Combinatorial optimality of membrane morphology and feedstock during microfiltration of bottle gourd juice

Highlights

• Membrane with 0.75 μm performed optimally in terms of combinatorial parameters.

• Moderate pressure (137.9 kPa) was optimal microfiltration process parameter.

• The flux recovery ratio of the optimized case was comparably high (80.90 to 86.97%).

• Cake filtration model was the best fit model to represent pertinent flux decline.

• Both 0.75 and 1 μm membranes possessed maximum reversible fouling resistances during microfiltration.

Abstract

The combinatorial optimality of membrane morphology and process parameters during dead end microfiltration of bottle gourd juice have been addressed in this article. Saw dust and kaolin based low cost ceramic membranes with varied morphology have been chosen to evaluate upon their microfiltration performance. For the chosen membranes, fresh, paper filtered and centrifuged juice samples were considered along with transmembrane pressure differential as process parameters. Combinatorial optimality was based on flux decline trends, fitness of fouling models, irreversible and reversible fouling data, irreversible permeation resistance and nutritional analysis of the permeate samples. An interesting feature of the article had been with respect to feed constitution playing a critical role in influencing the optimal choice of membrane morphology and transmembrane pressure differentials. Among all cases, paper filtered bottle gourd juice, 0.75 μm membrane and 137.9 kPa transmembrane pressure were found to be the best choice in terms of minimal irreversible fouling, lowest protein content, good clarity, good polyphenol and antioxidant activity in the permeate and appropriate flux.

Introduction

Wide varieties of nutritional compounds exist in fruits and vegetables. Fruits and vegetables contain phytochemicals which are effective to combat degenerative diseases (Cassano, Tasselli, Conidi, & Drioli, 2009; Rawson et al., 2011). The perishable nature of fruits and vegetables demands application of relevant food preservation principles (Kubde, Khadabadi, Farooqui, & Deore, 2010; Mondal, Biswas, & De, 2016; Shah, Seth, & Desai, 2010). Among various perishable vegetables, bottle gourd (Lagenaria siceraria) is promising for human consumption from perspective of abundance, wider consumption and medicinal benefits. Characterized to possess multiple vitamins, phytochemicals and bioactive constituents, bottle gourd has been used to prevent constipation, reduce body weight, treat insomnia, maintain body pH and as an antidote to certain poisons (Bhat & Sharma, 2016; Minocha, 2015; Mondal et al., 2016). With these benefits, bottle gourd juice products have become prominent nutritional beverage in India (Alum, 2019; Fame, 2019; Generic, 2019).

A conventional approach adopted to increase the shelf life of fruits and vegetable juices is to deploy a combination of mechanical extraction followed with centrifugation and low pressure concentration. Among these, while mechanical extraction is inexpensive and scalable, the other two down stream processes are highly expensive upon scale up. Therefore, these technologies can be suitably replaced with other inexpensive, scalable and compact processes. Among alternate process technologies, with short processing times, low temperature operation, scalability, compactness and ease of operation and lower cost, membrane clarification technology is promising. Other than these, membrane processing facilitates room temperature operation, elimination of external chemicals and higher productivity (Grampp, Schmitt, & Urlaub, 1978; Urošević, Povrenović, Vukosavljević, Urošević, & Stevanović, 2017). Further, it is important to note that microfiltration is an inevitable technology for those processing systems that eliminate centrifugation.

Polysulfone (PS), polyethersulphone (PES) and polyvinylidene fluoride (PVDF) membranes with pore sizes of 0.1 μm and 0.45 μm (PS) and 0.3 μm (PES and PVDF) were investigated for the retention of sugars from pineapple juice (De Carvalho, De Castro, & Da Silva, 2008). Laorko, Li, Tongchitpakdee, Chantachum, and Youravong (2010) used polysulfone membranes (0.1 and 0.2 μm) during pineapple juice microfiltration (MF) at a TMP of 1 bar and the authors concluded that membranes with 0.2 μm nominal pore size exhibited best performance in terms of flux, product quality and irreversible fouling (Laorko et al., 2010). Membranes with similar pore sizes were investigated by Mirsaeedghazi, Emam-Djomeh, Mousavi, Aroujalian, and Navidbakhsh (2010) using PVDF membranes (pore sizes 0.22 μm and 0.45 μm) for the clarification of pomegranate juice (Mirsaeedghazi et al., 2010). In summary, polymeric membranes characterized with narrow pore size distribution were extensively investigated. However, these membranes have several limitations associated to significant irreversible fouling and lower chemical resistance. On the contrary, low cost ceramic membranes can provide better combinations of lower irreversible fouling and higher chemical resistance.

