The Influence of Raw Materials on the Stability of Grisaille Paint Layers

Round 1

Reviewer 1 Report

Dear Editor, dear Authors,

the submitted research article presents the results of an experimental work aimed at evaluating the chemical and mechanical stability of grisaille according to composition or firing temperature. Although the topic is of interest in the conservation field, the introduction of the paper does not allow scholars from other fields to fully understand what grisailles are and the potentialities of the work. In particular, a more general public, reading “painting” might associate this word with canvas/panel paintings; likewise, the word “pigment” might be misled, confusing a more common art context with that of grisaille. Considering the aims of the journal, I would recommend reviewing the introduction to let a more general panel understand the premise and the aim of the work. Furthermore, results are shown in a non-always linear way and the discussion seems somehow full of gaps, an aspect that is a weakness in the chosen journal.

I would recommend the addition of a separate discussion paragraph, for a more complete overview of the results, considering further:

* that modern commercial slides have a very flat surface which might not promote good adhesion, whereas handcraft glass surfaces might be rougher; has this aspect been considered and substrate slabs been measured in their roughness? Moreover, commercial glass slides might have a coating over the surface or production residue: have slides been cleaned or somehow treated before the application of the grisaille (event though adhesion to this substrate revealed to be high)?

* has the wettability of the substrates been evaluated, especially with respect to the adhesion promoters employed? This aspect might play a role in the application of the preliminary grisaille. Please consider carrying out this measurement if possible and further discuss the data

* the use of modern and research-grade raw materials compared to more impure ones employed in the past, where contamination might have a role in degradation phenomena

* consider the influence of complete/limited molten BG in the samples

*roughness data and contact angle values: please further discuss differences according to firing temperature (600/650/700°C), presence of hematite PA/natural hematite/burned iron(/burned umber), presence of lead introduced with other components besides what already present in BG

* summarise the hydrophilic/hydrophobic feature of samples

* summarise reasons why burned iron acted similarly or not to hematite and burned umber

* summarise data according to adhesion promoters

* summarise data according to other components added

* add a comment about impurities in samples due to the crucible, as they represent almost 3% of the composition.

Following, are some questions and some suggestions to the authors to improve the composition.

Abstract

Line 24: chemical and/or mechanical stability?

Introduction

The authors considered mainly recent publications. Although reference 2 is largely cited and can be considered a great review- about grisaille, I recommend adding historical literature and recent analytical studies related to grisaille composition to better explain the choice of sample components, formulations, and firing conditions that have been employed.

Line 41: grisaille have limited shades of colour, being the substrate already coloured. Please highlight this aspect.

Line 44: besides citing ref.2, can the authors explain why the use of lead-rich formulations was so common, and, actually, essential?

Line 50: historically, in this context, is common to define “pigment” what is added to the grisaille to promote a different hue of the grisaille; nevertheless, the word “pigment” might be misleading by scholars not familiar with glasses and grisaille because pigments are insoluble components, dispersed in a binder as powder and always recognizable. The colour of glasses is generated in a different way than traditional paint. Furthermore, this introduced component can change their crystalline structure or chemical nature at firing temperatures, becoming something different. Please, stress what you considered as “pigment” or consider referring to this component throughout the text as compounds/minerals or another word.

Line 54: “painted on”, consider stretching this concept or changing the verb; possible suggestions “spread/applied on and shaped”, in reason of what reported before.

Line 56: to the reviewer's knowledge, wine and vinegar are not common binders, and neither are vehicle agents in paintings as this word is mainly related to panel/canvas paintings. Please, highlight they are common in this art technique, cite the historical bibliography, and make explicit this “etc”.

Line 58: in the case of grisaille, firing the preliminary grisaille, the nature of lead white changes completely from a carbonate structure to iron barysilite, thus it cannot be considered a pigment (according to handbook definition) or a colouring agent. Is PbO as well one of the products of the reaction?

Line 61: “all these changes”: it is not clear what the authors meant.

Line 63: “nevertheless” unsuitable adverb.

Line 64: “compatibility”: compatibility or adhesion?

Line 66: “this is visible”: what? Unclear what the authors meant.

