Comparative Studies on Flotation Performance of Saturated Fatty Acids and Unsaturated Fatty Acids Separated from Hogwash Oil

Round 1

Reviewer 1 Report

The manuscript shows the reasons why unsaturated fatty acids are better collectors for quartz flotation than saturated hydrocarbons. These include the differences in solubility and ionization. The authors found ways to isolate the different isomers from a hogwash oil, and conducted flotation tests, zeta-potential measurements, and FTIR studies. The species distribution diagrams are also useful. In these regards, the manuscript should be accepted for publication.

Some editorial changes are recommended, however.

1. "polyvalent" ions should be changed to "divalent" ions (p. 1, line 31)
2. SiO2 is misspelled. (Table 1)
3. The MFA is comprised of unsaturated fatty acids... (p.3, line 118)
4. reduced the negative zeta potential of... (p.6, line 185)
5. The authors show chemical shifts of the C=O bonds in Figure 7. But the changes are too small to be seen clearly. 
6. The authors claim that the UFA has 1.6 times higher RCOO-  concentrations than the SFA at pH > 10. It is difficult to see the difference in the species distribution diagram. (figure 6)   

Author Response

Response to reviewer 1 comments 

Dear sir or madam:

Point1: "polyvalent" ions should be changed to "divalent" ions (p. 1, line 31)

Response 1: Thanks so much for your advices. We would like to retain the expression “polyvalent ions” after careful consideration and discussion with other authors, because the Fe3+ also has a certain activated effect on quartz flotation.

Point2: SiO2 is misspelled. (Table 1)

Response 2: Thank you so much for your careful check. We feel so sorry for our carelessness. The label mistake has been corrected in the revised manuscript.

Point3: The MFA is comprised of unsaturated fatty acids... (p.3, line 118)

Response 3: I appreciate it for your careful check, the wrong one “comprised” has been changed to the right one “is comprised of”.

Point4: reduced the negative zeta potential of... (p.6, line 185)

Response 4: Thank you very much for suggestions. We totally agree that the description of "…reduced the negative zeta potential of…(p.6, line 185)" is much better.

Point5: The authors show chemical shifts of the C=O bonds in Figure 7. But the changes are too small to be seen clearly.

Response 5: Thank you very much for your professional comment. The absorption peak of the C=O here is compared to that of the original SFA and UFA (pure reagent). However, we did not show FT-IR spectrum of the pure SFA and UFA in the current work since we have published them in our previous work which entitled “the separation principle of hogwash oil and the flotation performance of desilication for industrial iron ore”. So, if you want to find them, I would definitely like provide them for your. In order to avoid data duplication, the FT-IR spectrum of the pure SFA and UFA was not displayed. In the corrected manuscript, we have cited our previous work to emphasize that it is compared with the FT-IR spectrum of the original SFA and UFA before adsorption. That is, the C=O peak of pure SFA and UFA is at 1709 cm-1, but after adsorbing on quartz surface, the C=O peak shifts to 1793 cm-1. Thus, the C=O group strengthened and the chemical adsorption should occur at the carboxyl group.

Point6: The authors claim that the UFA has 1.6 times higher RCOO- concentrations than the SFA at pH > 10. It is difficult to see the difference in the species distribution diagram. (figure 6)  

Response 6: Thank you for your suggestions. We would like to explain that the difference is not obvious because the vertical axis in Figure 6 represents the logarithm of concentration (log c).

Thank you so much!

Best!

Reviewer 2 Report

This manuscript reports an interesting experimental investigation of flotation of quartz with saturated (SFA) and unsaturated (UFA) long-chain carboxylic acids as collectors. The results look new and reliable, being worthy of publication in Minerals journal after some corrections. Scientifically, this reviewer has some doubt regarding the form of uptake of UFA and, especially SFA. The authors believe that both species are attached on activated quartz surface “through chemisorption and hydrogen bonding; however, the adsorption of UFA was much stronger and more favorable” (lines 20-21 in Abstract and then in the text). In my opinion, both FTIR and XPS spectra hint that the adsorbate is connected on calcium sites via carboxylic group rather than “hydrogen bonding”. Particularly, the IR band at 3420 cm^-1 attributed to OH vibrations on the surface is the same before and after adsorption of SFA and UFA. Meanwhile, the XPS found shifts both COOH and Ca 2p lines for UFA but not SFA. It is also noteworthy that the share of carboxylic group seems to be higher for SFA than UFA. I may guess that SFA form a kind of micelle or nanoparticles rather that adsorb as molecule or anions on the quartz surface. Nevertheless, interpretation of the mechanism is authors’ responsibility, and I don’t insist on this hypothesis.

