Al-Bearing Scorodite (Scorodite—Mansfieldite Series) from Hemerdon Ball Mine, Plympton, Tavistock District, Devon, United Kingdom: Single-Crystal X-Ray Diffraction, Chemistry and Vibrational Spectroscopy

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript provides a detailed characterization of Al-bearing scorodite from the Hemerdon Ball Mine through a multi-analytical approach. Overall, the research is technically sound and well-presented; however, several minor issues require attention to improve the manuscript's clarity. Consequently, I recommend a minor revision for this paper.

My specific comments are as follows:

1. Please reconcile the conflicting chemical formulas provided in the Abstract and Table 2A brief explanation regarding the discrepancies between EPMA and SCXRD results is necessary to clarify these compositional differences.
2. Itismentioned that Al-for-Fe substitution leads to a- and b-axis contraction but an unusual expansion of the c-axis. While distortion is mentioned, the authors should provide a more focused discussion on Fe(Al)-OH octahedral tilting or specific bond angle variations to better elucidate the physical cause of this c-axis expansion.
3. Pleaseensure that the visual clarity of all figures(e.g., Figure 1) is improved and that labels are consistently aligned to meet publication standards.
4. Table 1is currently overwhelmed by historical data from 1948. To make the findings more accessible, I recommend moving the raw comparison data to the supplement and instead using a simple binary plot to visually demonstrate the reported miscibility gap.
5. The spectral variations between ALS1 and ALS2 inRaman spectroscopyare attributed to crystal orientation. It would be beneficial to specify the exact orientations—such as the crystal axes relative to laser polarization—and explain the underlying reason why these orientations trigger such significant intensity shifts in the 810/814 cm^-1 region.
6. In Section 4.1, the author notes the substitution of P and S for As. The discussion should be expanded to address whether this anion-site replacement exerts a synergistic influence on Fe/Al octahedral occupancy or affects the mineral’s long-term stability in complex mining environments.

 

Author Response

Response Letter to Reviewer R1.

The authors appreciate your constructive review. We also hope that our answers to your questions satisfy your concerns. Below, we responded to each of your points addressed to us.

This manuscript provides a detailed characterization of Al-bearing scorodite from the Hemerdon Ball Mine through a multi-analytical approach. Overall, the research is technically sound and well-presented; however, several minor issues require attention to improve the manuscript's clarity. Consequently, I recommend a minor revision for this paper.

My specific comments are as follows:

1. Please reconcile the conflicting chemical formulas provided in the Abstract and Table 2A brief explanation regarding the discrepancies between EPMA and SCXRD results is necessary to clarify these compositional differences.

Response: We have only one crystal-chemical composition obtained by EPMA, shown in Table 1. The EPMA data are the right compositions considered in this work and represent, in fact, the second chemical data published after Allen et al 1948. The chemical composition shown in our section dedicated to SCXRD is deduced by modelling of Al-bearing structure and reflects the ”refinements performed using different Al contents (see now Figure 5, which replaced the former Table 4). The best result was achieved for 95% of Fe and 5% of Al, in agreement with the chemical analysis of the Al-bearing scorodite HBM (5% of Al)”. In Table 2 the first column was labeled “ Empirical Formula.

 

1. It is mentioned that Al-for-Fe substitution leads to a- and b-axis contraction but an unusual expansion of the c-axis. While distortion is mentioned, the authors should provide a more focused discussion on Fe(Al)-OH octahedral tilting or specific bond angle variations to better elucidate the physical cause of this c-axis expansion.

Response: We added: Lines 487  in Clean Manuscript: The refined structure of the Al-bearing compound (Figure 4) suggests a preference of Al³⁺ to occupy the octahedral sites alongside Fe³⁺, leading to a noticeable distortion of the Fe/Al-OH octahedron. This distortion is attributed to an expansion of the c unit-cell lattice occupied only by Fe-OH octahedron. A gradual contraction in the bond lengths of the a and b unit-cells supports the substitution of Al³⁺ for Fe^3+, resulting in the formation of an Fe/Al - OH octahedron.

