This work aims to assess the acclimation of microorganisms to a gradual increase of salinity and hydrocarbons, during the start-up of two moving bed membrane bioreactors (MB-MBRs) fed with saline oily wastewater. In both systems an ultrafiltration membrane was used and two types of carriers were employed: polyurethane sponge cubes (MB-MBRI) and polyethylene cylindrical carriers (MB-MBRII). A decreasing dilution factor of slops has been adopted in order to allow biomass acclimation. The simultaneous effect of salinity and hydrocarbons played an inhibitory role in biomass growth and this resulted in a decrease of the biological removal efficiencies. A reduction of bound extracellular polymeric substances and a simultaneous release of soluble microbial products (SMPs) were observed, particularly in the MB-MBRII system, probably due to the occurrence of a greater suspended biomass stress as response to the recalcitrance of substrate. On the one hand, a clear attachment of biomass occurred only in MB-MBRI and this affected the fouling deposition on the membrane surface. The processes of detachment and entrapment of biomass, from and into the carriers, significantly influenced the superficial cake deposition and its reversibility. On the other hand, in MB-MBRII, the higher production of SMPs implied a predominance of the pore blocking.

INTRODUCTION

The direct discharge of wastewater from washing of oil tankers (slops) into the sea involves a high environmental impact. Slops are mainly characterized by hydrocarbons and high salinity, although there are other pollutants such as oils and surfactants. The International Maritime Organization regulates and prevents marine pollution by means of the International Convention for the Prevention of Pollution from Ships, 1973, as modified by the Protocol of 1978 relating thereto (MARPOL 73/78). This regulation forces harbour authorities to implement wastewater treatments (Sun et al. 2010).

Biological treatment of slops is an alternative method to physicochemical treatment but it is hampered by high concentrations of salinity and total petroleum hydrocarbons (TPHs). Salinity causes plasmolysis resulting in cell death (Lay et al. 2010; Jang et al. 2013; Johir et al. 2013); however, halophilic microorganisms can be selected in highly saline environment (Woolard & Irvine 1995). On the other hand, hydrocarbons may inhibit the metabolism of microorganisms because of their toxicity towards bacteria (Abdollahzadeh Sharghi et al. 2014). Bearing in mind the above, a biomass acclimation to slop is required for any biological treatment (Kose et al. 2012).

In this scenario, membrane bioreactors (MBRs) are a consolidated advanced technology applied for the treatment of both saline wastewater (Di Bella et al. 2013) and high salinity wastewater contaminated by hydrocarbons (Pendashteh et al. 2012; Abdollahzadeh Sharghi et al. 2014). The MBR system has many advantages reported in the literature including, among others, high effluent quality and the possibility to treat slowly biodegradable substances (Stephenson et al. 2000). Indeed, the MBR systems are suitable for the treatment of slowly biodegradable and recalcitrant wastewater, thanks to the high sludge retention time (SRT). High SRT allows the enrichment of slowly growing bacteria and consequently, the establishment of a more diverse biocoenosis together with broader physiological and biological capabilities (Suárez et al. 2012), promoting the acclimation of the microorganisms to slowly biodegradable substrate.

Furthermore, by adding suspended carriers in the mixed liquor of a conventional activated sludge reactor, a hybrid moving bed biofilm reactor can be obtained (Sun et al. 2012). If an MBR system is characterized by the coexistence of suspended and attached microorganisms on carriers, it can be defined as a moving bed membrane bioreactor (MB-MBR) (Di Trapani et al. 2014). In this case, the overall biomass concentration (suspended and attached) is higher than in a conventional MBR and it is able to biodegrade the most recalcitrant compounds contained in slops (Yang et al. 2014). Nevertheless, like in all systems based on membrane technology, the above-mentioned schemes can present fouling problems whose minimization has represented a great challenge to this research field for several years (Meng et al. 2009; Guo et al. 2012).

The acclimation to salinity (Di Bella et al. 2013, 2014a; Di Trapani et al. 2014) and hydrocarbon removal in a saline environment by means of pre-acclimatized microorganisms (Kose et al. 2012; Pendashteh et al. 2012; Abdollahzadeh Sharghi et al. 2014) have been studied separately in the past. Unlike the previous works, the aim of this study is to evaluate the simultaneous acclimation of microorganisms from a conventional activated sludge plant, to a gradual increase of both salinity and hydrocarbons during the start-up of two MB-MBR systems. The key point and the novelty of the present work is represented by the treatment of real slops, not synthetic, deriving from washing of oil tankers, since very few studies on real slops are reported in the literature.

