Content uploaded by Yonghwi Kim
Author content
All content in this area was uploaded by Yonghwi Kim on Jul 15, 2020
Content may be subject to copyright.
Application of handheld laser-induced breakdown spectroscopy
(LIBS) to quantitative analysis of carbonate minerals
Yonghwi Kim1*, Cécile Fabre1, Jean Cauzid1
(*e-mail : yonghwi.kim@univ-lorraine.fr)
[1] GeoRessources Laboratory - Université de Lorraine, CNRS, Vandœuvre-lès-Nancy, France
This research has been conducted within the NEXT project (G.A. No. 776804-H2020-SC5-2017) and has been supported by funding from the EU’s Horizon 2020 research and innovation program.
Introduction Methodology
Element concentration Carbonate discrimination Mineral quantification
Conclusion and Upcoming work
● The spectral data obtained from portable LIBS device shows that emission lines are highly correlated with their elemental concentration, and the spectral ratios are
effectively applicable to the discrimination of carbonate minerals by comparing with elemental ratios calculated from portable XRF.
● For mixtures of two carbonate minerals, the spectral data showed a linear correlation, suggesting the possibility of in-situ quantitative analysis.
● An upcoming study will extend the investigations to two, three and four-phase mixtures in calcite-dolomite-ankerite-siderite space.
● The mineral quantitative analysis will examine the applicability of samples in the field by mixing carbonates with other mineral groups (e.g. silicates or sulfides).
Samples and Experimental setup
Ca-Mg-Fe+Mn ternary diagram
using carbonate references in this study.
Fe+Mn Mg
Ca
rhodochrosite
calcite
dolomite
ankerite
magnesite
siderite
2 mm
13 mm
30 μm
2345
6 7 8 9 10
11 12 13 14 15
1
10 Hz
• 5 Random locations
• 1 Cleaning shot
• 4 laser shots for each points
• Average of 4 data shots
Experimental protocol on pellets
with the handheld LIBS pixel mode.
● The H2020 NEXT (New Exploration Technologies) project aims at
developing new geo-models and novel sensitive exploration technologies,
and one task of the project focuses on obtaining rapid, reliable and efficient
data in the field within this frame.
● Numerous carbonates in the field occasionally show similar characteris-
tics not only in color, grain size and texture but also in geochemical aspects,
making it complicated to distinguish them with the naked eye.
● This study is to develop an optimized methodology using a handheld LIBS
analyser (Z-300, Sci-Aps ©) to quantify chemically and mineralogically
among the different carbonates in the field.
● To relate the changes in spectral characteristics with the mineral content of carbonate mixtures, we compared the ratio
using the intensity and area obtained from the LIBS spectrum with the mixture ratios between the two carbonate minerals.
● The calcite-ankerite mixture displayed a good correlation with Mg/Ca ratio. Both R2 values have higher than 0.97 and
MAAPE values showed 4% and 26% of intensity and integrated net area, respectively.
● In the calcite-dolomite mixture, these ratios showed a linear tendency due to low Fe content in both calcite and dolomite
endmember. The R2 and MAAPE values of the intensities showed the highest one of 0.995 and 0.08, respectively.
● To distinguish among carbonate minerals in
the field, different emission ratios are
compared with Fe# calculated from portable
XRF data.
● Mg/Ca ratio showed the linear tendencies
in both directions as the content of Mg and Fe
increased. However, this tendency makes it
difficult to distinguish among calcite, dolomite
and rhodochrosite.
● Mg/Fe ratio discriminated against rhodo-
chrosite, ankerite, and calcite, but dolomite
and magnesite are plotted in a similar position.
● Mg/Mn ratio distinguishes other minerals
except calcite (white) and rhodochrosite (pink).
This ratio can be considered the most suitable
variable for distinguishing carbonate minerals.
● In order to get a first understanding of the
relationship between the elemental concen-
trations on one hand and peak intensity or
area on the other hand, determination
coefficient (R2) and mean arctangent
absolute percentage error (MAAPE) are
used for cross-validation and forecast
accuracy, respectively. Ideally, R2 value
should be equal to 1, while MAAPE value is
equal to 0%.
● A calibration curve of Ca demonstrated a
linear correlation between LIBS data and Ca
concentration by exhibiting a higher R2 than
0.97 in both variables with the same MAAPE
= 22%.
● Although the integrated net area in Fe
showed the highest R2 value (0.98), the
MAAPE value displays the lowest (26%)
concurrently. However, the intensity shows a
tendency to be the opposite result, exhibiting
lower R2 values and better MAAPE values
(R2 = 0.95, MAAPE = 22%).
