Barbara K.
Kuhn
*ab,
Karin
Birbaum
a,
Yan
Luo
c and
Detlef
Günther
*a
aLaboratory of Inorganic Chemistry, ETH Zurich, 8093, Zurich, Switzerland. E-mail: guenther@inorg.chem.ethz.ch; Fax: +41 44 633 1071; Tel: +41 44 632 4687
bZürcherische Arbeitsgemeinschaft zur Erforschung und Bekämpfung der Staublungen in der Schweiz, Institute for Mineralogy and Petrography, 8092, Zurich, Switzerland. E-mail: barbara.kuhn@alumni.ethz.ch; Tel: +41 44 632 6581
cAnalysis and Air Quality Section, Air Quality Research Division, Atmospheric Science and Technology Directorate, Science and Technology Branch, Environment Canada, Ottawa, Ontario, Canada K1A 0H3
First published on 6th November 2009
Studies of the ablation rate, particle size distribution and signal intensities from NIST 610 and zircon 91500 were carried out using UV-ns LA-ICP-MS to investigate the limitations for non-matrix matched quantification of the Pb and U system. All major elements (Zr, Si, Hf) as well as Pb and U were measured in both matrices using different crater diameters. The results indicate two different laser ablation dominated sources of fractionation which prohibit non-matrix matched calibration for accurate age determination. Selective loss of Pb (crater diameter independent) and crater diameter dependent retention of U through phase change in the crater during the ablation of zircon are the two sources of fractionation which occur. However, at certain operating conditions the loss of Pb and the retention of U within the crater are equal. This leads to ages, which randomly agree with the reference age.
This study aimed to gain some insights into the behaviour of Pb and U during analysis. Various measurements were carried out to investigate the reported differences in fractionation of Pb and U for NIST 610 and zircon 91500. Therefore, ablation rates, particle size distributions and ICP-MS signal intensities were determined for both matrices using different crater diameters. Additionally, the number of ablated atoms was calculated for both matrices at the various crater diameters used for analysis and compared to the intensities. 206Pb and 238U concentrations in zircon were determined using all matrix elements (Si, Zr, Hf)11 as internal standards and NIST 610 as external standard. The ages were calculated based on the concentration ratios.
Laser system | ArF-excimer |
---|---|
Laser wavelength | 193 nm |
Repetition rate | 10 Hz |
Crater diameter | 20, 40, 60, 80 µm |
Incident pulse energy | 2.2 mJ |
Energy density on sample | 19.5 Jcm−2 |
ICP-MS model | Elan6100 DRC |
RF power | 1350 W |
Gas flows | |
Plasma (Ar) | 17.8 Lmin−1 |
Auxiliary (Ar) | 0.8 Lmin−1 |
Carrier (He) | 0.96 Lmin−1 |
Make-up (Ar) | 0.76 Lmin−1 |
Detector mode | Pulse counting + analogue |
Isotopes determined | 29Si, 91Zr, 178Hf, 202Hg, 204Pb, 206Pb, 207Pb, 208Pb, 232Th, 235U, 238U |
Dwell time | 10 ms |
Quadrupol settling time | ca. 3 ms |
Time/scan | 0.57 s/4 |
Data acquisition | 143 s (60 s blank, 60 s ablation) |
Particle size distributions were measured using a HS-LAS (Particle Measuring Systems, Denver, CO, USA) which provides the particle size information based on laser light scattering measurements. The instrument covers a size range between 65 nm and 1 µm. The aerosol flow rate through this instrument was adjusted internally to 0.2 L min−1.
The crater depth was estimated using an optical microscope (Olympus BX51, Olympus Optical Co. Europe, Hamburg) in reflected light mode to determine the ablation rate.
Each analysis sequence was started with four measurements of NIST 610 followed by eight measurements of zircon 91500 and ended by another four measurements of NIST 610. Data acquisition and reduction were based on the protocol described by Longerich et al.13
The numbers of ablated atoms were calculated for each crater diameter using the ablation rate, the crater diameter and the element concentration in the two matrices (table 2). These calculations were based on the assumption of a conical crater where a diameter to depth ratio of 1:1 leads to a reduction of the diameter at the bottom of the crater of 20%. For the ratio of the intensity [cps] per ablated atoms, we use the term ‘detection efficiency’ in this study.
We are well aware of the fact that the absolute concentrations of Pb and U are not homogeneously distributed in zircon 91500. However, the Pb/U ratio is constant14,15 which is of major interest for the interpretation of the results.