Also, a more detailed explanation may be read in the section: &4.2

 

1. Please ensure that the visual clarity of all figures(e.g., Figure 1) is improved and that labels are consistently aligned to meet publication standards.

Response: Figure 1 is now Figure 2, because a new Figure 1 was added with a photo of crystals of Al-bearing scorodite from ESM. Done!

 

1. Table 1 is currently overwhelmed by historical data from 1948. To make the findings more accessible, I recommend moving the raw comparison data to the supplement and instead using a simple binary plot to visually demonstrate the reported miscibility gap.

Response: We agree with this suggestion, but please also consider that good chemical analyses on mansfieldite are found only in the work of Allen et al 1948, and in this work (Al-bearing scorodite). We are talking about natural occurrences. Therefore, we prefer to keep Table 1. It is clear that the existence of a miscibility gap, but in a supposed figure, we’ll plot a few data.

 

1. The spectral variations between ALS1 and ALS2 in Raman spectroscopy are attributed to crystal orientation. It would be beneficial to specify the exact orientations—such as the crystal axes relative to laser polarization—and explain the underlying reason why these orientations trigger such significant intensity shifts in the 810/814 cm^-1 region.

Response: It is not so easy at a micro scale of optical microscope connected to Raman spectroscope. We intended to find a few crystallographic planes differently oriented under optical microscope.

 

1. In Section 4.1, the author notes the substitution of P and S for As. The discussion should be expanded to address whether this anion-site replacement exerts a synergistic influence on Fe/Al octahedral occupancy or affects the mineral’s long-term stability in complex mining environments.

Response: In this case, there is no influence. We deal with traces, estimated by EPMA. Uncertainties are higher than reported.

Reviewer 2 Report

Comments and Suggestions for Authors

The submitted manuscript concerns the study of Al-bearing scorodite from the Hemerdon Ball Mine and presents a dataset including electron microscopy, electron microprobe analysis, single-crystal X-ray diffraction, and infrared and Raman spectroscopy.
However, it is unclear whether the aim of the study is a straightforward characterization of Al-bearing scorodite from the Hemerdon Ball Mine, or a more general investigation of the effects of Al incorporation on the scorodite structure. With an appropriate interpretation and discussion—both of which are unfortunately lacking in the present form—the reported data could support a complete mineralogical characterization. Nevertheless, such a contribution would be of limited scientific interest and would not be suitable for a high-ranking journal such as Crystals.
If the intention is instead to address the influence of Al incorporation in the scorodite structure, a significantly more robust approach would be required. In particular, an adequate number of samples spanning a range of Al contents would be mandatory, together with a rigorous and comprehensive interpretation of a full analytical dataset. In this regard, the authors only present a comparison of spectroscopic data for “KW” scorodite and mansfieldite with the Hemerdon Ball Mine sample, without providing corresponding X-ray diffraction or EPMA data for these KW minerals.
Finally, the manuscript requires substantial improvement in both writing quality and overall organization. Additional detailed comments are provided in the attached annotated files.

Comments for author File: Comments.pdf

Author Response

Response Letter to Reviewer R2.

The authors appreciated your constructive review, and we thank you. We hope that our answers to your questions satisfy your concerns. Below, we responded to each of your points addressed to us.

The submitted manuscript concerns the study of Al-bearing scorodite from the Hemerdon Ball Mine and presents a dataset including electron microscopy, electron microprobe analysis, single-crystal X-ray diffraction, and infrared and Raman spectroscopy.

However, it is unclear whether the aim of the study is a straightforward characterization of Al-bearing scorodite from the Hemerdon Ball Mine, or a more general investigation of the effects of Al incorporation on the scorodite structure. With an appropriate interpretation and discussion—both of which are unfortunately lacking in the present form—the reported data could support a complete mineralogical characterization. Nevertheless, such a contribution would be of limited scientific interest and would not be suitable for a high-ranking journal such as Crystals.