Few researchers addressed the efficacy of ceramic membranes for juice clarification. Mosambi juice clarification with higher membrane pore sizes (0.77 μm to 1.54 μm) and moderate TMP's (41.4–165.5 kPa) were investigated by Nandi, Uppaluri, and Purkait (2011) (Nandi et al., 2011). Vladisavljević, Vukosavljević, and Veljović (2013) conducted clarification of raspberry juice using various combinations of membrane materials, configurations and morphology (Vladisavljević et al., 2013). Ultrafiltration using inorganic tubular membranes (molecular weight cut off (MWCO) ranging from 30 to 300 kDa), polysulfone hollow fiber membrane (MWCO 30 kDa) and multichannel ceramic microfiltration membrane (0.2 μm pore size) were deployed by the authors. Jegatheesan, Phong, Shu, and Aim (2009) clarified limed sugarcane juice (Jegatheesan et al., 2009) using alternate tubular ceramic membranes (0.02–0.1 μm pore size) at various TMP values (1–3 bar) and inferred best performance of 0.05 μm membranes with similar product quality for all cases. dos Santos, Scherer, Cassini, Marczak, and Tessaro (2016) carried out coupled MF (tubular zirconia membrane with 0.05 μm pore size) and UF (titania membrane with a MWCO of 20 kDa) of beet stalks extract to achieve 99.5% reduction in peroxidase activity and 99.9% reduction in turbidity (dos Santos et al., 2016).

Mondal et al. (2016) used hollow fiber polymeric membranes (42,000–127,000 kDa) to perform comparative study between ultrafiltration and microfiltration of bottle gourd juice and TMP of 104 kPa was found to be optimum based on the product quality and fouling (Mondal et al., 2016). However, detailed fouling assessment was not targeted for the chosen polymer membrane morphology for the evaluation of optimal permeate quality and fouling characteristics. In summary, available literature focused extensively towards optimality of membrane morphology and TMP for polymeric membranes for fruit juice clarification. Few literatures also provided similar methodologies for ceramic membrane based fruit juice clarification. Only one literature focused towards bottle gourd vegetable juice clarification and with polymeric membranes (Mondal et al., 2016). Membrane fouling generally occurs in forms of foulant adhesion/deposition (Chen et al., 2020; Teng et al., 2019; Teng et al., 2020) and thermodynamic filtration resistance of gel layer (Long et al., 2020; Teng et al., 2020; You et al., 2020). Model based irreversible and reversible fouling analysis was targeted by many research groups to identify optimality of feedstock using Hermia fouling models (Almandoz, Pagliero, Ochoa, & Marchese, 2010; Chakraborty, Das, & Uppaluri, 2020; Hermia, 1982; Nandi et al., 2011). For few cases, the alternate Bolton models were deployed (Bolton, LaCasse, & Kuriyel, 2006). In other words, given the significant surge in development of nutritional beverages using vegetable juices such as bottle gourd juice, membrane clarification research needs to be extended to gain useful insights into optimality of membrane morphologies, feed stock alternatives and TMPs.

Considering the above mentioned lacunae for the optimality of low cost ceramic membrane technology for bottle gourd juice clarification, this work addresses the efficacy of clay based ceramic membranes with diverse pore size distributions (0.45, 0.75 and 1 μm) for three different TMPs (103.4, 137.9 and 172.4 kPa) and three different types of feedstocks (fresh, paper filtered and centrifuged juice samples). The fouling analysis was considered by deploying conventional Hermia fouling models and modified empirical models proposed by Bolton et al., 2006 (Bolton et al., 2006). Permeate product characterization involved measurement of transmembrane flux and other parameters such as clarity, color, protein content and polyphenols content. Permeate samples characterized with higher combinations of transmembrane flux, polyphenols content, clarity and color are regarded as the bias to identify optimality of TMPs, feed stocks and membrane morphologies. Thus, the considered case study can enable visualization of tradeoffs associated to membrane morphology and transmembrane flux, irreversible fouling and permeate product quality. Thereby, the ultimate objective of the article is to define useful benchmarks for the clarification of bottle gourd juice using ceramic membranes.