![]() | ||
Fig. 1 Determination of ablation rate for zircon 91500 and NIST 610 using a 193 nm ArF-excimer laser. The number of pulses was selected according to typical analysis duration. The linear regression includes 4–6 individual measurements for the depth between 100 to 300 pulses with an RSD of 3 to 10%. Including the values for 600 pulses changes the slope for zircon by less than 3%. |
![]() | ||
Fig. 2 Particle size distribution measurements for zircon 91500 and NIST 610 at four different crater sizes showing bimodal distribution for both matrices. The measurements include the entire aerosol generated over a period of 60 s. |
![]() | ||
Fig. 3 Temporal changes of various intensity ratios during ablation of NIST 610 and zircon 91500. |
The U/Si ratio (figure 3c) decreases linearly in both matrices and shows a distinct crater diameter dependence for small crater diameters (20–40 µm).
The U/Zr intensity ratios show pronounced temporal changes during the first 15 s of zircon ablation (figure 3d) whereafter they remain constant. Furthermore, the U/Zr ratios (60 s) indicate no significant variation in dependence of crater diameter for both matrices.
The Si/Zr ratio strongly increases over time except for large crater diameters in zircon (60–80 µm, figure 3e). The temporal change varies with crater diameter. This is a similar behaviour as the U/Si ratio and indicates a correlation between U and Zr.
A linear increase is observed in the Pb/U ratio for both matrices (figure 3f). The temporal changes are more pronounced in NIST 610 than in zircon 91500. A crater diameter dependence is only seen for 20 µm. In zircon 91500, the low Pb concentration leads to noisy signal at this crater diameter. The 20 µm crater in NIST 610, shows a change of slope at approx. 25 s. At this time the width/depth ratio exceeds 1:3, which is still below the critical aspect ratio reported by Mank and Mason.17
Considering both matrices, the U/Si ratio and the Si/Zr ratio provide the highest variability with respect to crater diameter. This together with the temporal changes of the matrix element ratio Si/Zr indicate non-stoichiometric ablation of Si and Zr (U) in zircon.
a normalized to one second. | ||||
---|---|---|---|---|
NIST 610 | ||||
20 µm | 40 µm | 60 µm | 80 µm | |
29Si | 0.22E + 12 | 1.46E + 12 | 3.85E + 12 | 7.41E + 12 |
91Zr | 0.24E + 09 | 1.59E + 09 | 4.21E + 09 | 8.11E + 09 |
206Pb | 0.21E + 09 | 1.37E + 09 | 3.63E + 09 | 6.99E + 09 |
238U | 0.80E + 09 | 5.27E + 09 | 13.95E + 09 | 26.82E + 09 |
Zircon 91500 | ||||
20 µm | 40 µm | 60 µm | 80 µm | |
29Si | 0.16E + 12 | 0.90E + 12 | 2.24E + 12 | 4.20E + 12 |
91Zr | 0.41E + 12 | 2.27E + 12 | 5.65E + 12 | 10.57E + 12 |
206Pb | 0.05E + 09 | 0.27E + 09 | 0.66E + 09 | 1.24E + 09 |
238U | 0.22E + 09 | 1.22E + 09 | 3.05E + 09 | 5.70E + 09 |
Atom ratio (zircon 91500/NIST 610) | ||||
20 µm | 40 µm | 60 µm | 80 µm | |
29Si | 0.74 | 0.62 | 0.58 | 0.57 |
91Zr | 1696 | 1422 | 1340 | 1303 |
206Pb | 0.23 | 0.19 | 0.18 | 0.18 |
238U | 0.28 | 0.23 | 0.22 | 0.21 |
NIST 610 | ||||
20 µm | 40 µm | 60 µm | 80 µm | |
29Si | 0.21E + 06 | 0.86E + 06 | 1.70E + 06 | 2.92E + 06 |
91Zr | 6.45E + 03 | 27.28E + 03 | 56.87E + 03 | 99.33E + 03 |
206Pb | 14.93E + 03 | 60.23E + 03 | 124.61E + 03 | 209.91E + 03 |
238U | 0.08E + 06 | 0.35E + 06 | 0.73E + 06 | 1.26E + 06 |
Zircon 91500 | ||||
20 µm | 40 µm | 60 µm | 80 µm | |
29Si | 0.13E + 06 | 0.52E + 06 | 1.00E + 06 | 1.65E + 06 |
91Zr | 6.81E + 06 | 32.35E + 06 | 68.87E + 06 | 125.44E + 06 |
206Pb | 2.24E + 03 | 8.82E + 03 | 17.86E + 03 | 30.75E + 03 |
238U | 13.02E + 03 | 65.10E + 03 | 139.20E + 03 | 247.38E + 03 |
Measured cps ratio (zircon 91500/NIST 610) | ||||
20 µm | 40 µm | 60 µm | 80 µm | |
29Si | 0.63 | 0.61 | 0.59 | 0.57 |
91Zr | 1055 | 1186 | 1211 | 1263 |
206Pb | 0.15 | 0.15 | 0.14 | 0.15 |
238U | 0.16 | 0.18 | 0.19 | 0.20 |
Dividing the number of Si atoms ablated from zircon 91500 by the number of Si atoms ablated from NIST 610 leads to an ablated Si atom ratio of 0.74 for a crater diameter of 20 µm which changes to 0.57 for 80 µm. The measured Si intensity ratios between zircon 91500 and NIST 610 change from 0.63 at 20 µm to 0.57 at 80 µm. It can be seen, that at large crater diameters the intensity ratio is in good agreement with the ablated atom ratio (figure 4). The significant difference at the 20 µm crater (approx. 15%) could be caused by the uncertainty of volume estimation for the small crater diameter.