Response: The data presented support a complete mineralogical investigation: structural, morphological, crystal chemistry, and vibrational spectroscopy. Certainly, the presence of Al in the scorodite structure is an exotic part of this research.  Nevertheless, your opinion about the limited scientific interest in this topic is against that of others’ opinions from literature: “Its wide occurrence in comparison to other secondary arsenate minerals has led many to advocate it as an acceptable carrier for the immobilization of arsenic released during pyrometallurgical or hydrometallurgical processing of arsenic-containing ores as gold, copper, and uranium. Hence, considerable research has been carried out either looking into the experimental determination of its solubility.” See Arsenic Metallurgy Edited by R.G. Reddy and V. Ramachandran

If the intention is instead to address the influence of Al incorporation in the scorodite structure, a significantly more robust approach would be required. In particular, an adequate number of samples spanning a range of Al contents would be mandatory, together with a rigorous and comprehensive interpretation of a full analytical dataset. In this regard, the authors only present a comparison of spectroscopic data for “KW” scorodite and mansfieldite with the Hemerdon Ball Mine sample, without providing corresponding X-ray diffraction or EPMA data for these KW minerals.
Finally, the manuscript requires substantial improvement in both writing quality and overall organization. Additional detailed comments are provided in the attached annotated files.

Response: We are dealing with samples that crystallized in natural conditions. A scarce number of natural occurrences around the world of mansfieldite or Al-bearing scorodite are known. For a few synthetic samples, the internal structure was described in the literature. The analyses carried out on our samples and the data obtained are rigorously interpreted and discussed according to the results obtained. The KW scorodite and mansfieldite (collection Prof. Theo Kloprogge from the same site, HBM) were used for comparison of the molecular vibration obtained on Al-bearing scorodite, an intermediary member between scorodite and mansfieldite. We did not used XRD for KW samples, because our work is dedicated to Al- bearing scorodite. The main interest was to have a rigorous evaluation of the vibrational spectroscopy bands obtained by IR and RAMAN spectroscopy on Al-bearing scorodite. This is the reason why we used the samples KW.

Overall,  in terms of “writing quality and overall organization”, I’d like to mention that there are many differences between the English writing still and organization between each scientific area. For example, in mineralogy we do not need to give references for each mineral name discussed in our work.

RESPONSE from the corrected manuscript:

X-ray observations:

Include estimated standard deviations in the measurements.

Response: The standard deviations were included: a=8.92882(14); b=10.02217(14); c=10.30525(15)

Nonsense:

Response: We remove the keyword “unit cell lattice”

 CrysAlisPro software reference Agilent (2013). CrysAlisPro Software System, version XXX.YYY.ZZZ. Agilent Technologies UK Ltd, Oxford, UK

Response: in the text line 104 we rewrite for:

(CrysAlisPro software reference Agilent (2013). CrysAlisPro Software System, version V1.171.142.173a). Agilent Technologies UK Ltd, Oxford, UK)

Line 192 a more detailed discussion of the relationships is needed. include graphics properly showing the influence of chemistry on unit cells, based on values reported in literature and in table s3.

Line 192 we replaced by:

RESPONSE: The mineral crystallizes in the orthorhombic system (space group Pbca) with unit-cell parameters a = 8.92882(14) Å, b = 10.02217(14) Å, and c = 10.30525(15) Å, yielding a unit-cell volume intermediate (922.18(2) ų) ) between those reported for scorodite (931.5 ų) [9] and mansfieldite (868.7 ų) [18]. This intermediate volume is consistent with the partial substitution of Fe³⁺ by Al³⁺ at the octahedral site.

