以化學沉澱結合微過濾處理光電廢水中磷酸

行政院國家科學委員會專題研究計畫 成果報告

以化學沉澱結合微過濾處理光電廢水中磷酸 研究成果報告(精簡版)

計 畫 類 別 ： 個別型

計 畫 編 號 ： NSC 98-2221-E-011-019-

執 行 期 間 ： 98 年 08 月 01 日至 99 年 07 月 31 日 執 行 單 位 ： 國立臺灣科技大學化學工程系

計 畫 主 持 人 ： 劉志成

計畫參與人員： 碩士班研究生-兼任助理人員：謝超恩 碩士班研究生-兼任助理人員：蔣竣光

報 告 附 件 ： 出席國際會議研究心得報告及發表論文

處 理 方 式 ： 本計畫可公開查詢

中 華 民 國 99 年 09 月 20 日

1

行政院國家科學委員會補助專題研究計畫期末報告 行政院國家科學委員會補助專題研究計畫期末報告 行政院國家科學委員會補助專題研究計畫期末報告 行政院國家科學委員會補助專題研究計畫期末報告

※※※※※※※※※※※※※※※※※※※※※※※※

※ ※

※ 以化學沈澱結合微過濾處理光電廢水中磷酸 ※

※

※

※ ※

※※※※※※※※※※※※※※※※※※※※※※※※

計畫類別：
個別型計畫 □整合型計畫 計畫編號：NSC-98-2221-E-011-098

執行期間： 98 年 8 月 1 日至 99 年 7 月 31 日

計畫主持人：劉志成 共同主持人：

本成果報告包括以下應繳交之附件：

□赴國外出差或研習心得報告一份

□赴大陸地區出差或研習心得報告一份

X 出席國際學術會議心得報告及發表之論文各一份

□國際合作研究計畫國外研究報告書一份

執行單位：國立台灣科技大學化學工程技術系

中 華 民 國 99 年 09 月 20 日

2

行政院國家科學委員會專題研究計畫期末報告 行政院國家科學委員會專題研究計畫期末報告 行政院國家科學委員會專題研究計畫期末報告 行政院國家科學委員會專題研究計畫期末報告

以化學沈澱結合微過濾處理光電廢水中磷酸 計畫編號：NSC-98-2221-E-011-098 執行期限：98 年 8 月 1 日至 99 年 7 月 31 日 主持人：劉志成 國立台灣科技大學化學工程技術系

摘要

本實驗主要目的在於探討利用沈澱-微過濾程序去除與回收光電業蝕刻廢水中磷酸鹽之可行性。使用鈣 鹽作為沈澱劑，控制酸鹼值與濃度莫耳比於最適當操作條件下，分析廢水中磷酸鹽之移除率與分析沉 澱物組成，探討磷酸鹽回收再利用之潛力。並結合掃流微過濾進一步分離溶液中固體與液體。探討改 變沈澱過程中之酸鹼值與濃度莫耳比、過濾壓力、掃流速度與不同薄膜種類等變因，對濾速的影響。

結果顯示，控制酸鹼值在 8.5 至 10.5 之間，過量的鈣鹽可移除 96%以上的磷酸鹽。利用 X 光繞射 分析得知，沈澱物主要以結晶的氫氧基磷灰石(hydroxyapatite, HAP)及非結晶磷酸鹽(amorphous calcium phosphate, ACP)為優勢物種。然而因為蝕刻廢水中含有少量的氟，亦會有少量的氟化鈣與氟化磷灰石 (fluorapatite, FAP)存在於沈澱物中。此實驗使用電腦軟體 PHREEQC 計算理論的熱力學平衡，將模擬出 的值與實驗結果做比較。以微過濾處理經加鈣沈澱過後之廢水，可有效地分離溶液中固體與液體。以 不同條件沈澱後進行過濾，在酸鹼值為 8.5 時，有較明顯的堵塞現象。使用不同孔徑的薄膜進行過濾，

可發現較大的孔徑（0.45μm）雖然能有較高的初始濾速，但也會造成較嚴重的堵塞。而改變操作壓力 與掃流速度發現，當操作壓力與掃流速度越大時，可得到較高的初始濾速與穩定濾速。使用疏水性薄 膜時，發現改變操作壓力與掃流速度對穩定濾速的影響並不明顯，但所造成的堵塞現象較親水性薄膜 嚴重。在所有過濾程序中，濾餅是主要的過濾阻力。使用沈澱-微過濾程序可有效去除磷酸鹽、濁度與 總懸浮固體，不僅過濾後水質可達放流水標準，所生成之沈澱物也具有很高的潛力回收再利用。

關鍵詞：沈澱、微過濾、磷酸鈣、回收、光電、含磷廢水、PHREEQC

Abstract

The removal of phosphate and fluoride from thin-film transistor liquid crystal display (TFT-LCD) wastewater by a hybrid precipitation-microfiltration (MF) process was studied. Calcium salt was used to form precipitates, followed by crossflow MF for solid-liquid separation. The results showed that excess calcium could induce effective removal of phosphate and fluoride at pH 8.5 and 10.5. The dominant solids were hydroxyapatite (Ca5(PO4)3OH, HAP), amorphous calcium phosphate (ACP), fluorapatite (Ca5(PO4)3F, FAP), and calcium fluoride (CaF2). Precipitation conditions affected the MF, and more significant fouling was found at pH 8.5 than pH 10.5. Permeate analysis showed increased removal of phosphate and fluoride in MF, and effective removal of turbidity. The initial flux and steady state flux increased with increasing filtration pressure and crossflow velocity. The main fouling resistance was cake resistance. This study demonstrated that the hybrid precipitation-MF process could effectively remove phosphate and fluoride from wastewater, and produce filtrate and precipitates for potential recovery and reuse.

