Poly(octylmethylsiloxane)/oleyl alcohol supported liquid membrane for the pervaporative recovery of 1-butanol from aqueous and ABE model solutions

https://doi.org/10.1016/j.jiec.2012.07.022Get rights and content

Abstract

Supported liquid membranes (SLM) composed of oleyl alcohol (OA) and poly(octylmethylsiloxane) (POMS) in microporous flat sheets were investigated for the pervaporation of 1-butanol. Water contact angle and partition coefficients were used to characterize the SLMs. Optimum pervaporation performance was attained at 30 wt% OA SLM wherein high separation factor (α = 279) and total flux (JT = 95.9 g/(m2 h)) were achieved from 2.5% (w/v) 1-butanol/water feed, at 60 °C. In these conditions, 27% reduction in JT was observed when fed with model fermentation broths but remained highly selective toward 1-butanol (α = 76.4). SLM stability was demonstrated with <4% LM loss after operation.

Highlights

► Blended supported liquid membranes (SLMs) were developed for butanol recovery. ► Oleyl alcohol (OA) was used as the extractant and POMS as the stabilizing agent. ► Optimum pervaporation performance was observed at 30 wt% OA content SLM. ► The SLM showed high selectivity toward 1-butanol from fermentation broth. ► The above SLM exhibited excellent stability with minimal liquid membrane loss.

Introduction

Environmental concerns and the continuous depletion of fossil fuels have stimulated the development of technologies for the conversion of renewable resources into alternative transportation fuels, such as bio-butanol. 1-Butanol has a large potential as a fuel because it has superior properties than ethanol; it has lower Reid vapor pressure, higher energy content, higher blending capacity and higher hydrophobicity [1]. As a result, acetone–butanol–ethanol (ABE) fermentation has regained much attention in recent years for the industrial production of 1-butanol and other high value liquid fuels. But the low solvent concentration of ABE in fermentation broths (1–2%, w/v) makes the conventional solvent recovery methods (i.e. distillation) highly uneconomical, rendering the overall ABE process infeasible [2], [3].

Among the advanced solvent recovery technologies, pervaporation (PV) is considered as a cost-effective alternative as it is highly energy-efficient and conveniently generates concentrated organics from dilute solutions [4]. Several studies on the PV separation of 1-butanol have been reported using various types of membranes, such as: (1) porous polymeric membranes [5], [6], (2) dense polymeric membranes [7], [8], [9], [10], [11] (3) polymeric mixed matrix membranes (MMMs) with inorganic filler [12], [13], [14], [15], [16], and (4) supported liquid membranes (SLMs) [17], [18], [19].

Among these types of membranes, SLMs have shown exceptionally high selectivities and promising flux values in pervaporative recovery of alcohols. Aside from the generally lower mass transfer resistance offered by the liquid membrane (LM), the overall separation performance of SLMs is primarily governed by its affinity to the target solutes. Such LM–solute interactions are typically determined through the partition coefficient (Kp) parameter. For instance, it has been shown that oleyl alcohol (OA) with a moderately high 1-butanol Kp = 3 exhibited high 1-butanol flux (390 g/(m2 h)) and separation factor (180) as an SLM system for the pervaporation (30 °C) of 1–3.75 wt% 1-butanol feed [18]. Similarly, trioctyl amine-based SLM yielded 11 g/(m2 h) of 1-butanol flux and 275 separation factor for the pervaporation (54 °C) of 1.5 wt% of 1-butanol feed [19]. Both LMs (OA and TOA) showed exceptional performances but OA has the additional advantage over TOA and other organic extractants due to its biocompatibility with microorganisms (i.e. non-toxicity) and its relatively higher 1-butanol Kp [19], [20], [21]. Such features are important especially if the SLM-based PV is to be integrated with fermentation systems for enhanced 1-butanol yields. However, the main limitation of SLM relies on its unreliable mechanical stability due to the eventual detachment and loss of LM from the support matrix [22]. One way to minimize the physical deterioration of SLM is by blending the LM extractant with another component.

Although poly(dimethylsiloxane) (PDMS) is the most commonly preferred polymeric material for pervaporative organic separation, several studies have shown that poly(octylmethylsiloxane) POMS exhibits a slightly better separation performance than PDMS [22], [23], [24]. These results were corroborated by Bennett et al. [25] wherein the presence of longer chain alkyl groups on the siloxane backbone significantly enhances its affinity toward the organic compounds. It is clear from these studies that the potential of POMS for 1-butanol separation can also be exploited. The availability of POMS in liquid form as a silicone fluid presents its suitability for SLM applications. Moreover, the inherent hydrophobicity, viscosity of POMS and its miscibility with long-chain linear alcohols make it possible to obtain a potentially selective and more stable LM with simple blending of the liquid polymer with oleyl alcohol [22].

The present study focuses on the development and application of a highly selective SLM for the pervaporative recovery of 1-butanol from aqueous solutions and model ABE fermentation broths. The SLM is composed of blended OA and POMS as LM impregnated in a porous flat sheet support matrix (Celgard® 2400). The PV performances of SLMs with varied LM blend compositions were evaluated in terms of permeate mass flux, permeate concentration, separation factor and pervaporation separation index (PSI). Furthermore, the partitioning of 1-butanol in the LM phase and the hydrophobicity of the SLM were investigated at various OA contents in POMS through equilibrium liquid–liquid extraction experiments and contact angle measurements, respectively. LM losses incurred in the SLMs after PV operations were also measured as basis for SLM stability.