The crystal structure consists of corner-sharing AsO₄ tetrahedra and (Fe/Al)O₆ octahedra forming a three-dimensional framework. The AsO₄ tetrahedra are relatively rigid, as indicated by the narrow range of As–O bond distances (1.6818–1.6885 Å), whereas the octahedral site accommodates chemical variability. The refined occupancy (Fe₀.₉₅Al_0.05) indicates limited but significant substitution of Fe³⁺ by the smaller Al³⁺ cation.

This substitution has a direct structural effect: the smaller ionic radius of Al³⁺ leads to shorter M–O bond distances and a contraction of the octahedral framework. Consequently, the unit-cell volume decreases with increasing Al content, as observed in literature data for the scorodite–mansfieldite series. The unit-cell volume of the present sample plots between Fe-dominant scorodite and Al-dominant mansfieldite, consistent with its intermediate composition.

A comparison with literature values and those reported in Table S3 demonstrates a systematic relationship between composition and unit-cell parameters. This trend confirms that the substitution at the octahedral site controls both local geometry (M–O bond distances) and global structural parameters (unit-cell dimensions), while the AsO₄ tetrahedra remain largely unaffected.

 missing value line 196

Response: iii=1/2+x, 3/2-y, 1-z

 

Table 3 report for Fe coordination polyhedron mean bond value, quadratic elongation and bond angle variance values

 

Response: We changed the title of this table “ The Fe coordination polyhedron mean bond value, quadratic elongation and bond angle variance values are reported”

 

Substitute table 4 with a graph including Al content vs wr2 value

Response: We remove the table and show a graph Al content versus wR2value 

These crystal structure drawings are definitely not suitable for publication.

Response: We change the figure a and the figure b  

Figure 4. (A) Crystal packing of the Al-bearing scorodite from HBM along [100] direction showing the AsO₄ tetrahedra (dark blue color) linked with Fe/AlO₆ octahedra (green color). (B) Environment around Fe/Al and As atoms, where i, ii and iii are the symmetry operations i= x, 3/2-y, ½+z; ii= 1-x, 1-y, 1-z; iii=1/2+x, 3/2-y, 1-z.

 

Line 229 add details to this comparison

“…bond lengths Fe/Al–O5 = 2.1107(10) Å, and Fe/Al–O6 = 2.0446(10) Å. The As-O bond distances varied from 1.6818(10) to 1.6885(10) Å. These distances Fe/Al-O and As-O compare well with those reported for synthetic mansfieldite [18] and scorodite [9]. The referee ask to add details to this comparison”

Response: We substitute the paragraph in line 229 for this one

The Fe/Al–O bond distances in the studied structure (Fe/Al–O5 = 2.1107(10) Å; Fe/Al–O6 = 2.0446(10) Å) fall within the range typically reported for octahedrally coordinated Fe³⁺/Al³⁺ in arsenates. For example, in mansfieldite [18], Al–O distances are reported in the range 1.88–2.05 Å, while in scorodite [9], Fe–O distances typically span 2.061–2.125 Å.

Lines 411-415 were rewritten:

The lattice parameters calculated for the Al-bearing scorodite HBM were compared with the values reported for scorodite and synthetic mansfieldite (Table S3, ESD). The a and b unit-cell lattice parameters (8.93 Å and 10.02 Å) of Al- bearing scorodite HBM lie between scorodite [9] and synthetic mansfieldite [18], whereas the c unit-cell lattice parameter (10.30 Å) is larger than the values obtained for scorodite (10.04 Å) and synthetic mansfieldite (10.11 Å).

Both a and b unit-cell lattice parameters for synthetic mansfieldite are smaller because the Al³⁺ ionic radius (1.84 Å) is smaller than that of Fe³⁺ (2.04 Å). The b unit-cell lattice parameter (10.02 Å) is compressed in Al-bearing scorodite, confirming the substitution of Al for Fe accompanied by the formation of Fe/Al-OH octahedron. Conversely, the c unit-cell lattice parameter expanded (10.30 Å), being attributed only to Fe-OH octahedron.

