Evaluation of a Spatial Heterodyne Spectrometer for Raman Spectroscopy of Minerals

Round 1

Reviewer 1 Report

This is an excellent paper describing the theory of a spatial heterodyne Raman spectrometer (SHRS) and its application to the measurement of rocks and minerals.  The authors do a good job of citing the relevant literature.  SHRS is a very timely topic and will be of great interest to the readers of this journal.  The authors compare the SHRS to a bench top Raman system and a fiber optic handheld system.  The comparison studies are appropriate and the results adequately discussed with one small but important exception.  The SNR calculations as described are not the appropriate for comparison between the SHRS and dispersive Raman spectrometers because of the way the noise is distributed differently in the two types of systems, as pointed out by the authors.  It appears the authors used the ratio between the maximum of the Raman signal and the standard deviation of a background region to calculate the SNR.  This only works for a dispersive Raman spectrometer when the background is high compared to the Raman band intensity, and more importantly, it doesn't work at all for the SHRS because the noise is equally distributed.  Because of this background noise comes the sum of shot noise in all spectral regions, unlike the dispersive spectrometer.  A better way to compare the SNR between any two types of Raman spectrometers is to make a series of measurements (say 100 spectra) using each instrument and calculate the std. dev. of the band intensity (S-band).  The SNR is then the ratio of the band intensity divided by S-band.  This should be discussed in the paper and the SNR either recalculated if possible.  If this is not possible using the data on hand, it would be fine to discuss the limitations of the technique used for the comparison.  

Author Response

We thank the reviewer for the insightful comment. Our initial choice to use signal-to-background noise ratio (SN_bR) was motivated by the practice of using this quantity as a figure of merit for characterizing spectra, as well as by the fact that we have only a single measurement from benchtop spectrometer (hence, proper signal-to-noise calculations are impossible). Now, instead, we removed the part of the text where SN_bR is discussed (lines 239-242, 271-272 and 332 in the first version of the manuscript) and added the following text in Results and Discussion section:

The detection of the signal depends on the level of noise recorded and produced by the instrument. As discussed in the literature (e.g. \cite{thorne}), the noise distributions in a dispersive spectrometer and in an interferometry-based spectrometer are fundamentally different: for a source dominated by the shot noise, in the former case the noise is proportional to the square root of the signal at each particular pixel that is recording a certain wavelength, while in the latter case the noise in each pixel is the contribution of the noise from all wavelengths. For this reason, SHS acts as a contrast-enhancing instrument, because the bands with weak intensities are dominated by the noise, while the bands with high intensity benefit from the multiplexing characteristic. Signal-to-noise ratio (SNR) is calculated as the ratio of the average of the intensity at the measured Raman band to the standard deviation of the band intensity and it is presented in the caption of the figures of SHRS spectra. For each case, a series of ten spectra is used. The comparison of SNR for both spectrometers yielded similar results, which testifies that the spectrum was rather flat, which can mainly be attributed to the fluorescence and luminescence from the source.

As a result, a new reference ([16]) is added. The values for SNR are added in the captions of the figures.

Reviewer 2 Report

1.1 The idea of the SHRS Raman device has been clearly presented in the introduction, including technical schemes, the principle of operation and construction details.

Different commercial applications of this type of instrument, are worth considering. In case of proper technical maturity of the system it would be an interesting alternative for bench top Raman spectrometer. It’s especially interesting if we may operate on a Fourier transformed spectrometer without moving parts (when modified Michelson interferometer is used). The authors conclude that it’s a feature of the presented SHS Raman construction.

So in my opinion portable version of the Raman spectrometer of a satisfactory accuracy would be a highly appreciated tool for example in a rough preliminary gemmological analysis but also in commercial use in so-called special core analysis in a hydrocarbon prospecting companies practice. This type the device would be also a valuable tool for service companies for a screening analysis of mineral composition or organic matter, as a complementary for FTIR portable analysis of mineral composition. Assuming the possibility of miniaturization and further development of this tool it could be considered for borehole logging analysis.

It seems it is highly desirable to constantly improve construction details of the device. Nowadays the methods for organic matter maturity analysis with the use of Raman spectroscopy  are in a stage of dynamic development. The information of the presence and maturity of the organic matter in the borehole profile is crucial for the assessment of hydrocarbon reservoir resources.

1.2 In terms of mineral identification it is highly desirable to have accuracy and sensitivity allowing to register at least two or three band maxima for each mineral phase.

It must be stressed that in case of rock analysis for ore mining, as it is in case of hydrocarbon recovery, we usually have multi-phase systems. Even in a spatial resolution of tens of millimeters we usually observe at least a few mineral phases, while quartz being present in almost every core sample.

