Effect of Surfactants on Reverse Osmosis Membrane Performance

Round 1

Reviewer 1 Report

Dear Editor,

Thank you in advance for the opportunity of evaluating this paper. 

Herein, I submit my comments for initial submission of the manuscript entitled: “Effect of surfactants on Reverse Osmosis membrane performance”.

Comments for author File: Comments.pdf

Author Response

Responses to Reviewer 1 Comments

Herein, I submit my comments for initial submission of the manuscript entitled: “Effect of surfactants on Reverse Osmosis membrane performance”. The authors present the separation of various surfactant model aqueous solutions through reverse osmosis at laboratory scale using cellulose acetate and polyamide membranes. The performance of the membranes was evaluated by determining permeate flux and total organic carbon (TOC) rejection. I consider this manuscript suitable for publication. However, I came to a conclusion to encourage you to review it again. Please, use my comments listed below:

Response: First, the authors are very thankful to the reviewer for taking time to review our manuscript so thoughtfully and for judging this study as suitable for publication. The points raised by the reviewer will be addressed accordingly.

 

 

Point 1: The title seems like authors did the modification of the membrane with surfactants, instead of evaluating the use of RO in the removal of these pollutants from water. Then, I suggest reviewing and changing the title to one that describes what was done in the study.

Response 1: In our work we did not study the modification of the RO membrane with surfactants. We evaluated the use of RO membranes in the removal of these surfactants. Therefore, the present title “Effect of surfactants on Reverse Osmosis membrane performance” might reflect the purpose of the research.

 

 

Point 2: In introduction, lines 78 to 80 need a reference.

Response 2: Lines 78 to 80 of the original manuscript are a brief summary of the study of Archer et al. [27]. In the updated manuscript, the expression “ In their study,” was inserted in Line 79 in the revised manuscript.

 

 

Point 3: Equations must be referenced.

Response 3: We include the relevant references of the equations (1) and (2) in the revised manuscript. Lines 165 and 170.

No reference for equation (3) as it was proposed in this manuscript (inspired from equation (2))

 

 

Point 4: Further discussion of the results comparing with the literature is needed. Chapter 3 is just the presentation of the results and lacks, aside the comparison of the results with previous studies, in the physical interpretation of the obtained results.

 

Response 4: We agree on the reviewer’s point to discuss more our findings. In the revised manuscript, further discussion and comparison with previous study was included.

The following parts were added in the revised manuscript:

 

“For all surfactant concentration, the lowest permeate flux was obtained with cationic surfactant. Permeate flux behavior of PA membrane was decided by the interaction between surfactant and the membrane surface which is in accordance with previous study [33].”Lines 209-212

 

“For concentration range 100 -10 000 mg/L, different surfactants did not affect the permeate flux of CA membrane contrarily to the case of PA membrane. We assume that interactions between surfactant and CA membrane surface did not take place. Further discussions of interaction between RO membranes and different surfactants will be discussed in the following part.” Was added to the revised manuscript. Lines 222-227

 

“Similar behavior of permeate flux decline of CA and PA RO membranes was reported elsewhere [38]. In this study, the higher fouling rate of PA compared to CA membranes was attributed to its inherent surface roughness which could increase the surfactant adsorption onto the membrane surface in our study.” Line 264-267

 

“In this case, electrostatic attractions between hydrophilic groups of SO and CTAB occurred, and surfactant aggregate was formed. Our findings agree well with previous study [41].” Line 352-354

 

In addition, text was added to Figure 7 to better understand the interaction between RO membranes and surfactants.

 

 

Point 5: The conclusions must be reviewed, this section should present the meaningful findings, not a resume of the results.

Response 5: Following the recommendation of the reviewer, the conclusion part was reviewed. The conclusion was divided in three parts: first a summary of results followed by the meaningful findings then some recommendations. In the revised manuscript the following modifications was done:

“due to surfactant adsorption on the PA membrane surface.” Line 408-409

“Membrane surface morphology can also influence surfactant adsorption. Adsorption was more important at the rough surface of the PA membrane compared to the relatively smooth surface of CA membrane” was moved from Lines 376-378 in the original manuscript to Lines 421-423 in the revised manuscript.