MATERIALS AND METHODS

Bench-scale plants

The bench-scale plants were built adopting the same configuration of Di Bella et al. (2014a) (Figure 1) at the Laboratory of Sanitary and Environmental Engineering of Enna University (Italy). In both systems the microorganisms were adapted to hydrocarbons and salinity by progressively reducing the feed dilution factor between synthetic wastewater and real slop. The plants were characterized by a bioreactor of 14 L and by an ultrafiltration (UF) hollow fibre membrane module (Zee-Weed™01) having specific area equal to 0.093 m2 and nominal porosity of 0.04 μm. The permeate flux was kept at about 15 L m−2 h−1 and the hydraulic retention time was equal to about 18 h. The membranes were periodically backwashed (every 5 min for a time period of 1 min), by pumping a fraction of permeate back through the membrane modules, and were kept in a compartment separated from the rest of the reactor. For this purpose, a perforated wall separated the two compartments in order to avoid collisions of suspended carriers with UF modules. This wall was made of 2 mm thick plexiglass sheet and the size of the holes was such as to ensure the retention of the carriers within the MBBR compartment, separating them from the MBR zone. In the first reactor (named MB-MBRI) the carriers were Linpor®: polyurethane cubic sponges with a 14 mm side and a 1,000 m2 m−3 specific effective carrier surface area. In the second reactor (named MB-MBRII), AnoxKaldnesK1 was used: polyethylene cylindrical carriers with a 500 m2 m−3 specific effective carrier surface area. The filling ratio was kept at 30% for both systems. The experimental campaign lasted about 60 days and it was divided into three different phases, each of these characterized by a different ‘slop dilution factor’, in order to allow the acclimation of microorganisms:
  • Phase I (day 0–17): no slop addition;

  • Phase II (day 18–30): 5% slop volume;

  • Phase III (day 31–60): 10% slop volume.

Figure 1

Layout scheme of bench-scale plants: (a) MB-MBRI with Linpor® and (b) MB-MBRII with AnoxKaldnesK1.

Figure 1

Layout scheme of bench-scale plants: (a) MB-MBRI with Linpor® and (b) MB-MBRII with AnoxKaldnesK1.

Although nowadays there are no full-scale examples of slop treatment plant, the decision to operate up to a value of the dilution ratio of 10% was not problematic for the purpose of gradual biomass acclimation to the recalcitrant substrate.

In more detail, during Phase I the source of chemical oxygen demand (COD) in the synthetic wastewater was represented by sodium acetate. The ammonia was dosed with the amount strictly sufficient for heterotrophic metabolism. Starting from Phase II, slop was mixed with synthetic solution containing sodium acetate, ammonium chloride and potassium diphosphate, in order to maintain a COD near to values of 1,100–1,300 mg L−1 (high strength wastewater) and to ensure the minimum intake of C:N:P for bacterial metabolism in the ratio of 100:5:1 proportionately.

Slops were sampled from barges of an oil coastal deposit at the Augusta harbour (Sicily) and subjected to a preliminary treatment of de-oiling, prior to biological treatment, in order to remove a large amount of non-biodegradable oils and greases.

In both system, the inoculum was collected from a conventional activated sludge with an initial concentration of suspended solids of 3.5 g L−1. In Table 1, the main average features of slop are summarized.

Table 1

Wastewater inlet characterization

Parameter Units Slop Phase I Phase II Phase III 
COD mg L−1 1,566 1,095 ± 346 1,329 ± 200 1,303 ± 205 
TC mg L−1 428 459 ± 84 521 ± 41 533 ± 29 
IC mg L−1 38 79 ± 31 113 ± 32 174 ± 56 
TOC mg L−1 390 380 ± 54 408 ± 73 359 ± 84 
N-NH4 mg L−1 ≈ 0 115 ± 37 167 ± 66 155 ± 75 
Chlorides mg L−1 23,455 216 ± 5 1,446 ± 121 2,393 ± 260 
Bromides mg L−1 411 ≈0 21 ± 5 41 ± 8 
Sulphides mg L−1 3,541 ≈0 143 ± 17 263 ± 48 
TPH mg L−1 30 1.5 ± 0.2 3.1 ± 0.5 
Parameter Units Slop Phase I Phase II Phase III 
COD mg L−1 1,566 1,095 ± 346 1,329 ± 200 1,303 ± 205 
TC mg L−1 428 459 ± 84 521 ± 41 533 ± 29 
IC mg L−1 38 79 ± 31 113 ± 32 174 ± 56 
TOC mg L−1 390 380 ± 54 408 ± 73 359 ± 84 
N-NH4 mg L−1 ≈ 0 115 ± 37 167 ± 66 155 ± 75 
Chlorides mg L−1 23,455 216 ± 5 1,446 ± 121 2,393 ± 260 
Bromides mg L−1 411 ≈0 21 ± 5 41 ± 8 
Sulphides mg L−1 3,541 ≈0 143 ± 17 263 ± 48 
TPH mg L−1 30 1.5 ± 0.2 3.1 ± 0.5 

Although in Phase III the slop volume was higher than in Phase II, the COD was on average lower because of the significant variation of slop composition throughout all time periods.

