應用氧化鈣輔助蒸氣氣化將枝條轉生質能之潛力評估

Abstract

流體化床技術近年來因其透過氣化反應產生合成氣，具有高燃料利用潛力與實用性，逐漸被視為新興的廢棄物處理技術。本研究以臺灣大學校園中隨機採集並破碎處理的落葉與枯枝作為氣化原料，模擬農業廢棄物中之果樹枝條，並探討其在特定操作條件下的轉換潛力。本研究實驗採用反應曲面法（Response Surface Methodology，簡稱RSM）中的中心複合設計（Central Composite Design，簡稱CCD），選定兩項操作參數進行變因實驗。固定參數包括：儀器功率 1 kW、氣體收集時間 120 分鐘（作為反應時長）、最高反應溫度 700 °C、進氣量 2 mL/min，以及生質料重量20 g；非固定參數為氧化鈣/碳（CaO/C）比與蒸氣/生質物（S/B）比，對應變動條件為水蒸氣通入量 0–5 mL/min 與氧化鈣重量 0–25 g。本研究氣體分析部分採用氣相層析儀（GC-TCD 與 GC-MS）進行氣體成分鑑定，灰分以元素分析儀進行，並針對所得數據進行質量平衡計算、RSM 預測模型建構、生命週期評估（Life Cycle Assessment，簡稱LCA）與能源轉換潛力探討。 根據本研究實驗結果顯示，當 S/B 固定為 0.317 且調變 CaO/C 時，以 CaO/C = 1 的反應表現最佳，所得合成氣熱值達 0.43 kJ/g-biomass，且 H₂ 與 CH₄ 濃度皆為最高；若固定 CaO/C = 1 並調變 S/B 值，以 S/B = 0.423 時獲得最佳結果，熱值提升至每公克枝條生物質可產生約 0.68 kJ，且合成氣之H₂ 與 CH₄ 濃度亦為最高。關於RSM 模型方面，以 S/B 比與 CaO/C 比為自變數，合成氣中 CO、CO₂、CH₄ 與 H₂ 濃度為依變數進行建模。結果預測當 S/B = 0.41 與 CaO/C = 1.25 時，氣體產率最佳，每公克枝條生物質可產生合成氣含CO₂、CH₄、CO及H₂分別為1.7 mmol、0.2 mmol、0.1 mmol及1.3 mmol，對應合成氣熱值為 0.7 kJ/g-biomass。另一方面，本研究生命週期評估以中點衝擊與終點衝擊為兩大分析面向，針對三組數據進行比較（控制組、最佳 CaO/C 比組與最佳 S/B 比組）。結果指出，當 S/B = 0.423 且 CaO/C = 1 時，每產出 1 MJ 能量之環境衝擊最小。除水資源消耗與 CO₂ 排放外，另一關鍵衝擊來自於氧化鈣製程，因此催化劑使用量為影響環境負載之重要因素。根據終點衝擊結果顯示，本技術造成人類健康潛在風險主要來源為催化劑製程產生之毒性與空氣污染物排放，且對資源消耗與生態系影響之影響較低。 最後，本研究針對應用氧化鈣輔助蒸氣氣化將果樹枝條轉生質能之潛力進行評估，我國每年果樹枝條產量約 60 萬公噸，其中，以屏東縣產量最多；依本研究結果推估，屏東縣透過流體化床技術每年可產生約 32,600 GJ 能源。若以人口數計算各縣市之能源供應潛力，則以臺東縣最具發展潛力。全臺每年若採用此技術進行氣化處理，預估可減少約 210~4,254 公噸之化石燃料衍生 CO₂ 排放當量。綜合以上，經由優化的操作條件與催化劑添加，流體化床結合氧化鈣輔助蒸氣氣化技術可有效提升合成氣品質與反應效率，具有取代化石燃料之潛力。未來建議方向包括系統及設備設計最佳化、實驗放大規模與實地測試，以確實驗證其潛力並進而推廣之。