“CA membrane showed stable permeate flux during the separation of different surfactants at different concentration. We assume that no apparent interactions between surfactant and membrane surface took place.” Was added to the revised manuscript Lines 418-420.

“Through this research, better understanding of interactions between surfactant monomers and membrane surface was provided.  Despite the attractive water permeability of fresh PA membrane, permeate flux was deteriorated during the separation of nonionic and cationic surfactants aqueous solutions.  Satisfying results of permeate flux was observed with anionic surfactant.” Was added in Lines 424-428 in the revised manuscript.

“The high surfactant rejections gave RO membranes used in this research a promising potential in the treatment of surfactants contaminated wastewater compared to conventional methods of surfactant removal from wastewater. However, permeate flux decline due to surfactant adsorption can be a major drawback for scaling up this process. Thus, a proper choice of membrane material, depending on the type of surfactant discharged in the wastewater, is recommended in order to mitigate membrane fouling by surfactant adsorption. PA membrane is more effective in treating wastewater containing anionic surfactant. This type of membrane should be avoided while treating wastewater with cationic surfactant.  CA membrane is efficient in treating wastewater containing different types of surfactants.” Was added to added in the revised manuscript. Lines 430-439

“Hydrophilic, weak-charged with smooth surface RO membrane is suitable for treating surfactant-contaminated wastewater” was moved from Lines 396-397 in the original manuscript to Lines 440-441

 

 

Point 6: Is the implementation of the process economically viable? If so, in what situations? Thus, considering the cost of the membranes.

Response 6: Thank you for raising the economic feasibility issue.

 

At present, the RO membrane process for removal of surfactants from wastewater are in the basic study stage and the economics of the RO membrane process has not been studied well. Most common conventional methods for removal of surfactants from wastewater are biological and chemical by coagulation-flocculation or adsorption, or combined chemical and biological treatments using Advanced Oxidation Processes as pre-treatment methods. Compared to the conventional methods, membrane processes, especially RO membrane processes, provide high-quality treated water with low surfactant concentrations; RO membrane processes are simple and require less treatment energy, so the process is expected to be economically viable.

With the scarcity of water resources in arid and semi-arid regions, it is necessary to find alternative water sources. Reuse of the treated wastewater could be a promising alternative. This process is much cheaper than seawater desalination as sewage and industrial wastewater contain lower concentration of salt. With a proper choice of membrane material, we can mitigate membrane fouling, which lead to controlling the energy consumption and extending the life of membrane. To cover membrane and energy cost, reuse of the treated wastewater is important. This treated wastewater, due to its good quality, can be reused in many industrial processes with high added value such superconductors and chemical industries.

 

Based on the above questions, we have included in the revised manuscript the section “3.3 Process implementation and scalability” in Line 362 :

The following text was added to this new section in the revised manuscript:

 

“At present, the RO membrane process for removal of surfactants from wastewater are in the basic study stage and the economics of the RO membrane process has not been studied well. Most common conventional methods for removal of surfactants from wastewater are biological and chemical by coagulation-flocculation or adsorption, or combined chemical and biological treatments using Advanced Oxidation Processes as pre-treatment methods. Compared to the conventional methods, membrane processes, especially RO membrane processes, provide high-quality treated water with low surfactant concentrations as presented in this study; RO membrane processes are simple and require less treatment energy, so the process is expected to be economically viable.

With the scarcity of water resources in arid and semi-arid regions and competition between water usages (drinking, agriculture, industry, tourism,), it is necessary to find alternative water sources. Reuse of the treated wastewater could be a promising alternative. This process is much cheaper than seawater desalination as sewage and industrial wastewater contain lower concentration of salt which require lower operating pressure. Bai et al. economically evaluated the full-scale MF and RO process for wastewater treatment at the Changi NEWater Project in Singapore with production capacity of 228 000 m3/d [44]. They reported treatment cost between 0.08 and 0.15 $/ m3 . On the other hand, seawater desalination cost of the Shuqaiq 3 desalination plant in Saudi Arabia was  around 0.51 $/ m3  at a capacity of 380 000 m3/d [45].