There was no excess sludge wastage from both systems (SRT nearly infinite) during the first 30 days, in order to promote biomass acclimation and growth to saline conditions (Di Trapani et al. 2014). Afterwards, in order to minimize the differences in the cake layer formation on the membrane surface, it was decided to operate the two plants with the same biomass concentration, 4–5 g mixed liquor suspended solids (MLSS) L−1 in Phase I with synthetic wastewater and between 7–8 gMLSS L−1 in Phase II and III when real slops were fed to the systems. Therefore, the sludge withdrawals were carried out taking into account this specific aim. As a consequence, the two pilot plants were characterized by different values of the mixed liquor SRT, depending on the biokinetic behaviour of biomass in the two lines, resulting in an SRT close to 40 days for the MB-MBRI and about 45 days for the MB-MBRII.

Analytical methods

Throughout the period of study, the influent wastewater, the mixed liquor and the effluent permeate were sampled twice a week and analysed according to Standard Methods (APHA 2005). In particular, the following parameters were measured: MLSS, mixed liquor volatile suspended solids (MLVSS), COD, ammonia nitrogen (NH4-N), nitrate (NO3-N) and TPH concentration. The mixed liquor samples were firstly filtered through a 0.45 μm filter. The measurements of all anions were carried out by means of ionic chromatography (using ICS Dionex 1100®). The total organic carbon (TOC) was measured by a TOC-VCSH analyser that also provides the total carbon (TC) and the inorganic carbon (IC). The TPH concentration was measured using a gas chromatograph equipped with a flame ionization detector (Agilent 6890N®), after extraction of TPHs from samples with hexane. Periodically, samples of suspended carriers were taken and analysed for total solids (TS) content in order to evaluate the biofilm growth on carriers. For the details of the adopted procedure, the reader is referred to the literature (Di Trapani et al. 2013).

In the case of performance evaluation in terms of COD and TOC, in addition to the CODOUT and the TOCOUT of the permeate, the CODBIO and the TOCBIO were also evaluated. They are the fractions of soluble COD and TOC measured in the supernatant of mixed liquor samples obtained after centrifugation at 5,000 rpm for 10 min and filtering through a 0.45 μm filter paper. In this way it was possible to define two different performances: one referring to a biological system, which takes into account only the effect of biomass in the bioreactor prior to using membrane filtration (Equation (1)), and the other concerning a global removal due to biomass and membrane filtration (Equation (2)): 
formula
1
 
formula
2

Extracellular polymeric substances analysis

The total extracellular polymeric substances (EPST) are expressed as the sum of bound extracellular polymeric substances (EPSs) and soluble microbial products (SMPs), which represent the soluble portion of EPST, according to the following equation: 
formula
3
where the subscript symbol ‘P’ or ‘C’ indicates the relative content of proteins or carbohydrates respectively in the bound EPSs and SMPs. The SMPs were obtained by centrifugation at 5,000 rpm for 5 min while the bound EPSs were extracted by thermal extraction method (Zhang et al. 1999). Carbohydrates in the EPST were determined according to the phenol–sulphuric acid method with glucose as standard (Dubois et al. 1956). Proteins were determined by the Folin method with bovine serum albumin as standard (Lowry et al. 1951).

Resistances in series model

The membrane fouling was analysed by employing the resistance in series model which is funded on cake layer removal with ‘physical cleaning’. According to this model, the total resistance to filtration (RT) is defined by (Di Trapani et al. 2014): 
formula
4
where Rm is the intrinsic resistance of membrane, RPB is the irreversible resistance due to membrane pore blocking, RC,irr is the fouling resistance related to superficial cake deposition removable only by physical cleanings, RC,rev is the fouling resistance related to superficial cake deposition removable by ordinary backwashing. The RT can also be described by Darcy's law: 
formula
5
where TMP is the transmembrane pressure (Pa), μ the permeate viscosity (Pa·s), and J the permeate flux (m3 m−2 s−1). So Rm in Equation (4) was estimated by measuring the water flux and the TMP of ultrapure water with a new membrane module, using Equation (5). Regarding the physical cleaning, firstly it was necessary to extract the membrane from the reactor and secondly to wash it with ultrapure water; thus the cake layer on the membrane surface was removed according to the ‘manual water rinsing’ (Chang et al. 2001). Finally, the specific resistances to filtration were evaluated according to the following equations: 
formula
6
 
formula
7
 
formula
8
where RT1 and RT2 are the total resistances measured according to Equation (5) after physical cleaning in ultrapure water and into the bioreactor before cleaning, respectively.