Keywords: Fluoride; Microfiltration; Phosphate; Precipitation; TFT-LCD, Wastewater

1. Introduction

Discharge of phosphorus into the surface water causes eutrophication and the amount of phosphorus in domestic and industrial wastewater must be controlled. Various industries produce wastewater that contains high concentration of phosphorus and fluoride, such semiconductor, phosphoric acid processing, and fertilizer [1–4]. TFT-LCD manufacturing is one of the major industries in Taiwan, with production capacity ranked as one of the top countries in the world. The processes of TFT-LCD manufacturing include deposition, photolithography and etching. These processes require large quantity of water, and produce significant amount of wastewater. The wastewater generated from etching process typically contains high concentration of phosphate and fluoride Because of the environmental risks, it is imperative to find a cost-effective treatment process.

Fluoride and phosphate can simultaneously be removed by nanomagnetite aggregation process [5]. Cement paste has been used to remove fluoride and phosphate from semiconductor wastewater through precipitation and adsorption [3]. Very efficient removal of fluoride and phosphate by lime has been demonstrated for phosphogypsum leachate [1]. Selective precipitation of phosphate and fluoride has been shown feasible by calcium from synthetic wastewater [2], and by magnesium from semiconductor wastewater [4].

Phosphorus can be removed from wastewater by several techniques, such as biological treatment, chemical precipitation treatment, and crystallization treatment [6]. The efficiency of chemical precipitation process by the addition of lime is dependent on pH and the molar ratio of Ca:PO4 [7]. High concentration of fluoride is found among some industrial wastewater and its removal has been extensively studied using various technologies, such as adsorption [8,9], electrocoagulation [10], and precipitation [11]. Precipitates of calcium phosphate and calcium fluoride are usually very fine and coagulation-flocculation is needed for solid separation [2, 5, 11].

Phosphorus recovery is also critical since it is a non-renewable resource. Mostly, phosphorus is recovered as HAP (hydroxyapatite, Ca5(PO4)3OH) or MAP (struvite, MgNH4PO4。6H2O) that can be utilized in agriculture as fertilizer by precipitation or crystallization processes [6]. For example, phosphorus can be recovered as struvite using polymeric ligand exchanger from reverse osmosis concentrate [9]. Selective separation and recovery of phosphate by magnesium chloride to form struvite and bobbierite from semiconductor wastewater is feasible [4]. Separative recovery with lime of phosphate and fluoride from an acidic effluent has been demonstrated [12]. A hybrid flotation-microfiltration process has been developed to remove phosphate from water [13]. Lime softenin can improve crossflow microfiltration for treatment of phosphate-containing wastewater [14]. Membrane process is a very effective solid-liquid separation process that can remove colloids, microparticles, microorganisms, and macromolecules present in the suspension.

Chemical precipitation can remove phosphate and fluoride from wastewater and the following microfiltration can separate precipitates from water. The idea of integrating chemical precipitation with microfiltration (MF) is due to the potential benefits of better separation of solids and their recovery as a resource, and improved effluent quality. The objective this study was to remove phosphate and fluoride from TFT-LCD wastewater using a hybrid precipitation-MF process. Calcium chloride was used for chemical precipitation and the effects of pH and molar ratio on phosphate and fluoride removal were investigated. The effects of chemical precipitation conditions, membrane types, filtration pressure, and crossflow velocity on filtration flux and permeate quality were investigated in the MF experiments.

2. Materials and methods

The wastewater was collected from a TFT-LCD manufacturing company in Taiwan. It was sampled from the equalization tank of the inorganic wastewater. The pH was 4.72 and conductivity was 424 µS/cm. Its solid content was very low, with total suspended solid (TSS) of 0.58 mg/L and turbidity of 14.4 NTU. It contained 187.6 mg/L of phosphate (PO43-

), 19.53 mg/L of fluoride (F^-), 34.15 mg/L of sulfate (SO42-

), and 28.06 mg/L of nitrate (NO3-

) as analyzed by ion chromatography (DIONEX-1000). The regeneratant and eluent used was 1 mM sulfuric acid and 6 mM sodium carbonate (Na2CO3) and sodium hydrogen carbonate (NaHCO3), respectively. Each sample was first filtered through a 0.45 µm membrane (PVDF), and then filled into a 2 mL brown glass bottle before analysis. The other componets included 38.29 mg/L of sodium, 3.22 mg/L of calcium, and 0.68 mg/L of aluminum as measured by inductively coupled plasma – atomic emission spectrometer (Horiba Jobin Yvon, ICP-AES JY2000).