 

With a proper choice of membrane material, membrane fouling could be mitigated leading to better control of energy consumption and extending the life of membrane. To cover membrane and energy cost, due to its good quality, treated wastewater can be reused in many industrial processes with high added value such superconductors and chemical industries [47]” Lines 369-393

 

 

Point 7: Is the process scalable?

Response 7: At laboratory scale in dead-end configuration, both RO membranes used in this research exhibited promising results. Further investigation of membrane behaviors should be performed in crossflow configuration. Also cleaning tests of especially polyamide membrane should be performed. From our findings, removal of different types of surfactants using RO membranes have a good potential for application on a commercial scale. Characterization of wastewater and interactions with membrane material is important to guarantee satisfactory results, mitigate membrane fouling, thus extend the lifetime of the membranes. We assume that RO process for surfactant removal is scalable.

 

In the revised manuscript, the following sentences were added:

“Regarding the scalability of RO process in the removal of surfactant from wastewater, due to promising surfactant removal results we assume that the RO process for surfactant treatment is scalable. However, further investigation in laboratory scale crossflow configuration should be performed. Also membrane cleaning tests should be studied beforehand. As reported in our study, characterization of surfactant contaminated wastewater, interactions with the RO membrane material are with high importance to guarantee good performance of the process and treatment cost.” Lines 394-400

 

Author Response File: Author Response.docx

Reviewer 2 Report

Please see comments attached. 

Comments for author File: Comments.pdf

Author Response

Responses to Reviewer 2 Comments

 

Halleb et al reported the effect of different surfactants on reverse osmosis membrane performance, three different types of surfactants from cationic, anionic and nonionic surfactants on the performance of Cellulose acetate and polyamide reverse osmosis membranes were investigated. The overall logic of the paper is reasonable, but major revisions are needed.

 

Point 1: In the introduction, authors should elucidate more background on why PA and CA were selected as RO membranes, why not other materials? On Page 3 Line120-130, only a short explanation for PA but no explanation for why CA was chosen as the RO membrane material.

 

Response 1: we acknowledge the importance of explaining the reason behind our choice of membrane materials.

 

CA membrane was invented 60 years ago and was mainly used for industrial processes and ultrapure water production. It is still used today because of its high chlorine resistance and mechanical stability.

PA membrane is the dominant commercial RO membrane due to its high water flux and rejection compared to CA membrane.

 

The following text was included in the revised manuscript:

“Because PA membrane is most common commercial RO membranes, and CA membrane is still used widely, and compared as control [29], both PA and CA membranes are also used in the present study.” Lines  92-94

 

“ Polyamide (PA) and Cellulose Acetate (CA) RO membranes were selected in this study as they are the most  commonly used in water purification and industrial process today.” Lines 117 to 119

 

“Loeb-Sourirajan invented the CA membrane 60 years ago. This membrane was widely used till 1980. Despite its relatively low permeate flux, CA membrane is still used today because of its high chlorine resistance and good mechanical stability [34] ”. Lines 133 to 136

 

Point 2 : Page 5 3.1 Line 181, should it be Table 3 instead of table 2? The table needs to be relabeled for each column, it is confusing, instead of ‘concentration’, I think it should be zeta potential at 50mg/L, 100mg/L…

Response 2: “Table 2” was changed by “Table 3”. Also, Table 3 was relabeled in the revised manuscript.