1.3 Due to the fluorescence problem, high-quality results are often hard to achieve. The fluorescence is often caused  by presence of clay minerals, which are common in metal ore and mud rock samples, to name few. This problem makes geological samples generally hard to acquire by the different Raman techniques.

1.4 The problem of sample preparation is also an issue since in case of field analysis we usually don’t have possibility to polish the sample. Proper focusing of the beam matters in terms of acquiring quality spectra, which is not always easy in a field practice of portable devices.

1.5 Regarding the results from Oceans Optics dispersive spectrometer, it seems that it’s not an appropriate tool to perform the rock analysis for selected samples (the background is very high and also almost no Raman bands could be registered by the device. I would consider if it’s worth to show these results, maybe it would be enough to comment on it in a text giving eg signal to noise parameter value as a comment as it was actually done.

1.6 The authors could also consider baseline correction for most of the spectra – and alternatively show the uncorrected version on some separated figure. In the figure selected spectra may be grouped, to make visible the baseline alternation effects as a subject to method improvement. it’s stated in the paper that it’s not only the result of fluorescence.

32-33 lines – Please, if it's possible, construct a separate phrase for the information concerning snail shell. Although in terms of chemical composition it is mostly composed of calcium carbonate, the mineral form is calcite and aragonite, so it will be more proper not to put it in a line with some other minerals names but maybe in a new sentence.

298-299 lines - If possible, please reconstruct this conclusion since the results for ore samples, representing multiphase (heterogeneous) rocks should be subject to improvement. (I completely understand it's not an easy one since probably some optical settings of the device must be changed). However according to the results, only one phase was recognizable, for copper ore with the use of only one Raman band. Two Raman bands could be acquired for nickel ore – this is usually not enough to unambiguously recognize the mineral phase.

Hence it should be clearly stated in the conclusions, that when using SHS Raman at the present stage of development, some problems occur with the multiphase rock samples characterisation.

-This kind of result may be useful in case of detecting  known impurity which must be recognized eg during some synthesis process - but may not be satisfactory in terms of unknown mineral composition evaluation.

Author Response

Referee 2

1.1 The idea of the SHRS Raman device has been clearly presented in the introduction, including technical schemes, the principle of operation and construction details.

Different commercial applications of this type of instrument, are worth considering. In case of proper technical maturity of the system it would be an interesting alternative for bench top Raman spectrometer. It’s especially interesting if we may operate on a Fourier transformed spectrometer without moving parts (when modified Michelson interferometer is used). The authors conclude that it’s a feature of the presented SHS Raman construction.

So in my opinion portable version of the Raman spectrometer of a satisfactory accuracy would be a highly appreciated tool for example in a rough preliminary gemmological analysis but also in commercial use in so-called special core analysis in a hydrocarbon prospecting companies practice. This type the device would be also a valuable tool for service companies for a screening analysis of mineral composition or organic matter, as a complementary for FTIR portable analysis of mineral composition. Assuming the possibility of miniaturization and further development of this tool it could be considered for borehole logging analysis.

It seems it is highly desirable to constantly improve construction details of the device. Nowadays the methods for organic matter maturity analysis with the use of Raman spectroscopy  are in a stage of dynamic development. The information of the presence and maturity of the organic matter in the borehole profile is crucial for the assessment of hydrocarbon reservoir resources.

1.2 In terms of mineral identification it is highly desirable to have accuracy and sensitivity allowing to register at least two or three band maxima for each mineral phase.

It must be stressed that in case of rock analysis for ore mining, as it is in case of hydrocarbon recovery, we usually have multi-phase systems. Even in a spatial resolution of tens of millimeters we usually observe at least a few mineral phases, while quartz being present in almost every core sample.

1.3 Due to the fluorescence problem, high-quality results are often hard to achieve. The fluorescence is often caused  by presence of clay minerals, which are common in metal ore and mud rock samples, to name few. This problem makes geological samples generally hard to acquire by the different Raman techniques.

1.4 The problem of sample preparation is also an issue since in case of field analysis we usually don’t have possibility to polish the sample. Proper focusing of the beam matters in terms of acquiring quality spectra, which is not always easy in a field practice of portable devices.

> We thank the Reviewer for comments 1.1–1.4, which include several practical issues in potential fields of application of SHRS. Based on these comments we added to the Introduction:

As being able to deliver information on both, inorganic and organic constituents of samples, portable Raman devices have great potential for applications in gemmology, geological field studies, drill core logging, and hydrocarbon prospection. Due to their small size and robustness, spatial heterodyne Raman spectrometers are good candidates for such applications. Indeed, spatial heterogeneity and complex compositions of according samples are demanding, and fluorescence emission by either organic sample constituents or luminescent minerals are typical interferences in Raman spectra in general, which hamper some applications. Nevertheless, if a Raman device can be technically adapted to a certain application, e.g. by using an optimised excitation wavelength, such issues might be reduced, and a spatial heterodyne Raman spectrometer employing an ultraviolet laser source has already demonstrated to enable blocking of the major part of long-wavelength fluorescence emission by optical filters. [6] In general, spatial heterodyne Raman spectrometers are characterized by a high level of customizability…

1.5 Regarding the results from Oceans Optics dispersive spectrometer, it seems that it’s not an appropriate tool to perform the rock analysis for selected samples (the background is very high and also almost no Raman bands could be registered by the device. I would consider if it’s worth to show these results, maybe it would be enough to comment on it in a text giving eg signal to noise parameter value as a comment as it was actually done.