RESULTS AND DISCUSSION

Biological aspects

As aforementioned, both MB-MBR plants were inoculated with an MLSS concentration almost equal to 3.5 g L−1. Although both systems were inoculated with the same MLSS concentration, in the MB-MBRI the amount of suspended solids was always less than in the MB-MBRII because an amount of suspended biomass was entrapped within the porosity of the spongy carriers and this enabled the subsequent growth of the attached biomass on the polyurethane sponge cubes (Feng et al. 2012). On the other hand, the polyethylene cylindrical carriers in the MB-MBRII system showed a negligible biofilm formation during the whole period, as reported in Table 2. Indeed, in this condition and with the simultaneous presence of salt and hydrocarbons, these carriers require a pre-acclimation of the biological film from dispersed microorganisms, not aggregated in flocs. Moreover, the ‘entrapment’ operated by the cubic sponge provides a positive effect for the biofilm attachment and a lower suspended biomass mineralization.

Table 2

MLSS, MLVSS, attached biomass and food/microorganisms ratio (F/M) in MB-MBRI and MB-MBRII systems for each phase

Parameter Units Phase I
 
Phase II
 
Phase III
 
MB-MBRI MB-MBRII MB-MBRI MB-MBRII MB-MBRI MB-MBRII 
MLSS gSS L−1 4.0 ± 0.8 4.7 ± 0.9 6.7 ± 0.4 7.5 ± 0.6 7.8 ± 0.7 8.3 ± 1.8 
MLVSS gSS L−1 3.2 ± 0.8 3.7 ± 0.9 5.6 ± 0.5 6.3 ± 0.6 6.3 ± 0.4 6.1 ± 1.2 
Attached biomass gTS L−1 6.1 ± 0.6 7.7 ± 0.5 <1 9.7 ± 0.9 1.1 ± 0.5 
F/M gCOD gMLSS−1 d−1 0.65 ± 0.08 0.56 ± 0.07 0.47 ± 0.04 0.42 ± 0.03 0.40 ± 0.03 0.38 ± 0.02 
Parameter Units Phase I
 
Phase II
 
Phase III
 
MB-MBRI MB-MBRII MB-MBRI MB-MBRII MB-MBRI MB-MBRII 
MLSS gSS L−1 4.0 ± 0.8 4.7 ± 0.9 6.7 ± 0.4 7.5 ± 0.6 7.8 ± 0.7 8.3 ± 1.8 
MLVSS gSS L−1 3.2 ± 0.8 3.7 ± 0.9 5.6 ± 0.5 6.3 ± 0.6 6.3 ± 0.4 6.1 ± 1.2 
Attached biomass gTS L−1 6.1 ± 0.6 7.7 ± 0.5 <1 9.7 ± 0.9 1.1 ± 0.5 
F/M gCOD gMLSS−1 d−1 0.65 ± 0.08 0.56 ± 0.07 0.47 ± 0.04 0.42 ± 0.03 0.40 ± 0.03 0.38 ± 0.02 

