N- and F- species. Thus it would be more facile to be utilized as the silica source since high purity of silica could be obtained. In the present investigation, attempts have been made to evaluate the chemical composition of the optoelectronic waste powder as well as to recover the supernatant and the sediment into valuable resources. The supernatant is used as the silica source for the synthesis of MCM-41.

3.2 Experimental

3.2.1 Alkali extraction of silicate supernatant from optoelectronic waste powder The extraction of silica was carried out through alkali fusion treatment [99]. In a typical process, the waste powder and NaOH powder were thoroughly mixed at a weight ratio of 1:1.2 and fused at 550 oC for 1 h. The received fused product was then mixed with DI water at a weight ratio of 1:5 with continuous stirring for 24 h. The resulting mixture was then centrifuged to separate the sediment for further characterization of its components. And the supernatant was also analyzed by ICP-MS (SCIEX ELAN 5000- Inductively Coupled Plasma-Mass Spectrometer) and utilized for the synthesis of MCM-41.

3.2.2 Hydrothermal synthesis of MCM-41(PWP) using waste silicate supernatant Mesoporous silica materials of MCM-41 were synthesized by hydrothermal treatment method using either waste derived silica or pure silica source as the precursor solutions. Cetyltrimethylammonium bromide (CTAB) was employed as the

structure-directing template in the synthesis. For the optoelectronic waste derived MCM-41, the molar composition of the gel mixture was 1 SiO2 : 0.2 CTAB : 0.89

H2SO4 : 120 H2O, where the SiO2 precursor source was obtained from the supernatant

of optoelectronic waste powder extraction process. In a typical synthesis procedure, 50 ml of supernatant was firstly stirred vigorously for 30 min. Then, the pH of the solution was brought down to 10.5 using 4N H2SO4 followed by further stirring to

form a gel. After that 4.6 g of CTAB (dissolved in 16 ml of DI water) was added drop by drop into the above mixture and the combined mixture was stirred for three additional hours. The resulting gel mixture was transferred into a Teflon coated autoclave and kept in an oven at 145oC for 36 h. After cooling to room temperature, the resultant solid was recovered by filtration, washed with DI water and dried in an oven at 110oC for 6 h. Finally, the organic template was removed via a muffle furnace in air at 550oC for 6 h. The MCM-41 material synthesized from optoelectronic waste powder (PWP) was named as MCM-41(PWP).

3.2.3 Analysis and characterization

Powder X-ray diffraction patterns of calcined mesoporous adsorbents were recorded by X-ray diffractometer (Bruker D8 SSS) equipped with nickel-filtered CuK ( = 1.5405 Å) radiation. The diffractograms of the mesoporous samples were recorded in the 2 range of 1-50o _{with a scanning speed of 1 degree per minute. The}

specific surface area, pore volume and average pore diameter (BJH method) of the samples were measured by N2 adsorption-desorption isotherms at -196 oC using a

surface area analyzer (Micromeritics, ASAP 2000). All the samples were degassed for 6 hours at 350 oC under vacuum (10-6 mbar) prior to the adsorption experiments. The

29_{Si MAS-NMR spectra of the as-synthesized mesoporous silica were recorded at}

room temperature using a BRUKER DSX 400 WB NMR spectrometer. The elemental analysis and morphologies of the materials was observed by energy-dispersive X-ray spectroscopy in a scanning electron microscope (SEM/EDS, HITACHI-S4700). And transmission electron microscopy (TEM) images of the samples were observed with a JEOL JEM 1210 TEM instrument operated at 120 keV, with the samples (5-10 mg) ultrasonicated in ethanol and then dispersed on carbon film supported by copper grids (200 mesh).

3.3 Results and discussion

3.3.1 Characterization of optoelectronic waste powder

The results of EDS and FTIR analyses for the optoelectronic waste powder are shown in Figure 3.1, and the weight percentages of each element in the optoelectronic waste powder are listed in Table 3.1. As can be seen from Figure 3.1(a), the four observed elements of Si, O, N and F in the optoelectronic waste powder were expected since the solid waste was derived from the CVD process in which the three gaseous reactants were SiH4, NF3 and NH3. Furthermore, the FTIR spectrum of

optoelectronic waste powder shown in Figure 3.1(b) exhibited significant absorption bands at 3336, 3139, 1400, 1080, 960, 742 and 482 cm-1_{. The bands at 3336 and 3139}

cm-1 _{are related to N-H stretching, while 742 and 482 cm}-1 _{can be assigned to}

SiF62-[100]. Besides, the bands at 1400, 1080 and 960 cm-1 can be assigned to the

NH4+ [100], Si-O-Si and Si-OH stretching vibrations, respectively [101]. Considering

the above results, the primary components in optoelectronic waste powder could be the admixture of (NH4)2SiF6 and SiO2.

The XRD analysis was then carried out to further identify the crystalline components in the optoelectronic waste powder with the result depicted in Figure 3.2. Strong (NH4)2SiF6 diffraction peaks were identified [102]. In contrast, the diffraction

peak of silica was not observed from the XRD pattern, suggesting that SiO2 was only

present in little amount and it might be trapped in the (NH4)2SiF6 lattice. However

the precise weight percentages of (NH4)2SiF6 and SiO2 could not be obtained from Table 3.1 since the EDS data only provided a rough estimate of the elemental content, while the ICP-MS could only detect the Si content in the optoelectronic waste powder.