There are a number of minor points which should be corrected.

Line 85: “Collectors were … diluted to 2% solution”. Please explain this, the concentrations of the solutions used are actually much lower.

Line 100: “50 mg/mL CaCl₂ and then treated with 100 mg/mL…” Should it be 50 mg/L and 100 mg/L?

Line 110-111  “The sputtering source was Al Kα X-rays” is incorrect.

Section 3.4. FT-IR analyses. It is more common to use terms “band” and “vibration”, etc. but not “peak” for transmission FTIR spectra; no “peaks” are seen in Fig.7.

Again, please note that the band at 3420 cm^-1 is roughly the same in all the spectra.

Section 3.5. XPS analysis.

Please reduce the number of significant digits for concentrations and binding energies, e.g., there should be 13.8 or 14% and 284.6 eV instead of “13.83 (284.65)” and so forth in Table 3 and throughout the text.

Lines 239 and 243: “Ca (2p, 2.27%, 346.52 eV) adsorption” should be corrected to something like “Ca adsorption (Ca2p_3/2 at 346.5 eV, concentration of 2.3 at.%)…”. By the way, do the concentrations of Ca differ by 7 times for the SFA- and UFA-treated samples? Why? This is not obvious from the survey spectra presented.

Line 262-264: “The components of C-OH bonds were found in both samples treated with SFA and UFA and their binding energies of were almost same. This indicated that the adsorptions of both SFA and UFA involved hydrogen bonding.”  This statement is invalid as alcohol C-OH group at BE 286.8 eV is not the same as O=COH in carboxylic groups.

Figure 10. The designation “CO=O” is wrong (leaving aside the idea of hydrogen bonding).

There are a number misprints, for example in lines 100, 263, etc.

This manuscript reports an interesting experimental investigation of flotation of quartz with saturated (SFA) and unsaturated (UFA) long-chain carboxylic acids as collectors. The results look new and reliable, being worthy of publication in Minerals journal after some corrections. Scientifically, this reviewer has some doubt regarding the form of uptake of UFA and, especially SFA. The authors believe that both species are attached on activated quartz surface “through chemisorption and hydrogen bonding; however, the adsorption of UFA was much stronger and more favorable” (lines 20-21 in Abstract and then in the text). In my opinion, both FTIR and XPS spectra hint that the adsorbate is connected on calcium sites via carboxylic group rather than “hydrogen bonding”. Particularly, the IR band at 3420 cm^-1 attributed to OH vibrations on the surface is the same before and after adsorption of SFA and UFA. Meanwhile, the XPS found shifts both COOH and Ca 2p lines for UFA but not SFA. It is also noteworthy that the share of carboxylic group seems to be higher for SFA than UFA. I may guess that SFA form a kind of micelle or nanoparticles rather that adsorb as molecule or anions on the quartz surface. Nevertheless, interpretation of the mechanism is authors’ responsibility, and I don’t insist on this hypothesis.

There are a number of minor points which should be corrected.

Line 85: “Collectors were … diluted to 2% solution”. Please explain this, the concentrations of the solutions used are actually much lower.

Line 100: “50 mg/mL CaCl₂ and then treated with 100 mg/mL…” Should it be 50 mg/L and 100 mg/L?

Line 110-111  “The sputtering source was Al Kα X-rays” is incorrect.

Section 3.4. FT-IR analyses. It is more common to use terms “band” and “vibration”, etc. but not “peak” for transmission FTIR spectra; no “peaks” are seen in Fig.7.

Again, please note that the band at 3420 cm^-1 is roughly the same in all the spectra.

Section 3.5. XPS analysis.

Please reduce the number of significant digits for concentrations and binding energies, e.g., there should be 13.8 or 14% and 284.6 eV instead of “13.83 (284.65)” and so forth in Table 3 and throughout the text.

Lines 239 and 243: “Ca (2p, 2.27%, 346.52 eV) adsorption” should be corrected to something like “Ca adsorption (Ca2p_3/2 at 346.5 eV, concentration of 2.3 at.%)…”. By the way, do the concentrations of Ca differ by 7 times for the SFA- and UFA-treated samples? Why? This is not obvious from the survey spectra presented.

Line 262-264: “The components of C-OH bonds were found in both samples treated with SFA and UFA and their binding energies of were almost same. This indicated that the adsorptions of both SFA and UFA involved hydrogen bonding.”  This statement is invalid as alcohol C-OH group at BE 286.8 eV is not the same as O=COH in carboxylic groups.