Although the growth of biomass was different, the organic matter biological removals were similar in both systems. This is probably due to the reduced levels of salinity and hydrocarbons, up to a maximum of about 2.4 g L−1 of chloride and 3 mg L−1 of TPHs, respectively. More specifically, both plants reached COD biological removal efficiencies in the range of 72–77%. Nevertheless, the combined effect of salinity and hydrocarbons appears to have a significant effect in terms of TOC biological removal, which is reduced from approximately 91% to 77% for the MB-MBRI and from about 92% to 71% for the MB-MBRII, passing from phase I to Phase III. Despite there being an important influence on the reliability of COD measurements in presence of salinity (Di Bella et al. 2014b), the disagreement between the trends of COD and TOC is probably due also to the different species of compounds in the bulk of the mixed liquor (produced by cellular lysis and biological stress) that are more recalcitrant and with a low propensity to further biological oxidation. In fact, in terms of total biological–physical removal efficiencies, with the contribution of UF, the COD and TOC removal reached high and homogeneous values in both systems, in the ranges of 90%–95% for COD and 95%–98% for TOC, respectively. Thus this circumstance confirms the important effect exercised by the membrane filtration towards the dissolved organic compounds. Moreover, as the literature studies show that the metabolic inhibition produced by salinity occurs at higher saline concentrations compared to those of the present study (Sun et al. 2010; Di Bella et al. 2013), it is probable that the inhibitory effect on biomass in Phases II and III is mainly due to the effect of hydrocarbons (or other compounds) contained in the real slops. In particular, the decrease of the biological removal efficiency, defined in terms of TOC, can be related to the inability of the heterotrophic strains to biodegrade the slops, given an inadequate biomass acclimation to TPHs (Kose et al. 2012). The previous considerations are confirmed by a negligible removal efficiency of TPHs, up to about 8% and 5% for MB-MBRI and MB-MBRII, respectively. This slight difference is probably due to phenomena such as physical adsorption and entrapment of hydrocarbons within the porosity of the spongy carriers in MB-MBRI plant. Table 3 shows the average removal efficiencies for both bench-scale plants.

Table 3

Average values and standard deviations of concentrations and removal efficiencies in terms of COD, TOC, NH4-N and TPH, in MB-MBRI and MB-MBRII systems for each phase

Parameter   Units MB-MBRI
 
MB-MBRII
 
Phase I Phase II Phase III Phase I Phase II Phase III 
COD Influent mg L−1 1,095 ± 346 1,329 ± 200 1,303 ± 205 1,095 ± 346 1,329 ± 200 1,303 ± 205 
Supernatant mg L−1 285 ± 49 372 ± 62 300 ± 29 252 ± 47 385 ± 83 300 ± 41 
Permeate mg L−1 175 ± 97 146 ± 46 65 ± 41 131 ± 78 120 ± 42 65 ± 52 
ηBIO 74 ± 8 72 ± 3 77 ± 5 77 ± 8 71 ± 1 77 ± 9 
ηTOT 84 ± 12 89 ± 7 95 ± 4 88 ± 9 91 ± 3 95 ± 5 
TOC Influent mg L−1 380 ± 54 408 ± 73 359 ± 84 380 ± 54 408 ± 73 359 ± 84 
Supernatant mg L−1 34 ± 12 45 ± 37 83 ± 23 30 ± 11 53 ± 26 100 ± 25 
Permeate mg L−1 8 ± 6 20 ± 3 21 ± 4 8 ± 5 12 ± 4 21 ± 5 
ηBIO 91 ± 5 89 ± 4 77 ± 2 92 ± 5 87 ± 3 72 ± 1 
ηTOT 98 ± 2 95 ± 1 94 ± 2 98 ± 1 97 ± 2 94 ± 1 
NH4-N Influent mg L−1 115 ± 37 167 ± 66 155 ± 75 115 ± 37 167 ± 66 155 ± 75 
Supernatant mg L−1 3 ± 1 3 ± 1 2 ± 1 2 ± 1 3 ± 2 5 ± 1 
Permeate mg L−1 2 ± 1 2 ± 1 2 ± 1 2 ± 1 2 ± 1 4 ± 1 
ηBIO 97 ± 1 98 ± 1 99 ± 1 98 ± 1 98 ± 1 97 ± 1 
ηnitrified 48 ± 2 96 ± 3 98 ± 1 34 ± 4 92 ± 1 97 ± 1 
TPH Influent mg L−1 – 1.5 ± 0.2 3.1 ± 0.5 – 1.5 ± 0.2 3.1 ± 0.5 
Permeate mg L−1 – 1.5 ± 0.2 2.8 ± 0.4 – 1.5 ± 0.2 2.9 ± 0.4 
ηTOT – <1 8 ± 2 – <1 5 ± 1 
Parameter   Units MB-MBRI
 