 

Figure 3.1 (a) EDS spectrum and (b) FTIR spectrum of the optoelectronic waste powder. 2 theta (degree) 10 20 30 40 50 60 70 80 Inte nsi ty (counts) 0 200 400 600 800 1000 1200 1400 1600 1800 (NH4)2SiF6

Figure 3.2 XRD pattern of the optoelectronic waste powder.

Wavenumber (cm-1) 500 1000 1500 2000 2500 3000 3500 4000 A b so rba n ce ( a .u ) N-H NH4+ Si-O-Si SiF6 2-3139 1400 1080 742 482 SiF6 2-N-H 3336 960 Si-OH (a) (b)

The weight percentages of (NH4)2SiF6 and SiO2 presented in the optoelectronic

waste powder were determined using TGA weight loss and differential thermo-gravimetric (DTG) analyses. It can be seen from Figure 3.3(a) that the optoelectronic waste powder sample showed an initial weight loss at <150 oC, which could be ascribed to the evaporation of the physically adsorbed water on the surface of the materials. As the temperature exceeded 150 oC, there was a significant weight loss for the optoelectronic waste powder at around 237 oC as clearly observed from

Figure 3.3(b). Mel'nichenko et al. [103] investigated the mechanism of the solid-phase reaction between (NH4)2SiF6 and SiO2, and proposed that (NH4)2SiF6 could be

thermally decomposed between 220-300 oC in the presence of SiO2. Generally, the

following chemical reactions are expected to take place when (NH4)2SiF6 was heated

up to 900 oC [102]: (NH4)2SiF6  C o 220 _SiF 4 (g) + 2NH4F (s) (1) NH4F(s) 850 oC NH4F (g) (2)

Considering the above reactions, (NH4)2SiF6 would be completely decomposed

during the thermal treatment up to 900 oC. However, it is noted that there was still 15 wt. % of residues remained after heating up the optoelectronic waste powder to 900 oC. This is probably due to the presence of SiO2 since it is thermally stable up to

900 oC. The total Si weight fraction (from (NH4)2SiF6 and SiO2) of 20.4% calculated

by the TGA/DTG result was very close to the Si mass fraction of 22.4% detected by the ICP-MS result shown in Table 3.1. Therefore, the primary components in optoelectronic waste powder were determined to be around 85% of (NH4)2SiF6 and

Temperature (

o

C)

0 100 200 300 400 500 600 700 800 900

We

ight rem

ained (%)

0 10 20 30 40 50 60 70 80 90 100 TGA Temperature (oC) 0 100 200 300 400 500 600 700 800 Mas s C h an ge Rate (% /m in ) -25 -20 -15 -10 -5 0 5 DTG (a) (b)

3.3.2 Characterization of sediment and supernatant liquid after extraction

Figure 3.4 shows the EDS spectrum of the sediment and its corresponding XRD pattern. It can be found from Figure 3.4(a) thatNa and F were the primary elements in the sediment, with negligible Si and O species, which was also confirmed by the result of EDS analysis shown in Table 3.1. The residual SiO2 could be due to

insufficient NaOH amounts applied in the extraction process. The major component of NaF presented in the sediment was also demonstrated by Figure 3.4(b). The five sharp diffraction peaks located at 34, 39, 56, 67 and 70.5o were in agreement with the NaF standard XRD peaks [104,105]. Besides, there was a broad peak centering at 22 o as observed from Figure 3.4(b), which can be ascribed to the amorphous SiO2 [106]. This

result is consistent with the EDS spectrum which showed that Si and O species were also presented in minor amounts.

The thermal behavior of the sediment was subsequently investigated using TGA and DTG analyses. It can be seen from Figure 3.5(a) that there was only around 5% weight loss for the sediment during the thermal treatment up to 900 oC. The DTG curve showed three distinguished peaks from room temperature to 900 oC (Figure 3.5(b)). The first step (<150 oC) is due to the loss of physically adsorbed water on the surface of the sediment and the second step (150-600oC) is attributed to the loss of chemically adsorbed water bonded to Si-OH through hydrogen bond [107] since the sediment contained only a little amount of SiO2. From 600 to 900 oC, the weight loss

is expected to be associated with the further condensation of the Si-OH groups from the amorphous SiO2 [106].

As a result, one can conclude that F was effectively captured by NaOH and the sodium fluoride (NaF) sediments were formed after the extraction process. The NaF purity was quite high (>90%) in the sediment as observed from the TGA/DTG data, which provides further possibility for reuse. Sodium fluoride is one of the well-known chemical compounds which are widely utilized in industries as the source of fluoride ion in diverse applications. In other words, the optoelectronic waste powder can be converted into two valuable resources, the supernatant as the silica precursor and the sediment of sodium fluoride.

 

Figure 3.4 (a) EDS spectrum and (b) XRD pattern of the sediment after silica extraction. Temperature (oC) 0 100 200 300 400 500 600 700 800 900 Weight rem a in ed ( % ) 95 96 97 98 99 100 TGA Temperature (oC) 0 100 200 300 400 500 600 700 800 900 Ma ss C h an g e R a te (% /m in ) -0.15 -0.12 -0.09 -0.06 -0.03 DTG (a) (b)

Figure 3.5 (a) TGA analysis and (b) DTG profile of the sediment.

2 theta (degree) 10 20 30 40 50 60 70 80 Intensity ( count s) 0 100 200 300 400 500 600 NaF Amorphous silica (a) (b)