1. Introduction
About 90% of the titanium dioxide (TiO2) and titanium metal (Ti) produced worldwide is derived from ilmenite (FeTiO3). Due to its ability to form light and resistant metallic alloys and being inert to humans, titanium has applications in different areas, such as prostheses in biomedicine, pigments in manufacturing paints and plastics, the aerospace industry, and nanotechnology [1, 2]. Due to its electrical and magnetic properties, ilmenite behaves like a semiconductor material, which is important in the production of chips in the growing electronics industry [3]. The current semiconductor crisis, amplified by the COVID-19 pandemic, brought negative economic and technological consequences, indicating the significance of developing the worldwide supply chain [4].
Deposits of ilmenite and other heavy minerals of economic interest are found in many coastal regions of the world [5, 6] and are associated with the evolution of coastlines. An example is the placer deposit on the middle coast of the coastal plain of Rio Grande do Sul (PCRS) formed about 5.6 thousand years ago during the last transgressive and regressive sea-level cycles [7]. This deposit has been known since the 1970s. Currently, the Rio Grande Mineração S.A. mining company holds the rights to the mining areas and seeks environmental license from the competent bodies [8].
Orbital sensors with adequate spectral and spatial resolutions make remote sensing a valuable tool in geological studies. The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) has 14 spectral bands (near, shortwave, and thermal infrared) five of which are in the thermal infrared region (bands 10 to 14), with a spatial resolution of 90 meters. The ASTER is one of the sensors on board the spacecraft Terra, which is the first satellite of the Earth Observing System (EOS) series, aimed at long-term observations of the Earth’s surface, biosphere, atmosphere, and oceans. It flies at an altitude of 705 km in a sun-synchronous polar orbit, crossing the equator at 10:30 h GMT-3, with an average duty cycle of 8% per orbit, and a revisit period of 16 days [9].
The thermal infrared region of the electromagnetic spectrum corresponds to the radiance emitted from surfaces. From the radiance measured in the sensor, the parameters of land surface temperature (LST) and land surface emissivity (LSE) can be recovered from the inversion of Planck’s Law. These parameters are important in remote sensing and geological and environmental studies [10, 11]. Particularly, LSE is a key parameter in the characterization of geological targets, since the main mineral groups such as silicates, sulfates, and some oxides, present spectral diagnostic features in the TIR region [12–15].
The area coverage of an ASTER scene is approximately 60 x 60 km, processed at the L1 level (radiance recorded on the sensor), and stored at the LP-DAAC (Land Process Distributed Active Archive Center). ASTER higher-level L2 products are available on demand via Earth Data Search—NASA, such as AST_05 (LSE) and AST_08 (LST), generated from AST_09T (surface radiance TIR) [9]. The accuracy of these products depends on the accuracy of the atmospheric profiles used. Atmospheric parameters of temperature, water vapor, ozone, and aerosols are inputs to the MODTRAN® radiative transfer model. The AST_05 contains the pixel-by-pixel surface emissivity measurements, produced using the TES (Temperature and Emissivity Separation) algorithm [16, 17]. In geological remote sensing, the AST_05 emissivity product makes it possible to obtain the distribution of a given target (mineral) through hyperspectral image classification techniques practically and accurately [10, 18, 19].
In emission spectroscopy, molecular vibrational movements produce diagnostic spectral features in minerals [20]. Spectroscopy allows the use of reference spectra (endmembers) in comparison with spectra of pixels from orbital sensors [21, 22]. Some minerals, such as silicates and oxides, have a diagnostic feature in the field of TIR [12, 23, 24]. In TIR, the fundamental vibration of the quartz Si-O molecule produces a diagnostic spectral feature, between 8 and 9.5 μm, known as Reststrahlen bands [25, 26]. Ilmenite presents a rectilinear spectrum in the optical domain of the electromagnetic spectrum (VIS-SWIR) allowing it identifiable in the thermal region of the electromagnetic spectrum [27, 28]. Obtaining mineral signatures from the VIS to the TIR regions can be performed in the field, laboratory, or even by accessing available spectral libraries, such as ASTER, JPL (Jet Propulsion Laboratory), and JHU (Johns Hopkins University) [29, 30]. The μFT-IR Model 102F ultraspectral radiometer is used in the field and laboratory for retrieving temperature and emissivity variables from radiance measurements of targets [10, 31, 32]. To perform spectroscopy, it is important to know the composition of the analyzed sample, and for that, scanning electron microscopy is a technique capable of providing information regarding the degree of mineral purity of the sample [33, 34].
The development of orbital sensors and the advancements in image processing/classification algorithms have expanded the science of spectroscopy and geological remote sensing [35]. The Spectral Angle Mapper (SAM) algorithm, even though the spatial resolution of the image is a limitation, is a practical and fast method of mapping by comparing spectra [36, 37]. The Linear Spectral Unmixing (LSU) algorithm is a subpixel analysis model that overcomes such limitations to evaluate the proportion of the target of interest, considering the pixel value as a linear combination of the targets within the pixel [38–40]. Several studies on geological remote sensing have been developed using the SAM and LSU classification algorithms in recent years [5, 35, 41–43]. The current study calculated the surface distribution and concentration of ilmenite in the AST_05 product based on Spectral Angle Mapper (SAM) and Linear Spectral Unmixing (LSU) spectral classification methods and reference spectra (endmember) of ilmenite.
2. Methodology
The spectral signatures (from 8 to 14 μm) of ilmenite were obtained using a portable Fourier Transformed Infrared spectroradiometer (μFT-IR M102F) at the Geological Remote Sensing Laboratory (LabSRGeo) in the Federal University of Rio Grande do Sul (UFRGS), Brazil. In addition, the microscopic analysis of the samples was carried out in a Scanning Electron Microscope (SEM) JEOL JSM-6610LV at Laboratório de Geologia Isotópica (CPGq-IGEO-UFRGS) to estimate the degree of purity of the collected samples. The spectral signatures obtained were used as the reference spectra to identify the mineral target (ilmenite) from the surface emissivity product (AST_05) of the ASTER image, integrating the spaceborne and ground-based remote sensing approaches. The steps implemented are summarized in Fig 1.
[Figure omitted. See PDF.]
2.1. Study area and geological settings
The city of Mostardas (31°06′25″ S; 50°55′15″ W) is located on the middle coast of the coastal plain of Rio Grande do Sul state. It is limited to the east by the Atlantic Ocean and to the west by Patos Lagoon (Fig 2). The coastal plain of Rio Grande do Sul has a NE-SW orientation along 625 km of coastline, with areas having erosion and progradation. Heavy mineral deposits are found on the coast of the Rio Grande do Sul sector where the recessive barrier is dominant, on the extreme south coast near the city of Chuí, on the central coast of the state [7].
[Figure omitted. See PDF.]
The concurrent incidence of recessive barrier and transgressive dunes indicates a link between erosion, formation of transgressive dune fields, and concentration of heavy detrital minerals [44] as it occurs on the central coast, where the resort of Mostardas is located (sample collection site and study area of the present study). The evolution of the lagoon-barrier system, especially barriers III and IV, and the formation of the aeolian placer are related to the last two regressive-transgressive events of mean sea level during the Pleistocene (17.5 thousand years ago) and Holocene (5.6 thousand years ago) through the drowning and reworking of alluvial fan deposits and paleodrainage that cut the current continental shelf [45]. Barrier (IV) is predominantly regressive on the central coast, associated with erosion processes caused by the incidence of waves, storm surges, and the passage of cold fronts coming from the south and southeast, mainly in the autumn and winter months, and migration of dune deposits towards the continent by the northeasterly winds, mainly in the spring and summer months [46]. The deposit extends over barrier IV and anchors on barrier III. The study area is configured as an important mosaic of coastal ecosystems that harbors high biodiversity, classified as a Federal Conservation Unit for Full Protection (Lagoa do Peixe National Park), according to federal laws [47].
2.2. Collection and processing of heavy mineral samples
The mineral samples were collected at 4 sampling locations shown in Fig 2 (1. 31°9’12" S-050°48’32" W; 2. 31°8’54" S-050°49’1" W; 3. 31°8’31" S-050°49’10" W; 4. 31°7’9" S-050°50’7" W) in the dune field area, especially the frontal dunes, where the greatest accumulation of heavy minerals occurs. The analyses were carried out at the Center for Studies in Petrology and Geochemistry (CPGq), Institute of Geosciences, UFRGS.
First, the hydraulic separation of light and heavy minerals was carried out with the help of a pan disc. Then the light fraction was discarded (quartz and feldspar, while the heavy fraction after drying in an oven (~180°C / 30 min) was subjected to different amperages in the Frantz® Model L-1 isodynamic magnetic separator for concentration of the heavy mineral species based on their magnetic susceptibilities.
Magnetite and ilmenite are the first minerals separated from the fraction due to their high magnetic powers. These two species are difficult to distinguish from one another due to the similarity of color, texture, and magnetic power. In the present work, an amperage of 0.1 A for magnetite and 0.3 A for ilmenite was used with subsequent microanalysis of the samples in electron microscopy. After that, mineral concentrates were mounted in epoxy resin disks and polished for observation at SEM.
2.3. Analysis in scanning electron microscopy
For analyzing the mineral composition of the ilmenite concentrate, slides were made for microanalysis using a scanning electron microscope (SEM) JEOL JSM-6610LV. The SEM provides an image of the atomic density, from the emission and backscattering of an electron beam, a characteristic emission of X-rays. The higher the atomic density, the brighter the grain will appear in the image [48].
From BSE (backscattering electrons) digital images in grayscale and x-ray maps, it is possible to detect the presence of minerals rich in iron (Fe) and titanium (Ti) in the scanned area of the sample. Then, 33 grains of each sample (fractions 0.1 and 0.3 A) were selected for quantitative microanalysis. In the selected grains, a punctual scan was performed to obtain the characteristic X-rays using an EDS detector (energy dispersive spectrometer). A quantitative microanalysis provides the identification of the constituent elements as well as the concentration of these elements. The electron acceleration voltage used was 15 KeV, spot size 55, and working distance 11.2 mm, for a period of 30 s for the EDS analyses. Microanalysis allowed the selection of the most representative magnetic fraction of the mineral ilmenite for thermal spectroscopy analysis later in the laboratory.
The quantitative microanalysis allowed the classification of the ilmenite grains according to the degree of mineral alteration [49]. Primary ilmenites have a low degree of alteration, titanium oxide (TiO2) content between ~48–53%, and Ti/(Ti+Fe) ratio ≤ 0.5. As weathering acts on the grain, there is a hydrated ilmenite stage with a TiO2 concentration between ~53–60% and a Ti/(Ti+Fe) ratio between 0.5 and 0.6. The continuous process of ilmenite alteration continues with the formation of pseudo-rutile with TiO2 content ~60–70% and Ti/(Ti+Fe) ratio between 0.6 and 0.7. Following the alteration process, leucoxene is formed with a TiO2 content > 70% and a Ti/Ti+Fe ratio between 0.7 and 0.9. The alteration of ilmenite has as a final product the formation of rutile with TiO2 content > 90%. Variability in the TiO2 and FeO analyses due to impurities such as Al2O3 and SiO2 was eliminated by multiplying every analysis by 100 / [100—wt% (Al2O3 + SiO2)] [50].
2.4. Fourier-transform infrared spectroscopy analysis
After SEM analysis, the ilmenite concentrate sample was measured with a μFT-IR Model 102F spectroradiometer under controlled temperature (25°C) and luminosity in the LabSRGeo in UFRGS. With the use of μFT-IR, it is possible to calculate the emissivity and temperature of the target of interest (ilmenite samples in this case). The calibration of the device and atmospheric correction (removal of downwelling radiance L↓λ) were performed by using a reference gold plate (emissivity = 0.04) [51].