MB-MBRII
 
Phase I Phase II Phase III Phase I Phase II Phase III 
COD Influent mg L−1 1,095 ± 346 1,329 ± 200 1,303 ± 205 1,095 ± 346 1,329 ± 200 1,303 ± 205 
Supernatant mg L−1 285 ± 49 372 ± 62 300 ± 29 252 ± 47 385 ± 83 300 ± 41 
Permeate mg L−1 175 ± 97 146 ± 46 65 ± 41 131 ± 78 120 ± 42 65 ± 52 
ηBIO 74 ± 8 72 ± 3 77 ± 5 77 ± 8 71 ± 1 77 ± 9 
ηTOT 84 ± 12 89 ± 7 95 ± 4 88 ± 9 91 ± 3 95 ± 5 
TOC Influent mg L−1 380 ± 54 408 ± 73 359 ± 84 380 ± 54 408 ± 73 359 ± 84 
Supernatant mg L−1 34 ± 12 45 ± 37 83 ± 23 30 ± 11 53 ± 26 100 ± 25 
Permeate mg L−1 8 ± 6 20 ± 3 21 ± 4 8 ± 5 12 ± 4 21 ± 5 
ηBIO 91 ± 5 89 ± 4 77 ± 2 92 ± 5 87 ± 3 72 ± 1 
ηTOT 98 ± 2 95 ± 1 94 ± 2 98 ± 1 97 ± 2 94 ± 1 
NH4-N Influent mg L−1 115 ± 37 167 ± 66 155 ± 75 115 ± 37 167 ± 66 155 ± 75 
Supernatant mg L−1 3 ± 1 3 ± 1 2 ± 1 2 ± 1 3 ± 2 5 ± 1 
Permeate mg L−1 2 ± 1 2 ± 1 2 ± 1 2 ± 1 2 ± 1 4 ± 1 
ηBIO 97 ± 1 98 ± 1 99 ± 1 98 ± 1 98 ± 1 97 ± 1 
ηnitrified 48 ± 2 96 ± 3 98 ± 1 34 ± 4 92 ± 1 97 ± 1 
TPH Influent mg L−1 – 1.5 ± 0.2 3.1 ± 0.5 – 1.5 ± 0.2 3.1 ± 0.5 
Permeate mg L−1 – 1.5 ± 0.2 2.8 ± 0.4 – 1.5 ± 0.2 2.9 ± 0.4 
ηTOT – <1 8 ± 2 – <1 5 ± 1 

Although the nutrient removal is not the main goal of this investigation, the analysis of biological processes of nitrogen removal is important in order to complete the study of the general biomass acclimation.

In Phase I, the total nitrification (NH4-Nnitrified) performance was mainly lower than Phase II and III because the inoculum was collected from a conventional activated sludge plant where the amount of autotrophic species usually is moderate. In MB-MBRI the nitrification efficiency was higher than MB-MBRII, since the presence of attached biomass inside the cubic polyurethane sponges contributed to the growth of autotrophic bacteria able to oxidize ammonium.

This effect was lower for MB-MBRII where the biofilm contribution, at the end of Phase I, was very negligible. On the other hand, as discussed above, the amount of ammonia added to the synthetic wastewater was chosen in order to be sufficient for heterotrophic metabolism and consequently all ammonium was used for metabolism and no ammonia was left for nitrification (NH4-Nbio).

In Phase II and III, according to recent studies (Kose et al. 2012; Abdollahzadeh Sharghi et al. 2014; Di Bella et al. 2014a, b; Di Trapani et al. 2015), the presence of salt or hydrocarbons modified the conventional metabolic balance between C and N, and a lower mass of nitrogen per gram of C occurred to complete biological removal (C:N:P = 100:3:1 and not 100:5:1). Thus, in the Phase II and III, after metabolic use, there was ammonium that nitrified to NO3-N. In this context, it is important to underline that the autotrophic biomass had not undergone any metabolic inhibition. Obviously, due to the combinations of the effects, no worsening in the ammonium removal efficiencies occurred and practically all ammonium was removed or nitrified in both systems.

Concerning the EPSs production, the analysis shows a decrease of the specific bound EPSs in both plants during the whole experimentation (Figure 2(a)). More specifically, the concentration of bound EPSs decreased from 211 mg gMLVSS−1 to about 123 mg gMLVSS−1 for the MB-MBRI, and from 221 mg gMLVSS−1 to 153 mg gMLVSS−1 for the MB-MBRII. These similar trends suggested a reduction of the biological activity of the suspended microorganisms, linked to the increase of the slop volume from one phase to another.
Figure 2

Average values of bound EPSs (a) and SMPs (b) in MB-MBRI and MB-MBRII systems for each phase.

Figure 2

Average values of bound EPSs (a) and SMPs (b) in MB-MBRI and MB-MBRII systems for each phase.