 

Lines 574-586 were rewritten:

 

The substitution of Al³⁺ for Fe³⁺ affects the unit-cell parameters of Al-bearing scorodite HBM. Specifically, the a and b unit-cell parameters decrease with the substitution of Al³⁺ for Fe³⁺ and formation of an AlO₆ octahedron in synthetic mansfieldite (a = 8.82 Å and b = 9.82 Å), while the c unit-cell remains unchanged, as in scorodite (c = 10.04 Å). Conversely, the c unit-cell increases at 10.30 Å in Al-bearing scorodite HBM because both Fe and Al share the same octahedral position, enhancing distortion of the newly formed Fe/Al-OH octahedron. However,

the a and b unit-cell lattice parameters decrease gradually, indicating the substitution of Al for Fe with the formation of the Al-OH octahedron.

The lattice structure of Al-bearing scorodite HBM becomes slightly distorted by the formation of the Fe/Al-OH octahedron, which leads to a contraction in the a and b unit-cell lattice parameters and an enlargement in the c unit-cell lattice parameter by the formation of the Fe-OH octahedron.  This structural distortion likely impedes further substitution of Al³⁺ for Fe³⁺, creating a miscibility gap as noted by several authors. Consequently, a solid-solution series between scorodite and the Al-end-member scorodite (i.e., mansfieldite) is questionable.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

I found only a few minor issues which are marked in the manuscript.  One of them is formatting n the reference section that got a bit messed up in the second half.  None of these are big issues and can be fixed easily.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The revision of the submitted manuscript, although improved in some parts, still presents some issues to be clarified.

• A large number of the reviewer’s previous comments appear to have been overlooked, and the authors are expected to carefully reconsider and address them. As an example, the reported ionic radii (row 439) of 1.84 Å for Al³⁺ and 2.04 Å for Fe³⁺ are somewhat unexpectedly larger than the standard values given by Shannon (1976). The authors should clarify the origin of these values and provide a justification for their use.
• Although the authors state (see their response) that scorodite is of current interest, they fail to clearly articulate the novelty of their contribution in the context of the existing literature. As a result, the manuscript does not convincingly demonstrate what it really adds to the current knowledge on this research topic.
• Classical studies on Hobart Butte (e.g., Denning, 1943; Allen et al., 1948) performed through chemical and optical studies, maintain the presence of aluminium-bearing scorodite and intermediate members of the scorodite–mansfieldite series.
  So, a major concern relates to the use of the scorodite KW and mansfieldite KW samples as reference materials in the IR and Raman sections. Although these samples play a key role in the interpretation of the spectroscopic data, no EPMA chemical characterization is reported for them, either in the main text or in the Supplementary Data. I was not able to find in the manuscript the provenance of KW samples; in the supplementary it is stated they come from Hemerdon. If so report it in the proper position in the manuscript.
• Partial results only of a structural study are reported, but it is not clearly stated whether the approach is ab initio or based on the refinement of an existing structural model. This point should be clarified. By the way, have the positions of hydrogen atoms been determined and included in the model?
  I was also unable to retrieve the CIF file from the CCDC repository, and the authors should verify its accessibility or provide it directly as supplementary material.
• At least three distinct structural studies can be found in literature, namely from Kitahama (1975), Hawthorne (1976) and Xu (2007). It is worth noticing that the results of these studies are follow different space group orientations possibile for Pbca space group.
  The authors should justify why they compare their data to Kitahama data only, dropping the more complete study by Hawthorne (1976) and the more recent study by Xu (2007). A complete comparison should preferable instead.
• The indexing of X-ray powder pattern (supplementary) is not consistent with the reported unit cell. Please index the pattern according to the single crystal study cell.
• Furthermore, the authors should explicitly report the relationship between the atomic coordinates and the orientation of the unit-cell axes in their model with respect to the previous crystal structureal studies. This would eliminate the possibility that the reported and unexplained expansion of the c-axis is in fact an artifact arising from a wrong comparison of crystallographic orientations.