Then the emissivity of the sample was calculated according to Eq 1 given below:(1)
Where, Lλ is the spectral radiance, and Bλ(Ts) refers to the Planck equation given by Eq 2.(2)Where C1 and C2 are constants (C1 = 1.191 x 108 W μm4 sr-1 m-2, C2 = 1.439 x 104 μm K).
The removal of the downwelling radiance L↓λ occurred before and after the sample measurements. The accuracy of the downwelling radiance retrieval determines the accuracy of obtaining the sample’s emissivity and temperature measurements. The equipment was calibrated for forty-five minutes to obtain greater accuracy in emissivity measurements [26].
For the calibration process, distinctive temperatures for the cold blackbody (below the ambient temperature) and the hot blackbody (above the sample temperature) were chosen. The assumed calibration range was 10°C to 40°C. Samples of mineral concentrates were heated in an oven to approximately 60°C. Then the temperature and emissivity measurements of the ilmenite and quartz samples were obtained, generating the reference spectra of these minerals.
2.5. ASTER imagery
ASTER sensor data from the AST_05 emissivity product, processing level 2, acquired on demand through NASA’s Earth Data Search was used in the present study. In the AST05 products, the pixel-by-pixel surface emissivity is calculated using the temperature and emissivity separation (TES) method for the thermal infrared bands (5 bands each having 90 m spatial resolution) [16]. The spectral range of ASTER bands used are 8.125–8.475 μm (Band 10), 8.475–8.825 μm (Band 11), 8.925–9.275 μm (Band 12), 10.25–10.95 μm (Band 13), and 10.95–11.65 μm (Band 14), respectively.
The accuracy of the AST_05 products will depend on that of the AST_09T surface radiance product, which uses the MODTRAN® radiative transfer model to remove atmospheric effects [9]. The absolute accuracy of AST_05 is 0.05–0.1, and the relative accuracy is 0.005 (unitless). The date and time of the study area’s image acquisition was on 29/01/2014 at 13:36h GMT-3.
2.6. Digital image processing
To classify the AST_05 emissivity product image, the Spectral Angle Mapper (SAM) and Linear Spectral Unmixing (LSU) algorithms were used. For this analysis, the emissivity spectrum of the ilmenite sample obtained using μFT-IR Model 102F was used. The signatures were resampled for the response function of the thermal bands of the ASTER sensor (Fig 3). With this, the SAM allows determining or not the presence of a material of interest in the images from the previous knowledge of its spectrum [36, 37] while LSU allows estimating the abundance of ilmenite in pixels previously classified by SAM [38–40].
[Figure omitted. See PDF.]
2.6.1. Spectral Angle Mapper (SAM).
SAM is an algorithm of supervised classification pixel by pixel for digital images that is used to calculate similarities between spectra contained in the library of interest and those contained in the image (endmember vs. pixel). Despite the pioneering works that were developed in the 1990s (e.g. Van der Meer et al. [52]), SAM is still widely being used in geological mapping [53]. Input endmembers can be obtained in the field, in the laboratory, and from spectral libraries (i.e., reference endmembers), or extracted from pixels of the image itself (i.e., image endmembers). To be comparable, however, the spectra must have the same wavelength range. The similarity is calculated by estimating the mean angle of separation (in radians) between spectra pairs (reference endmember vs. pixel). The smaller the mean angle (Ɵ), the higher the similarity will be. Spectra are treated as vectors within an n-dimensional space, where n is the number of bands as shown in Eq 3.(3)Where cos-1 is the angle between the two spectra.
As the output, the SAM algorithm generates a rule image in grayscale for each endmember, in which digital number (DN) values express the mean separation angle formed between the reference spectrum (endmember) and the spectrum present in the pixel, weighted between 0 and 1. The closer to zero, the greater the similarity and the lighter the pixels. Thus, the output is also a binary image, in which pixels with angle values within a threshold determined by the user will be displayed in white (high correspondence), and outside the threshold (threshold) in black (low correspondence).
2.6.2. Linear Spectral Unmixing (LSU).
LSU is an algorithm that expresses the spectral unmixing analysis model in which the spectrum measured at the pixel is disintegrated into its constituents (fractions, with each fraction referring to each endmember) and abundances at the subpixel level [38–40]. Spectral unmixing models are widely used in geological studies with consistent results with the development of spectral unmixing analysis methods [54]. In LSU, the spectral response of each pixel per band is a linear combination of the spectral responses of its constituents, according to Eq 4.(4)Where rkis the spectral response of the pixel in the band k; sk is the spectral response of the endmember in the k band; fk is the endmember proportion; and i = indeterminate portion.
The number of equations for each pixel corresponds to the number of bands in the image. In the LSU model applied to the AST_05 image, the gray level recorded by the pixel is the linear combination of the emissivity of each of the pure targets (endmembers). In the AST_05 multispectral images, the five TIR bands will produce five linear equations (Eq 4) with n unknowns (endmembers). Each equation has two restrictions: 1) the value of the fraction must be between 0 and 1, and 2) the sum of the fractions and the indeterminate portion must be equal to 1.
3. Results
3.1. Electron microscopy analysis
From the processing of the field samples (hydraulic separation between light and heavy minerals, and magnetic separation between heavy minerals) the minerals were concentrated according to the degree of magnetism. Fractions of 0.1 A and 0.3 A were responsible for removing and concentrating magnetite and ilmenite from other heavy minerals. Fractions 0.1 and 0.3 A were analyzed using SEM and digital images were generated from backscattered electrons (BSE), in grayscale (Fig 4A) and characteristic x-ray map in RGB colors (Fig 4B) for qualitative analysis of minerals. The gray shades in the BSE image (Fig 4A) show the compositional contrast related to the atomic density of the grains, which is lighter when the atomic number is higher. The characteristic x-ray map from a previous study by the authors [55] (Fig 4B) collected by the EDS detector indicates the presence of iron (in red) and titanium (in green) in the grains. Based on this, it was possible to verify the prominent abundance of minerals rich in titanium in the scanned area of the sample.
[Figure omitted. See PDF.]
Digital images obtained in electron microscopy: a) the composition spectrum from backscattered electrons (BSE), and b) characteristic X-ray map with an EDS detector based on Hallal et al. [55].
In the samples, 33 grains of each fraction (0.1 and 0.3 A) were selected to obtain spectra for microanalysis of the elemental composition of the grains (Fig 4) with an EDS sensor. Thus, we obtained the most representative fraction of the mineral ilmenite. Fraction 0.3 A presented 78.8% of ilmenite (Table 1), while fraction 0.1 A presented 45.5% of ilmenite and 42.4% of magnetite. The separation between magnetite and ilmenite was partially achieved by increasing the amperage in the magnetic separator (from 0.1 to 0.3 A), with ilmenite being slightly less magnetic compared to magnetite. Qualitative microanalysis by EDS was performed for 66 grains from samples 0.1 A and 0.3 A. The spectrum below (Fig 5) is an example of ilmenite, composed of 50.6% TiO2, 37.9% of FeO, and 2.2% of MgO.
[Figure omitted. See PDF.]
[Figure omitted. See PDF.]
Ilmenite (FeTiO3) undergoes a mineral alteration process that involves the fortification of titanium in the grain as iron is leached from the crystalline structure. Elements such as Mg and Mn replace iron and Al and Si are adsorbed as impurities in the weathering process. This allows the classification of ilmenite according to the TiO2 concentration and the ratio between titanium and iron [Ti/(Ti+Fe)]. Primary ilmenite is the nomenclature given to grains with a low degree of alteration, with TiO2 between 48–53% and Ti/(Ti+Fe) ratio ≤ 0.5. The spectrum (Fig 5) is of primary ilmenite with TiO2 content of 50.6% and Ti/(Ti+Fe) ratio = 0.5.
Of the total of 66 microanalyses, 41 ilmenites were identified, and classified according to the degree of alteration, according to the scatter plots below (Fig 6). For the Ti/(Ti+Fe) ratio, the values were between 0.44 and 0.78 (Fig 6A). For the TiO2 content, the grains range from 44.5% to 69.8% (Fig 6B). This shows the continuous process of alteration that occurs in ilmenites during and after deposition, going from primary ilmenite with a low degree of alteration. With a reduction in iron content (Fe+2) and an increase in water content in the molecule, the so-called hydrated ilmenite appears as an intermediate phase between primary and pseudorutile. The continuation of the process leads to the transformation of pseudorutile into leucoxene. Transformation into leucoxene raises the impurity content. The average concentration of TiO2 in the analyzed ilmenite grains was 53%. The predominant ilmenite class was primary, with 20 spectra (48.8%) identified. Then, the most frequent class was hydrated ilmenite, with 10 grains (24.4%). Then pseudorutile with 7 spectra (17.1%), and finally leucoxene with 4 spectra (9.7%) (Table 2).
[Figure omitted. See PDF.]
[Figure omitted. See PDF.]
A microanalysis of the main impurity elements found in ilmenite was also performed (Table 3). Scatter plots showed the variation of these elements as the TiO2 content and degree of change increased (Fig 7). The concentration of magnesium oxide (MgO) varied between 0 and 3.4%, with an average value of 1.1%. A higher concentration of MgO as a minor element indicates an ilmenite with a lower degree of alteration as shown in the scatter plot (Fig 7A). For the concentration of manganese oxide (MnO), the minerals showed values between 0 and 5.8% (Fig 7B), with an average value of 0.9%. The increase in MnO concentration indicates a stage of alteration in ilmenite as this element replaces leached iron in the crystalline structure. The values found for the concentration of silica (SiO2) are between 0 and 6.5%, with an average value of 0.5%. Aluminum oxide (Al2O3) presented concentration values varied between 0 and 8.3%, with an average value of 0.8% (Fig 7C). Among the 41 ilmenite minerals analyzed, 34 do not contain SiO2 and 24 do not contain Al2O3. In these grains, grains are without the impurity of Al2O3 and SiO2 as there are no occurrences of Al2O3 and SiO2. Iron leaching decreases the volume of the grain, causing micro-fractures through which these impurities are incorporated. The occurrence of MgO and MnO occurs alternately in the ilmenite structure. They occur in place of iron by the similarity of atomic radius, filling the space in the octahedron (Fig 7D). The concentration of the oxides that make up the analyzed ilmenites is summarized in Table 3.
[Figure omitted. See PDF.]
[Figure omitted. See PDF.]
3.2. Emissivity measurements in the laboratory
In possession of the 78.8% ilmenite concentrate (0.3 A sample) the emissivity reading of the ilmenite sample was performed. The measurement was carried out using a μFT-IR Model 102F spectroradiometer under controlled temperature conditions (~25°C) and luminosity in LabSRGeo at UFRGS. The spectral curve for the ilmenite sample in the TIR domain (between 8 and 14 μm) shows a maximum emissivity behavior in the first wavelengths (~8–12.5 μm). An absorption peak is observed at approximately 12.7 μm, which is attributed to the vibration of the bonds of the ilmenite molecule (FeTiO3). The minimum emission feature is known as Restrahlen [56]. After the most intense absorption peak at 12.7 μm, a second absorption feature of lesser intensity occurs at approximately 12.9 μm, and after returning to emissivity close to 1 from 13.2 μm (Fig 8).
[Figure omitted. See PDF.]
The ilmenite hyperspectral signature was resampled concerning the central value and bandwidth for the response function of the thermal infrared bands of the ASTER data and to be compatible with the bands of the emissivity product image AST_05. In other words, it was resampled in the spectral intervals that correspond to the TIR bands (b10, b11, b12, b13, b14) of the ASTER sensor. The reading values performed by the μFT-IR Model 102F spectroradiometer, which correspond to the beginning and end intervals of a given spectral band, are resampled by the median function, generating a value corresponding to that spectral interval.
3.3. Mineral mapping of ilmenite in the coastal plain of Southern Brazil
False color composite (RGB 3:2:1) was performed for the study site, based on the ASTER L1B image. Band 3 (0.78–0.86 μm) of the ASTER sensor comprises the NIR wavelength, where vegetation has a greater spectral response. In the false-color composition performed, band 3 was inserted in the red channel of the RGB, with the vegetation highlighted in red in the image (Fig 9).