The chemical precipitation experiments were conducted in batch-mode by a jar-test apparatus. The measured volume of stock solution of calcium chloride (0.51 M CaCl2‧2H2O) was added to the beaker that contained 1L of wastewater in each beaker. Molar ratio of calcium to phosphate and fluoride, ([Ca⁺²]: [PO43-

]:

[F^-]) was controlled at 1.5:1:0.7 and 2.5:1:0.7, respectively. The pH was controlled at 8.5 and 10.5, respectively. After adding calcium chloride, the wastewater was rapidly mixed at 200 rpm for 3 minutes and the pH was adjusted to desired value using 1N NaOH or 1N HCl. The speed was then adjusted to 50 rpm for 30 minutes, and the pH of the mixture was determined. The wastewater was subject to 30-min sedimentation or MF process afterwards for subsequent separation of solid from water.

. To determine particle size distribution of precipitates, 500 mL of wastewater after subject to chemical precipitation was measured by a small angle light scattering instrument (Malvern Mastersizer 2000). Crystal structure of precipitates was analyzed by an X-ray diffractometer (Rigaku, D/Max-RC) after precipitates were dried, ground into powder form, and made as a pellet. The second method was by wet chemical analysis, in which a portion of precipitate was separated from 200 mL of solution by filtration. The solid was dried in a dessicator with silica gel under room temperature for 48 hour, weighed and then dissolved in 5 mL of 2N HNO3, and transferred to a 250-mL volumetric flask. The volumetric flask was filled up to 250 mL with distilled water. The solution was mixed with a stirring bar for 1 hour, and analyzed for Ca, PO4 and F concentration with aforementioned analytical procedures.

The saturation indices and the distribution of aqueous species under equilibrium conditions as affected by calcium concentration and pH were simulated by the PHREEQC Interactive 2.8 software. The possible species considered in modeling included hydroxyapatite (Ca5(PO4)3OH, HAP), amorphous calcium phosphate (ACP), octacalcium phosphate (Ca4H(PO4)3‧2.5H2O, OCP), monetite (CaHPO4, DCPA), brushite (CaHPO4‧2H2O, DCPD), fluorapatite (FAP), and calcium fluoride (CaF2).

After precipitation, the suspension was subject to MF for 100 min. A schematic diagram of the MF system used in the study is shown in Figure 1. The wastewater was stirred in a slurry tank to ensure that there is no solids settled in the tank. The temperature of the suspension was maintained at 25⁰C. It was pumped by a peristaltic pump, and a rotameter was used to monitor crossflow velocity. Mixed cellulose ester hydrophilic (MCE) membranes with pore size of 0.22 and 0.45 µm was used, respectively. The membrane was immersed in water and alcohol, respectively, to ensure the maximum wettability. The filtration procedure was repeated

with filtration pressure at 0.5 and 1.0 bar, and crossflow velocities at 0.48 m/s for laminar flow (Reynolds number of 1,618) and 0.96 m/s for turbulent flow (Reynolds number of 3,234), respectively. The basic model used for determining filtration resistance occurring during permeates transport through porous membrane is the Darcy’s law [15]:

Rt

P dt

dV J A

µ

= ∆

≡ 1

(Eq. 1)

where J is the permeate flux at temperature T (m³/m²s), A the area of membrane (m²), V the volume of filtration (m³), t the time (s), △P filtration pressure (Pa), µ viscosity of water at temperature T (Pa‧s), and Rt total resistance or hydraulic resistance (m^-1). Total resistance consists of some resistances [14]:

Rt = Rm + Rf + Rc (Eq. 2) where Rm : membrane resistance (m^-1),

Rf : fouling resistance (m^-1), Rc : cake resistance (m^-1).

First, the pure water flow through the membrane filter module was run for 15 minutes under a given filtration pressure and crossflow velocity, and the membrane resistance (Rm) could be calculated from the pure water flux. The pure water flow was switched to the wastewater sample with the identical operation condition for 100 minutes. Thus, the total resistance (Rt) could be calculated from the measured flux. As soon as the filtration of wastewater was terminated, it was switched instantaneously to pure water and the cake layer would be swept away. The filtration of pure water was continued for 15 minutes, and the fouled membrane resistance (Rm + Rf) can be obtained. The value of resistance of fouling (Rf) could be calculated.

3. Results and discussion

3.1 Chemical precipitation

Calcium chloride was chosen because of its higher solubility and less amount of sludge than lime.

Experimental results of residual concentration of phosphate and fluoride as affected by molar ratio and pH are shown in Table 1. The chemical precipitation of phosphate and fluoride was significantly affected by molar ratio of calcium to phosphate and fluoride (Ca:PO4:F), and pH. It was found that removal efficiency of PO4

increased with increasing molar ratio. When Ca:PO4:F increased from 1.5:1:0.7 to 2.5:1:0.7, the residual PO4

concentration decreased from 30.87 mg/L (84 % removal) to 8.11 mg/L (96 % removal) at pH 8.5. The residual fluoride concentration was 13.26 mg/L (32 % removal) and 13.13 mg/L (33 % removal), respectively.

Removal efficiency for both PO4 and F increased further at pH 10.5. Results showed that excess calcium was needed to increase supersaturation in solution as a driving force for precipitation for PO4 and F [2,7,11].

