透過您的圖書館登入
IP:13.58.112.1
  • 學位論文

管式直接接觸薄膜蒸餾放大規模之模擬及實驗探討

Simulation and experiment study on scale-up of tubular direct contact membrane distillation

指導教授 : 莊清榮
本文將於2025/08/15開放下載。若您希望在開放下載時收到通知,可將文章加入收藏

摘要


薄膜蒸餾在海水淡化及廢水處理方面,近年來持續受到產學界的關注。而管式模組其單位模組體積下可具有較高之膜表面積,應用於工業界中較具有發展潛力。本研究利用PVDF管式模組進行直接接觸薄膜蒸餾(DCMD)實驗,探討兩不同長度模組之通量差異,並改變不同操作參數進行討論,另一方面則以MATLAB建立管式模組DCMD之通量模擬程式,與文獻實驗數據及本研究之實驗結果進行比較,並進一步以模擬程式預估在不同膜管長下的通量及熱效率等。 在DCMD實驗中,當模組為30及60cm時,進料溫度從50°C升至70°C,通量皆上升了約2.2倍,且觀察到60cm模組整體通量較30cm模組大,這與由熱質傳學理所預估的趨勢相反,推測為兩模組之膜管封裝排列方式有差異,且長模組內部膜管受流體流動會造成較大擺動振幅,降低膜面溫度與濃度極化。比較同向流與逆向流兩種流動方式對通量之影響,兩管長之實驗結果顯示,逆向流之實驗結果皆較同向流者高了約3.9%。 為了要驗證本研究所建之模擬程式的可行性,先依據文獻上既有的操作參數代入本模擬程式進行預估,數篇文獻之實驗結果與模擬預估值之相對差距都小於5%。至於本研究之實驗與模擬預測結果相比,30cm之模組兩者之相對差距僅為2%,至於60cm長之模組,兩者相對差距達59%。另外針對進料端鹽濃度與通量關係之模擬,在進料為3.5及15wt%NaCl時,30cm模組者之模擬與實驗的相對差距為12及43%,乃因模擬無法有效預測高濃度進料時膜面濃度極化所致。 確認了模擬程式的可行性後,利用模擬程式探討管式模組之規模放大時之通量變化,預測結果當管長從10cm上升至100cm,進料溫度範圍為50~70°C,進料流量固定為1.25 L/min,通量下降了46~55%,而熱效率值下降了3~9% ;當填充密度從5%上升至50%,通量下降了23~25%,而熱效率值上升了3~4%。最後比較同向流與逆向流兩種流動方式對通量之影響,模擬結果顯示當管長低於20cm時,兩者通量幾乎相同,管長越長時,兩者通量差距越大。

關鍵字

薄膜蒸餾 模擬 管式模組

並列摘要


In recent years, membrane distillation has continued to receive attention from the industry and academia in seawater desalination and wastewater treatment. Due to tubular module can have a higher membrane surface area per unit module volume, so it has more potential in scale-up for industrial applications. In this study, direct contact membrane distillation (DCMD) experiments were carried out using PVDF tubular modules with two different lengths. The fluxes under different operating parameters were analyzed. On the other hand, a simulation program for DCMD flux with tubular module was established and verified first by comparing with the experimental data from literatures. The simulation program was further used to compare the experimental results of this study and also estimate the flux and thermal efficiency under different membrane tube lengths. In the DCMD experiment with feed temperature rises from 50 °C to 70 °C, both the modules of 30 and 60cm lengths give almost the same flux increases by about 2.2 times. It is noted that the flux obtained with 60cm module is larger than that of the 30cm module. This phenomenon is contrary to the trend predicted by the theory of heat and mass transfer. It is speculated that the packaging arrangement of the two modules is different and fluid flow will make the long membrane tube have a large swing, reducing temperature and concentration the polarization of the membrane surface. Comparing the effects of co-current and countercurrent flow on flux, the experimental results of the two modules showed that countercurrent flow gives about 3.9% higher flux than those of co-current flow. In order to verify the feasibility of the simulation program built in this study, the operating parameters from some literatures were used for simulation. The relative difference between the experimental results of these literatures and the estimated value of this simulation is less than 5%. As for the comparison between the experimental and simulation prediction results of this study, the relative difference for 30cm modules is only 2%, but for the 60cm modules the difference is 59%. In addition, when discussing the effect of feed salt concentration on flux, the difference between simulation and experiment was obvious. The average difference of 30cm module is 12% and 43% when the feed concentration is 3.5 and 15wt% NaCl. This is due to the inability of simulations to effectively predict concentration polarization during high-concentration feeds. In this study, the established simulation program was further used to analyze the flux change for the scale-up of the tubular module. The simulation results showed that when the feed flow is fixed at 1.25 L/min and the feed temperature is 50°C, the increase of tube length from 10 to 100cm will lead 46% decrease of the average flux and the thermal efficiency will decreases from 31% to 29%. While when the feed is at 70°C, the flux decreases by 55% and the thermal efficiency decreases from 44% to 40%. The simulation results show that when the module length is less than 20 cm, the flux is almost the same. The longer the module length, the larger the flux gap between the two.

參考文獻


Ali, A., Quist-Jensen, C. A., Macedonio, F., & Drioli, E. (2016). Optimization of module length for continuous direct contact membrane distillation process. Chemical Engineering and Processing: Process Intensification, 110, p. 188–200.
Ali, A., F. Macedonio, E. Drioli, S. Aljlil & O.A. Alharbi(2013). Experimental and theoretical evaluation of temperature polarization phenomenon in direct contact membrane distillation. Chemical Engineering Research and Design, 91(10), p. 1966–1977.
Alkhudhiri, A., & Hilal, N. (2018). Membrane distillation—Principles, applications, configurations, design, and implementation. Emerging Technologies for Sustainable Desalination Handbook, p. 55–106.
Alrehaili, O., F. Perreault, S. Sinha & P. Westerhoff (2020). Increasing net water recovery of reverse osmosis with membrane distillation using natural thermal differentials between brine and co-located water sources: Impacts at large reclamation facilities. Water Research, 116134–.
Anvari, A., A. A. Yancheshme, K. M. Kekre & A. Ronen (2020). State-of-the-art methods for overcoming temperature polarization in membrane distillation process: A review, Journal of Membrane Science, 118413–.

延伸閱讀