[Figure omitted. See PDF.]
Middle coast of the state of Rio Grande do Sul, Brazil.
The ilmenite signature obtained in the spectral library version 7 implemented in ENVI® [57], presents a low reflectance between Visible and SWIR wavelengths, with a homogeneous and straight line below 0.1. However, although not very pronounced, the presence of a spectral feature is observed between 0.5 and 1 μm (Fig 10). The presence of this feature indicated the possibility of using VIS and SWIR data for mapping this mineral but limited to the signal-to-noise ratio that can mask the response of this target against other components of the pixel. The ilmenite spectrum, with no significant feature in the VIS and SWIR, increases its confusion with other dark targets of low reflectance, such as organic matter, that occur in the dune field.
[Figure omitted. See PDF.]
Through the classification of the AST_05 image, the distribution of ilmenite for the area of interest was mapped. This result was sequentially generated using the Spectral Angle Mapper (SAM) and Linear Spectral Unmixing (LSU) algorithms. With this, it was possible to identify the presence of the ilmenite endmember (Fig 11) and to quantify the proportion of this endmember in each pixel of the scene, thus obtaining the distribution and concentration of ilmenite in the dune field of the central coast of the state of Rio Grande do Sul.
[Figure omitted. See PDF.]
SAM products for ilmenite in the mid-coast of Rio Grande do Sul, Brazil a) rule image b) binary image.
As a SAM product, a grayscale rule image is generated, in which the value of the digital numbers of the pixels (weighted between 0 and 1) expresses the angular difference between the spectra reference versus pixel. Lighter pixels with values closer to zero indicate greater similarity between the pair of compared spectra, while dark pixels closer to 1 indicate less correspondence (Fig 11A). In addition, the algorithm employs threshold classification, generating a binary image indicating the presence or absence of the endmember in the respective tested pixel, represented by white (presence) and black (non-presence), as in the image below (Fig 11B). Based on the presence of pixels classified by SAM, areas of interest were filtered for the application of LSU, and the abundance of ilmenite within these pixels was estimated.
The estimate of maximum ilmenite concentration for the study area was 29.6% of the sediment, represented by pixels in hot colors, while values of lower concentration of the mineral are represented by pixels in cold colors. The distribution of ilmenite indicates that the field of transgressive dunes on the central coast near the Mostardense resort, concentrates a significant amount of the mineral, with values ranging between about 10% and 25% of the sediment composed of ilmenite (Fig 12).
[Figure omitted. See PDF.]
The relationship between the spectral signatures of the pixels and the reference signature generated the ilmenite distribution and concentration map. In the graph represented by Fig 13, the distance between a point (representative of a pixel) and the diagonal line indicates the similarity between the spectra (pixel vs. reference). The location of the pixel over the diagonal line would indicate a coincidence between the spectral response of the pixel and the reference. This shows the occurrence of ilmenite in the pixel qualitatively. The ilmenite concentration in the pixel is estimated by the values of the modulus of the vectors, represented on the x and y axes of the graph. The mineral concentration varied between 0 and 29.6%, with the rest of the composition present in the indeterminate pixel (Fig 13).
[Figure omitted. See PDF.]
4. Discussion
The focus of this study was to map the surface distribution and concentration of ilmenite for the middle littoral portion of the coastal plain of Rio Grande do Sul. For this purpose, we used Spectral Angle Mapper (SAM) and Linear Spectral Unmixing (LSU) techniques applied to the AST_05 surface emissivity product.
The ilmenite deposit in the study area spreads from the frontal dunes to the transgressive dune field that anchors to the shore of the Lagoa dos Patos. For this dune field, we found ilmenite concentration values of up to 29.6% within the pixel (Fig 12), with the rest being indeterminate (up to 70.4%). The reference spectral signature used in the classification algorithms was obtained in the laboratory with μFT-IR emissivity measurements of the ilmenite sampled from the study area. The signature showed a diagnostic feature at approximately 12.7 μm caused by molecular vibration of the FeTiO3 bond (Fig 8).
The ilmenite sample used in the μFT-IR was previously evaluated for purity in electron microscopy analysis. The degree of purity estimated by microanalysis was 78.8% (Table 1). We also performed an initial analysis of the quality of the ilmenite deposit and found a low degree of alteration in these minerals, with a predominance of primary ilmenites (Fig 6), indicating economic potential [50].
4.1. Electron microscopic analysis
Scanning electron microscopy and microanalysis of X-ray characteristics in mineral studies allow the identification of the elemental composition, impurities, fracture zones, and pores of the grains [58, 59]. Previous works on the alteration of ilmenite in coastal deposits were carried out in South Africa [60], Brazil [61], Australia [50], and India [62]. Primary ilmenite corresponds to the purest grain, with less weathering and higher market value.
The ilmenite alteration process involves iron oxidation (Fe2+) and oxygen incorporation in the formation of pseudorutile (Fe2Ti3O9) [63]. Between the primary ilmenite and pseudorutile phases, an intermediate phase called hydrated ilmenite occurs, with a decrease in Fe2+ content and an increase in water content. The formation of pseudorutile implies a reduction of approximately 6% in the volume of the grain, causing microfractures and porosities. Continuing the alteration leads to the transformation of pseudorutile to leucoxene. The leucoxene phase has higher structural water content and impurities such as Mg2+, Mn2+, SiO2, and Al2O3. Due to impurities, leucoxene has no market value. In the final stage of alteration, titanium enrichment occurs, while iron is leached and leads to the formation of rutile [64]. Our results obtained from the microanalysis by applying the nomenclature given in Frost et al. [63] indicate the predominance of primary ilmenite grains (Fig 5 and Table 2) with a low degree of weathering (Fig 6).
The characteristic x-ray image (Fig 4B) and x-ray microanalysis of the grains (Fig 5) showed that the magnetic separation between magnetite and ilmenite was possible, with an increase in amperage, from 0.1 to 0.3 A. Because it is more magnetic, the magnetite was retained in the magnet in the first pass at 0.1 A, concentrating the ilmenite in the second pass at 0.3 A (Table 1). The alteration of the ilmenite causes a change in the magnetic susceptibility because of the leaching of iron and the incorporation of impurities, the magnetism decreases [65, 66].
In spectroscopic analysis, the construction of spectral libraries is preceded by electron microscopy to verify the purity of the sample to be measured [20, 27]. The data measured from the collected samples indicated a concentration of 78.8% of ilmenite in the sample (Table 1) with the change of amperage in the magnetic separator. Scanning electron microscopy and x-ray microanalysis allowed us to infer the stage of alteration of the ilmenite and the degree of purity of the sample used in the sequence of the work, with spectroscopy measurements in the laboratory.
4.2. Emissivity measurements in the laboratory
TIR spectroscopy has been used for understanding the spectral features of minerals since the 1960s. More recently, Laakso et al. [67] studied the spectral features of rare earth oxides in TIR, and Hecker et al. [21] studied the application of TIR spectroscopy for mapping feldspar by remote sensing. Christensen et al. [27] and Xie et al. [20] highlight the importance of determining the composition and purity of the sample and the descriptive care in the preparation and acquisition of spectra.
It is in the TIR where minerals of silicate groups and some oxides present spectral features [20, 24]. The spectral features result from the selective absorption of photons that depend on the elemental composition and crystalline structure of the mineral [68]. Absorption causes fundamental molecular vibrations, known as the Restrahlen feature, characterized by minimum emissivity values (maximum reflectance) [23, 69]. Christensen et al. [27] measured emissivity spectra of oxides such as ilmenite, hematite, magnetite, and rutile, identifying that the spectral features of these oxides occur at longer wavelengths (above 10 μm).
In the laboratory, under controlled conditions of temperature and luminosity, we took temperature and emissivity measurements of the ilmenite sample using μFT-IR. The spectral signatures we obtained identify the Restrahlen feature at 12.7 μm (Fig 8). Minor impurities, such as those found in the process of alteration of the ilmenite in the study area (Fig 7), do not affect the spectrum of the dominant mineralogy in the TIR. However, they can change the color of the sample and cause variations in the VIS spectrum [12]. Spectral libraries of ASTER, JPL (Jet Propulsion Laboratory), JHU (Johns Hopkins University), and USGS (United States Geological Survey), containing thousands of spectra of mineral species, from Visible to TIR in standard data format, describe the purity, size of the particle, the origin of the sample and the spectrometer provided a quick and practical resource for geological mapping and eliminate the tedious sample collection and laboratory measurement steps [29].
However, in this work, we chose to create the spectral library itself; according to Beddel et al. [70], better results in geological mapping can be expected from specific samples from the study area. To perform measurements with the μFT-IR, we considered the stabilization time and thermal balance of the equipment to avoid noise in the signal and obtain better accuracy, as recommended by Käfer et al. [26], highlighting the quality of high spectral resolution, and the level and portability of μFT-IR for validation of multispectral orbital data in TIR.
Ilmenite is an opaque mineral that does not present a diagnostic spectral feature in the VNIR-SWIR, with rectilinear spectra and low reflectance (Fig 10A) [27]. The opacity of this mineral is due to charge transfers between Fe2+, Fe3+, O2-, and Ti4+ ions. Ilmenite has a rhombohedral trigonal crystalline system with the elements iron and titanium alternating in layers of ions (Fe2+ and Ti4+) bonded to oxygen atoms (MO6). In addition to the metal-ligand Ti-O and Fe-O bonds, Ti-Fe intermetallic interaction also occurs [71].
The emission spectroscopy technique applied in our work acquires the spectrum in a similar way to orbital remote sensing data and allows a comparison between data. The spectral library we created for ilmenite served as input data for algorithms SAM and LSU to classify the distribution of this mineral in the placer deposit on the coastal plain of Rio Grande do Sul.
4.3. Mapping of ilmenite in the coastal plain of Southern Brazil
For the identification and quantification of the distribution of ilmenite on the dune field surface of the study area, we performed the integration of ground-based remote sensing data (i.e., μFTIR thermal spectroradiometer data) with orbital remote sensing data (i.e., surface imaging data obtained by the ASTER). Similar approaches were employed by Ducart et al. [72] for mapping iron oxides and clay minerals in Serra dos Carajás, northern Brazil, by Abubakar et al. [42] in identifying clay minerals associated with geothermal vents in Nigeria, and by Rejith et al. [73] to map heavy minerals in India.
The use of thermal infrared (wavelength between 8 μm– 14 μm) to investigate targets on the Earth’s surface has increased in recent decades [10]. Because of its spectral resolution with five bands positioned in the TIR, the ASTER sensor is extensively utilized in geological studies [53]. Furthermore, advances in digital processing software and image classification algorithms such as SAM and LSU are capable of generating geological maps with accuracy and agility at the subpixel level using ASTER data [37], as in the work on identification of hydrothermal alteration zones associated with Lead (Pb)–Zinc (Zn) mineralization [41, 74], hydrothermal alteration minerals and the presence of copper deposits in Iran [75], the assemblage of hydrothermal alteration minerals and listvenite formation, in Antarctica [54].
Only a few studies utilized remote sensing and digital image processing for mapping heavy mineral deposits in coastal areas. Among these, Chandrasekar et al. [5] used SAM to map coastal heavy minerals in India, showing the potential of applying hyperspectral remote sensing techniques combined with multispectral imaging in mapping heavy mineral deposits in the coastal zone; Ekanayake et al. [76] used Hyperion hyperspectral imaging data and digital processing to map ilmenite in Sri Lanka; Gazi et al. [77] mapped the presence of heavy minerals off the coast of Bangladesh using SAM and Landsat-8 sensor OLI data; Dhinesh et al. [78] used Landsat-7 ETM+ imagery for mapping paleochannels showing heavy mineral deposits in India; Rejith et al. [73] mapped the occurrence of heavy minerals in Kerala coast (India). Rejith et al. [73] used hyperspectral analysis techniques applied to Landsat multispectral imagery to reduce the dimensionality of the data and SAM classifier to map the occurrence of ilmenite in India.