Furthermore, observing the specific average concentrations of the SMPs (Figure 2(b)), the MB-MBRI showed a slight increase of concentrations in the range of 19–21 mg gMLVSS−1 from Phase I to Phase III. In contrast in the MB-MBRII the SMPs increased from 24 mg gMLVSS−1 in Phase I to 40 mg gMLVSS−1 in Phase III. This indicates that the MB-MBRI system was biologically more robust to cope with metabolic stress, since it was constituted by both suspended and attached biomass. Conversely, in the MB-MBRII system a biomass stress occurred in response to the simultaneous increase of salinity (Kose et al. 2012; Johir et al. 2013; Di Bella et al. 2014a; Di Trapani et al. 2014) and hydrocarbon concentration, resulting in a release of organic intracellular constituents as SMPs. The different phenomena confirm that the immobilized bacteria (that are present in MB-MBRI but not in MB-MBRII) are more resistant to toxicity than are suspended bacteria.

Fouling analysis

Regarding the evolution of the total resistance to filtration (RT) (Figure 3(a)), it was higher in the MB-MBRII system because suspended biomass was characterized by a higher EPST concentration and this caused a considerable increase of mixed liquor hydrophobicity (to up to 90%). Accordingly, the fouling rate (FR) (Figure 3(b)) was lower in MB-MBRI.
Figure 3

Total resistance to filtration in MB-MBRI and in MB-MBRII (a); FR in MB-MBRI and in MB-MBRII (b); specific fouling resistances in MB-MBRI (c) and in MB-MBRII (d) for each phase.

Figure 3

Total resistance to filtration in MB-MBRI and in MB-MBRII (a); FR in MB-MBRI and in MB-MBRII (b); specific fouling resistances in MB-MBRI (c) and in MB-MBRII (d) for each phase.

Nevertheless, during the Phase II until the 49th day of Phase III, there was a slight increase of FR for the MB-MBRI plant with respect to the MB-MBRII due to the partial detachment of biomass from polyurethane sponge cubes that altered the permeability of the cake layer. Probably it was mainly due to both the shear stress caused by the reciprocal impacts of the carriers and a partial inhibition of the biofilm. Afterwards, from the 49th day until the end, in the MB-MBRI system the FR was again lower than in the MB-MBRII system, due to a new entrapment of biomass (Feng et al. 2012) inside the polyurethane sponge cubes.

In terms of specific resistances, regarding the MB-MBRI system (Figure 3(c)), in Phase I the RC,rev was 0.22×1012 m−1 and RC,irr was 2.03×1012 m−1. In Phase II the probably partial detachment of biofilm mainly resulted in an increase of the RC,rev and RC,irr to 0.97×1012 m−1 and 1.03×1012 m−1, respectively. This was confirmed in Phase III (until the 49th day) in which there was a further increase of the reversible and irreversible cake until about 3.28×1012 m−1 for RC,rev and 4.84×1012 m−1 for RC,irr. After, the values of RC,rev and RC,irr decreased because new biomass was trapped inside the carriers.

Concerning the MB-MBRII system (Figure 3(d)), in Phase I the suspended biomass entirely impacted against the membrane fibres, with a subsequent predominance of RC,irr close to 5.96×1012 m−1. This value was higher than in MB-MBRI because there was not the contribution of ‘entrapment’ inside the carriers. During Phase II, the increase of SMPs, and also of hydrophobicity (93%), implied a swelling of the cake with a decrease of the irreversible cake and an increase of RPB until about 3.75×1012 m−1. In Phase III, for the first time the cake became more compact with a consequent increase of RC,rev and RC,irr to around 1.46×1012 m−1 and 4.13×1012 m−1, respectively. Subsequently, at the end of Phase III a further increase of SMPs, together with a further increase of mixed liquor hydrophobicity (94%), resulted in an additional swelling of cake with a consequent decrease of reversible and irreversible cake and an increase of RPB to around 4.96×1012 m−1.

CONCLUSIONS

This technical note defined two important aspects, which were not investigated in the literature, concerning the treatment of saline oily wastewater:

  1. the conventional activated sludge can be adapted to a simultaneous hydrocarbons and salinity increase in the influent;

  2. for an MB-MBR, the analysis of general performance, EPSs production and fouling mechanisms suggested that the best carriers for biofilm are the cubic sponges, since they guaranteed an important ‘entrapment effect’ that fixed the suspended biomass and allowed the subsequent attached biomass growth.

Nevertheless, the TPH removal requires a longer acclimation period in order to be removed in a satisfactory manner. Despite this, the MB-MBR with polyurethane sponge cubes resulted in the best potential the TPH removal.

ACKNOWLEDGEMENTS

This research was funded by the National Operational Programme for Research and Competitiveness 2007–2013, PON 01_01844, as a part of the project ‘SIBSAC – An integrated system for sediments remediation and high salinity marine wastewater treatment’ (Unique Project Code: CUP B71C11000520005).