Comments on the Quality of English Language

The revision of the submitted manuscript, although improved in some parts, still presents some issues to be clarified.

• A large number of the reviewer’s previous comments appear to have been overlooked, and the authors are expected to carefully reconsider and address them. As an example, the reported ionic radii (row 439) of 1.84 Å for Al³⁺ and 2.04 Å for Fe³⁺ are somewhat unexpectedly larger than the standard values given by Shannon (1976). The authors should clarify the origin of these values and provide a justification for their use.
• Although the authors state (see their response) that scorodite is of current interest, they fail to clearly articulate the novelty of their contribution in the context of the existing literature. As a result, the manuscript does not convincingly demonstrate what it really adds to the current knowledge on this research topic.
• Classical studies on Hobart Butte (e.g., Denning, 1943; Allen et al., 1948) performed through chemical and optical studies, maintain the presence of aluminium-bearing scorodite and intermediate members of the scorodite–mansfieldite series.
  So, a major concern relates to the use of the scorodite KW and mansfieldite KW samples as reference materials in the IR and Raman sections. Although these samples play a key role in the interpretation of the spectroscopic data, no EPMA chemical characterization is reported for them, either in the main text or in the Supplementary Data. I was not able to find in the manuscript the provenance of KW samples; in the supplementary it is stated they come from Hemerdon. If so report it in the proper position in the manuscript.
• Partial results only of a structural study are reported, but it is not clearly stated whether the approach is ab initio or based on the refinement of an existing structural model. This point should be clarified. By the way, have the positions of hydrogen atoms been determined and included in the model?
  I was also unable to retrieve the CIF file from the CCDC repository, and the authors should verify its accessibility or provide it directly as supplementary material.
• At least three distinct structural studies can be found in literature, namely from Kitahama (1975), Hawthorne (1976) and Xu (2007). It is worth noticing that the results of these studies are follow different space group orientations possibile for Pbca space group.
  The authors should justify why they compare their data to Kitahama data only, dropping the more complete study by Hawthorne (1976) and the more recent study by Xu (2007). A complete comparison should preferable instead.
• The indexing of X-ray powder pattern (supplementary) is not consistent with the reported unit cell. Please index the pattern according to the single crystal study cell.
• Furthermore, the authors should explicitly report the relationship between the atomic coordinates and the orientation of the unit-cell axes in their model with respect to the previous crystal structureal studies. This would eliminate the possibility that the reported and unexplained expansion of the c-axis is in fact an artifact arising from a wrong comparison of crystallographic orientations.

Author Response

RESPONSE LETTER TO REVIEWER 2

We thank you very much for taking the time to review this manuscript again. We sincerely appreciate all your constructive comments and suggestions! According to your suggestions, we have made corrections to the relevant part of the manuscript.

The responses are listed below in blue normal font for your comments and suggestions.

The revision of the submitted manuscript, although improved in some parts, still presents some issues to be clarified.

 

• A large number of the reviewer’s previous comments appear to have been overlooked, and the authors are expected to carefully reconsider and address As an example, the reported ionic radii (row 439) of 1.84 Å for Al³⁺ and 2.04 Å for Fe³⁺ are somewhat unexpectedly larger than the standard values given by Shannon (1976). The authors should clarify the origin of these values and provide a justification for their use.

 

Response: We agree and thank you for your remark. Certainly, the ionic radius of Al³⁺ is 0.535 Å, and for Fe³⁺ is 0.645 Å. The values used were for the atomic radius, at least for Al³⁺ (1.84 Å). Anyway, we replaced with the ionic radii.

•  
• Although the authors state (see their response) that scorodite is of current interest, they fail to clearly articulate the novelty of their contribution in the context of the existing literature. As a result, the manuscript does not convincingly demonstrate what it really adds to the current knowledge on this research topic.

 

Response: The Al-scorodite is the current research topic, poorly known in literature, and some previous studies are mentioned in our work. Scorodite is just used for comparison in our vibrational spectroscopy study, not as a current interest.

•  
• Classical studies on Hobart Butte (e.g., Denning, 1943; Allen et , 1948) performed through chemical and optical studies, maintain the presence of aluminium-bearing scorodite and intermediate members of the scorodite– mansfieldite series.

So, a major concern relates to the use of the scorodite KW and mansfieldite KW samples as reference materials in the IR and Raman sections. Although these samples play a key role in the interpretation of the spectroscopic data, no EPMA chemical characterization is reported for them, either in the main text or in the Supplementary Data. I was not able to find inthe manuscript the provenance of KW samples; in the supplementary it is stated they come from Hemerdon. If so report it in the proper position in the manuscript.

 

Response: Right, Allen et al. published only chemical analyses on mansfieldite and Al-bearing scorodite. This work brings new data on the crystal shape, the first accurate EPMA data, structural data obtained by X-ray single crystal, and a complete study dedicated to infrared and Raman vibrational spectroscopy only on Al-bearing scorodite. We have explained in the Materials section (line 95-97) why we used scorodite KW (catalog number 030107073; J. Theo Kloprogge) and mansfieldite KW (catalog number 8472; J. Theo Kloprogge). Both minerals are from the personal collection of Prof. Kloprogge used in this work for comparison of vibrational spectroscopy with our results obtained. Provenience: Hobart Butte Mine and Mount Cobalt Mine, Selwyn District, Mount Isa–Cloncurry area, Queensland (see lines 95-97).

 

Partial results only of a structural study are reported, but it is not clearly stated whether the approach is ab initio or based on therefinement of an existing structural model. This point should be clarified. By the way, have the positions of hydrogen atoms been determined and included in the model?

Response: Sim!

I was also unable to retrieve the CIF file from the CCDC repository, and the authors should verify its accessibility or provide it directly as supplementary material.

Response: The CIF File was also added to the supplementary material. We attached also to this response letter.

• At least three distinct structural studies can be found in literature, namely from Kitahama (1975), Hawthorne (1976) and Xu (2007). It is worth noticing that the results of these studies are follow different space group orientations possibile for Pbca space group.

The authors should justify why they compare their data to Kitahama data only, dropping the more complete study byHawthorne (1976) and the more recent study by Xu (2007). A complete comparison should preferable instead.

Response: The scorodite structure was determined by Kitahama et al. Subsequent corrections for the O - H bond lengths and H-atom positions were made by Hawthorne. Location of the Fe(III) arsenate dihydrate and the H atoms with satisfactory bond lengths and angles were obtained by Xu.

In conclusion, our work is more complex and discusses the structure of an intermediary member between scorodite and mansfieldite, and explains the location of Al in the octahedron. This is well explained in the section dedicated to the structure of Al-bearing scorodite.

• The indexing of X-ray powder pattern (supplementary) is not consistent with the reported unit Please index the pattern according to the single crystal study cell.

Response: The d(hkl) from Table S1 corresponds to data obtained by X-ray powder diffraction, not from SCXRD. The unit cell was calculated from SCXRD data. We added to ESM also the SCXRD pattern. There is no difference; the same crystal was used.

Furthermore, the authors should explicitly report the relationship between the atomic coordinates and the orientation of the unit-cell axes in their model with respect to the previous crystal structure studies. This would eliminate the possibility that thereported and unexplained expansion of the c-axis is in fact an artifact arising from a wrong comparison of crystallographic orientations.

Response: We thank the reviewer for this important observation. In the present study, the structure refinement was performed in the standard Pbca setting, as in previous studies of scorodite and mansfieldite (Xu et al. 2007; Harrison 2000). Therefore, the observed differences in the unit-cell dimensions, including the slight expansion of the c-axis in the Al-bearing scorodite sample, do not arise from a permutation or misassignment of crystallographic axes, but reflect genuine variation

Author Response File: Author Response.pdf