Journal of Membrane Science 328 (2009) 11–14
Contents lists available at ScienceDirect
Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci
Rapid communication
Biogas, membranes and carbon dioxide capture Eric Favre ∗ , Roda Bounaceur, Denis Roizard Laboratoire des Sciences du Génie Chimique, Nancy-Université, CNRS, 1 rue Grandville, BP 541, 54001 Nancy, France
a r t i c l e
i n f o
Article history: Received 11 March 2008 Received in revised form 8 September 2008 Accepted 7 December 2008 Available online 13 December 2008 Keywords: Biogas Carbon dioxide Capture Energy Membranes Process
a b s t r a c t Biogas, which consists primarily of methane, can be obtained through the biological transformation of a large variety of organic wastes, and has drawn an increased interest within a framework of renewable energy sources. The use of gas permeation membranes for upgrading biogas (i.e., for removing carbon dioxide and hydrogen sulfide from biomethane) has been abundantly investigated and already displays practical industrial applications. A different concept, based on the use of carbon dioxide/nitrogen selective membranes, is presented in this study. The key issue is to achieve carbon capture and storage (CCS) from the flue gas of a power plant where raw biogas, together with oxygen-enriched air, is used as fuel. A CO2 concentrated flue gas is obtained, and as a result, gas permeation membranes can easily achieve carbon capture targets for post-combustion situations (90% CO2 purity and 90% capture ratio). A simulation study has shown that, based on this strategy, it is possible to acquire a very low overall energy requirement for carbon dioxide capture (less than 1 GJ per ton of recovered CO2 ). Such a remarkable performance can be achieved as soon as membranes show a CO2 /N2 selectivity in the range of 50; a target value already reported for several polymeric materials. © 2008 Elsevier B.V. All rights reserved.
1. Introduction The production of energy from renewable sources is one of several means of reducing the emissions of greenhouse gases (GHG). Moreover, this topic has been paid an increasing attention given the restrictions on fossil fuel uses. Apart from carbon dioxide, methane, which is produced in natural environments as soon as anaerobic degradation of organic substrates takes place, is one of the greenhouse gases that significantly contributes to the greenhouse effect, with a global warming potential 23-fold that of carbon dioxide [1]. The production of methane through the microbial degradation of organic wastes, which leads to so-called biogas (i.e., a wet mixture of CH4 and CO2 with hydrogen sulfide and hydrocarbons as minor compounds), is appealing. This strategy does indeed offer the possibility of using a large variety of renewable feedstocks (food and agricultural wastes, manure, crop residues, municipal and industrial wastewater, landfill, etc.) for an alternative production of gaseous fuel. Furthermore, the biogas technology, which is one of the oldest processes used for the treatment of industrial wastes and the stabilization of sludge, is a mature technique that is already employed on a large scale [2]. Several analyses have also shown that biomethane production through energy crops can lead to larger yields in Joule per hectare as compared to for instance ethanol or biodiesel [3]. Consequently, several countries (e.g. India,
∗ Corresponding author. Tel.: +33 3 83 17 53 90; fax: +33 3 83 32 29 75. E-mail address:
[email protected] (E. Favre). 0376-7388/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2008.12.017
Germany, Austria, Sweden) have initiated ambitious research and development programmes on biomethane production, at various power output levels (typically ranging from 0.1 to 5 MW). In terms of use, dried and desulfurised biomethane can be directly fuelled on-site, without CO2 separation, into stationary block heat and power plants [3]. An upgrading process can also be applied, in order to increase the calorific value, minimize corrosion problems and modify combustion characteristics to a pseudo natural gas quality; in that case, CO2 and H2 S have to be removed, and 90% methane content is often aimed [4–6]. Upgrading is also mandatory when biomethane production has to be connected to a pipe-line for network distribution (such as a gas station for transport applications). Membrane processes have been identified as one of the most effective means for upgrading biogas and this topic has already received considerable attention [7]. Taking into account the increased development of so-called carbon capture and storage (CCS) frameworks [1], the present communication proposes a completely different approach. The key concept consists in exploring the possibility of capturing carbon dioxide in the flue gas of a biomethane power plant, in order to comply to the requirements of a CCS constraint framework. Should this possibility be feasible from a technico-economical point of view, a negative GHG emission would result since carbon from a renewable source would not be released in the atmosphere but rather stored underground (in geological formations or deep saline aquifers, for instance). Such a possibility would obviously be of great interest in the portfolio of global GHG emissions reduction [1]. From the carbon capture process point of view, a 90% carbon dioxide
12
E. Favre et al. / Journal of Membrane Science 328 (2009) 11–14