In the present study, the AST_05 emissivity product image was chosen, which represents a fast and accurate way to obtain the LSE directly. The AST_05 products were validated in studies that used pseudo-invariant targets such as the Algodones Dunes, United States [79, 80], and Cidreira dunes, southern Brazil [19], in addition to the documentation in the sensor manual [9]. Rowan et al. [81] evaluated the AST_05 emissivity data when analyzing the lithological compositions of the Mordor complex, Australia. With reference spectra obtained from the image itself applied to the SAM algorithm; the authors were able to detect the variations in the silica content of the rocks. Aboelkhair et al. [82] identified granitic rocks of the albite type in the desert of Egypt using AST_05 image with emissivity spectra of laboratory samples and the JPL/NASA spectral library for rocks and minerals.
With this approach exposed, we identified the presence of ilmenite in the dune field of the study area by remote sensing using the SAM algorithm (Fig 11) on the AST_05 image and the ilmenite signature obtained by us in the laboratory (endmember) (Fig 8). The SAM classifier algorithm has been successfully applied in geological mapping over the past decades. Girouard et al. [83] applied the SAM method to images from different sensors (i.e., QuickBird and TM/Landsat-5). According to the authors [82], regardless of the low spatial resolution of the Landsat TM (30 m), the sensor was more effective in differentiating rocks due to its higher spectral resolution, with bands positioned in the SWIR region. Chandrasekar et al. [5] analyzed heavy mineral deposits in India, with reflectance data from the ETM/Landsat-7 sensor. Using “pure pixel” spectra obtained from the image and then applied to the SAM algorithm, the authors [5] were able to map the presence of zircon, garnet, and monazite. However, ilmenite did not rank well due to its low reflectance. Honarmand et al. [79] highlighted the potential of using TIR data from the ASTER sensor applied to the SAM algorithm for mapping silicic rocks of hydrothermal alteration. The spectral resolution of the ASTER sensor (specifically bands 10 and 12 in the TIR) detects the diagnostic feature of the Si-O vibration and makes it possible to identify the shift of the feature to longer wavelengths when high silicate composition presents.
To quantify the surface distribution over the dune field and its estimated concentration for ilmenite, we used the LSU model, which involves the acquisition of the reference spectrum, reduced to its characteristic bands (i.e., endmember) and a multispectral image containing the same bands collected pixel-by-pixel to produce an estimate of its proportion in the image [52]. In the model, we inserted the spectrum of the ilmenite sample that we obtained in the laboratory using μFT-IR (Fig 8). The estimated proportion of ilmenite focused on an interval from 0% to 29.6% in the pixels, and most of the pixels presented values between 10 and 15% of ilmenite (Figs 12 and 13).
The LSU is a linear model of spectral mixing that, when considering the endmembers spatially separated within the pixel, generates subpixel images (lower resolution pseudopixels) with the position and proportion of each endmember (fraction images) [84]. Applying a linear model (LSU) to spatially mixed targets, as in the case of a sedimentary deposit of heavy minerals, allowed us to obtain the proportion of the target of interest (i.e., ilmenite), but not its position estimated within the pixel. Abubakar et al. [42] used the SAM and LSU algorithms on SWIR data from the ASTER sensor, proving to be able to identify clay minerals (i.e., kaolinite, illite, monotremalite, and muscovite), hydrothermal alteration products indicators of geothermal systems in Nigeria. Gabr et al. [43] explored the potential of VNIR-SWIR data from the ASTER sensor to detect hydrothermal alteration zones that could be possible indicators of gold mineralization. The authors extracted endmembers of interest from the image itself (i.e., alunite, montmorillonite, kaolinite, goethite, hematite, limonite, and jarosite), applied to the spectral mixture model (LSU) and the SAM algorithm to locate the areas of high probability of gold mineralization. Hosseinjani and Tangestani [85], from ASTER data (from the VNIR and SWIR bands), identified the distribution and sub-pixel quantification of groups of hydrothermal alteration minerals (i.e., kaolinite-sericite, pyrophylite-alunite, and chlorite-calcite-epidote) in southeastern Iran. These groups of minerals are indicators of copper deposits. To map them, the authors [85] used endmembers extracted from the images and the LSU algorithm, which was able to discriminate them with a certainty of 82%. Also, according to the authors, the iterative processes of selection of characteristic bands of the targets of interest are relevant, since the results of the spectral mixture analysis are highly dependent on the selection of the endmembers.
In the present study, we chose the TIR, as the ilmenite response is greater in the thermal domain of the electromagnetic spectrum, with a higher signal-to-noise ratio, significantly increasing the accuracy of the estimates, in addition to reducing methodological steps and the computational cost involved. The present methodology can be used to estimate the purity of heavy mineral deposits in coastal areas. This is important as many coastal areas underwent contamination due to anthropogenic activities [86]. Furthermore, the radioactive risk associated with coastal sand samples, such as those (226Ra, 232Th, and 40K) associated with Egyptian black sand [87], can also be analyzed using remote sensing methods.
According to Dillenburg et al. [7] on C14 dating of the basal layer of peat, the heavy mineral deposit of the middle coast of the coastal plain of Rio Grande do Sul began to form in the last 1000 years and is still in the process of formation. It is believed that these minerals originated from the igneous and metamorphic rocks of the Sul-Riograndense Shield, in which paleochannels of drainage cut the current continental shelf [45]. According to the authors, with the rise in sea level and drowning of the channels, the heavy minerals were reworked and deposited with the evolution of the coastal barrier by the action of waves and winds on the coast. Similar kinds of complex geological formations have been observed in various coastal areas, such as in the Abu Murat area in Egypt [88]. Dhinesh et al. [78] used Landsat ETM+ imagery for mapping paleochannels, and the study was able to identify such geological features associated with the formation of deposits of heavy minerals such as ilmenite, magnetite, garnet, and zircon in a concentration of 10% of the sediment in the coastal zone of southern India.
The need for energy transition as well as reduced CO2 emissions, and the increasing demand for titanium-rich minerals consider new placer deposits an opportunity [89]. Magnetic properties of ilmenite are capable of increasing the energy efficiency of reactors powered by oxygen combustion [90]. This includes prospecting for lunar deposits by remote sensing using thermal infrared (20–40 μm), where oxides such as ilmenite show prominent spectral features [91]. Subasinghe et al. [6] highlight global growth in demand for heavy minerals, with forecast changes in global production and increases in ilmenite prices by 2030. The authors also address the importance of the mining industry following the United Nations’ sustainable development goals to overcome ecological and socioeconomic challenges involving mineral exploration. Ahmed et al. [92] proposed a technique for exploring heavy minerals on the coast of Bangladesh in which the excavation follows the angle of incidence of waves, to optimize exploration under areas with the highest concentration of heavy minerals, avoiding greater remobilization of sediment.
Ilmenite is a strategic mineral, with increasing demands, in which few countries hold a large part of the production of this ore. Therefore, it becomes important to discover new sources of heavy minerals and diversify the production chain. In this sense, our work innovates in mapping the ilmenite deposit on the coast of southern Brazil, by combining ground-based and satellite remote sensing techniques in the TIR. The creation of its spectral library obtained from laboratory measurements from local samples allowed the effective mapping of ilmenite. Our results give perspective to the use of TIR in mapping coastal deposits of heavy minerals. Integrating field spectrometer data and satellite remote sensing techniques with digital image processing is a promising tool for mapping heavy mineral (ilmenite) deposits in coastal zones using TIR data from the ASTER sensor.
Our work made a shallow quantification of the distribution of the material, which is important in the segment of the present study, a subsurface analysis with local soundings. The relevance of this note is given by the fact already pointed out by Dillenburg et al. [7] who point to a trend of subsurface accumulation greater than superficial for heavy detrital mineral assemblages in coastal deposits.
5. Conclusions
From the outcome of this study, the importance of thermal remote sensing in supporting geological prospecting can be underlined. From the integration of field spectrometry data and ASTER satellite imagery, together with analyses by electronic microscopy and digital image processing, the characterization and spatialization of the ilmenite concentration on the surface of the central coast of the state of Rio Grande do Sul in southern Brazil has been carried out. The laboratory measurements with μFT-IR identified spectral features of ilmenite between 8 and 14 μm. The construction of its own spectral library with ilmenite sampled in the study area, and resampled emissivity values for the thermal bands of the ASTER sensor allowed inferring ilmenite concentration between approximately 0 and 30% per pixel along the dune field.
The AST_05 emissivity product image uses SAM and LSU classification algorithms, dispensing the application of atmospheric correction methods and obtaining the surface emissivity parameter from thermal radiance. Scanning electron microscopy allows considering the ilmenite sample with a satisfactory degree of purity (~80%). An initial analysis of the alteration stage of the ilmenite from the study area was also carried out; minerals with a low degree of weathering were found, indicating the good quality of the deposit. For future works, new spectral laboratory measurements need to be carried out for the mineral magnetite, which together with ilmenite are the main species of heavy minerals in the study area, to expand the minerals of the local spectral library and analyze the behavior of the spectral features. We considered the analysis of the entire assemblage of heavy minerals to characterize the placer deposit on the middle coast of Rio Grande do Sul better.
Supporting information
S1 File. Inclusivity in global research.
https://doi.org/10.1371/journal.pone.0314238.s001
(DOCX)
References
1. 1. Liao C, Li Y, Tjong SC. Visible-Light Active Titanium Dioxide Nanomaterials with Bactericidal Properties. Nanomaterials. 2020; 10(1): 124. pmid:31936581
* View Article
* PubMed/NCBI
* Google Scholar
2. 2. Kroczek K, Turek P, Mazur D, Szczygielski J, Filip D, Brodowski R, et al. Characterization of Selected Materials in Medical Applications. Polymers. 2022; 14(8): 1526.
* View Article
* Google Scholar
3. 3. Kim T, Choi CH, Hur JS, Ha D, Kuh BJ, Kim Y, et al. Progress, Challenges, and Opportunities in Oxide Semiconductor Devices: A Key Building Block for Applications Ranging from Display Backplanes to 3D Integrated Semiconductor Chips. Advanced Materials. 2023; 35: 2204663.
* View Article
* Google Scholar
4. 4. Pihl I, Lind J, Henkow S (2022) How do Swedish companies handle disruptions in global supply chains stemming from the semiconductor crisis? Bachelor’s Thesis in Business Administration. Jonkoping International Business School, Jonkoping, Sweden.
5. 5. Chandrasekar N, Mujabar PS, Rajamanickam GV. Investigation of heavy-mineral deposits using multispectral satellite data. International Journal of Remote Sensing. 2011; 32: 8641–8655.
* View Article
* Google Scholar
6. 6. Subasinghe CS, Ratnayake AS, Roser B, Sudesh M, Wijewardhana DU, Attanayake N, et al. Global distribution, genesis, exploitation, applications, production, and demand of industrial heavy minerals. Arab J Geosci, 2022; 15: 1616.
* View Article
* Google Scholar
7. 7. Dillenburg SR, Tomazelli LJ, Barboza EG. Barrier evolution and placer formation at Bojuru Southern Brazil. Marine Geology. 2004; 203: 43–56.
* View Article
* Google Scholar
8. 8. RGM; HAR; CPEA. RIMA—Relatório de Impacto Ambiental: Projeto Retiro. São José do Norte, April 2014.
9. 9. Abrams M, Yamaguchi Y, Yuan K (2002) Users handbook, version 3.
* View Article
* Google Scholar
10. 10. Tang H, Li ZL (ed.). Quantitative remote sensing in thermal infrared: theory and applications. Springer, 2014.
11. 11. Sekertekin A, Bonafoni S. Land Surface Temperature Retrieval from Landsat 5, 7, and 8 over Rural Areas: Assessment of Different Retrieval Algorithms and Emissivity Models and Toolbox Implementation. Remote Sensing. 2020; 12(2): 294.