Modeled residual concentrations of PO4 and F at pH 8.5 were lower than experimental results. The model predicts that calcium reacts with phosphate to form hydroxyapatite (HAP) and fluorapatite (FAP) at molar ratio of 1.5:1:0.7, and HAP at molar ratio of 2.5:1:0.7. In fact, other precipitates of calcium phosphate, such as amorphous calcium phosphate (ACP) may also be produced [16]. That was probably slight difference was

found between the modeled and experimental results.

When using calcium salt to treat wastewater that contains phosphate and fluoride, the interactions of phosphate and fluoride have been well documented [1-3, 17]. The calcium salt could react with fluoride and phosphate to form precipitates of HAP, FAP, and CaF2. From both simulated and experimental results, a certain fraction of fluoride was removed. The results were in agreement with previous results that higher than stoichiometric amount of calcium is required to ensure effective removal of fluoride, and calcium tends to react with phosphate preferentially than fluoride under alkaline pH [17]. Figure 2 shows XRD patterns of precipitates formed at different pH and different molar ratio. It was found that the dominant phase of precipitate was dependent on the pH. For example, at pH 8.5 and molar ratio of 1.5:1:0.7, HAP and FAP were the two dominant phases found; while the formation of CaF2 was confirmed as the molar ratio increased to 2.5:1:0.7 at pH 8.5. Broad peaks typical of amorphous materials were found among precipitates formed at pH 10.5 regardless of molar ratio. The probable precipitate formed was amorphous calcium phosphate (ACP).

Data from wet chemical analysis of precipitates are depicted in Figure 3. The normalized concentration ratio of Ca, PO4, and F in the precipitates as affected by initial molar ratio and pH is compared with PO4

concentration set as 1. It was found that the ratio of Ca to PO4 in precipitate was 1.63 at initial molar ratio of 1.5:1:0.7 and was independent of pH. It was approximately equal to the stoichiometric ratio of HAP (Ca/P:

1.67). It implied that HAP was the dominant phase with traces of other precipitates, such as CaF2 and OCP (Ca/P: 1.33). When at molar ratio of 2.5:1:0.7 the ratio of Ca to PO4 in precipitate was 1.88 and 1.98 at pH 8.5 and 10.5, respectively. This ratio was higher than HAP, and the possible explanation was that Ca(OH)2 was formed and contributed to Ca fraction in the precipitates. In addition, the ratio in precipitate at pH 10.5 was higher than that at pH 8.5, probably because ACP was formed. Since HAP is known to have slow reaction rate and a number of species, such as amorphous calcium phosphate (ACP) and octacalcium phosphate (OCP) act as precursors of HAP [16]. This again confirmed that three precipitates, HAP, FAP, and CaF2, were formed.

Figure 4 shows the distribution of particle size (number-based) under different conditions. The particles were mostly very fine when formed at pH 8.5 regardless of molar ratio. The median particle size was 0.72 and 0.71 μm, respectively. Larger particles were formed when at pH 10.5, and median particle size was 8.74 and 2.46 μm, respectively at two different molar ratio. The volume-based median particle size was > 21.2 µm under different precipitation conditions.

3.2 Microfiltration of precipitated wastewater

MF experiments were first conducted under laminar flow region (us=0.48 m/s) and filtration pressure of 0.5 bar. Table 2 shows the filtration flux under various precipitation conditions. When at molar ratio (Ca:PO4:F) of 1.5:1:0.7, the initial flux and steady state flux was 1,710 and 254 L/m²h at pH 8.5; and 2,610 and 225 L/m²h at pH 10.5, respectively. At molar ratio of 2.5:1:0.7, the initial flux and steady state flux was 2,250 and 330 L/m²h at pH 8.5; 2,790 and 285 L/m²h at pH 10.5, respectively. The flux loss (%) was calculated by the difference in pure water flux of original membrane and of used membrane after washing. It was found that the flux loss at pH 8.5 was higher than those at pH 10.5. It was probably caused by membrane fouling of fine (< 1 μm) calcium phosphate or calcium fluoride precipitates at pH 8.5, as smaller particles tend to produce higher fouling than larger particles in crossflow filtration [18]. The specific resistances, including Rm, Rf, and Rc, derived from MF of wastewater are depicted in Figure 5. It shows that cake

resistance was the main one causing flux decline regardless of the chemical precipitation conditions. The total resistance at pH 10.5 was higher than at pH 8.5, probably because of higher amount of precipitates formed at pH 10.5.

The filtrate quality is compared with the supernatant after chemical precipitation and sedimentation for 30 min. Table 3 shows turbidity and conductivity in effluent of precipitation-sedimentation and of hybrid precipitation-MF process. Effluent quality of hybrid precipitation-MF was all better than that of precipitation-sedimentation. Excellent removal of turbidity was found after precipitation-MF, and the filtrate contained very low turbidity (< 0.33 μm). The MF process also had limited removal efficiency (24% to 31%) of conductivity. Table 4 shows the removal efficiency of phosphate and fluoride after precipitation-sedimentation and after precipitation-MF. Again, higher removal efficiency of phosphate and fluoride was found for precipitation-MF. It could be due to the longer reaction time in MF, or due to the concentration polarization near membrane surfaces that is beneficial to precipitation.