purity and a 90% capture ratio are often proposed as minimal targets [1,8]. A key issue associated with this strategy corresponds to the overall energy requirement of the capture process, which, ideally, should not exceed 2 GJ per ton of recovered carbon dioxide in order to be competitive [9]. Currently, the best available technologies for carbon dioxide capture in post-combustion processes (such as chemical absorption in amines) typically require 3–4 GJ per ton [1,10]. Another strategy for post-combustion consists in achieving combustion in pure oxygen, thereby preventing the dilution effect exerted by nitrogen. In such a case, a final requirement in the range of 0.9 GJ per ton of oxygen (imposed by the stoichiometric combustion equation) is often reported [1,10]. For a natural gas plant, this data corresponds to 1.8 GJ per ton of recovered carbon dioxide. It will be shown that, thanks to the novel approach presented below, a much lower overall energy requirement is potentially achievable for biogas power plants. 2. Rationale The starting point for the exploratory analysis proposed in this study makes use of the following statement: employing membranes as a carbon capture process is difficult in post-combustion due to the low carbon dioxide content, typically 4–20%, as well as the low total pressure that prevails in flue gases [11,12]. It has to be kept in mind, however, that membrane processes show a very high parametric sensitivity towards the CO2 content of the feed gas. The specific energy requirement logically decreases when a high CO2 volume ratio (typically 30% or more) has to be treated. To that respect, the fact that biomethane already contains a significant amount of CO2 can be seen as beneficial since a large content of CO2 can be expected in the dry flue gas (i.e., after methane combustion and drying). Table 1 summarises some examples of biogas compositions from various sources. Nevertheless, when biogas combustion is achieved in air, the dilution effect exerted by nitrogen might result in a CO2 content in the flue gas that is still too low for a sufficiently efficient membrane capture. In order to circumvent this drawback, a classical solution would be to achieve combustion with an oxygen-enriched mixture. Oxycombustion (i.e., fuel combustion in pure oxygen), which is one of the most promising technologies for post-combustion CCS [10], typically corresponds to this situation. For biogas applications, the use of an oxygen-enriched mixture presents additional advantages in terms of improvement of energy efficiency and combustion performances. Oxygen/nitrogen separation by membranes has for instance been tested in diesel motors in order to take advantage of this effect [13]. Based on the two previous arguments, i.e., the need to achieve a high CO2 content in the flue gas and the possible use of an oxygen-enriched mixture, a systematic analysis of the energy
Table 1 Typical biogas compositions in volume fractions (from Refs. [2–7]). Biogas constituents
Municipal wastewater
Landfill
Agricultural residues
CH4 CO2 N2 O2 H2 O
60 33 1 0 6
45 32 17 2 4
68 26 1 0 5
requirement of a hybrid process that combines carbon capture and oxygen-enriched air production on a biogas power plant has been investigated. It is important to stress at this stage that a whole range of oxygen concentrations has been screened, from air to pure oxygen (i.e., oxycombustion), in order to possibly identify an optimal set of operating conditions. Fig. 1 summarises the different process parts that were taken into account in the simulation study: (i) Obviously, a biogas composition first had to be defined. According to Table 1, a binary CH4 /CO2 mixture was postulated for the sake of simplicity. The methane content was tuned between 0.5 (low quality biogas) and 1 (pure methane, i.e., natural gas combustion) in the simulation study. (ii) Cryogeny was selected as the oxygen-enriched air production process (noted 1 in Fig. 1), since this technology is known to offer the best performances for such applications [10,14,15]. Moreover, the energy requirement of this process has been reported over the whole range of oxygen purity (i.e., from 0.21 to 0.99 volume fraction) [15,16]. (iii) The flue gas composition can be computed as soon as the biogas composition and the volume fraction of oxygen have been fixed (steps (i) and (ii)). An ideal combustion process, governed by the stoichiometric equation, was postulated for that purpose. (iv) A drying operation was performed on the flue gas, similar to that carried out in post-combustion capture frameworks [8]. (v) A carbon dioxide capture process based on a carbon dioxide selective membrane module was finally selected (noted 4 on Fig. 1). The energy requirement of a gas permeation membrane module could indeed be easily computed, through well-established models [11,17]. Two selectivity values were used: 50 and 200. The first enabled the evaluation of the performances of already reported membrane materials [12], and the latter corresponded to more efficient materials. A defined CO2 purity (noted y and fixed at 0.9) and a target carbon dioxide recovery ratio (noted R and fixed at 0.9) are required in order to respect the classical guidelines in CCS [1,9]. The feed mixture of the membrane module contains a carbon dioxide mole fraction noted x , which corresponds to the dry flue gas composition issued from the biogas combustion. This value is imposed by
Fig. 1. A hybrid carbon capture process combining an oxygen-enriched air production unit (1), a combustion process (2), a drying step (3) and a carbon dioxide capture process (4). x represents the methane mole fraction in the raw biogas, y the oxygen mole fraction which is injected in the combustion chamber, and x the CO2 mole fraction in the dry flue gas. The CO2 mole fraction y , which is recovered for transportation and sequestration, as well as the capture ratio (R) is usually fixed at 0.9.
