Drying Oil and Natural Varnishes in Paintings: A Competition in the Metal Soap Formation
Round 1
Reviewer 1 Report
The two main weaknesses of the manuscript are i) the lack of representative (or even simplified) compositions of the various materials / reagents and chemical reactions occurring in order to give an accurate description of the chemistry and kinetics; ii) the lack of precise attribution of FT-IR spectra and of the precise mention of the references used.
Add a representative / simplified formula for metallic soap in the summary and in the introduction.
The main expected / demonstrated reaction should be given.
Lines 68-71: potassium is involved but the active ions are oxonium ions and protons which replace potassium by ion exchange in the smalt (see for example Solid State Ionics 179, 2008, 2142-2154).
Page 3, lines 97-99: the authors should go into more detail and formulate the reactions. The size of the potassium ions is the same as that of the oxonium ion (H3O +) which promotes ion exchange in particular for potassium silicate glasses (see above the Solid State Ionics reference and the references cited in the article). References must be added: for example line 99 “Bibliography suggests that metal soaps in smalt painted …. pigment”; no reference is given. Etc.
Figure 1: decrease the size of the figure but increase the size of the labels; replace the comma by a point for the separation between the unit and the decimal number; label the main and most characteristic peaks or number the peaks and add a table listing the assignments and references / arguments used for the assignment: nu OH, nu aromatic CH, nu CH2, nu CH3, nu C = O, nu Si- O, etc. The addition of the spectra of pure smalt and pure oil will increase the interest and the didactic character of the work. The fact that the work focuses on the simplified reaction between KOH and acids should be better emphasized in the abstract and introduction.
Figure 1 and Figure 2 should be merged into a single figure for a better comparison of the spectra, in particular to properly compare the replacement of the smart-oil mixture by a KOH-oil.
Lines 132-135: as proved by the strong increase of the nu OH (H2O) band, a counterpart delta OH increases, centred at 1620 cm-1; this point should be considered. The decrease of the ca. 3080 cm-1 peak (arom cycles) gives a good signal to follow the advancement of the reactions. This point should be discussed. A table comparing the peaks of the figures with literature (e.g. DOI: 10.1016/S0304-386X(03)00137-3) should be added. As Figure 1 the drawing of the figures must be improved (label size, dot in place of coma for decimal number, etc.). The spectrum of pure acid should be added for a better comparison.
Water vapour and CO2 gas signals (narrow peaks at ca. 2250, 1650 cm-1, etc.) increase in Figures 3 to 7; why? Add comments also about the sequence of narrow peaks in Figures 5 to 7 between 1300 and 1100 cm-1.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This manuscript reports the evaluation of metal soaps formation in presence of palmitic and abietic acid and potassium hydroxide by means of FTIR spectroscopy. The results illustrate the interaction between carboxylates and different types of acids. The acids showed different reactivity and ability to produce carboxylates. Although the study is carried out in a more simplified condition compared to the reactions that occur in a painting, the results are very interesting and highlight the role that the acids in natural resins used for the preparation of varnishes can play in the formation of metal soaps. It would be interesting to compare these results with data collected from historical paintings.
I think potassium hydroxide can easily form soaps, and I was wondering if the same experiment has been carried out using other types of pigment (like zinc oxide or lead white), and if the Authors obtained the same results as with potassium hydroxide. I think it would be interesting to add these data to complete the study.
I have few comments to be addressed below:
Line 15: change “painting” to “paint”
Line 24: I suggest removing “conservative” before lifetime.
Line 42: I suggest using “efflorescence” instead of “exudations”.
Line 48-50: I think it is worthy mentioning that metal soaps are often not soluble in traditional solvents used in conservation.
Line 55: please add “particles” after pigment.
Line 57: please use “however”, instead of “but”.
Line 58-60: this sentence is not clear and very long, please try to shorten it.
Line 60: please add references here.
Line 62: please add “can” before “come”.