Line 68-71: unclear sentence.

Materials and methods

The authors should report components manufacturing.

Can the authors explain:

* why was the choice of Cu and Fe for the model grisaille followed by changes of colouring agents?

* the selection of other components and the effect on the properties/features or workability of the final materials of their addition

* why has the firing temperature been changed only for one CA:BG ratio?

Line 78: the word “variable” might be misleading since the word mainly refers to a physical variable; in this case, authors can talk about changes in composition with change of main components ratio, selection of a different component, addition of a third one or change of firing temperature.

Line 81: it is suggested to introduce the rocaille already in the introduction

Line 86/89: do authors know approximately the amount of carbon in the steel and the average size of the particles of copper and iron oxides after both grinding times?

Line 131: characteristics that were not evaluated for raw materials; it is suggested rephrasing.

Line 134: Wolbers’ method focuses on painted surfaces intended as tempera/oil paintings on wooden panels/canvases mainly; please highlight this aspect and discuss the authors’ choice of the method, especially because water, ethanol and acetone normally do not solubilize glass and water has been used for contact angle measurements. Add references related to the use of Wolbers’ method to grisaille/glass substrates besides what has been already mentioned.

Line 137: add XRF analytical parameters.

Line 139-140/140-143: unclear, please consider rephrasing.

Line 161: water drop volume?

Line 163: how contact angles were measured?

Line 167: were colourimetric analyses recorded in SCI or SCE mode? Were they carried out in the same areas of roughness and wettability evaluations?

Line 180: new line for value 2.

Table 1: indicate manufacturing, unit of the ratio (volume, weight, …), the concentration of the gum Arabic in water, the amount of the “other materials”; burned lead and burned lead and tin: for these compounds, it is not specified how they have been treated.

Since umber is introduced as burned, please be consistent with this name throughout the text.

Figure 1: please consider caption rephrasing.

Results and discussion

3.1

Which is the LOD of the XRF spectrometer for light elements such as Na?

Have XRF analyses been carried out on colouring agents, at least for those bought as pigment?

Line 220: supplementary materials or appendix?

Line 220 and line 225/226: “only” and “instead of more complex”: please justify these statements.

Line 222: “however” unsuitable adverb.

Line 231: I would suggest mentioning already in the text hydrocerrusite, explicating in brackets that this is a hydrated form of cerrusite.

Table 3, line 3: “Potash-lime glass” reports Na in its synthetic composition.

3.2.1

Probably XRD data of samples should be introduced as a single paragraph to discuss easier further analytical data. Table 5 introduces crystallographic results in the colourimetric section.

Lines 244-249: it is not a result, but part of the introduction of the work. Moreover, lead white is clearly not a pigment in this context, because of a different generation process of the grisaille hue.

Figure 2: what about standard deviation? Moreover, consider reducing the size of the model symbol to pin it with higher precision in the colourimetric space.

Line 257: Figure S12: it is known that in oxidative environment cuprite changes to tenorite, this is then promoted by higher temperatures. Besides that, has in general been considered how crystalline structures have changed during the firing process in terms of lattice, phases and particles/domains size?

3.2.2

Please, consider adding in brackets experimental values obtained when discussing the sample, to let the reading of the results more immediate.

Lines 266-270: have the authors considered that the whole process can be considered handcraft, thus particle size and amount and evaporation of the adhesion promoter might influence the preliminary-grisaille distribution? This, not considering that the following firing step still changes the surface feature. Was roughness measured before and/or after the firing step?

Lines 270-271: unclear what the authors meant.

Line 272: “is similar” does this means that the results are of the same order? Just reading, the average of the roughness can be indicated as ca 7µm±5µm, a standard deviation of great range. Moreover, if values are similar, how is it possible to discuss differences among samples? Only four samples showed lower roughness values than the model, considering as model values also samples obtained at different firing temperatures (an aspect which has not been discussed).

Lines 278-281: as so, did BG completely melt when added in CA:BG 1:1 ration or, better, is it possible to be sure that was the evaluation of the roughness not affected by a limited molten BG?

Lines 288-290: why this statement?