Section snippets

Reagents and materials

Reagents used were purchased from various companies: OA (60%) was from Acros-Organics (NJ, USA), POMS (viscosity = 600–1000 cSt; density@25 °C = 0.91 g/cm3) was procured from ABCR GmbH & Co. (KG, Karlsruhe, Germany), 1-butanol (>99.0%) was supplied by Showa Chemical Co. Ltd. (Japan), HPLC grade acetone was from J. T. Baker (NJ, USA), HPLC grade absolute ethanol was obtained from Fisher Scientific (USA), methanol was from Merck KGaA (Darmstadt, Germany) whereas butyric acid (>99%) and phosphoric acid

Effect of OA content on 1-butanol Kp in LM/water system

The 1-butanol Kp with different OA/POMS compositions tested at 0.5%, 1.5% and 2.5% (w/v) initial 1-butanol concentrations are shown in Fig. 2. Results reveal that 1-butanol Kp increased linearly concurrent with an increase of OA content at 25 °C while minimal differences were observed at different initial 1-butanol concentrations.

To explain these results, Hansen solubility parameter (HSP) distance (Ra) was calculated between different species in the system using Eq. (8) [31]:Ra12=4(δd,1δd,2)2+(

Conclusions

Blended OA/POMS SLMs were developed and tested for the pervaporative recovery of 1-butanol from model solutions. Among the SLMs with varied OA/POMS compositions, the 30 wt% OA/POMS SLM exhibited the best performance for the pervaporation of aqueous 1-butanol solution. The PV performance of the SLM was improved at higher operating temperature but more remarkably, when feed concentration was increased. Highest 1-butanol flux and separation factor were obtained when the SLM was operated at 60 °C.

Acknowledgment

This research was supported by the Priority Research Centers Program (Project No. 2012-0006693) and by the National Research Foundation of Korea (NRF) (Project No. 2012R1A2A1A01009683) funded by the Korea government Ministry of Education, Science and Technology (MEST).

References (43)

  • N. Qureshi et al.

    Journal of Membrane Science

    (2001)
  • P. Shao et al.

    Journal of Membrane Science

    (2007)
  • K.W. Boddeker et al.

    Journal of Membrane Science

    (1990)
  • F. Liu et al.

    Separation and Purification Technology

    (2005)
  • A.G. Fadeev et al.

    Journal of Membrane Science

    (2001)
  • N. Qureshi et al.

    Journal of Membrane Science

    (1999)
  • A. Jonquieres et al.

    Journal of Membrane Science

    (1997)
  • J. Huang et al.

    Journal of Membrane Science

    (2001)
  • H. Zhou et al.

    Separation and Purification Technology

    (2010)
  • H. Zhou et al.

    Separation and Purification Technology

    (2011)
  • A. Thongsukmak et al.

    Journal of Membrane Science

    (2007)
  • G.M. Nisola et al.

    Chemosphere

    (2010)
  • O. Trifunovic et al.

    Separation and Purification Technology

    (2006)
  • P. Sampranpiboon et al.

    Journal of Membrane Science

    (2000)
  • M. Bennett et al.

    Journal of Membrane Science

    (1997)
  • P.R. Brookes et al.

    Journal of Membrane Science

    (1995)
  • E.A. Fouad et al.

    Journal of Membrane Science

    (2008)
  • O. Trifunovic et al.

    Journal of Membrane Science

    (2003)
  • D. Hofmann et al.

    Polymer

    (1997)
  • N.M. Kocherginsky et al.

    Separation and Purification Technology

    (2007)
  • S.Y. Li et al.

    Biotechnology Progress

    (2011)
  • Cited by (31)

    • Aqueous two-phase extraction to separate alpha-ketoglutaric acid from fermentation broth by combining crystallization

      2022, Journal of Industrial and Engineering Chemistry
      Citation Excerpt :

      A strategy of ATPS combined with crystallization was developed to separate succinic acid [24], and the purity of succinic acid was 90.58%, which provides a reference for the separation of α-KG. Although there are many studies on the separation of bioproducts from model fermentation systems [25,26], the existence of impurities and byproducts in the actual biosynthesis system might affect the distribution behavior and purity of target products [27], which needs to be further explored. In the latest work, α-KG was effectively separated from biotransformation broth by salting-out extraction combined with solar-driven distillation [28].

    • Development of correlations between deasphalted oil yield and Hansen solubility parameters of heavy oil SARA fractions for solvent deasphalting extraction

      2022, Journal of Industrial and Engineering Chemistry
      Citation Excerpt :

      Moreover, because these correlations only require the HSP value of the feedstock, they can be applied to the liquid–liquid extraction process, wherein thermodynamic parameters are difficult to measure. The HSP has been used in theoretical analysis of phase equilibrium in various separation technologies [40–43]. A new correlation for DAO yield from heavy oil was developed applying modified REDs with investigating the effects of operating variables.

    • Aqueous acidified ionic liquid pretreatment for bioethanol production and concentration of produced ethanol by pervaporation

      2019, Journal of Industrial and Engineering Chemistry
      Citation Excerpt :

      As shown in Fig. 7, the ethanol flux increased from 50.1 g/m2h to 90.8 g/m2h, and water flux increased from 229 g/m2h to 651 g/m2h as temperature increased from 20 °C to 60 °C. These results are in agreement with those in literatures [19,55,60–62]. The pervaporation performance can also be observed based on the effects of the operating temperature on the separation factor.

    View all citing articles on Scopus
    View full text