* View Article
* Google Scholar
12. 12. Lane MD, Bishop JL (2009) Mid-infrared (Thermal) Emission and Reflectance Spectroscopy. In: Bishop JL, Bell JF III, Moersch JE (Eds.) Remote Compositional Analysis, p. 42–67.
* View Article
* Google Scholar
13. 13. Vicente LE, Filho CRS. Identification of mineral components in tropical soils using reflectance spectroscopy and advanced spaceborne thermal emission and reflection radiometer (ASTER) data. Remote Sensing of Environment. 2011; 115: 1824–1836.
* View Article
* Google Scholar
14. 14. Adams JB, Smith MO, Johnson PE. Spectral mixture modelling: a new analysis of rock and soil types at the Viking Lander I site. Journal of Geophysical Research. 1986; 91: 8098–8112.
* View Article
* Google Scholar
15. 15. Käfer PS, Rolim SBA, Diaz LR, Rocha NS, Iglessias ML, Rex FE. Comparative analysis of split-window and single-channel algorithms for land surface temperature retrieval of a pseudo-invariant target. Bol. Ciênc. Geod. 2020; 26 (02): e2020008.
* View Article
* Google Scholar
16. 16. Gillespie A, Rokugawa S, Matsunaga T, Cothern JS, Hook S, Kahle ABA. Temperature and Emissivity Separation Algorithm for Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Images. IEEE Transactions on Geoscience and Remote Sensing. 1998; 36: 1113–1126.
* View Article
* Google Scholar
17. 17. Miao X, Zhang Y, Zhang J, Zhou X. Temperature and Emissivity Smoothing Separation with Nonlinear Relation of Brightness Temperature and Emissivity for Thermal Infrared Sensors.
* View Article
* Google Scholar
18. 18. Hulley G, Baldridge A (2013) Validation of Thermal Infrared (TIR) Emissivity Spectra Using Pseudo-Invariant Sand Dune Sites. In: Kuenzer C, Dech S (Eds.). Thermal infrared remote sensing: sensors, methods, applications. Springer, p. 515–537.
19. 19. Rolim SBA, Veettil BK, Käfer PS, Grondona AEB, Iglesias MJ, Diaz LR, et al. Comparison of Emissivity Retrieval Methods from ASTER Data Using Fourier-Transform Infrared Spectroscopy. Acta Geophysica. 2020; 68: 1867–1876.
* View Article
* Google Scholar
20. 20. Xie B, Mao W, Peng B, Zhou S, Wu . Thermal-Infrared Spectral Feature Analysis and Spectral Identification of Monzonite Using Feature-Oriented Principal Component Analysis. Minerals. 2022; 12(5): 508.
* View Article
* Google Scholar
21. 21. Hecker C, Der Meijde MV, Van Der Meer FD. Thermal infrared spectroscopy on feldspars—Successes, limitations and their implications for remote sensing. Earth-Science Reviews. 2010; 103: 60–70.
* View Article
* Google Scholar
22. 22. Alonso K, Bachmann M, Burch K, Carmona E, Cerra D, de los Reyes R, et al. Data Products, Quality and Validation of the DLR Earth Sensing Imaging Spectrometer (DESIS). Sensors. 2019; 19(20): 4471. pmid:31618940
* View Article
* PubMed/NCBI
* Google Scholar
23. 23. Laukamp C, Rodger A, LeGras M, Lampinen H, Lau IC, Pejcic B, et al. Mineral Physicochemistry Underlying Feature-Based Extraction of Mineral Abundance and Composition from Shortwave, Mid and Thermal Infrared Reflectance Spectra. Minerals. 2021; 11(4): 347.
* View Article
* Google Scholar
24. 24. Savitri KP, Hecker C, van derMeer FD, Sidik RP. VNIR-SWIR infrared (imaging) spectroscopy for geothermal exploration: Current status and future directions. Geothermics. 2021; 96: 102178.
* View Article
* Google Scholar
25. 25. Hiesinger H, Helber J, Alemanno G, Bauch KE, D’Amore M, Maturilli A, et al. Studying the Composition and Mineralogy of the Hermean Surface with the Mercury Radiometer and Thermal Infrared Spectrometer (MERTIS) for the BepiColombo Mission: An Update. Space Sci Rev. 2020; 216: 110.
* View Article
* Google Scholar
26. 26. Käfer PS, Rolim SBA, Iglessias ML, Rocha NS, Diaz LR. Land surface temperature retrieval by LANDSAT 8 thermal band: applications of laboratory and field measurements. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2019; 12: 2332–2341.
* View Article
* Google Scholar
27. 27. Christensen PR, Banfield JL, Hamilton VE, Howard DA, Lane MD, Piatek JL, et al. A Thermal emission spectral library of rock-forming minerals. Journal of Geophysical Research. 2000; 105: 9735–9739.
* View Article
* Google Scholar
28. 28. Chen L, Wei G, Liu K. Influences of reaction and thermal conditions on microstructural evolutions of ilmenite oxygen carriers for chemical looping combustion. Fuel Processing Technology. 2022; 228: 107139.
* View Article
* Google Scholar
29. 29. Baldridge AM, Hook SJ, Grove CI, Rivera G. The ASTER Spectral Library Version 2.0. Remote Sensing of Environment. 2009; 113: 711–715.
* View Article
* Google Scholar
30. 30. Krupnik D, Khan S. Close-range, ground-based hyperspectral imaging for mining applications at various scales: Review and case studies. Earth-Science Reviews. 2019; 198: 102952.
* View Article
* Google Scholar
31. 31. Gao L, Cao L, Zhong Y, Jia Z. Field-Based High-Quality Emissivity Spectra Measurement Using a Fourier Transform Thermal Infrared Hyperspectral Imager. Remote Sensing. 2021; 13(21): 4453.
* View Article
* Google Scholar
32. 32. Gerace A, Eon R, Kleynhans T, Raqueno N, Webber CM, Ely P, et al. (2022) In-flight performance of the Multi-band Uncooled Radiometer Instrument (MURI) thermal sensor. Remote Sensing of Environment 279: 113086.
* View Article
* Google Scholar
33. 33. Alqaheem Y, Alomair AA. Microscopy and Spectroscopy Techniques for Characterization of Polymeric Membranes. Membranes. 2020; 10(2): 33. pmid:32102383
* View Article
* PubMed/NCBI
* Google Scholar
34. 34. Vanderbruggen A, Gugala E, Blannin R, Bachmann K, Serna-Guerrero R, Rudolph M Automated mineralogy as a novel approach for the compositional and textural characterization of spent lithium-ion batteries. Minerals Engineering. 2021; 169: 106924.
* View Article
* Google Scholar
35. 35. Rejith RG, Sundararajan M, Gnanappazham L, Loveson VJ. Satellite-based spectral mapping (ASTER and landsat data) of mineralogical signatures of beach sediments: a precursor insight. Geocarto International. 2020; 37: 2580–2603.
* View Article
* Google Scholar
36. 36. Noori L, Pour AB, Askari G, Taghipour N, Pradhan B, Lee C-W, et al. Comparison of Different Algorithms to Map Hydrothermal Alteration Zones Using ASTER Remote Sensing Data for Polymetallic Vein-Type Ore Exploration: Toroud–Chahshirin Magmatic Belt (TCMB), North Iran. Remote Sensing. 2019; 11(5): 495.
* View Article
* Google Scholar
37. 37. Moud MF, van Ruitenbeek F, Hewson R, van der Meijde M. An Approach to Accuracy Assessment of ASTER Derived Mineral Maps. Remote Sensing. 2021; 13(13): 2499.
* View Article
* Google Scholar
38. 38. Keshava V, Mustard JF. Spectral unmixing. IEEE Processing Magazine. 2002; 2: 44–57.
* View Article
* Google Scholar
39. 39. Ghosh G, Kumar S, Saha SK. Hyperspectral Satellite Data in Mapping Salt-Affected Soils Using Linear Spectral Unmixing Analysis. J Indian Soc Remote Sens. 2012; 40: 129–136.
* View Article
* Google Scholar
40. 40. Yu J, Chen D, Lin Y, Ye S. Comparison of linear and nonlinear spectral unmixing approaches: a case study with multispectral TM imagery. International Journal of Remote Sensing. 2017; 38: 773–795.
* View Article
* Google Scholar
41. 41. Sekandari M, Masoumi I, Pour AB, Muslim AM, Hossain MS, Misra A. ASTER and WorldView-3 satellite data for mapping lithology and alteration minerals associated with Pb-Zn mineralization. Geocarto international. 2022; 37(6): 1782–1812.
* View Article
* Google Scholar
42. 42. Abubakar AH, Hashim M, Pour AB. Spectral Mineral Mapping for Characterization of Subtle Geothermal Prospects Using ASTER Data. Journal of Physics Conference Series. 2017; 852: 012024.
* View Article
* Google Scholar
43. 43. Gabr S, Ghulam A, Kusky T. Detecting areas of high-potential gold mineralization using ASTER data. Ore Geology Reviews. 2010; 38: 59–69.
* View Article
* Google Scholar
44. 44. Martinho TC, Dillenburg SR, Hesp P. Wave energy and longshore sediment transport gradients controlling barrier evolution in Rio Grande do Sul, Brazil. Journal of Coastal Research. 2009; 25: 285–293.
* View Article
* Google Scholar
45. 45. Weschenfelder J, Corrêa ICS, Aliotta S, Baitelli R. Paleochannels related to late Quaternary sea-level changes in Southern Brazil. Brazilian Journal of Oceanography. 2010; 58: 35–44.
* View Article
* Google Scholar
46. 46. Calliari LJ, Speranski N, Boukareva I. Stable focus of wave rays as a reason of local erosion at the southern Brazilian coast. Journal of Coastal Research. 1998; 26: 19–23.
* View Article
* Google Scholar
47. 47. Brasil. Lei n° 9.985, de 18 de julho de 2000. Institui o Sistema Nacional de Unidades de Conservação da Natureza e dá outras providências. Available at: https://www.planalto.gov.br/ccivil_03/leis/l9985.htm.
48. 48. Reed SJB (2006) Electron Microprobe Analysis and Scanning Electron Microscopy in Geology. 2nd Ed. Cambridge: UK, p. 232.
49. 49. Rahman MJJ, Pownceby MI, Rana MS. Distribution and characterization of heavy minerals in Meghna River sand deposits, Bangladesh. Ore Geology Reviews. 2022; 143: 104773.
* View Article
* Google Scholar
50. 50. Pownceby MI. Alteration and associated impurity element enrichment in detrital ilmenites from the Murray Basin, southeast Australia: a product of multistage alteration. Australian Journal of Earth Sciences. 2010; 57: 243–258.
* View Article
* Google Scholar
51. 51. D&P Instruments (2009) Model 102 Win2K Instruction Manual. Simsbury, CT.
52. 52. Van Der Meer F, Vazquez-Torres M, Dijk PMV. Spectral Characterization of Ophiolite Lithologies in the Troodos Ophiolite Complex of Cyprus and its Potential in Prospecting for Massive Sulphide Deposits. International Journal Remote Sensing. 1997; 18: 1245–1257.
* View Article
* Google Scholar
53. 53. Abrams M, Yamaguchi Y. Twenty years of ASTER contributions to lithologic mapping and mineral exploration. Remote Sensing. 2019; 11: 1394.
* View Article
* Google Scholar
54. 54. Pour A, Park Y, Crispini L, Läufer A, Kuk Hong J, Park TYS, et al. Mapping listvenite occurrences in the damage zones of Northern Victoria Land, Antarctica using ASTER satellite remote sensing data. Remote Sensing. 2019; 11(12): 1408.