Line 66-67: please change as follows: In this work the carboxylates formation and kinetics have been studied by Fourier transform infrared (FTIR) spectroscopy, working in transmission and with an attenuated total reflectance (ATR) accessory, exploiting two representative acids of painting oils and natural varnishes (palmitic and abietic acid respectively) and potassium hydroxide (KOH) as cation source.
Line 70-72: I think in this part it should be specified that smalt is one of the most reactive pigment in metal soaps formation in presence of drying oils.
Line 75: was it used a metal soaps with K cation in this case too?
Line 79: it is not very clear the procedures followed by the Authors. I suggest explaining better the preparation of the samples. For example, in line 88, the Authors refer to “mixtures”. It is not clear the composition of these mixtures. Did you mix palmitic acid with abietic acid and potassium hydroxide? Or did you mix each single acid with potassium hydroxide? Please add information about quantities and concentrations of the compounds in the mixtures.
Line 91: why did you stop recording the spectra after 700 minutes?
Line 93: did you normalise the spectra collected by ATR-FTIR?
Line 94: change “of add” with “to add”.
Line 96-102: I suggest moving this paragraph in the introduction (line 70-72), and adding bibliographic references.
Figure 1: remove a) and b) below the figures.
Line 109: please change “od” with “of”.
Line 114-115: I think this information about the pigments needs to be moved to the Materials and Methods session. How was the paint casted on the silicon wafer?
Line 119: I think it is the composition of the binders (type of fatty acid and concentration), rather than the grinding of the pigment that affects the broadening of the carboxylates band.
Line 124: add bibliographic references.
Figures: I suggest removing the arrows to point out the band related to the carboxylate, and using a line that starts at a specific wavenumber and crosses every spectra in each figure to better highlight changes in a specific band. In addition, I think the Authors should remove the absorbance values on the y axis.
In the spectra in figures 1 and 2, the intensity of the peak at about 1760 cm-1 looks quite stable, instead of decreasing, due to the formation of carboxylates. How do you explain this result?
Line 144 and 147: chance “1664” with “1564”.
Line 149: I suggest replacing with “even in paints with natural resins..”.
Line 153: remove “these”.
Line 155: change “hystorical” with “historical”.
Line 140-160: this paragraph is not very clear. The sentence (at line 155) should start a new paragraph.
Figure 5: I suggest changing it, in order to avoid the overlapping of the main figure with the inset.
Figure 6 and 7: I suggest adding the time of collection of the spectra close to the spectra, as in the previous figures, instead of in the captions.
Line 171: what do you mean with “prepared with a defect of KOH in order to be sure that no KOH was available”? please add information about the ratio of the compounds in these mixtures.
Line 181: please change as follows: The palmitic acid does not seem able to steal the cation of the salts based on abietic acid in a significant amount.
Line 188: how do you justify this behaviour? Does this have to do with the acid dissociation constants of the two acids?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The manuscript has been improved significantly. However, I believe that tagging and assigning major IR peaks or adding a table listing the assignment of all peaks giving information relevant to the work is needed. The authors' response (information is available in the literature) is irrelevant, a manuscript should be consistent and an assignment is required to support the discussion.
Correct cm-1 with -1 superscript.
Author Response
The table with positions and assignments has been added and introduced in the text.
All the cm-1 have been checked.
Thank you again for your suggestions.
Reviewer 2 Report
I think the qualitity of the manuscript improved after the revisions, and I would like to thank the Authors for the consideration of my suggestions.
I only have one more comment: I think the Authors should better specify in the discussion of the results their hypothesis about the competition among abietic and palmitic acid and the ability to snatch the cations from already formed fatty acid carboxylates. As you pointed out the question cannot be fully explained by only considering the pKa, but this factor might play a role, probably together with pH, for example.
Author Response
A table with positions and main assignments has been added as strongly requested by reviewer 1.
At the moment we don't have a consistent hypothesis. It's very difficult even to approach the problem from the pH point of view since we worked in solid/dry conditions and the two acids are almost insoluble in water.
Nevertheless, we added some observations.
Thank you again for your precious suggestions.