E. Favre et al. / Journal of Membrane Science 328 (2009) 11–14
13
the biomethane and oxygen-enriched stream compositions. With this set of constraints, membrane selectivity (50 or 200), permeate purity (0.9) and carbon dioxide recovery (0.9) being imposed, a target pressure ratio ( ) can be computed through the numerical resolution of the governing equations of the membrane module (single module case, cross-plug flow) [11]. The energy requirement of the membrane module can thus be determined based on the classical adiabatic compression of a single stage compressor on the feed side, which makes use of the adiabatic expansion coefficient of the mixture (). The feed compression was selected in the analysis in order for the carbon dioxide, which was recovered at the permeate side, to be at atmospheric pressure. This choice enabled a direct comparison of the energy requirement with the chemical absorption case (i.e., at the exit of the stripper). The potential gain in energy of a vacuum pumping strategy for the membrane module operation on the permeate side [11], as opposed to upstream compression, has not been considered here. In a first step, biogas composition being fixed (0.5 < x < 0.95), the oxygen mole fraction produced by cryogeny is imposed (0.21 < y < 1). Assuming a perfect combustion process, the quantity of oxygen needed for a given quantity of methane present in the biogas to burn can be calculated. From this, the energy requirement for oxygen production can be computed. The increasing specific energy requirement for an increased oxygen content (y) is expressed as: EO2 = −0.12563 + 0.45962y + 0.68487y2
(1)
where EO2 is expressed in GJ per ton of oxygen [15]. Additionally, x , the carbon dioxide mole fraction in the dry flue gas, is determined as a function of the methane mole fraction in the raw biogas (x) and the oxygen mole fraction (y) through: x =
1 + (1 − x/x) 1 + (1 − x/x) + 2(1 − y/y)
(2)
At this stage, the energy requirement for the membrane capture step can be computed. Since membrane selectivity (˛), permeate purity (y ), feed composition (x ) and capture ratio (R) are fixed, a single pressure ratio value ( ) is obtained [11]. The overall energy requirement of the process ET , expressed in GJ per ton of recovered CO2 , can finally be determined according to the following expression, which takes into account both the oxygen production (EO2 ) and the carbon dioxide capture contributions: E=
EO2 R
2+
−1
T ( )
(−1/)
10−3
−1
Rx 44
(3)
3. Results and discussion Fig. 2 summarises the simulation results for the two membrane selectivity data as functions of the oxygen content in the feed stream (y) and the carbon dioxide content of the biogas (x). According to this approach, an optimal oxygen purity, leading to a minimal overall energy requirement, can be clearly determined for each biogas composition. The validity of the concept, based on the association of raw biogas combustion with oxygen-enriched air, is thus confirmed within a CCS framework from an overall energy requirement point of view. The minimal overall energy requirement, expressed as the O2 mole fraction in the feed mixture, approximately ranged between 0.4 and 0.5. Increasing membrane selectivity from 50 to 200 induced a slight shift in the optimal oxygen purity location. A more selective membrane material moderately decreases the overall energy requirements (Fig. 2a and b); for example, with a 95% methane content in biogas (bold line in Fig. 2), a minimal energy requirement of 1.4 GJ per ton of CO2 and an optimal oxygen mol fraction (y) of 0.62 is obtained with a selectivity (˛) of 50, while the corresponding data are 1.2 GJ per ton of CO2 and 0.52 respectively
Fig. 2. The overall energy requirement of a hybrid process (E), in GJ per ton of recovered CO2 as a function of the oxygen mole fraction in the feed stream of the biogas combustion unit (y). Each curve corresponds to a given methane volume fraction in the raw biogas (x). From the upper to the lower curves: bold: x = 0.95, dotted bold: x = 0.8, dotted: x = 0.7, solid line: x = 0.5. The capture ratio (R) and carbon dioxide purity (y ) have been fixed at 0.9 and a single stage cross-plug flow membrane module is used. (a) Membrane selectivity (˛): 50. (b) Membrane selectivity (˛): 200.