* View Article
* Google Scholar
55. 55. Hallal GP, de Almeida Espinoza JM, Veettil BK, Porcher CC, Oliveira Righi da Silva M, Beatriz Alves Rolim S. Mapping heavy mineral deposits on the coast of the state of Rio Grande do Sul (Brazil) using orbital and proximal remote sensing. PLoS ONE. 2024; 19(9): e0309043. pmid:39240841
* View Article
* PubMed/NCBI
* Google Scholar
56. 56. Martin AC, Emery JP, Loeffler MJ. Spectral effects of regolith porosity in the mid-IR–Forsteritic olivine. Icarus. 2022; 378: 114921.
* View Article
* Google Scholar
57. 57. United States Geological Society (USGS, 2021) Mineral Commodity Summaries 2021: U. S. Geological Survey, p. 200.
58. 58. Asadi P, Beckingham LE. Intelligent framework for mineral segmentation and fluid-accessible surface area analysis in scanning electron microscopy. Applied Geochemistry. 2022; 143: 105387.
* View Article
* Google Scholar
59. 59. Han S, Lӧhr SC, Abbott AN, Baldermann A, Farkaš J, McMahon W, et al. Earth system science applications of next-generation SEMEDS automated mineral mapping. Front. Earth Sci. 2022; 10: 956912.
* View Article
* Google Scholar
60. 60. Ramakokovhu MM, Olubambi PA, Mbaya RKK, Mojisola T, Teffo ML. Mineralogical and Leaching Characteristics of Altered Ilmenite Beach Placer Sands. Minerals. 2022; 10: 1011–1022.
* View Article
* Google Scholar
61. 61. Gonçalves CC, Braga PFA. Heavy Mineral Sands in Brazil: Deposits, Characteristics, and Extraction Potential of Selected Areas. Minerals. 2019; 9(3):176.
* View Article
* Google Scholar
62. 62. Rao DS, Vijayakumar TV, Prabhakar S, Bhaskar Raju G, Ghosh TK. Alteration Characteristics of Ilmenites from South India. Journal of Minerals & Materials Characterization & Engineering. 2005; 4: 47–59.
* View Article
* Google Scholar
63. 63. Frost MT, Grey IE, Harrowfield IR, Mason K. The Dependence of Alumina and Silica Contents on the Extent of Alteration of Weathered Ilmenites from Western Australia. Mineralogical Magazine. 1983; 47: 201–208.
* View Article
* Google Scholar
64. 64. Meinhold G. Rutile and its Applications in Earth Sciences. Earth-Science Reviews. 2020; 102: 1–28.
* View Article
* Google Scholar
65. 65. Mehdilo A, Irannajad M, Rezai B. Chemical and mineralogical composition of ilmenite: Effects on physical and surface properties. Minerals Engineering. 2015; 70: 64–76.
* View Article
* Google Scholar
66. 66. Liu C, Wang W, Wang H, Zhu C, Ren B. A Review on Removal of Iron Impurities from Quartz Mineral. Minerals. 2023; 13(9): 1128.
* View Article
* Google Scholar
67. 67. Laakso K, Turner DJ, Rivard B, Sanchez AA. The Long-Waveinfrared (8–12 μm) spectral features of selected rare earth element—Bearing carbonate, phosphate and silicate minerals. International Journal of Applied Earth Observation and Geoinformation. 2019; 76: 77–83.
* View Article
* Google Scholar
68. 68. Zhang L, Yang J, Zhao X, Sun F, Zuo X, Nan J. Small-molecule surface-modified bismuth-based semiconductors as a new class of visible-light-driven photocatalytic materials: Structure-dependent photocatalytic properties and photosensitization mechanism. Chemical Engineering Journal. 2020; 380: 122546.
* View Article
* Google Scholar
69. 69. Salisbury JW, Wald A. The Role of Volume Scattering in Reducing Spectral Contrast of Reststrahlen Bands in Spectra of Powdered Minerals. ICARUS. 1992; 96: 121–128.
* View Article
* Google Scholar
70. 70. Bedell RL, Rivard B, Browning D, Coolbaugh M (2017) Thermal Infrared Sensing for Exploration and Mining–An Update on Relevant Systems for Remote Acquisition to Drill Core Scanning. Spectral Geology and Remote Sensing. In: Proceedings of Exploration 17: Sixth Decennial International Conference on Mineral Exploration. p. 881–897.
71. 71. Agui A, Uozumi T, Mizumaki M, Kaambre T. Intermetallic Charge Transfer in FeTiO3 Probed by Resonant Inelastic Soft X-Ray Scattering. Physical Review. 2009; 79: 92402–092405.
* View Article
* Google Scholar
72. 72. Ducart DF, Silva AM, Toledo CLB, Assis LM. Mapping iron oxides with Landsat-8/OLI and EO-1/Hyperion imagery from the Serra Norte iron deposits in the Carajás Mineral Province, Brazil. Brazilian Journal of Geology. 2016; 46: 331–349.
* View Article
* Google Scholar
73. 73. Rejith RG, Sundararajan M, Venkatesan S, Mohammed-Aslam MA (2021) Remote sensing for exploring heavy mineral deposits: a case study of Chavara and Manavalakurichi deposits, southwest coast of India. In: Remote Sensing of Ocean and Coastal Environments (pp. 177–188). Elsevier.
* View Article
* Google Scholar
74. 74. Ghorbani A, Honarmand M, Shahriari H, Hassani MJ. Regional scale prospecting for non-sulphide zinc deposits using ASTER data and different spectral processing methods. International Journal of Remote Sensing. 2019; 40(23): 8647–8667.
* View Article
* Google Scholar
75. 75. Yousefi M, Tabatabaei SH, Rikhtehgaran R, Pour AB, Pradhan B. Application of Dirichlet process and support vector machine techniques for mapping alteration zones associated with porphyry copper deposit using ASTER remote sensing imagery. Minerals. 2021; 11(11): 1235.
* View Article
* Google Scholar
76. 76. Ekanayake EMMB Vithana SSP, Ekanayake EMHEB Rathnayake ARMAN, Abeysekara AMR Oorloff TSJ, et al. Mapping ilmenite deposit in Pulmudai, Sri Lanka using a hyperspectral imaging-based surface mineral mapping method. Journal of the National Science Foundation of Sri Lanka. 2019; 47: 271–284.
* View Article
* Google Scholar
77. 77. Gazi MY, Tafhim KT, Ahmed MK, Islam MA. Investigation of heavy-mineral deposits using multispectral satellite imagery in the eastern coastal margin of Bangladesh. Earth Sci. Malaysia. 2019; 3(2): 16–22.
* View Article
* Google Scholar
78. 78. Dhinesh S, Udayaganesan P, Ramakrishnan D, Rajani R, Xavier Ignaci Muthu J. An integrated strategy for the exploration of palaeofluvial placer deposits. Applied Geomatics. 2021; 13: 165–177.
* View Article
* Google Scholar
79. 79. Honarmand M, Ranjbar H, Shahabpour J. Application of Principal Component Analysis and Spectral Angle Mapper in the Mapping of Hydrothermal Alteration in the Jebal-Barez Area, Southeastern Iran. Resource Geology. 2012; 62: 119–139.
* View Article
* Google Scholar
80. 80. Mushkin A, Gillespie AR, Abbott EA, Batbaatar J, Hulley G, Tan H, et al. Validation of ASTER emissivity retrieval using the Mako airborne TIR imaging spectrometer at the Algodones dune Field in Southern California, USA. Remote Sensing. 2020; 12(5), 815.
* View Article
* Google Scholar
81. 81. Rowan LC, Mars JC, Simpson CJ. Lithologic mapping of the Mordor, NT, Australia ultramafic complex by using the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). Remote sensing of Environment. 2005; 99: 105–126.
* View Article
* Google Scholar
82. 82. Aboelkhair H, Ninomiya Y, Watanabe Y, Sato I. Processing and Interpretation of ASTER TIR Data for Mapping of Rare-Metal-Enriched Albite Granitoids in the Central Eastern Desert of Egypt. Journal of African Earth Sciences. 2010; 58: 141–151.
* View Article
* Google Scholar
83. 83. Girouard G, Bannari A, Harti AE, Desrochers A (2004) Validated Spectral Angle Mapper Algorithm for Geological Mapping: comparative study between Quickbird and Landsat-TM. Anais ISPRS XX, Geo-Imagery Bridging Continents, p.12–23.
* View Article
* Google Scholar
84. 84. Quintano C, Fernandez-Manso A, Shimabukuro YE, Pereira G. Spectral Unmixing. International Journal of Remote Sensing. 2012; 33: 5307–5340.
* View Article
* Google Scholar
85. 85. Hosseinjani M, Tangestani MH. Mapping alteration minerals using sub-pixel unmixing of ASTER data in the Sarduiyeh area, SE Kerman, Iran. International Journal of Digital Earth. 2011; 4: 487–504.
* View Article
* Google Scholar
86. 86. Seif RA, Ene A, Zakaly HMH, Sallam AM, Taalab SA, Fnais MS, et al. Distribution of Heavy Metals along the Mediterranean Shoreline from Baltim to El-Burullus (Egypt): Consequences for Possible Contamination. Minerals. 2024; 14(9):931.
* View Article
* Google Scholar
87. 87. Zakaly HMH, Awad HA, Abbasi A, Almousa N, El-Salam LMA, Mostafa AMA, et al. (2024) Radioactive and mineralogical assessment of mediterranean black sands: a systematic analysis and health risk evaluation. J Radioanal Nucl Chem. 2024; 333, 1937–1947.
* View Article
* Google Scholar
88. 88. Awad HA, El-Leil IA, Granovskaya NV, Ene A, Tolba A, Kamel M, et al. Petrotectonic and Geochemistry of Abu Murrat Ilmenite Bearing Gabbro, NE Desert, Egypt. Romanian Journal of Physics. 2022; 67: 808.
* View Article
* Google Scholar
89. 89. WIPO (2022) Production of titanium and titanium dioxide from ilmenite and related applications. Geneva: World Intellectual Property Organization.
90. 90. Faust R, Lamarca I, Schaefer A, Lind F, Knutsson P. Magnetic properties of ilmenite used for oxygen carrier-aided combustion. Fuel. 2023; 340: 127593.
* View Article
* Google Scholar
91. 91. Ciazela J, Bakala J, Kowaliński M, Pieterek B, Steslicki M, Ciazela M, et al. Lunar Ore Geology and feasibility of ore mineral detection using a far-IR spectrometer. Frontiers in Earth Science. 2023; 11: 1190825.
* View Article
* Google Scholar
92. 92. Ahmed MT, Monir MU, Rajib M, Sarkar SM, Hasan Y, Quamruzzaman C, et al. Heavy minerals identification and extraction along coastal sediments using placer mining technique. Journal of Sedimentary Environments. 2023; 8(1): 81–95.
* View Article
* Google Scholar
Citation: Prates Hallal G, Porcher CC, Veettil BK, de Almeida Espinoza JM, Beatriz Alves Rolim S (2025) Distribution of ilmenite minerals in placer deposits along the middle coast of Southern Brazil using spaceborne and ground-based remote sensing. PLoS ONE 20(2): e0314238. https://doi.org/10.1371/journal.pone.0314238
About the Authors:
Gabriel Prates Hallal
Roles: Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft
Affiliation: Laboratório de Sensoriamento Remoto Geológico, Centro Estadual de Pesquisas em Sensoriamento Remoto e Meteorologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brasil
Carla Cristine Porcher
Roles: Methodology
Affiliation: Centro de Estudos em Petrologia e Geoquímica, Instituto de Geologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brasil
Bijeesh Kozhikkodan Veettil
Roles: Supervision, Writing – review & editing
E-mail: [email protected]
Affiliations: Laboratory of Ecology and Environmental Management, Science and Technology Advanced Institute, Van Lang University, Ho Chi Minh City, Vietnam, Faculty of Applied Technology, School of Technology, Van Lang University, Ho Chi Minh City, Vietnam
ORICD: https://orcid.org/0000-0003-4158-4578
Jean Marcel de Almeida Espinoza
Roles: Methodology
Affiliation: Laboratório de Física Experimental,Instituto Federal de Educação Ciência e Tecnologia de Santa Catarina, Caçador, Brasil
ORICD: https://orcid.org/0000-0002-7933-2897
Silvia Beatriz Alves Rolim
Roles: Supervision, Writing – review & editing
Affiliation: Laboratório de Sensoriamento Remoto Geológico, Centro Estadual de Pesquisas em Sensoriamento Remoto e Meteorologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brasil
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
[/RAW_REF_TEXT]
1. Liao C, Li Y, Tjong SC. Visible-Light Active Titanium Dioxide Nanomaterials with Bactericidal Properties. Nanomaterials. 2020; 10(1): 124. pmid:31936581
2. Kroczek K, Turek P, Mazur D, Szczygielski J, Filip D, Brodowski R, et al. Characterization of Selected Materials in Medical Applications. Polymers. 2022; 14(8): 1526.