● Mn shows very high R2 values higher than
0.99 for both variables, but MAAPE values
vary diversely, exhibiting the lowest value
(90%) in the integrated net area, and the
highest value (57%) in the intensity.
● Although the emission line of Mg used in
this study is correlated with Mg contents,
spectral interference with Mn emission lines
has been identified. Thus, the calibration
curves for Mg were made by excluding
rhodochrosite.
● Similar R2 values for Mg were shown with
respect to both LIBS parameters around
0.88. MAAPE values, however, displayed a
diverse tendency compared to the R2 value,
which was 29% and 77%, respectively.
LIBS analysis
Element concentration
LIBS spectra portable XRF
corrected by AAS
Mineral quantification
calcite - ankerite
calcite - dolomite
ankerite - dolomite
Coefficient determination (R2)
Mean Arctangent Absolute
Percentage Error (MAAPE)
Mass mixing ratio
75:25 / 50:50 / 25:50
Ca: 31%, Mg <LOD
calcite
dolomite
Ca: 19%, Mg: 14%
Calibration curves vs. concentration in natural carbonate standards. The spectral ratio using Mg, Fe and Mn lines
with Fe# calculated from pXRF data.
wt.% (corrected pXRF)
wt.% (corrected pXRF)
0 10 20 30 40
0
3000
6000
9000
12000
15000
18000
LIBS intensity (a.u)
calcite
ankerite
dolomite
magnesite
siderite
rhodochrosite
Ca I (558.9 nm)
R2 : 0.97
MAAPE : 22%
0 10 20 30 40
0
1000
2000
4000
3000
LIBS (net area)
Ca I (558.9 nm)
R2 : 0.97
MAAPE : 22%
LIBS intensity (a.u)
Mg I (309.7 nm)
-4 0
200
0
400
600
800
4 8 12 16 20 24 28
calcite
ankerite
dolomite
magnesite
siderite
rhodochrosite
R2 : 0.89
MAAPE : 29%
LOD
LOD
LIBS (net area)
0
5
10
15
20
25
-4 0 4 8 12 16 20 2824
Mg I (309.7 nm)
R2 : 0.88
MAAPE : 77%
LIBS intensity (a.u)
Fe I (495.8 nm)
0
2000
4000
6000
0 8 16 24 32 40 48
calcite
ankerite
dolomite
magnesite
siderite
rhodochrosite
R2 : 0.95
MAAPE : 22%
LIBS (net area)
Fe I (495.8 nm)
0
300
600
900
0 8 16 24 32 40 48
R2 : 0.98
MAAPE : 26%
LIBS intensity
Mn I (478.3 nm)
0
5000
10000
15000
0 8 16 24 32 40 48
calcite
ankerite
dolomite
magnesite
siderite
rhodochrosite
R2 : 0.99
MAAPE : 57%
LIBS (net area)
Mn I (478.3 nm)
0
600
1200
1800
2400
0 8 16 24 32 40 48
R2 : 0.99
MAAPE : 90%
Fe# (Fe/[Ca+Mg+Fe+M n])
0.0 0.2 0.4 0.6 0.8 1.0
Mg/Fe (LIBS intensity)
0
1
2
3
4
0.0 0.2 0.4 0.6 0.8 1.0
Mg/Mn (LIBS intensity)
Fe increase
Fe increase
Fe increase
Mg increase
Mg increase
Mg increase
0
2
4
6
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.5
1.0
1.5
2.0
Mg/Ca (LIBS intensity)
calcite
ankerite
dolomite
rhodochrosite
siderite
magneiste
calcite volume (%) calcite volume (%)
Mg/Ca (LIBS net area)
Mg/Ca (LIBS intensity)
Mg/Ca (LIBS net area)
Mg/Ca (LIBS intensity)
0.00
0.00
0.02
0.04
0.06
0.00
0.00
0.02
0.04
0.06
0.08
0.02
0.04
0.06
0.01
0.02
0.03
0.04
calcite - ankerite
(R2: 0.988, MAAPE: 4%)
calcite - ankerite
(R2: 0.970, MAAPE: 26%)
calcite - dolomite
(R2: 0.995, MAAPE: 8%)
calcite - dolomite
(R2: 0.917, MAAPE: 42%)
100 75 50 25 0
100 75 50 25 0
100 75 50 25 0
100 75 50 25 0