 

The following text was also added in this section in the revised manuscript:

“The negative zeta potential for Tw 20 is due to the association of hydroxide ions with the hydrophilic head of the surfactant [36]. Zeta potential slightly changed with the surfactant concentration. The zeta potential of the SO + CTAB mixture was -16. 7 ± 1.8 mV. The mixture of anionic and cationic surfactants results the formation of aggregates. The negative zeta potential of this mixture indicated that the formed aggregates are negatively charged.” Lines 191-197

 

 

Point 3: For 3.1.1, authors only described the change of permeate flux, but what are the reasons for these changes need to be discussed in detail. Also, if you listed the Figures in the main document, then you need to refer to each figure in the text, not only one generalized sentence like on Page 5 line 187.

Response 3: Following the comment raised by the reviewer, part 3.1.1 was reviewed and more discussion on the reasons of fluxes changes as function of time was included. The main reason of the changes is the interactions between the surfactant and the PA membrane surface. We assume that no interactions for the case of the CA membrane. In depth discussion was presented in part 3.1.2.

In the revised manuscript, the following modifications were made:

 

“As shown in Figure 2 (a-h), more separation time in needed to collect 10 % permeate during separation of cationic surfactant.” Added to the revised manuscript. Lines 205-206

 

“For all surfactant concentration, the lowest permeate flux was obtained with cationic surfactant. Permeate flux behavior of PA membrane was decided by the interaction between surfactant and the membrane surface which is in accordance with previous study [33].” Added to the revised manuscript. Lines 209-212

 

“For the case of CA membrane, permeate fluxes were almost constant for all feed composition. The permeate flux of CA membrane did not change with time confirming that the steady state was reached from the beginning of the separation test.” Was moved from Lines 210-212 in the original manuscript to Lines 219-222 in the revised manuscript.

 

“For concentration range 100 -10 000 mg/L, different surfactants did not affect the permeate flux of CA membrane contrarily to the case of PA membrane. We assume that interactions between surfactant and CA membrane surface did not take place. Further discussions of interaction between RO membranes and different surfactants will be discussed in the following part.” Was added to the revised manuscript. Lines 222-227

 

 

Point 4:  Figures should be referred to in order, 3.1.2 Line 209 figure 4 was first described, and then jumped back to Figure 3.

Response 4: We revised this point in the updated manuscript.

 

Point 5: It should be possible to obtain the cross-section of two membranes through microscopic characterizations instead of cartoons, for example, SEM should be able to look at the cross section of these membranes.

Response 5: Following the suggestion of the reviewers, SEM cross section characterization of both RO membrane was performed.

The following text was included in the revised manuscript:

 

“Surface morphology (SEM) of fresh PA and CA membranes was characterized using SEM imaging (TM-1000 Miniscope, Hitachi High Technologies, Tokyo, Japan), at an accelerating voltage of 15 kV.  Samples of the fresh flat sheet RO membranes were manually cross-sectioned using clean scissors and the edge of the cut section were attached perpendicularly to a metal stubs using double-sided adhesive tapes.” Lines 139-143.

Also, the schematic representation was replaced by obtained cross section SEM images of both RO membranes and Figure 6 was updated.

 

Point 6: It is unclear how many times did the authors repeat these experiments, however, none of the Figures from 2-5 and Figure 8-9 have error bars. These results will be more persuasive with error bars.

 

Response 6: We acknowledge the concern of the reviewer regarding the repetition of separation experiments. Due to the wide range of surfactant concentration, the experiments were performed once. Regarding the permeate flux analysis, relative flux (RF) was used in this study to remove the difference of initial pure water flux of the RO membranes at each separation test.

In the revised manuscript, we included “ Due to the wide range of surfactant concentration, separation experiments were performed one time.” Lines 156-157

 

Point 7:  Figure 7 is the most important figure among all, but it is not self-explanatory, the authors should consider how to better way to present these interactions between RO membranes and different surfactants, text can be added to the scheme if necessary.

 Response 7: We agree that Figure 7 is not self-explanatory, in the revised manuscript text was added to the scheme to explain the type of interaction between the surfactant and the membrane surface.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The manuscript has improved and can be published in its present form.

Reviewer 2 Report

All comments were addressed, suitable for publication.