Elsevier

Chemical Engineering Journal

Volume 305, 1 December 2016, Pages 19-27
Chemical Engineering Journal

Humic acids removal in a photocatalytic membrane reactor with a ceramic UF membrane

https://doi.org/10.1016/j.cej.2015.10.024Get rights and content

Highlights

Abstract

The investigations on the influence of TiO2 photocatalyst (Aeroxide® P25) loading, feed cross-flow velocity (CFV) and feed water composition on the fouling of ceramic TiO2 ultrafiltration membrane and removal of humic acids (HA) in a photocatalytic membrane reactor (PMR) are presented. Moreover, the stability of the ceramic membrane after 400 h of operation in the PMR is discussed. In case of model solution of HA in ultrapure water it was revealed that by an increase of TiO2 loading from 0.5 to 1.5 g/dm3 the membrane fouling can be avoided. Alkaline conditions (pH 9) had a negative influence on the permeate flux and humic acids removal, whereas at pH 3 and pH 6.5 no membrane fouling was noticed. In the presence of HCO3, SO42− and HPO42− at both low and high concentrations the membrane fouling was more severe than in the absence of these species, what was attributed to a lower efficiency of HA removal in the feed. Addition of Ca2+ and Mg2+ to the feed containing the inorganic anions allowed to improve the efficiency of HA removal and reduce membrane fouling. After 400 h of PMR operation a deterioration of membrane separation properties was observed what was attributed to the abrasive action of photocatalyst particles.

Introduction

A photocatalytic membrane reactor (PMR) is a system coupling photocatalysis and a membrane separation in one unit. When applied to water or wastewater treatment, the role of photocatalysis is decomposition and mineralization of organic pollutants to H2O, CO2 and mineral salts, while a membrane enables separation of the photocatalyst from the treated solution thus creating a possibility of its further reuse [1]. Furthermore, the membrane could act as a barrier for the molecules present in the reaction medium, both initial compounds and products or by-products which are formed during the decomposition [1].

Humic acids (HA) constitute a vast majority of natural organic matter (NOM). They are commonly present in natural water and strictly affect its taste and color. Moreover, HA can contribute to biofouling in pipelines carrying potable water. What is more important humic acids are precursors of trihalomethanes (THMs) which are very toxic and carcinogenic compounds formed during disinfection and chlorination of drinking water [2], [3]. Humic acids are substances characterized by a high level of carboxyl (–COOH) and phenolic hydroxyl (–OH) groups attached to aromatic rings. Since they are a very complex mixture they cannot be represented by any single formula [4]. Although their detailed structure is unknown, numerous investigations have been undertaken to elucidate HA composition [5], [6]. In Fig. 1 a hypothetical structure of HA proposed by Schulten [6] is presented.

The degradation of HA can proceed through a variety of pathways [4]. It was reported [7], [8] that the reactions of hydroxyl radicals with NOM include: (i) the addition of radical dotOH to double bonds (e.g. Cdouble bondC), (ii) the abstraction of H-atom, which yields carbon-centered radicals and (iii) the gaining of an electron by radical dotOH radical from an organic substituent. The radical dotOH addition reactions to the HA unsaturated groups can result in formation of organic by-products such as ketones, alcohols, esters, ethers and carboxylic acids [7]. Some examples of by-products identified during oxidation of HA are shown in Fig. 2 [7], [9], [10].

Photocatalysis coupled with membrane separation could be a very promising method of NOM removal from water. One of the most commonly used photocatalysts is TiO2 [11], mainly due to its high activity as well as physical and chemical stability under various conditions. When membrane issue is considered, the ceramic membranes seem to be especially advantageous for the application in PMRs. They are resistant to the severe conditions prevailing during photocatalytic process e.g. hydroxyl radicals, UV light or extreme pH values, opposite to the polymer ones.