In recent years, fluidized bed technology has emerged as a promising waste treatment approach due to its high fuel utilization efficiency and practicality in syngas production via gasification. This study investigates the conversion potential of branches randomly collected and shredded from the campus of National Taiwan University, serving as a simulated representation of fruit tree branches in agricultural waste, under controlled operating conditions. A Central Composite Design (CCD) under the framework of Response Surface Methodology (RSM) was employed to evaluate the effects of two key operating parameters. The fixed experimental parameters included a reactor power of 1 kW, gas collection time of 120 minutes (as the reaction duration), a maximum reaction temperature of 700 °C, an inlet gas flow rate of 2 mL/min, and a biomass feedstock weight of 20 g. The variable parameters were the calcium oxide to carbon ratio (CaO/C) and the steam to biomass ratio (S/B), corresponding to steam flow rates of 0–5 mL/min and CaO of 0–25 g, respectively. Gas composition was analyzed using gas chromatography (GC-TCD and GC-MS), while ash content was determined via elemental analysis. The experimental data were further utilized to conduct mass balance calculations, develop RSM-based predictive models, perform life cycle assessment (LCA), and evaluate the overall energy conversion potential. Experimental results indicated that when the steam to biomass (S/B) ratio was fixed at 0.317 and the CaO/C ratio varied, the optimal performance was observed at CaO/C = 1. Under this condition, the resulting syngas exhibited the highest heating value of 0.43 kJ/g, along with the highest concentrations of H₂ and CH₄. Conversely, when the CaO/C ratio was fixed at 1 and the S/B ratio was varied, the best outcome was achieved at S/B = 0.423, yielding an increased heating value of 0.68 kJ/g-biomass and similarly elevated H₂ and CH₄ concentrations. The mass balance analysis revealed a relatively low overall mass recovery rate of approximately 30–40%, which is likely attributed to unquantified carbon content in low-chain hydrocarbons and tar, as well as the loss of ash particles carried out of the reactor by the gas stream during the reaction process. With respect to the RSM model, the steam to biomass ratio (S/B) and CaO/C ratio were selected as independent variables, while the concentrations of CO, CO₂, CH₄, and H₂ in the syngas were treated as dependent variables. The predictive model suggested that the optimal gas yield occurs at S/B = 0.41 and CaO/C = 1.25, with the corresponding gas concentrations of CO₂: 1.7 mmol/g-biomass, CH₄: 0.2 mmol/g, CO: 0.1 mmol/g, and H₂: 1.3 mmol/g. Under these conditions, the predicted syngas heating value reaches 0.7 kJ/g-biomass. On the other hand, the life cycle assessment (LCA) in this study was conducted using both midpoint and endpoint approaches to compare three experimental conditions: the control group, the optimal CaO/C ratio group, and the optimal S/B ratio group. The results indicated that the combination of S/B = 0.423 and CaO/C = 1 yielded the lowest environmental impact per MJ of energy produced. In addition to water consumption and CO₂ emissions, a significant contributor to environmental burden was identified as the calcium oxide production process, highlighting catalyst consumption as a key influencing factor. According to the endpoint analysis, this technology poses the greatest potential risk to human health compared to impacts on resource depletion and ecosystems, primarily due to the toxicity and air pollutants emitted during catalyst manufacturing. Furthermore, this study assessed the potential of converting fruit branches into bioenergy via CaO-assisted steam gasification. In Taiwan, annual fruit branches is estimated at approximately 600,000 tonnes, with Pingtung County contributing the largest share. Based on the findings, it is estimated that this county could generate approximately 32,600 GJ of energy annually using fluidized bed gasification. When normalized by population to assess regional energy supply potential, Taitung County was identified as having the highest development potential. If this technology were implemented nationwide, the estimated annual reduction in fossil fuel–derived CO₂-eq emissions could range from 211 to 4,254 tonnes. In conclusion, with optimized operating conditions and catalyst addition, fluidized bed gasification can significantly enhance syngas quality and reaction efficiency, demonstrating considerable potential as an alternative to fossil fuels. Future work should focus on equipment optimization, scale-up studies, and field trials to validate its practical applicability and support broader deployment.