3.3 Effects of filtration pressure and crossflow velocity

Wastewater precipitated at pH 10.5 and molar ratio of 2.5:1:0.7 was used to study how operation conditions affected filtration. When under laminar region with crossflow velocity of 0.48 m/s, the initial flux was 2,790 and 4,050 L/m²h, respectively, at filtration pressure of 0.5 and 1.0 bar; while the steady state flux was 285 and 315 L/m²h, respectively (Table 5). The initial and steady state fluxes increased with filtration pressure because higher filtration pressure induces higher drag force toward membrane, higher initial flux, and more rapid cake deposition [15]. When under turbulent flow condition with crossflow velocity of 0.98 m/s, the initial flux was 2,790 and 4,500 L/m²h, respectively, at filtration pressure of 0.5 and 1.0 bar; while the steady state flux was 345 and 390 L/m²h, respectively. When compared with laminar flow condition, the increased crossflow velocity resulted in higher steady state flux. It was because high crossflow velocity sweeps away more effectively the cake deposited on the membrane and maintains a relatively higher flux [19].

Figure 6 shows that cake resistance was the most significant resistance under all conditions. The total resistance decreased with increasing crossflow velocity, whereas it increased with increasing pressure. The increased filtration pressure led to increase in cake deposition and higher cake compressibility [15]. However, the increased crossflow velocity induced higher shear stress which was effective in reducing the cake deposition. Table 6 shows that values of the specific cake resistance (α) and cake mass per unit filtration area (wc) increased with increasing filtration pressure and decreased with increasing crossflow velocity. It was because of cake compression and higher amount of particles retained with the membrane [20]. The higher value of specific cake resistance and cake mass per unit filtration area under laminar flow condition than that under turbulent flow condition indicated the higher amount of and more compact cake was deposited on the membrane surface under laminar flow condition. The decrease of specific cake resistance was related to the fact that pressure drop across the filter cake increases as the crossflow velocity increases [19]. The decrease of cake mass might be caused by high crossflow velocity that reduced the particles attached on the membrane surface.

3.4 Effects of different membrane

Filtration experiments were repeated using a MCE membrane with pore size of 0.45 µm. Results under

various operation conditions are shown in Table 7. The effects of filtration pressure and crossflow velocity on the membrane showed similar trends with those of the membrane with pore size of 0.22 µm. In general, the higher flux was found when using membrane with larger pore size. However, higher flux loss was found as well. The steady state flux increased with increasing filtration pressure and crossflow velocity. In addition, it was found that the flux loss decreased when increasing crossflow velocity or filtration pressure. The cake resistance was still the dominant one that caused flux decline. The increased crossflow velocity could reduce the total resistance slightly. The increased filtration pressure not only resulted in higher steady state flux but also higher cake resistance.

3.5 Overall assessment

A large amount of wastewater containing phosphate and fluoride is produced in TFT-LCD manufacturing.

From experimental results shown above, both phosphate and fluoride could be effectively removed using chemical precipitation as long as molar ratio and pH were properly controlled. The introduction of MF had several advantages when compared with conventional processes. MF has smaller foot print, and higher removal efficiency of phosphate and fluoride was found when compared with conventional processes. More effective solid-water separation was obtained without adding coagulant, and the filtrate could potentially be reclaimed and reused since it did not contain organic matters and its turbidity was low. The precipitates could potentially be reused as raw material in fertilizer industry.

4. Conclusions

A hybrid precipitation-MF process was utilized to remove phosphate and fluoride from TFT-LCD wastewater. The effects of molar ratio and pH on chemical precipitation were studied and compared with model predictions. The effects of operation pressure and crossflow velocity on filtration flux, permeate quality, and membrane fouling characteristics were examined in MF experiments. The following results were obtained:

1. Molar ratio (Ca:PO4:F) and pH had significant effect on the precipitation reactions of calcium with phosphate and fluoride. Higher than stoichiometric amount of calcium is required to ensure effective removal of phosphate and fluoride, and calcium tended to react with phosphate preferentially than fluoride at alkaline pH. In addition, higher removal efficiency for both PO4 and F was found at pH 10.5. The model predicts the formation of HAP, FAP, and CaF2, and was confirmed by instrumental analysis and wet chemical analysis.

2. The very fine (< 1 μm) particles were formed at pH 8.5, which led to higher fouling resistance and higher flux loss. The initial flux and steady state flux increased with increasing filtration pressure and crossflow velocity. The main fouling resistance was cake resistance among all MF experiment. The specific cake resistance and cake mass per unit filtration area value increased with increasing filtration pressure and decreased with increasing crossflow velocity.

3. The hybrid precipitation-MF process showed higher removal efficiency of turbidity, phosphate, and fluoride. Advantages of the hybrid precipitation-MF process included effective phosphate and fluoride removal from the wastewater, effective solid/water separation, smaller foot print, and potential reuse of HAP and FAP.

Table1 Residual phosphate and fluoride as affected by molar ratio and pH

pH 8.5 pH 10.5

Molar ratio

(Ca:PO4:F) Residual [PO4] (mg/L)

Residual [F]

(mg/L)

Residual [PO4] (mg/L)

Residual [F]

(mg/L)

1.5 : 1 : 0.7 30.87 (15.86)* 13.26 (8.62) 6.93 (15.17) 10.81 (8.65)

2.5 : 1: 0.7 8.11 (< 0.01) 13.13 (4.04) 1.67 (< 0.01) 9.98 (4.07)

* Number in bracket represents modeled results.