REFERENCES

REFERENCES
APHA
2005
Standard Methods for the Examination of Water and Wastewater
.
American Public Health Association/American Water Works Association/Water Environment Federation
,
Washington, DC
,
USA
.
Chang
I. S.
Bag
S. O.
Lee
C. H.
2001
Effects of membrane fouling on solute rejection during membrane filtration of activated sludge
.
Process Biochemistry
36
(
8–9
),
855
860
.
Di Bella
G.
Di Trapani
D.
Freni
G.
Torregrossa
M.
Viviani
G.
2014a
Analysis of biomass characteristics in MBR and MB-MBR systems fed with synthetic wastewater: influence of a gradual salinity increase
.
Chemical Engineering Transactions
38
,
445
450
.
Di Trapani
D.
Christensson
M.
Torregrossa
M.
Viviani
G.
Ødegaard
H.
2013
Performance of a hybrid activated sludge/biofilm process for wastewater treatment in a cold climate region: influence of operating conditions
.
Biochemical Engineering Journal
77
,
214
219
.
Di Trapani
D.
Di Bella
G.
Mannina
G.
Torregrossa
M.
Viviani
G.
2015
Effect of C/N shock variation on the performances of a moving bed membrane bioreactor
.
Bioresource Technology
189
,
250
257
.
Dubois
M.
Gilles
K. A.
Hamilton
J. K.
Rebers
P. A.
Smith
F.
1956
Colorimetric method for determination of sugars and related substances
.
Analytical Chemistry
28
(
3
),
350
356
.
Guo
W.
Ngo
H. H.
Li
J.
2012
A mini-review on membrane fouling
.
Bioresource Technology
122
,
27
34
.
Johir
M. A. H.
Vigneswaran
S.
Kandasamy
J.
BenAim
R.
Grasmick
A.
2013
Effect of salt concentration on membrane bioreactor (MBR) performances: detailed organic characterization
.
Desalination
322
,
13
20
.
Kose
B.
Ozgun
H.
Ersahin
M. E.
Dizge
N.
Koseoglu-Imer
D. Y.
Atay
B.
Kaya
R.
Altinbas
M.
Sayili
S.
Hoshan
P.
Atay
D.
Eren
E.
Kinaci
C.
Koyuncu
I.
2012
Performance evaluation of a submerged membrane bioreactor for the treatment of brackish oil and natural gas field produced water
.
Desalination
285
,
295
300
.
Lowry
O. H.
Rosebrough
N. J.
Farr
A. L.
Randall
R. J.
1951
Protein measurement with the Folin phenol reagent
.
Journal of Biological Chemistry
193
(
1
),
265
275
.
MARPOL 73/78
International Convention for the Prevention of Pollution from Ships, 1973, as modified by the protocol of 1978 relating thereto. http://www.imo.org
.
Meng
F.
Chae
S. R.
Drews
A.
Kraume
M.
Shin
H. S.
Yang
F.
2009
Recent advances in membrane bioreactors (MBRs): membrane fouling and membrane material
.
Water Research
43
(
6
),
1489
1512
.
Pendashteh
A. R.
Abdullah
L. C.
Fakhru'L-Razi
A.
Madaeni
S. S.
Zainal Abidin
Z.
Awang Biak
D. R.
2012
Evaluation of membrane bioreactor for hypersaline oily wastewater treatment
.
Process Safety and Environmental Protection
90
(
1
),
45
55
.
Stephenson
T.
Judd
S.
Jefferson
B.
Brindle
K.
2000
Membrane Bioreactors for Wastewater Treatment
.
IWA Publishing, London
.
Sun
C.
Leiknes
T.
Weitzenböck
J.
Thorstensen
B.
2010
Development of a biofilm-MBR for shipboard wastewater treatment: the effect of process configuration
.
Desalination
250
(
2
),
745
750
.
Sun
C.
Leiknes
T.
Fredriksen
R. H.
Riviere
E.
2012
Comparison of membrane filtration performance between biofilm-MBR and activated sludge-MBR
.
Desalination and Water Treatment
48
(
1–3
),
285
293
.
Woolard
C. R.
Irvine
R. L.
1995
Treatment of hypersaline wastewater in the sequencing batch reactor
.
Water Research
29
(
4
),
1159
1168
.
Zhang
X.
Bishop
P. L.
Kinkle
B. K.
1999
Comparison of extraction methods for quantifying extracellular polymers in biofilms
.
Water Science and Technology
39
(
7
),
211
218
.