Siloxane removal using silicone–rubber membranes
Highlights
► Siloxane permeabilities in PDMS are between 9000 (L2) and 50,000 Barrer (D5) at 20 °C. ► Siloxane–methane selectivity in PDMS is between 13 (L2) and 76 (D5). ► Permeabilities were found to be slightly lower in 4-end than in 3-end operation. ► 4-End is significantly more energy-efficient than 3-end operation. ► Air can be used as sweep gas without reaching explosive limits.
Introduction
Volatile siloxanes are considered to be the most undesired trace components of biogas [1], [2]. If the biogas is used for combustion, the siloxanes form silicon dioxide, leading to particulate formation in the gas engine. This may results in abrasion, changes in heat transfer and emission problems. A common value imposed by gas engine manufacturers is 5 mg of silicon per normal cubic metre of methane [3]. In an idealised biogas containing 50 vol.% of methane, this equals 2.5 mgSi Nm−1 corresponding to roughly 1 ppm of hexamethyldisiloxane (L2) or 0.4 ppm decamethylcyclopentasiloxane (D5), for instance. Numerous authors have conducted research on siloxane removal (e.g. [4], [5], [6], [7], [8]). A summary of the general topic is presented by Dewil et al. [9], a review on siloxane abatement techniques can be found in Ajhar et al. [10].
The majority of methods studied so far implies the screening of adsorbents, mainly activated carbon, silica gel, molecular sieves or alumina. In practice, the use of activated carbon adsorbers is predominant. The biggest drawback of this method is that the medium can usually not be efficiently regenerated. A membrane-based process, however, can be operated continuously. This work investigates the viability of membrane gas separation by using a commercially available silicone–rubber membrane. Silicone itself is a long-chained siloxane and by far the most common rubbery membrane material used. Apart from a recently described membrane contactor absorption process for biogas purification using water [11], no detailed investigations involving membranes for siloxane removal are known to the authors. The following volatile methylsiloxanes (VMS) were considered: hexadimethylsiloxane (L2), octamethyltrisiloxane (L3), decamethyltetrasiloxane (L4), hexamethylcyclotrisiloxane (D3), octamethylcyclotetrasiloxane (D4) and decamethylcyclopentasiloxane (D5). Permeability measurements were conducted using a carbon dioxide feed flow containing siloxane concentrations typical for landfill gas. The siloxane separation performance of the module was then assessed with a synthetic biogas mixture in 3-end (vacuum) and 4-end (sweep) operation. The temperature dependence of siloxane permeance was also determined.
Section snippets
Experimental
The employed PDMS membrane module, PermSelect XA1M-10 cm2 (Fig. 1) consisted of 32 parallel, dense hollow fibres without any porous support layer. When the module was purchased in the beginning of 2010, the dimensions of the fibres as stated by the producer were 167 μm inner diameter and 237 μm outer diameter, resulting in a wall thickness of 35 μm [12]. First experiments with pure gases indicated that the fibres were thicker than stated. According to an e-mail communication with the manufacturer
Pure gas permeance: carbon dioxide, methane, nitrogen
Pure gas permeance of carbon dioxide was determined threefold at 20, 30 and 40 °C in 3-end operation by applying a bore-side feed flow at roughly 1 barabs and measuring the permeate via a soap bubble flow metre. The shell-side of the module was kept at a vacuum pressure of around 200 mbar. No distinguishable permeance dependence on temperature was noticeable. Pure carbon dioxide permeance amounted to roughly = 0.134 Nm3 h−1 m−2 bar−1 ± 2%RSD, assuming a membrane area of 7.05 cm2 (based on the inner
Conclusions
A mixture of VMS in landfill gas-typical concentrations and carbon dioxide was fed to a commercially available, laboratory-scale PDMS membrane. Measurements with carbon dioxide as sweep gas revealed siloxane permeabilities between roughly 9000 (L2) and 50,000 Barrer (D5) at 20 °C. The ideal selectivity towards carbon dioxide was in the range of approximately 4 (L2) to 23 (D5) and towards methane between 13 (L2) and 76 (D5). The permeabilities of all studied siloxanes decreased with rising
Acknowledgement
The authors like to thank the Deutsche Bundesstiftung Umwelt (DBU) for funding (Project No. 23278).
References (22)
- et al.
Removal of siloxanes in biogases
J. Hazard. Mater.
(2001) - et al.
Energy use of biogas hampered by the presence of siloxanes
Energy Convers. Manage.
(2006) - et al.
Siloxane removal from landfill and digester gas – a technology overview
Bioresour. Technol.
(2010) - et al.
Suitability of Tedlar (R) gas sampling bags for siloxane quantification in landfill gas
Talanta
(2010) - VDI, Measurement of landfill gases – Measurements in the gas collection system, in: K.d.L.i.V.u.D.-N. KRdL (Ed.), Beuth...
- et al.
Volatile organic silicon compounds: the most undesirable contaminants in biogases
Water Sci. Technol.
(2008) - DVGW, Nutzung von Gasen aus regenerativen Quellen in der öffentlichen Gasversorgung, in: DVGW G 262 (A), Bonn,...
- et al.
Siloxane removal from biogas by biofiltration: biodegradation studies
Clean Technol. Envir.
(2008) - et al.
Biogas pre-upgrading by adsorption of trace compounds onto granular activated carbons and an activated carbon fiber-cloth
Water Sci. Technol.
(2009) - et al.
Purification of biogases from siloxanes by adsorption: on the regenerability of activated carbon sorbents
Energy Fuels
(2009)
Removal of siloxane from digestion gas of sewage sludge
Bioresour. Technol.
Cited by (38)
Siloxanes removal from biogas and emerging biological techniques
2021, Emerging Technologies and Biological Systems for Biogas UpgradingImpurities in biogas: Analytical strategies, occurrence, effects and removal technologies
2020, Biomass and BioenergyHigh-solid anaerobic digestion of sewage sludge: challenges and opportunities
2020, Applied EnergyDesiloxanation process of biogas using an amorphous iron hydroxide-based adsorbent: A comparison between laboratory and field-scale experiments
2020, Chemical Engineering Research and DesignCitation Excerpt :There are growing interests focused on the control of the siloxanes in anaerobic digestion gas as well as in landfill gas because it increases the costs of maintenance and then induces cost problems due to metal silicate formation such as calcium silicate (CaSiO3) during methane (CH4) combustion (Ajhar et al., 2010; Nam et al., 2013; Ricaurte Ortega and Subrenat, 2009). Biofiltration system (Accettola et al., 2008), gas separation membrane (Ajhar et al., 2012), decomposition over solid oxides (Finocchio et al., 2008), and adsorption into porous materials like activated carbon (Nam et al., 2013; Ricaurte Ortega and Subrenat, 2009; Alba et al., 2014; Gong et al., 2015), zeolite (Ricaurte Ortega and Subrenat, 2009; Montanari et al., 2010), silica (Nam et al., 2013; Ricaurte Ortega and Subrenat, 2009; Montanari et al., 2010; Jafari et al., 2016), activated alumina (Nam et al., 2013), and sewage sludge (Matsui and Imamura, 2010), are recently studied to control the siloxanes in anaerobic digestion gas and landfill gas. The most widely used method to control siloxane concentrations is adsorption on activated carbon (Ajhar et al., 2010; Nam et al., 2013; Ricaurte Ortega and Subrenat, 2009; Alba et al., 2014; Gong et al., 2015) but nowadays many efforts are put on metal hydroxide-based adsorbents (Lee et al., 2017; Magnone et al., 2018).