when a selectivity of 200 is used. Decreasing the methane content decreases the energy requirement and the optimal oxygen content. From a practical point of view, the search and the experimental evaluation of membrane materials for CO2 /N2 separation with a CO2 /N2 selectivity of 50 or more can be seen as interesting for the hybrid situation investigated in this study, which corresponds to concentrated carbon dioxide feed mixtures (with typical CO2 mole fractions of 0.4–0.65). Unfortunately, experimental membrane permeation data are, under such feed conditions, scarce. Last but not least, when taking 1.8 GJ per ton of CO2 as a classical energy requirement for oxycombustion fuelled by natural gas [10], the hybrid process proposed in this study could lead to an impressive decrease of the overall energy requirement, i.e., down to 0.8 GJ per ton of CO2 .
14
E. Favre et al. / Journal of Membrane Science 328 (2009) 11–14
This figure could be achieved provided that the carbon dioxide volume fraction in the raw biogas was of 30% or more, a range that can be found in a large number of situations (Table 1). 4. Conclusions and prospects The objective of the present study was to explore the potential interest of a hybrid process for the post-combustion carbon dioxide capture of a biomethane power plant. A simulation was carried out for a plant combining a cryogenic oxygen-enriched air production (O2 purity from 21 to 99%) and a membrane post-combustion capture. It was shown that such a hybrid process could lead to an overall energy requirement of less than 1 GJ per ton of recovered carbon dioxide, provided that: (i) A carbon dioxide volume fraction of 30% or more was found in the biomethane feedsource. Remarkably, this value corresponds to a large majority of biogas compositions. (ii) An oxygen purity of approximately 40% was used. This result differed significantly from that in the only study that has proposed the use of pure oxygen for a combined electricity and liquid carbon dioxide production from landfill gas [20]. It should be stressed that this level of oxygen purity corresponds to the compositional range for which gas separation membranes can compete with other technologies for air separation [13,18,19,21]. (iii) A membrane with a CO2 /N2 selectivity of 50 or more was used in a single stage cross-plug flow module. This type of selectivity performance has already been reported by several authors on equivalent mixtures, particularly thanks to block copolymers [12]. The concept of a hybrid process that is based on a combination of an oxygen-enriched air production step and a membrane capture step for biogas use can thus be seen as promising. Moreover, it offers the possibility of achieving negative GHG emission performances. More generally, the study confirms that gas permeation membranes can play a decisive role in carbon capture processes as soon as concentrated carbon dioxide streams are to be treated. It is also important to note that recent studies suggest that carbon capture processes should be applied to medium-scale combustion installations (i.e., 1–100 MW) where membranes could play a decisive role [22]. Biogas installations belong to this category. It is obvious that further studies, including the experimental validation of the proposed concept, have to be carried out. First, the results that we reported are solely considering the energy requirement for CO2 separation and the optimum occurs at the lowest methane concentration under consideration. It has to be noted that the amount of power actually produced will also go down with decreasing methane content. Furthermore, the stoichiometric conditions that have been postulated in this note do not comply with engines and gas turbines because of temperature control requirements. NOx production is particularly critical and an excess of gas or a recirculation is often needed in order to minimize these emissions. These considerations call for a rigorous analysis of the overall power production as a function of methane content; an experimental investigation of the hybrid process, including membrane integration is currently under progress. It is hoped that the present research efforts will stimulate an interest in this direction that, up until now, has remained unexplored. Acknowledgements The Centre National de la Recherche Scientifique is gratefully acknowledged for financial support (Phycap and Cocase grants from