Table 2 Flux under filtration pressure of 0.5 bar and crossflow velocity of 0.48 m/s

Ca:PO4:F pH

Initial flux (L/m²h)

Steady state flux (L/m²h)

Pure water flux (L/m²h)

Pure water flux (used membrane)

(L/m²h)

Flux loss (%)

1.5:1:0.7 8.5 1,710 254 4,051 1,050 74

1.5:1:0.7 10.5 2,610 225 3,946 3,033 23

2.5:1:0.7 8.5 2,250 330 4,160 887 79

2.5:1:0.7 10.5 2,790 285 5,547 3,355 40

Table 3 Effluent quality of various treatment methods and conditions

pH Ca:PO4:F Method Turbidity

(NTU)

Conductivity (μs/cm) Precipitation-sedimentation 3.2 1,049 1.5:1:0.7

Precipitation-MF 0.33 796 Precipitation-sedimentation 3.01 1,327 8.5

2.5:1:0.7

Precipitation-MF 0.32 956

Precipitation-sedimentation 3.94 1,168 1.5:1:0.7

Precipitation-MF 0.33 806 Precipitation-sedimentation 3.42 1,468 10.5

2.5:1:0.7

Precipitation-MF 0.31 1,084

Table 4 Residual concentration of phosphate and fluoride of various treatment methods and conditions pH Ca:PO4:F Method [PO4]Res

(mg/L)

[F] Res (mg/L) Precipitation-sedimentation 27.85 (85 %)^* 13.26 (31 %) 1.5:1.0:0.7

Precipitation-MF 21.89 (88 %) 11.90 (38 %) Precipitation-sedimentation 7.77 (96 %) 11.04 (42 %) 8.5

2.5:1.0:0.7

Precipitation-MF 6.92 (96 %) 8.83 (54 %)

Precipitation-sedimentation 6.93 (96 %) 13.13 (31 %) 1.5:1.0:0.7

Precipitation-MF 4.72 (97 %) 10.51 (45 %) Precipitation-sedimentation 1.67 (99 %) 9.98 (48 %) 10.5

2.5:1.0:0.7

Precipitation-MF 1.46 (99 %) 9.12 (52 %)

* Number in bracket represents removal percentage.

Table 5 Flux under different operation conditions (precipitation conditions: Ca:PO4:F = 2.5:1:0.7, pH 10.5)

Us (m/s) TMP (bar)

Initial flux (L/m²h)

Steady state flux (L/m²h)

Pure water flux (L/m²h)

Pure water flux (used membrane)

(L/m²h)

Flux loss (%)

0.48 0.5 2,790 285 5,547 3,355 40

0.48 1.0 4,050 315 8,856 7,712 13

0.96 0.5 2,790 345 4,770 4,080 14

0.96 1.0 4,500 390 9,199 7,134 22

Table 6 The specific cake resistance, cake mass, and cake resistance under different operation condition (0.22 μm MCE membrane)

Crossflow velocity (m/s)

Filtration pressure (bar)

Cake mass (kg/m²)

Specific cake resistance

(m/kg)

Cake resistance

(1/m)

0.48 0.5 0.016 3.65E+13 5.78E+11

0.48 1.0 0.017 6.32E+13 1.10E+12

0.96 0.5 0.013 3.67E+13 4.78E+11

0.86 1.0 0.016 5.36E+13 8.73E+11

Table 7 Flux under different operation conditions when using a 0.45 µm MCE membrane (precipitation conditions: Ca:PO4:F = 2.5:1.0:0.7, pH 10.5)

Crossflow velocity (m/s)

Filtration pressure (bar)

Initial flux (L/m²h)

Steady state flux (L/m²h)

Flux loss (%)

0.48 0.5 2,050 270 92

0.48 1.0 7,740 360 46

0.96 0.5 4,950 420 48

0.96 1.0 5,310 405 22

References

[1] P. Battistoni, E. Carniani, V. Balboni, P. Tonabuoni, Chemical-physical

pretreatment of phosphogypsum leachate, Ind. Eng. Chem. Res. 45(2006) 3237-3242.

[2] B. Grzmil, J. Wronkowski, Removal of phosphates and fluorides from industrial wastewater, Desal. 189(2006) 261-268.

[3] J.Y. Park, H.J. Byun, W.H. Choi, W.H. Kang, Cement paste column for

simultaneous removal of fluoride, phosphate, and nitrate in acidic wastewater, Chemosphere 70(2008) 1429-1437.

[4] Warmadewanthi and J.C. Liu, Recovery of phosphate and ammonia as struvite from semiconductor wastewater, Sep. Purif. Technol. 64(2009) 368-373.

[5] A. Eskandarpour, K. Sassa, Y. Bando, H. Ikuta, K. Iwai, M. Okido, S. Asai, Creation of nanomegnetite aggregated iron oxide hydroxide for magnetically removal of fluorid and phosphate from wastewater, Iron Steel Instit. Japan 47(2007) 558-562.

[6] L.E. De-Bashan, Y. Bashan, Recent advances in removing phosphorus from wastewater and its future use as fertilizer (1997-2003). Water Res. 38((2004) 4222-4246[7] K. Hosni, S. Ben Moussa, A. Chachi, M.