3. Kim T, Choi CH, Hur JS, Ha D, Kuh BJ, Kim Y, et al. Progress, Challenges, and Opportunities in Oxide Semiconductor Devices: A Key Building Block for Applications Ranging from Display Backplanes to 3D Integrated Semiconductor Chips. Advanced Materials. 2023; 35: 2204663.
4. Pihl I, Lind J, Henkow S (2022) How do Swedish companies handle disruptions in global supply chains stemming from the semiconductor crisis? Bachelor’s Thesis in Business Administration. Jonkoping International Business School, Jonkoping, Sweden.
5. Chandrasekar N, Mujabar PS, Rajamanickam GV. Investigation of heavy-mineral deposits using multispectral satellite data. International Journal of Remote Sensing. 2011; 32: 8641–8655.
6. Subasinghe CS, Ratnayake AS, Roser B, Sudesh M, Wijewardhana DU, Attanayake N, et al. Global distribution, genesis, exploitation, applications, production, and demand of industrial heavy minerals. Arab J Geosci, 2022; 15: 1616.
7. Dillenburg SR, Tomazelli LJ, Barboza EG. Barrier evolution and placer formation at Bojuru Southern Brazil. Marine Geology. 2004; 203: 43–56.
8. RGM; HAR; CPEA. RIMA—Relatório de Impacto Ambiental: Projeto Retiro. São José do Norte, April 2014.
9. Abrams M, Yamaguchi Y, Yuan K (2002) Users handbook, version 3.
10. Tang H, Li ZL (ed.). Quantitative remote sensing in thermal infrared: theory and applications. Springer, 2014.
11. Sekertekin A, Bonafoni S. Land Surface Temperature Retrieval from Landsat 5, 7, and 8 over Rural Areas: Assessment of Different Retrieval Algorithms and Emissivity Models and Toolbox Implementation. Remote Sensing. 2020; 12(2): 294.
12. Lane MD, Bishop JL (2009) Mid-infrared (Thermal) Emission and Reflectance Spectroscopy. In: Bishop JL, Bell JF III, Moersch JE (Eds.) Remote Compositional Analysis, p. 42–67.
13. Vicente LE, Filho CRS. Identification of mineral components in tropical soils using reflectance spectroscopy and advanced spaceborne thermal emission and reflection radiometer (ASTER) data. Remote Sensing of Environment. 2011; 115: 1824–1836.
14. Adams JB, Smith MO, Johnson PE. Spectral mixture modelling: a new analysis of rock and soil types at the Viking Lander I site. Journal of Geophysical Research. 1986; 91: 8098–8112.
15. Käfer PS, Rolim SBA, Diaz LR, Rocha NS, Iglessias ML, Rex FE. Comparative analysis of split-window and single-channel algorithms for land surface temperature retrieval of a pseudo-invariant target. Bol. Ciênc. Geod. 2020; 26 (02): e2020008.
16. Gillespie A, Rokugawa S, Matsunaga T, Cothern JS, Hook S, Kahle ABA. Temperature and Emissivity Separation Algorithm for Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Images. IEEE Transactions on Geoscience and Remote Sensing. 1998; 36: 1113–1126.
17. Miao X, Zhang Y, Zhang J, Zhou X. Temperature and Emissivity Smoothing Separation with Nonlinear Relation of Brightness Temperature and Emissivity for Thermal Infrared Sensors.
18. Hulley G, Baldridge A (2013) Validation of Thermal Infrared (TIR) Emissivity Spectra Using Pseudo-Invariant Sand Dune Sites. In: Kuenzer C, Dech S (Eds.). Thermal infrared remote sensing: sensors, methods, applications. Springer, p. 515–537.
19. Rolim SBA, Veettil BK, Käfer PS, Grondona AEB, Iglesias MJ, Diaz LR, et al. Comparison of Emissivity Retrieval Methods from ASTER Data Using Fourier-Transform Infrared Spectroscopy. Acta Geophysica. 2020; 68: 1867–1876.
20. Xie B, Mao W, Peng B, Zhou S, Wu . Thermal-Infrared Spectral Feature Analysis and Spectral Identification of Monzonite Using Feature-Oriented Principal Component Analysis. Minerals. 2022; 12(5): 508.
21. Hecker C, Der Meijde MV, Van Der Meer FD. Thermal infrared spectroscopy on feldspars—Successes, limitations and their implications for remote sensing. Earth-Science Reviews. 2010; 103: 60–70.
22. Alonso K, Bachmann M, Burch K, Carmona E, Cerra D, de los Reyes R, et al. Data Products, Quality and Validation of the DLR Earth Sensing Imaging Spectrometer (DESIS). Sensors. 2019; 19(20): 4471. pmid:31618940
23. Laukamp C, Rodger A, LeGras M, Lampinen H, Lau IC, Pejcic B, et al. Mineral Physicochemistry Underlying Feature-Based Extraction of Mineral Abundance and Composition from Shortwave, Mid and Thermal Infrared Reflectance Spectra. Minerals. 2021; 11(4): 347.
24. Savitri KP, Hecker C, van derMeer FD, Sidik RP. VNIR-SWIR infrared (imaging) spectroscopy for geothermal exploration: Current status and future directions. Geothermics. 2021; 96: 102178.
25. Hiesinger H, Helber J, Alemanno G, Bauch KE, D’Amore M, Maturilli A, et al. Studying the Composition and Mineralogy of the Hermean Surface with the Mercury Radiometer and Thermal Infrared Spectrometer (MERTIS) for the BepiColombo Mission: An Update. Space Sci Rev. 2020; 216: 110.
26. Käfer PS, Rolim SBA, Iglessias ML, Rocha NS, Diaz LR. Land surface temperature retrieval by LANDSAT 8 thermal band: applications of laboratory and field measurements. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2019; 12: 2332–2341.
27. Christensen PR, Banfield JL, Hamilton VE, Howard DA, Lane MD, Piatek JL, et al. A Thermal emission spectral library of rock-forming minerals. Journal of Geophysical Research. 2000; 105: 9735–9739.
28. Chen L, Wei G, Liu K. Influences of reaction and thermal conditions on microstructural evolutions of ilmenite oxygen carriers for chemical looping combustion. Fuel Processing Technology. 2022; 228: 107139.
29. Baldridge AM, Hook SJ, Grove CI, Rivera G. The ASTER Spectral Library Version 2.0. Remote Sensing of Environment. 2009; 113: 711–715.
30. Krupnik D, Khan S. Close-range, ground-based hyperspectral imaging for mining applications at various scales: Review and case studies. Earth-Science Reviews. 2019; 198: 102952.
31. Gao L, Cao L, Zhong Y, Jia Z. Field-Based High-Quality Emissivity Spectra Measurement Using a Fourier Transform Thermal Infrared Hyperspectral Imager. Remote Sensing. 2021; 13(21): 4453.
32. Gerace A, Eon R, Kleynhans T, Raqueno N, Webber CM, Ely P, et al. (2022) In-flight performance of the Multi-band Uncooled Radiometer Instrument (MURI) thermal sensor. Remote Sensing of Environment 279: 113086.
33. Alqaheem Y, Alomair AA. Microscopy and Spectroscopy Techniques for Characterization of Polymeric Membranes. Membranes. 2020; 10(2): 33. pmid:32102383
34. Vanderbruggen A, Gugala E, Blannin R, Bachmann K, Serna-Guerrero R, Rudolph M Automated mineralogy as a novel approach for the compositional and textural characterization of spent lithium-ion batteries. Minerals Engineering. 2021; 169: 106924.
35. Rejith RG, Sundararajan M, Gnanappazham L, Loveson VJ. Satellite-based spectral mapping (ASTER and landsat data) of mineralogical signatures of beach sediments: a precursor insight. Geocarto International. 2020; 37: 2580–2603.
36. Noori L, Pour AB, Askari G, Taghipour N, Pradhan B, Lee C-W, et al. Comparison of Different Algorithms to Map Hydrothermal Alteration Zones Using ASTER Remote Sensing Data for Polymetallic Vein-Type Ore Exploration: Toroud–Chahshirin Magmatic Belt (TCMB), North Iran. Remote Sensing. 2019; 11(5): 495.
37. Moud MF, van Ruitenbeek F, Hewson R, van der Meijde M. An Approach to Accuracy Assessment of ASTER Derived Mineral Maps. Remote Sensing. 2021; 13(13): 2499.
38. Keshava V, Mustard JF. Spectral unmixing. IEEE Processing Magazine. 2002; 2: 44–57.
39. Ghosh G, Kumar S, Saha SK. Hyperspectral Satellite Data in Mapping Salt-Affected Soils Using Linear Spectral Unmixing Analysis. J Indian Soc Remote Sens. 2012; 40: 129–136.
40. Yu J, Chen D, Lin Y, Ye S. Comparison of linear and nonlinear spectral unmixing approaches: a case study with multispectral TM imagery. International Journal of Remote Sensing. 2017; 38: 773–795.
41. Sekandari M, Masoumi I, Pour AB, Muslim AM, Hossain MS, Misra A. ASTER and WorldView-3 satellite data for mapping lithology and alteration minerals associated with Pb-Zn mineralization. Geocarto international. 2022; 37(6): 1782–1812.
42. Abubakar AH, Hashim M, Pour AB. Spectral Mineral Mapping for Characterization of Subtle Geothermal Prospects Using ASTER Data. Journal of Physics Conference Series. 2017; 852: 012024.
43. Gabr S, Ghulam A, Kusky T. Detecting areas of high-potential gold mineralization using ASTER data. Ore Geology Reviews. 2010; 38: 59–69.
44. Martinho TC, Dillenburg SR, Hesp P. Wave energy and longshore sediment transport gradients controlling barrier evolution in Rio Grande do Sul, Brazil. Journal of Coastal Research. 2009; 25: 285–293.
45. Weschenfelder J, Corrêa ICS, Aliotta S, Baitelli R. Paleochannels related to late Quaternary sea-level changes in Southern Brazil. Brazilian Journal of Oceanography. 2010; 58: 35–44.
46. Calliari LJ, Speranski N, Boukareva I. Stable focus of wave rays as a reason of local erosion at the southern Brazilian coast. Journal of Coastal Research. 1998; 26: 19–23.
47. Brasil. Lei n° 9.985, de 18 de julho de 2000. Institui o Sistema Nacional de Unidades de Conservação da Natureza e dá outras providências. Available at: https://www.planalto.gov.br/ccivil_03/leis/l9985.htm.
48. Reed SJB (2006) Electron Microprobe Analysis and Scanning Electron Microscopy in Geology. 2nd Ed. Cambridge: UK, p. 232.
49. Rahman MJJ, Pownceby MI, Rana MS. Distribution and characterization of heavy minerals in Meghna River sand deposits, Bangladesh. Ore Geology Reviews. 2022; 143: 104773.
50. Pownceby MI. Alteration and associated impurity element enrichment in detrital ilmenites from the Murray Basin, southeast Australia: a product of multistage alteration. Australian Journal of Earth Sciences. 2010; 57: 243–258.