Programme Energie of the CNRS). The authors also thank reviewers of this manuscript for their useful comments and suggestions. Notation E R T x x y y
overall energy requirement [GJ per ton of CO2 ] carbon dioxide recovery ratio perfect gas constant (8.314 J mol−1 K−1 ) temperature (K) methane mole fraction in raw biogas carbon dioxide mole fraction in dry flue gas oxygen mole fraction in the oxygen-enriched air stream carbon dioxide mole fraction in the permeate side of membrane module
Greek letters ˛ membrane separation factor adiabatic gas expansion coefficient pressure ratio References [1] O. Davidson, B. Metz, Special Report on Carbon Dioxide Capture and Storage, International Panel on Climate Change, Geneva, Switzerland (2005) (www. ipcc.ch). [2] S. Yadvika, T.R. Sreekrishnan, S. Kohli, V. Rana, Enhancement of biogas production from solid substrates using different techniques. A review, Biores. Technol. 95 (2004) 1. [3] D. Antoni, V.V. Zverlov, W.H. Schwartz, Biofuel from microbes, Appl. Microbiol. Biotechnol. 77 (2007) 23. [4] S.A. Stern, B. Krshnakumar, S.G. Charati, W.S. Amato, A.A. Friedman, D.J. Fuess, Performance of a bench scale membrane pilot plant for the upgrading of biogas in a wastewater treatment plant, J. Membr. Sci. 151 (1998) 63. [5] R. Rautenbach, W. Dahm, Oxygen and methane enrichment. A comparison of module arrangements in gas permeation, Chem. Eng. Technol. 10 (1987) 256. [6] R. Rautenbach, K. Welsch, Treatment of landfill gas by gas permeation. Pilot plant results and comparison to alternatives, J. Membr. Sci. 87 (1994) 107. [7] W.J. Schell, Use of membranes for biogas treatment, Energy Prog. 3 (1983) 96. [8] A. Aspelund, K. Jordal, Gas conditioning. The interface between CO2 capture and transport, Int. J. Greenhouse Gas Cont. 1 (2007) 343. [9] P. Deschamps, P.A. Pilavachi, Research and development actions to reduce CO2 emissions within the European Union, Oil Gas Sci. Technol. 59 (2004) 323. [10] D. Singh, E. Croiset, P.L. Douglas, M.A. Douglas, Techno-economic study of CO2 capture from an existing coal-fired power plant: MEA scrubbing vs O2 /CO2 recycle combustion, Energy Conv. Manage. 44 (2003) 3073. [11] R. Bounaceur, N. Lape, D. Roizard, C. Vallières, E. Favre, Membrane processes for post-combustion carbon dioxide capture: a parametric study, Energy 31 (2006) 2220. [12] E. Favre, Carbon dioxide recovery from post combustion processes: can gas permeation membranes compete with absorption? J. Membr. Sci. 294 (2007) 50. [13] G.R. Rigby, H.C. Watson, Application of membrane gas separation to oxygen enrichment of diesel engines, J. Membr. Sci. 87 (1994) 159. [14] F. Notaro, Advances in ambient temperature air separation. Refrigeration Science and Technology Proceedings, in: Proceedings of the Meeting on Air Separation Technology, International Institute of Refrigeration, Munich, Germany, 1996, ISBN 2-903633-86X. [15] G. Göttlicher, The Energetics of Carbon dioxide Capture in Power Plants, US Department of Energy, National Energy Technology Laboratory, 2004. [16] A. Tuberga A., Oxygen from a jointed PSA and cryogenic process. Refrigeration Science and Technology, in: Proceedings of the Meeting of Commission A3, International Institute of Refrigeration, 1996, ISBN 2-903633-45-2. [17] R. Zolandz, G.K. Fleming, Gas permeation applications, in: W.S. Ho, K.K. Sirkar (Eds.), Membrane Handbook, Van Nostrand Reinhold, New York, 1992. [18] R.W. Baker, Future directions of membrane gas separation technology, Ind. Eng. Chem. Res. 41 (2002) 1393. [19] B.D. Bhide, A. Stern, A new evaluation of membrane process for the oxygen enrichment of air. II. Effects of economic parameters and membrane properties, J. Membr. Sci. 62 (1991) 37. [20] V.A. Malik, S.L. Lerner, D.L. MacLean, Electricity, methane and liquid carbon dioxide production from landfill gas, Gas Sep. Purif. 1 (1987) 77. [21] S.G. Kimura, W.R. Browall, Membrane oxygen enrichment. I. Demonstration of membrane oxygen enrichment for natural gas combustion, J. Membr. Sci. 29 (1986) 69. [22] CO2 capture from medium scale combustion installation, International Energy Agency (2007) Report # 2007/7.