Ben Amor, The removal of PO43- by

calcium hydroxide from synthetic wastewater: optimization of the operating conditions, Desal. 223(2008) 337-343.

[8] L. Lv, J. He, M. Wei, D.G. Evans, Z. Zhou, Treatment of high fluoride concentration water by MgAl-CO3 layered double hydroxides: Kinetic and equilibrium studies, Water Res. 41(2007) 1534-1542.

[9] M. Kumar, M. Badsuzzaman, S. Adham, J. Oppenheimer, Beneficial phosphorus recovery from reverse osmosis (RO) concentrate of an integrated membrane system using polymeric ligand exchanger (PLE), Water Res. 41 (2007) 2211-2219.

[10] N. Drouiche, N. Ghaffour, H. Lounici, N. Mameri, A. Maallemi, H. Mahmoudi, Electrochemical treatment of chemical mechanical polishing wastewater: removal of fluoride – sludge characteristics – operating cost, Desal. 223(2008) 134-142.

[11] M.F. Chang and J.C. Liu, Precipitation removal of fluoride from semiconductor wastewater, J. Environ.

Eng. 133(2007) 419-425.

[12] M. Gouider, M. Feki, S. Sayadi, Separative recovery with lime of phosphate and fluoride from an acidic effluent containing H3PO4, HF and/o HeSiF6, J. Hazard Mater. 170(2009) 962-968.

[13] E. N. Peleka, P. P. Mavros, D. Zamboulis, K. A. Matis, Removal of phosphates from water by a hybrid flotation-membrane filtration cell, Desal. 198 (206) 198-207.

[14] J. Zhang, Y. Sung, Q. Chang, X. Liu, G. Meng, Improvement of crossflow microfiltration performances for treatment of phosphorus-containing wastewater, Desal. 194 (2006) 182-191.

[15] K.J. Hwang, C.Y. Liao, K.L. Tung, Analysis of particle fouling during microfiltration by use of blocking models. J. Mem. Sci. 287((2007) 287-293.

[16] S. V. Dorozhkin, Calcium orthophosphates. J. Mater. Sci. 42((2007) 1061-1095.

[17] Warmadewanthi and J.C. Liu, Selective separation of phosphate and fluoride from semiconductor wastewater, Wat. Sci. Tech. 59(2009) 2047-2053.

[18] J. Altmann, S. Ripperger, Particle deposition and layer formation at the crossflow microfiltration. J.

Mem. Sci. 124(1997) 119-128.

[19] A.A. McCarthy, P.K. Walsh, G. Foley, Experimental techniques for quantifying the cake mass, the cake and membrane resistances and the specific cake resistance during crossflow filtration of microbial suspension. J. Mem. Sci. 201(2002) 31-45.

[20] H. Connell, J. Zhu, A. Bassi, Effect of particle shape on crossflow filtration flux. J. Mem. Sci. 153(1999) 121-139

Figure 1 Schematic diagram of the MF experimental apparatus

Figure 2 XRD of precipitate formed by CaCl2 under various conditions

Figure 3 Wet chemical analysis of precipitates as affected by molar ratio and pH (normalized ion ratio in precipitates, [PO4]=1)

Figure 4 Particle size distributions (number-based) under different precipitation conditions

Figure 5 Resistance under various chemical precipitation conditions

Figure 6 Resistance under various operation conditions using 0.22 µm MCE membrane

出國參加國際會議報告

台灣科技大學化工系 劉志成 教授 2010.09.13

一 會議背景：

第六屆永續水環境國際研討會(6^th International Conference on Sustainable Water Environment)是起源於台灣，由台灣大學環境工程研究所蔣本基教授所主 辦主辦的國際研討會，歷經其他在新加坡、日本、與韓國等不同國家的主辦，本 次第六屆輪由美國德拉瓦州德拉瓦大學土木與環境工程研究系黃金寶講座教授 負責籌劃與主辦，其他也包括台灣、新加坡、韓國、日本、中國、英國等許多著 名學者聯合主辦。本人是第一次參加非台灣主辦的此國際研討會，蒙國科會經費 補助參加經費補助機票款，以及註冊費生活費等，是難得經驗。

二 會議內容：

本次研討會除了在地美國學者之下，台灣共有超過二十人報名參加，分別為 中興大學環境工程研究系林明德主任等四人，台灣大學環境工程研究所蔣本基教 授等三人，嘉南藥理科技大學甘其詮教授等三人，成功大學環境工程研究系林財 富主任，及交通大學黃志彬教授等多人，可以說是少見的盛會。總共約 200 名國 際學者分別來自韓國、日本、法國、加拿大、澳大利亞、荷蘭、俄羅斯、越南、