Application of a photocatalytic membrane reactor for humic acids removal was first described by Lee et al. [12]. They used cellulose acetate UF membrane to separate TiO2 photocatalyst during water treatment. The authors investigated the influence of the decomposition of HA on the permeate flux under various cross flow velocities (CFV). They concluded that although humic acids were not completely mineralized by photocatalysis under the applied conditions, their photodegradation helped to enhance the UF flux, as they were transformed to less adsorbable compounds. Molinari et al. [13] compared two PMR systems equipped with polyethersulfone or sulfonated polyethersulfone nanofiltration (NF) membranes applied for humic acids, patent blue dye and 4-nitrophenol removal. They found that the quality of permeate in the PMR was lower than during simple NF what was attributed to the formation of small molecular size by-products and intermediate species during the photodegradation. They concluded that in order to select a suitable membrane, rejection should be determined during operation of the photoreactor. Another group of researchers [14] carried out the investigations on the hybrid photocatalysis – membrane process with application of a polyvinylidene fluoride membrane for treatment of real natural surface water containing low concentrations of NOM. They considered the influence of pH value and photocatalyst loading on HA decomposition rate [14]. Fu et al. [15] used a slurry submerged laboratory-scale photocatalysis–ultrafiltration reactor (PUR) to examine the effect of feed pH, TiO2 concentration and UV light intensity on the degradation of fulvic acids [15]. Similar system was applied by Choo and co-workers [16]. The authors stressed the role of adsorption and desorption of HA on the surface of TiO2 photocatalyst particles and ferrihydrite adsorbents, which were used in order to enhance the removal efficiency [16]. The effectiveness of the PMR with a PVDF UF membrane was evaluated by Patsios et al. [3]. The researchers determined the influence of different pH values and two humic acids concentrations on HA decomposition [3].

The literature review revealed that a majority of works is focused on the effectiveness of organics removal in the PMRs equipped with polymeric membranes. There are only a few studies reporting NOM removal in PMRs with ceramic photocatalytic membranes. Syafei et al. [17] described removal of NOM and performance of a PMR with a photocatalytic flat sheet ceramic membrane. Yu et al. [18] evaluated the influence of HA concentration on membrane fouling and treatment efficiency in a system with a tubular alumina ultrafiltration membrane and TiO2 photocatalyst-coated polypropylene beads. However, to the best of our knowledge, in the literature survey there are no reports on slurry PMRs with ceramic membranes for HA removal. In fact, the literature data on ceramic membranes applied in PMRs with a suspended photocatalyst are in general very scarce [19], [20], [21].

In the present work the investigations on the removal of humic acids in a photocatalytic membrane reactor equipped with a tubular TiO2 ultrafiltration (UF) membrane are presented and discussed. The influence of the parameters such as feed cross-flow velocity (CFV), photocatalyst loading, feed water composition and pH on the permeate flux and membrane fouling as well as the permeate quality during HA removal was determined. Moreover, the stability of the ceramic TiO2 membrane after 400 h of operation in the PMR during treatment of HA solutions was evaluated.

Section snippets

Materials

A Filtanium tubular asymmetric TiO2 membrane (TAMI Industries, France) with the molecular weight cut-off (MWCO) of 100 kDa was used. The membrane length was 0.25 m and the external/internal diameters amounted to 10 mm/6 mm. The effective membrane area was 0.0047 m2. A commercially available TiO2 Aeroxide® P25 (Evonik, Germany) at a concentration of 0.5–2.0 g/dm3 was applied as a photocatalyst. Humic acids (Sigma Aldrich) at a concentration of 5 ± 0.5 mg TOC/dm3 were used as a model contaminant. In order

Influence of feed cross flow velocity on the permeate flux and HA removal in PMR

In the first stage of the research the experiments were conducted under the feed cross flow velocities ranging from 3 to 6 m/s and the transmembrane pressure of 0.1 MPa. The HA solution denoted as F1 (Table 1) was applied as a feed. The TiO2 concentration was 0.5 g/dm3. Fig. 4 presents changes of permeate fluxes in time of the experiments. Regardless of the CFV the permeate flux after an initial decrease lasting for ca. 90 min got stabilized. However, the cross flow velocity had a certain influence

Conclusions

The present investigations revealed that the effectiveness of HA removal was independent of the feed cross flow velocity. However, the CFV had a certain influence on the permeate flux. The least severe flux decline was observed at the highest CFV applied (6 m/s).

Optimization of TiO2 photocatalyst concentration is necessary from both the treatment efficiency and the permeate flux point of view. No decrease of permeate flux at 1.5 and 2.0 g TiO2/dm3 was observed as a result of effective HA removal

Acknowledgment

This research was supported by The National Science Center (Poland) under project No. 2011/03/B/ST5/01053.

References (23)

Cited by (75)

  • BiOCl as the role of cleaner for the fabrication of self-cleaning PES membranes based on the RTIPS method

    2024, Colloids and Surfaces A: Physicochemical and Engineering Aspects
  • Nanoceramics: fabrication, properties, and applications

    2024, Industrial Applications of Nanoceramics
View all citing articles on Scopus
View full text