磷為引起藻華之限制營養源，削減磷污染負荷為控制湖泊水質優養化之重要策略之一；脫硫渣(Desulfurization slag，DS)與轉爐石（basic-oxygen-furnace slag，BOF）為鋼鐵工業副產物且為廉價吸附劑。本研究探討DS與BOF兩種爐石除磷效能，研究目的為:(1)比較DS與BOF除磷效能、(2)操作參數對DS與BOF除磷之影響、及(3)DS與BOF除磷之管柱動力學。以磷酸二氫鉀(KH2PO4)配製含磷人工廢水，操作參數包含DS與BOF添加量、接觸時間、初始磷濃度等，採等溫吸附實驗、管柱實驗評估DS與BOF除磷效能。 實驗結果顯示DS與BOF於水中會溶出鈣離子且提高pH，與磷形成磷酸鈣Ca5(PO4)3(OH) (s) 沉澱物沉澱。因DS含鈣量大於BOF含鈣量， DS之單位磷吸附量約為BOF之3.2-4.9倍，顯示DS之磷去除效果較BOF更佳。DS與BOF對磷之吸附符合Freundlich 等溫吸附，DS之n值為4.95(大於1)，對磷屬有利吸附。相對地，BOF之n值小於1，對磷屬不利吸附。磷初始濃度為100-200 mg /L且EBCT為10 min時，管柱實驗結果顯示DS於停止點之吸附量約為BOF之1.3-2.6倍。EBCT由20 min降至10 min時，DS之吸附量則由506 mg-P /g降至373 mg-P /g，EBCT愈短，則到達停止點時間愈快，且吸附量愈小。管柱動力學顯示DS之最大吸附量q0為BOF之1.55- 1.69倍。綜合上述，DS除磷性能優於BOF。 Abstract: Phosphate is a limiting-nutrient to cause algal bloom. Desulfurization (DS) and basic-oxygen-furnace (BOF) slags are valuable by-products from steel industry and cost-effective adsorbates. This study evaluates the removal performance of phosphate from aqueous solution by DS and BOF slags. The aim of this study are (1) to compare the phosphate removal performance between DS and BOF, (2) to investigate the effect of operation parameters of the removal of phosphate, and (s) to evaluate the removal of phosphate by DS and BOF with column kinetics. All experiments were conducted by adsorption isotherm and column tests. The operational parameters included dosage of slags, contact time (or empty bed contact time, EBCT) and initial phosphate concentration. Both DS and BOF slags sampled from The China steel company. Furthermore, the synthetic phosphate-containing wastewater was prepared from potassium phosphate monobasic (KH2PO4). The result shows that both DS and BOF released Ca ions and induced pH of slag solution. The released Ca ions could react with phosphate to form Ca5(PO4)3(OH)(s) precipitates to remove phosphate. The released Ca ions of DS were larger than for BOF. The amount of adsorbed phosphate of DS was about 3.2-4.9 times of that for BOF. This implies that the performance on phosphate removal by DS was better that that by BOF. The removal of phosphate by DS and BOF followed the Freundlich adsorption isotherm model. The DS was favorable adsorption for phosphate, because the n value of DS was 4.9 (larger than 1). By contrast, due to n value being 0.34 (less than 1), BOF was unfavorable adsorption for phosphate. Furthermore, at initial phosphate concentration of 100-200 mg/L and EBCT of 10 min, the column test results show that the amount of adsorbed phosphate of DS was about 1.3-2.6 times of that for BOF. The amount of adsorbed phosphate decreased from 506 mg-p/g to 373 mg-p/g, as EBCT decreased from 20 min to 10 min. The shorter the EBCT is the faster to reach operational terminate point and the smaller amount of adsorbed phosphate. Column kinetics shows that the maximum amount of adsorbed phosphate, qo value, of DS was 1.55-1.69 times of that for BOF. It is concluded that performance on phosphate removal of DS was better than that of BOF.