中國、阿根廷、英國等六大洲。本人 7 月 28 日清晨抵達，7 月 298 日與 30 日為 主要議程，其他則參訪附近區域等。會議首日是 7 月 29 日，一早開幕典禮，大 會演講是美國國科會相關工程學門召集人 Dr. Paul Bishop(辛辛那提大學土木與 環境工程研究系)提出再生年源與有限水資源必須掛勾(Coupled)處理與管理的大 角度概念，以及永續水環境技術之研究，他以宏觀角度強調研究的重要，以及相 關科技的潛在價值與極限，算是非常有力的內容，且十分具創意。接著是大會安 排兩場主題演講，以及接連同時五個不同議程與主題的(包括氣候變遷(Climate Change) 、新興污染物(Emerging Contaminants) 、整合式水資源管理(Integrated water Management)、水資源節約與再利用(Water Conservation and Reuse)、與水資 源基礎建設(Water Infrastructure)等研討。本次會議的特色之一是連續二天全天均 為口頭發表論文，總數約 130 篇，外加 60 篇璧報論文。因為將議題清楚分別為 五個不同議程與主題，大部份演講者多著重自己新近突破，課題聚焦清楚，相同 背景者多可以容易充分理解，本人大多參與水資源節約與再利用，亦覺得學習良 多受益無窮。第二天晚上為大會晚宴，認識新朋友，並共進晚餐，現場許多各國 年輕研究生也參加，談笑風生，真是豐富輕鬆的一晚!

我報告的是”隧道工程廢水處理與再利用”，可惜並非國科會補助之計劃，過 程尚稱順利。值得提到的是，薄膜程序在水與廢水處理看起來更加受到重視。週 六(07/31)中午則應邀至黃金寶教授家共用午餐打總共約有 40 名各國學者參加大 家喝酒吃當季螃蟹天南地北交換意見可以說也是很寶貴的經驗與許多同行學者

增加彼此瞭解也是一大助益。整個三天議程結束好像一轉眼工夫，本人有始有 終，從頭到尾積極聽講，並且參與討論，只在會議結束後，才安排赴紐約觀賞棒 球賽與其他私人行程二天，於 08/03 搭機飛返台北。

三 會議心得：

這一次是大會安排非常完善細心的國際研討會，議程很早確定，所有論文以 一定格式書寫，會議進行緊湊，討論議題非常聚焦，相較其他專業領域經常性國 際學術交流的經驗，本人覺得收獲頗豐，本人執行國科會分離程序之計畫，此行 有同行研究學者切磋，實在受益很大。下一屆已經確定由中國科學研究院與清華 大學主辦。

無研發成果推廣資料

98 年度專題研究計畫研究成果彙整表

計畫主持人：劉志成 計畫編號：98-2221-E-011-019- 計畫名稱：以化學沉澱結合微過濾處理光電廢水中磷酸

量化

成果項目 ^{實際已達成}

數（被接受 或已發表）

預期總達成 數(含實際已

達成數)

本計畫實 際貢獻百

分比

單位

備 註 （ 質 化 說 明：如 數 個 計 畫 共 同 成 果、成 果 列 為 該 期 刊 之 封 面 故 事 ...

等）

期刊論文 0 0 100%

研究報告/技術報告 0 0 100%

研討會論文 0 0 100%

論文著作 篇

專書 0 0 100%

申請中件數 0 0 100%

專利 已獲得件數 0 0 100% 件

件數 0 0 100% 件

技術移轉

權利金 0 0 100% 千元

碩士生 1 0 100%

博士生 0 0 100%

博士後研究員 0 0 100%

國內

參與計畫人力

（本國籍）

專任助理 0 0 100%

人次

期刊論文 1 1 100%

研究報告/技術報告 0 0 100%

研討會論文 0 0 100%

論文著作 篇

專書 0 0 100% 章/本

申請中件數 0 0 100%

專利 已獲得件數 0 0 100% 件

件數 0 0 100% 件

技術移轉

權利金 0 0 100% 千元

碩士生 2 2 100%

博士生 0 0 100%

博士後研究員 0 0 100%

國外

參與計畫人力

（外國籍）

專任助理 0 0 100%

人次

其他成果

(

本人於研究休假期間應邀於 University of Delaware 土木與環境工程系演講, 即以此研究為主題

成果項目 量化 名稱或內容性質簡述

測驗工具(含質性與量性) 0

課程/模組 0

電腦及網路系統或工具 0

教材 0

舉辦之活動/競賽 0

研討會/工作坊 0

電子報、網站 0

科 教 處 計 畫 加 填 項

目 計畫成果推廣之參與（閱聽）人數 0

國科會補助專題研究計畫成果報告自評表

請就研究內容與原計畫相符程度、達成預期目標情況、研究成果之學術或應用價 值（簡要敘述成果所代表之意義、價值、影響或進一步發展之可能性） 、是否適 合在學術期刊發表或申請專利、主要發現或其他有關價值等，作一綜合評估。

1. 請就研究內容與原計畫相符程度、達成預期目標情況作一綜合評估

■達成目標

□未達成目標（請說明，以 100 字為限）

□實驗失敗

□因故實驗中斷

□其他原因 說明：

2. 研究成果在學術期刊發表或申請專利等情形：

論文：■已發表 □未發表之文稿 □撰寫中 □無 專利：□已獲得 □申請中 ■無

技轉：□已技轉 □洽談中 ■無 其他：（以 100 字為限）

於 Separation and Purification Technology 發表論文一篇

3. 請依學術成就、技術創新、社會影響等方面，評估研究成果之學術或應用價 值（簡要敘述成果所代表之意義、價值、影響或進一步發展之可能性）（以 500 字為限）

1.結合沉澱與薄膜程序是去除與回收磷的技術創新 2.學術上之論文未來應有一定影響力