> As the differences between the results of SHRS and the used Ocean Optics spectrometer can be much better understood from graphs, we decided to leave Figures 11–13 as they are. The addition is in the captions of the figures, where we provide the signal-to-noise ratio measured and calculated by a statistics of ten measurements.

1.6 The authors could also consider baseline correction for most of the spectra – and alternatively show the uncorrected version on some separated figure. In the figure selected spectra may be grouped, to make visible the baseline alternation effects as a subject to method improvement. it’s stated in the paper that it’s not only the result of fluorescence.

> It is a common practice to post-process Raman spectra using various routines, such as smoothing, scaling, deconvolution, and baseline correction, with the aim to improve the results for both quantitative and quantitative presentation. The only post-processing applied to SHRS spectra presented in the main part of the text was scaling (normalization to unity), because the other routines did not improve their presentation, as demonstrated with Figure A1. For high resolution spectra, such as those obtained by SHRS, deconvolution is not necessary, because Raman bands are quite wide and well resolved. Baseline correction is necessary for quantitative measurements, but this is not the case in our work. Nevertheless, a combination of baseline correction and smoothing using a gentle Savitzky-Golay filter (polynomial of the third order and length of five data) is shown in the Figure A1. The applied baseline correction was based on the algorithm proposed in reference . The difference between the spectra is negligible and did not yield better results.

FIGURE A1 (see attachment)

Figure A1: Raw spectra (blue, below) and post-processed spectra (orange, above, shifted for unit value for clarity). Applied post-processing was baseline correction and smoothing.

Detailed recommendations for changes

32-33 lines – Please, if it's possible, construct a separate phrase for the information concerning snail shell. Although in terms of chemical composition it is mostly composed of calcium carbonate, the mineral form is calcite and aragonite, so it will be more proper not to put it in a line with some other minerals names but maybe in a new sentence.

> We changed the according sentence into

…calcite [5]; gypsum, quartz, and calcite in rocks, the latter also in a snail shell [6].

298-299 lines - If possible, please reconstruct this conclusion since the results for ore samples, representing multiphase (heterogeneous) rocks should be subject to improvement. (I completely understand it's not an easy one since probably some optical settings of the device must be changed). However according to the results, only one phase was recognizable, for copper ore with the use of only one Raman band. Two Raman bands could be acquired for nickel ore – this is usually not enough to unambiguously recognize the mineral phase.

Hence it should be clearly stated in the conclusions, that when using SHS Raman at the present stage of development, some problems occur with the multiphase rock samples characterisation.

-This kind of result may be useful in case of detecting  known impurity which must be recognized eg during some synthesis process - but may not be satisfactory in terms of unknown mineral composition evaluation.

> We agree with the reviewer’s comment on the difficulty of identifying minerals based on a small number of Raman bands. For this, we added to the Results and Discussion:

…modest increase of this background level.

Another issue is the detection of only 1–2 Raman bands in each SHR spectrum. Only some reasons for this are specific to SHRS, as the notch filters used in this version of the setup restrict the Raman shift range to > 400 cm^-1 and thus, exclude characteristic low-frequency modes of some minerals (compare the spectrum of quartz in Ref. [6], showing only one band for the same reason). As calcite, quartz, and forsterite are minerals commonly occurring in rocks, the spectra assignment can be considered unambiguous, even if mainly based on the most prominent Raman bands. In the case of calcite, the weak mode at 713 cm^-1 enables an undoubted discrimination from other calcium carbonate polymorphs (e.g., aragonite) and carbonates of other elements (e.g., siderite), which exhibit their most prominent Raman mode at a similar wavenumber. In Raman spectra of unexpected and unknown compounds, more bands in each acquisition would be necessary for their identification.

Similar to the collection of the control dataset…

and the following adjustment is done in the Conclusion section:

            …, but better than that of the dispersive spectrometer, based on the visibility of the bands on the spectra. Nevertheless, the fact that only a small number of bands is recorded shows that the SHRS needs further adjustments to reach the spectral range with a small Raman shift in order to be applied for unambiguos identification of unknown or multiphase mineral samples. Similarly to the benchtop Raman spectrometer…

Author Response File: Author Response.pdf