Figure 10. The designation “CO=O” is wrong (leaving aside the idea of hydrogen bonding).

There are a number misprints, for example in lines 100, 263, etc.

Author Response

Response to reviewer 2 comments  

Dear sir or madam:

Point1: Line 85: “Collectors were … diluted to 2% solution”. Please explain this, the concentrations of the solutions used are actually much lower.

Response 1: Thanks so much for your professional comment. The concentration of 2% is wrong and it should be 0.2%. It has been corrected in the manuscript.

Point2: Line 100: “50 mg/mL CaCl₂ and then treated with 100 mg/mL…” Should it be 50 mg/L and 100 mg/L?

Response 2: Thanks very much for your careful check. The mistake has been corrected in the revised manuscript.

Point3: Line 110-111 “The sputtering source was Al Kα X-rays” is incorrect.

Response 3: The sentence “The sputtering source was Al Kα X-rays…” has been rewritten as “The X-ray source was monochromatized Al Kα (1486.7 eV)…”.

Point4: Section 3.4. FT-IR analyses. It is more common to use terms “band” and “vibration”, etc. but not “peak” for transmission FTIR spectra; no “peaks” are seen in Fig.7. Again, please note that the band at 3420 cm^-1 is roughly the same in all the spectra.

Response 4: According to your advice, we have corrected the description of “peak” in the revised manuscript. We also agree that the proof of hydrogen bond is not sufficient. It is true that the positions of hydrogen bond and -OH bond in the infrared spectrum are very similar. The vibration around 3420 cm-1 in the FTIR spectra of quartz treated by SFA and UFA is very similar, but compared with the infrared spectrum of pure quartz, there are weak differences, including position and shape. You know, hydrogen bonds are difficult to detect directly and there is no effective way to prove the existence of hydrogen bonds so far.

Point5: Section 3.5. XPS analysis. Please reduce the number of significant digits for concentrations and binding energies, e.g., there should be 13.8 or 14% and 284.6 eV instead of “13.83 (284.65)” and so forth in Table 3 and throughout the text.

Response 5: Thank you for the advice. According to your suggestions, the number of significant digits has been reduced and unified in the revised manuscript including the text and table 3.

Point6: Lines 239 and 243: “Ca (2p, 2.27%, 346.52 eV) adsorption” should be corrected to something like “Ca adsorption (Ca2p_3/2 at 346.5 eV, concentration of 2.3 at.%)…”. By the way, do the concentrations of Ca differ by 7 times for the SFA- and UFA-treated samples? Why? This is not obvious from the survey spectra presented.

Response 6: According to your suggestion, the description of “Ca (2p, 2.27%, 346.52 eV) adsorption” has been corrected to “Ca adsorption (Ca2p_3/2 at 346.5 eV, concentration of 2.3 at. %)”, also the unit of concentration “%” has been changed as “at.%” in this section. The concentrations of Ca for SFA-treated samples is about 7 times higher than that of UFA-treated samples, because the adsorption of UFA on activated quartz surface is much stronger and more favorable, so more UFA molecules covers on Ca ions, resulting in a lower content of Ca ions detected. The Ca concentrations of the SFA- and UFA-treated samples are low and the peaks in both survey spectra are weak, so the 7 times difference between them is not obvious. But if the image is enlarged, it will be found that the Ca peak of UFA treated sample is much weaker.

Point7: Line 262-264: “The components of C-OH bonds were found in both samples treated with SFA and UFA and their binding energies of were almost same. This indicated that the adsorptions of both SFA and UFA involved hydrogen bonding.”  This statement is invalid as alcohol C-OH group at BE 286.8 eV is not the same as O=COH in carboxylic groups.

Response 7: After discussion with other authors, we agree that the expression of " C-OH group" is incorrect and should be " O=COH", which has been corrected in both manuscript and Fig.9. And the statement of “involved hydrogen bonding” has been changed to “may involve hydrogen bonding”. We would hope to retain the conclusion about hydrogen bonding.

Point8: Figure 10. The designation “CO=O” is wrong (leaving aside the idea of hydrogen bonding).

Response 8: Thank you so much for careful check. The right structure should be “O=CO” and it has been corrected in Fig. 10.

Point9: There are a number misprints, for example in lines 100, 263, etc.

Response 9: We have corrected the misprints in lines 100 and 263 and carefully checked the whole manuscript again.

Thank you very much!

Best!