Lines 299-305: please rephrase and think to anticipate this part at the beginning of the paragraph.

Figure 3: please consider displaying data not as listed values in a table but in a graph (histogram) and divide them into groups, according to table 1.

3.2.3

Lines 314-316: why does this happen?

Line 321: wettability and water penetration are ruled by different parameters; please consider rephrasing.

Lines 321-330: unclear data discussion

Line 324: “unsuitable firing temperature as in the 700°C sample”: this statement introduces further results without explaining the concept which is being discussed.

Line 327: here PA and natural hematite are considered, but what about burned iron in this context, which is as well hematite?

Figure 4: standard deviation?

3.3

Line 341: most of the samples were classified as 1 not because they showed similar results as the model, but because of the classification method chosen

Line 350-352: so, is there a connection between this adhesion behaviour and contact angles/roughness?

Lines 353-356: actually, everything added to the base mixture CA:BG alters the ratio/balance between these two components

Lines 371-375: is the mechanical action with the cotton swab or the solvent the cause of the removal?

Lines 376-377: mostly, this is not a result of the cleaning test

Line 395: porosity has not been tested on samples

Lines 397-400: please, discuss this sentence in a historical context

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

I found this article about the stability of grisaille very interesting and informative. It is clearly written and clearly structured, and it provided me with a good overview of what can affect the durability of grisaille. The authors' approach appears sounds and their results are presented in a suitable manner.

Author Response

See the attached document

Author Response File: Author Response.docx

Reviewer 3 Report

The paper reports relevant contribution by investigating how raw materials can influenciate the grisaille paint layers´ stability. The methodology proposed included the characterization of the raw materials used in the grisaille execution as well as manufacturing factors that can influenciate it´s color appearance and stability. The cleaning test seems to be a good addition since it could play a decision rule in further interventions related to this works of art typology.  

I recommend the publication of this manuscript after minor revisions:

1.       Please re-organize XRD and colorimetric sections (paragraphs 252-262 are related to XRD results).

2.       Check if it is possible to set together tables 4 and 5 and please clarify their context on the main text

3.       Please remove the sentence “The color of grisailles depends on coloring agents” (line 387)

4.       Suggestion: discuss how aging tests can contribute to understand the grisailles stability – further research?

Author Response

See the attached document

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Dear authors,

Thank you for the time you spent answering my questions, I appreciated it. Your manuscript has been much improved, and the introduction and the discussion are now much clear. What has been added to the conclusion helps the reader to have a general understanding of the results obtained.

Just a few more suggestions:

Line 65-66: in the answer you gave me, you stated that the role of the other materials historically added is still uncertain: I would specify that it is like that.

Line 217-218: I would talk about “cleaning” and not “chemical cleaning” of the stained-glass windows since a cleaning treatment is not supposed to alter the chemistry of the object.

Line 240: they “would not influence” instead of “will not”

Paragraph 3.2.3.: some samples (e.g., burned umber, alumina, …) showed an average within the range of the model, despite a very narrow SD. This aspect is not only dependent on the handcrafting process, thus I would not say that these values would not be considered for evaluation and interpretation, since they are valuable results and have their potential. Moreover, throughout the paragraph, I would be consistent in indicating or not the SD in the text. Fig.3a: “burned umber” instead of “umbra”. Be careful with the yellow colour.

Line 387-388: “can also act” or “can become a protective layer”, since the rise of grisaille has an art/aesthetic aim at first

Fig.4b: “burned” umber, samples 600°C/700°C inserted twice. If possible, highlight subgroups as in the previous graph

Line 465: “unbalanced VOLUME between base glass and colouring agent particles”

Since colourimetric analyses were carried out in SCI mode, there might be a correspondence between the L* value and roughness. It is a pity wettability tests were not performed on the same mock-ups where the morphology of the surface was evaluated since roughness and contact angles measured can be correlated to each other. Another suggestion for future steps of research is the evaluation of the contact angle of the vehicle with respect to the substrate, as the affinity between these materials might affect the homogeneity of the application of the grisaille over the surface.  

Author Response

See the document attached

Author Response File: Author Response.docx