51. D&P Instruments (2009) Model 102 Win2K Instruction Manual. Simsbury, CT.
52. Van Der Meer F, Vazquez-Torres M, Dijk PMV. Spectral Characterization of Ophiolite Lithologies in the Troodos Ophiolite Complex of Cyprus and its Potential in Prospecting for Massive Sulphide Deposits. International Journal Remote Sensing. 1997; 18: 1245–1257.
53. Abrams M, Yamaguchi Y. Twenty years of ASTER contributions to lithologic mapping and mineral exploration. Remote Sensing. 2019; 11: 1394.
54. Pour A, Park Y, Crispini L, Läufer A, Kuk Hong J, Park TYS, et al. Mapping listvenite occurrences in the damage zones of Northern Victoria Land, Antarctica using ASTER satellite remote sensing data. Remote Sensing. 2019; 11(12): 1408.
55. Hallal GP, de Almeida Espinoza JM, Veettil BK, Porcher CC, Oliveira Righi da Silva M, Beatriz Alves Rolim S. Mapping heavy mineral deposits on the coast of the state of Rio Grande do Sul (Brazil) using orbital and proximal remote sensing. PLoS ONE. 2024; 19(9): e0309043. pmid:39240841
56. Martin AC, Emery JP, Loeffler MJ. Spectral effects of regolith porosity in the mid-IR–Forsteritic olivine. Icarus. 2022; 378: 114921.
57. United States Geological Society (USGS, 2021) Mineral Commodity Summaries 2021: U. S. Geological Survey, p. 200.
58. Asadi P, Beckingham LE. Intelligent framework for mineral segmentation and fluid-accessible surface area analysis in scanning electron microscopy. Applied Geochemistry. 2022; 143: 105387.
59. Han S, Lӧhr SC, Abbott AN, Baldermann A, Farkaš J, McMahon W, et al. Earth system science applications of next-generation SEMEDS automated mineral mapping. Front. Earth Sci. 2022; 10: 956912.
60. Ramakokovhu MM, Olubambi PA, Mbaya RKK, Mojisola T, Teffo ML. Mineralogical and Leaching Characteristics of Altered Ilmenite Beach Placer Sands. Minerals. 2022; 10: 1011–1022.
61. Gonçalves CC, Braga PFA. Heavy Mineral Sands in Brazil: Deposits, Characteristics, and Extraction Potential of Selected Areas. Minerals. 2019; 9(3):176.
62. Rao DS, Vijayakumar TV, Prabhakar S, Bhaskar Raju G, Ghosh TK. Alteration Characteristics of Ilmenites from South India. Journal of Minerals & Materials Characterization & Engineering. 2005; 4: 47–59.
63. Frost MT, Grey IE, Harrowfield IR, Mason K. The Dependence of Alumina and Silica Contents on the Extent of Alteration of Weathered Ilmenites from Western Australia. Mineralogical Magazine. 1983; 47: 201–208.
64. Meinhold G. Rutile and its Applications in Earth Sciences. Earth-Science Reviews. 2020; 102: 1–28.
65. Mehdilo A, Irannajad M, Rezai B. Chemical and mineralogical composition of ilmenite: Effects on physical and surface properties. Minerals Engineering. 2015; 70: 64–76.
66. Liu C, Wang W, Wang H, Zhu C, Ren B. A Review on Removal of Iron Impurities from Quartz Mineral. Minerals. 2023; 13(9): 1128.
67. Laakso K, Turner DJ, Rivard B, Sanchez AA. The Long-Waveinfrared (8–12 μm) spectral features of selected rare earth element—Bearing carbonate, phosphate and silicate minerals. International Journal of Applied Earth Observation and Geoinformation. 2019; 76: 77–83.
68. Zhang L, Yang J, Zhao X, Sun F, Zuo X, Nan J. Small-molecule surface-modified bismuth-based semiconductors as a new class of visible-light-driven photocatalytic materials: Structure-dependent photocatalytic properties and photosensitization mechanism. Chemical Engineering Journal. 2020; 380: 122546.
69. Salisbury JW, Wald A. The Role of Volume Scattering in Reducing Spectral Contrast of Reststrahlen Bands in Spectra of Powdered Minerals. ICARUS. 1992; 96: 121–128.
70. Bedell RL, Rivard B, Browning D, Coolbaugh M (2017) Thermal Infrared Sensing for Exploration and Mining–An Update on Relevant Systems for Remote Acquisition to Drill Core Scanning. Spectral Geology and Remote Sensing. In: Proceedings of Exploration 17: Sixth Decennial International Conference on Mineral Exploration. p. 881–897.
71. Agui A, Uozumi T, Mizumaki M, Kaambre T. Intermetallic Charge Transfer in FeTiO3 Probed by Resonant Inelastic Soft X-Ray Scattering. Physical Review. 2009; 79: 92402–092405.
72. Ducart DF, Silva AM, Toledo CLB, Assis LM. Mapping iron oxides with Landsat-8/OLI and EO-1/Hyperion imagery from the Serra Norte iron deposits in the Carajás Mineral Province, Brazil. Brazilian Journal of Geology. 2016; 46: 331–349.
73. Rejith RG, Sundararajan M, Venkatesan S, Mohammed-Aslam MA (2021) Remote sensing for exploring heavy mineral deposits: a case study of Chavara and Manavalakurichi deposits, southwest coast of India. In: Remote Sensing of Ocean and Coastal Environments (pp. 177–188). Elsevier.
74. Ghorbani A, Honarmand M, Shahriari H, Hassani MJ. Regional scale prospecting for non-sulphide zinc deposits using ASTER data and different spectral processing methods. International Journal of Remote Sensing. 2019; 40(23): 8647–8667.
75. Yousefi M, Tabatabaei SH, Rikhtehgaran R, Pour AB, Pradhan B. Application of Dirichlet process and support vector machine techniques for mapping alteration zones associated with porphyry copper deposit using ASTER remote sensing imagery. Minerals. 2021; 11(11): 1235.
76. Ekanayake EMMB Vithana SSP, Ekanayake EMHEB Rathnayake ARMAN, Abeysekara AMR Oorloff TSJ, et al. Mapping ilmenite deposit in Pulmudai, Sri Lanka using a hyperspectral imaging-based surface mineral mapping method. Journal of the National Science Foundation of Sri Lanka. 2019; 47: 271–284.
77. Gazi MY, Tafhim KT, Ahmed MK, Islam MA. Investigation of heavy-mineral deposits using multispectral satellite imagery in the eastern coastal margin of Bangladesh. Earth Sci. Malaysia. 2019; 3(2): 16–22.
78. Dhinesh S, Udayaganesan P, Ramakrishnan D, Rajani R, Xavier Ignaci Muthu J. An integrated strategy for the exploration of palaeofluvial placer deposits. Applied Geomatics. 2021; 13: 165–177.
79. Honarmand M, Ranjbar H, Shahabpour J. Application of Principal Component Analysis and Spectral Angle Mapper in the Mapping of Hydrothermal Alteration in the Jebal-Barez Area, Southeastern Iran. Resource Geology. 2012; 62: 119–139.
80. Mushkin A, Gillespie AR, Abbott EA, Batbaatar J, Hulley G, Tan H, et al. Validation of ASTER emissivity retrieval using the Mako airborne TIR imaging spectrometer at the Algodones dune Field in Southern California, USA. Remote Sensing. 2020; 12(5), 815.
81. Rowan LC, Mars JC, Simpson CJ. Lithologic mapping of the Mordor, NT, Australia ultramafic complex by using the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). Remote sensing of Environment. 2005; 99: 105–126.
82. Aboelkhair H, Ninomiya Y, Watanabe Y, Sato I. Processing and Interpretation of ASTER TIR Data for Mapping of Rare-Metal-Enriched Albite Granitoids in the Central Eastern Desert of Egypt. Journal of African Earth Sciences. 2010; 58: 141–151.
83. Girouard G, Bannari A, Harti AE, Desrochers A (2004) Validated Spectral Angle Mapper Algorithm for Geological Mapping: comparative study between Quickbird and Landsat-TM. Anais ISPRS XX, Geo-Imagery Bridging Continents, p.12–23.
84. Quintano C, Fernandez-Manso A, Shimabukuro YE, Pereira G. Spectral Unmixing. International Journal of Remote Sensing. 2012; 33: 5307–5340.
85. Hosseinjani M, Tangestani MH. Mapping alteration minerals using sub-pixel unmixing of ASTER data in the Sarduiyeh area, SE Kerman, Iran. International Journal of Digital Earth. 2011; 4: 487–504.
86. Seif RA, Ene A, Zakaly HMH, Sallam AM, Taalab SA, Fnais MS, et al. Distribution of Heavy Metals along the Mediterranean Shoreline from Baltim to El-Burullus (Egypt): Consequences for Possible Contamination. Minerals. 2024; 14(9):931.
87. Zakaly HMH, Awad HA, Abbasi A, Almousa N, El-Salam LMA, Mostafa AMA, et al. (2024) Radioactive and mineralogical assessment of mediterranean black sands: a systematic analysis and health risk evaluation. J Radioanal Nucl Chem. 2024; 333, 1937–1947.
88. Awad HA, El-Leil IA, Granovskaya NV, Ene A, Tolba A, Kamel M, et al. Petrotectonic and Geochemistry of Abu Murrat Ilmenite Bearing Gabbro, NE Desert, Egypt. Romanian Journal of Physics. 2022; 67: 808.
89. WIPO (2022) Production of titanium and titanium dioxide from ilmenite and related applications. Geneva: World Intellectual Property Organization.
90. Faust R, Lamarca I, Schaefer A, Lind F, Knutsson P. Magnetic properties of ilmenite used for oxygen carrier-aided combustion. Fuel. 2023; 340: 127593.
91. Ciazela J, Bakala J, Kowaliński M, Pieterek B, Steslicki M, Ciazela M, et al. Lunar Ore Geology and feasibility of ore mineral detection using a far-IR spectrometer. Frontiers in Earth Science. 2023; 11: 1190825.
92. Ahmed MT, Monir MU, Rajib M, Sarkar SM, Hasan Y, Quamruzzaman C, et al. Heavy minerals identification and extraction along coastal sediments using placer mining technique. Journal of Sedimentary Environments. 2023; 8(1): 81–95.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2025 Prates Hallal et al. This is an open access article distributed under the terms of the Creative Commons Attribution License: http://creativecommons.org/licenses/by/4.0/ (the “License”), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Titanium oxide is of fundamental strategic importance in the global market as it is used as a raw material by several industries, such as medical prostheses, paints, pigments, and, more recently, electronic chips. The main source of titanium oxide is ilmenite, a mineral deposited in many coastal areas of the world, including the state of Rio Grande do Sul in Southern Brazil in its central coastal plain, under specific morphodynamic conditions. Some geological targets, such as mineral oxides, show distinct thermal spectral features. The present study evaluated the surface concentration of ilmenite in Southern Brazil using thermal spectroscopy (μFT-IR). The emissivity spectral signatures of pure ilmenite between 8 and 14 μm were determined and some indicative features were identified. The obtained emissivity spectrum has been employed as a reference for the Spectral Angle Mapper (SAM) and Linear Spectral Unmixing (LSU) image classification algorithms. An image from the Advanced Spaceborne Thermal Emission Radiometer (ASTER) sensor (AST_05 emissivity product) was used to recognize the occurrence and assess the richness of the ilmenite. The outcomes of the present study indicated pixels with ilmenite concentration between 0 and 29.6%, with the highest concentration occurring under the transgressive dune field. In contrast, a lower concentration is found in the backshore. To obtain the degree of purity of the ilmenite, a quantitative microanalysis of the samples was conducted in a scanning electron microscope (SEM), and the results indicated that 80% of the minerals were ilmenite. Qualitative microanalysis showed that ilmenite is in the primary alteration phase, with a low degree of weathering and a lower concentration of impurities. Integrated techniques for analyzing multispectral and hyperspectral data in the thermal infrared were able to identify and map minerals rich in titanium oxide (ilmenite) quickly, effectively, at low cost, and non-destructively.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer