The Application of Portable X-ray Fluorescence (pXRF) for Elemental Analysis of Sediment Samples in the Laboratory and Its Influencing Factors

Round 1

Reviewer 1 Report

This manuscript discusses the conditions for Portable X-ray Fluorescence (pXRF). I am quite satisfied with this manuscript. The figures are beautiful, the evidence is good, and the discussion is correct. My only suggestion is to reduce the text in the Conclusion. Only the important results of this manuscript are listed.

• This manuscript has fewer linguistic problems.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This is good paper which emphasises the need understanding factors which impact on results, as instrumentation becomes more accessible.

line 113-115 - base on what operations? this was not clear?

117 'entrusted to a professional laboratory' this needs explained, what did it involved, what was this lab? what type of instrumentation - this is a continued theme throughout the manuscript and needs to be checked and reviewed

120 - what acids for the four acid digestion?

table 1 - why Hatu in bold?

table 2/3 - whats the shading for? what units are the data - no detail has been provided in the text?

table 3 - instead of par - suggest putting the sizes.

there are alot of statement about 'tend to more reliable', these statement need to be strengthened and quantified

figure 2 - axis required, why is the stdev not part of the mean, relabel par to sizes.

it is stated that the reliability is increased with detection time - figure 3 - do these results really show this? it is not a line graph

section 3.6 is very difficult to follow, and despite reading through several times, it is difficult to understand the graphs, which is not helped by no axis. it is commented upon several times in the text about minimum values for the instrument, this will be different for both instruments - can this be added as a supplentary?

instead of 'laboratory analysis' it need to be stronger to state ICP-MS - line 245

conclusions - detection set to 80-120 but hard to see in text where this was supported?

the conclusion also mentioned absolute values - but where were these in text? this is confusing as stated 20-30% values but what does that mean?

it was not until the conclusion that a fuller understanding of what you were attempting to do came through, however, this is confused and more attention is needed on the laboratory data? statistical data from the instruments is required to show instrument performance

good - but the main problem is the statements to  professional laboratory, what instruments - more information required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

P3L63: Why is mineral resource exploration itself is a high-risk work.

General comment: The advantages of ‘Portable’ XRF are normally claimed to be the rapid analysis of samples in the field by placing the instrument directly in contact with the material to be tested. Immediate availability of the results can then show significant benefits in directing a field sampling programme and in the interpretation of field data. In this mode of operation, the instrumentation is normally referred to as ‘hand-held’. For convenience and flexibility, portable XRF instrumentation normally comes with a ‘lab stand’ mode of operation, which appears to correspond to the way the authors have used instrumentation in this paper. However, in this mode, operation of portable XRF instrumentation is very similar to conventional laboratory based energy dispersive XRF instrumentation. Few if any of the specific advantages of PXRF are being used in this laboratory mode use. To avoid misleading readers, it is recommended that the title and introduction are revised to make clear that this paper uses PXRF instrumentation only in the laboratory and not the field.

P3L109/110 To what extent did the Cu/Zn ore program (and related matrix correction parameters) match the composition of the stream sediment samples being tested?

As well as preparing the four batches of sieved material for PXRF analysis, it is recommended that the authors consider preparing a compressed powder pellet as well and to include these data when comparing the Par1-Par4 results.

The purpose of paragraph 3.1 and Figure 1 is unclear. PXRF has the potential to detect all the elements in the periodic table but is restricted by a restricted detection limit capability compared to techniques like ICP-MS. The actual power of the technique in exploration studies may be its ability to detect Pd, Ag and Cd, which make no real impact in Fig 1. A more general comment about the detection limit capability of PXRF may be more appropriate.

Paragraph 3.2. Before comparing PXRF results with ICP data, it is important to validate the reliability of the PXRF results themselves to check that the selected PXRF calibration programme is appropriate for the materials being analysed and gives accurate datas. I would have expected, therefore, that the authors would have selected one or more reference materials that have matching matrices and analysed them as compressed powder pellets so that reference and analysed values can be compared with PXRF measurement results. I would expect that this would show that reliable measurement of lower atomic number elements (e.g., Mg, Al, Si, P, S, Cl) cannot be made by PXRF so that there is no point in comparing such data with ICP results in Table 2.

Table 2 and associated text: I am not a statistical expert but am surprised that the authors claim that elements with a correlation coefficient as low as 0.5 can be considered ‘relatively’ reliable. Are there published references that confirm this claim? It would be useful for the authors to include in the paper a few graphs of PXRF versus ICP data to demonstrate the reliability of their measurement results other than by correlation coefficient alone. It is wrong to say that for elements where the correlation coefficient is negative,  the pXRF analysis results of these elements are relatively unreliable. Such elements cannot be reliably measured by PXRF.

Table 3. I would like to see this table revised as follows. (1) Only include data for those elements that have been shown to be reliably measured by PXRF (by comparison with a matrix matched reference materials). (2) Concentrate the interpretation on elements that show a high correlation coefficient (>0.9) in Table 2 (i.e., when compared with ICP results). Otherwise there are too many variables that could affect the PXRF measurement results to allow clear interpretation of the data. I would not expect the PXRF results for Al listed in Table 3 to be reliable.

Paragraph 3.4 This paragraph appears to be about the reliability of the PXRF method and so should be presented before results of PAR1-PAR4. It is a bit confusing why these data have been included since it is well known that XRF results are affected by how fine the test material has been crushed.

I am not clear that the rest of the paper adds much to an understanding of PXRF performance in the analysis of stream sediments. In my view, the novel aspect of the paper is the reliability of the technique in measuring stream sediments of differing size fractions. However, such an assessment cannot be made without first evaluating the reliability of the PXRF results themselves.

The paper is well written in English.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

comments have been fully addressed

Author Response

Thank you for reviewing our paper and your approval of the revisions we made.

Reviewer 3 Report

Introduction: It would be useful to explain that PXRF is being used as a laboratory analytical technique and not for the in situ measurement of stream sediments in the field.

P2L90 “The 90 Hongliutan study area is located deep in the Kunlun mountain……. “ It would be sensible here to mention that PXRF is not able to measure Li and to summarise the pathfinder elements that might reveal a Li anomaly.

Table 3 It would be useful if the figure caption explained what the shaded areas represented (I realise that this is also mentioned in the text).

P5L176 “In addition, Table 3 also shows 176 that the correlation between the Zr content measured by pXRF in samples of various particle sizes and the laboratory analysis results is low (i.e., the correlation coefficient is less 178 than 0.5), but the pXRF analysis results of medium particle size (par 2 and par3) are relatively reliable.” For information, the authors might like to note the WIKI interpretation of the magnitude of correlation coefficients:

WIKI: Correlation coefficient: A statistic used to show how the scores from one measure relate to scores on a second measure for the same group of individuals. A high value (approaching +1.00) is a strong direct relationship, values near 0.50 are considered moderate and values below 0.30 are considered to show weak relationship. A low negative value (approaching -1.00) is similarly a strong inverse relationship, and values near 0.00 indicate little, if any, relationship

I cannot find any reference to Figure 2 in the text.

Table 4 Again it would be useful if the Figure caption explained the meaning of the shaded areas.

P6L211+ It would be clearer to use “count time” not “detection time”.

Figure 4. Once again, it would be useful if the figure caption summarised the meaning of the different lines (I realise that this is also explained in the text). In the text, explain briefly how the data were standardised.

Author Response

Please see the attachment

Author Response File: Author Response.pdf