![]() | ||
Fig. 4 Ablated atom ratios and intensity ratios (zircon 91500/NIST 610) for different crater diameters. |
The same calculations for Zr demonstrate that the intensity ratio (zircon 91500/NIST 610) increases with increasing crater diameter while the ablated atoms ratio decreases. In contrast to Si, the two ratios do not converge completely at 80 µm. This indicates that the Zr remains at the ablation site in zircon (20 µm: 1.69 times difference between cps and ablated atoms, 80 µm: 1.08). The difference becomes smaller when ablating larger crater diameters. The changes measured for 238U are identical to those measured for Zr.
The ablated atoms ratio of 206Pb decreases from 20 to 40 µm craters and remains stable over all other crater diameters. Surprisingly, the intensity ratio of 206Pb between the two matrices is the most stable ratio and not correlated to any other matrix element. In contrast, Zr and U are well correlated in their intensity ratios but negatively correlated to the ablated atoms ratio.
Fig. 5 summarizes the differences in the ablated atoms and intensities (detection efficiency). The detection efficiency for Pb, Zr and U indicates losses during the ablation of zircon. Si shows only at 20 µm a 10% difference between the two matrices. Overall, the most significant differences between the two matrices were found for the detection efficiencies at 20 µm, which is partly related to the uncertainty of the volume determination.
![]() | ||
Fig. 5 Detection efficiency ratio (zircon 91500/NIST 610) for different crater diameters for an ablation time of 40 s. Detection efficiency is calculated as intensity ratio divided by the ablated atoms ratio. |
crater size | concentration | |||||
---|---|---|---|---|---|---|
internal standard | Si | Zr | Hf | |||
206Pb | 238U | 206Pb | 238U | 206Pb | 238U | |
literature14 | 14.9 | 79 | 14.9 | 79 | 14.9 | 79 |
20 µm | 11.1 ± 0.5 | 55 ± 2 | 15.5 ± 0.7 | 75 ± 2 | 15.5 ± 0.8 | 76 ± 3 |
40 µm | 11.1 ± 0.3 | 64 ± 2 | 13.7 ± 0.6 | 77 ± 4 | 14.4 ± 0.8 | 81 ± 5 |
60 µm | 11.2 ± 0.2 | 68 ± 2 | 12.8 ± 0.3 | 76 ± 2 | 13.8 ± 0.4 | 82 ± 3 |
80 µm | 11.8 ± 0.2 | 72 ± 2 | 12.7 ± 0.3 | 76 ± 2 | 12.8 ± 0.3 | 77 ± 2 |
Calculating the concentrations using Zr as internal standard leads to 238U concentrations which were within 0.7% independent of the crater diameter. The 238U concentrations were 5 % lower than the reference value, whereas the 206Pb concentrations changed in dependence on crater diameter from +8 % (20 µm) to −14% (80 µm). Similar results were obtained using Hf as internal standard.
Using Si as internal standard, concentrations for Zr and Hf approach the stoichiometric composition of the zircon towards larger crater diameters (table 6).
crater size | concentration | |
---|---|---|
Zr | Hf | |
literature | 465 432 | 5 895 |
20 µm | 343 100 ± 6 200 | 4 320 ± 90 |
40 µm | 390 300 ± 9 900 | 4 700 ± 180 |
60 µm | 417 900 ± 9 300 | 4 920 ± 150 |
80 µm | 447 700 ± 6 200 | 5 610 ± 80 |
crater size | age | ||
---|---|---|---|
internal standard | Si | Zr | Si for Pb |
Zr for U | |||
20 µm | 1237 ± 17 | 1256 ± 17 | 914 ± 16 |
40 µm | 1044 ± 16 | 1061 ± 17 | 878 ± 30 |
60 µm | 1001 ± 14 | 1017 ± 14 | 893 ± 9 |
80 µm | 986 ± 11 | 1002 ± 11 | 940 ± 11 |
In contrast to Zr, Hf and U retained within the crater, Si and Pb are removed out of the crater. Fig. 5 demonstrates that the sampling efficiency (ablation, transport and detection) of Si is the same in both matrices (40–80 µm). Fig. 5 and table 5 indicate a loss of Pb (21–26%) in zircon. For the 80 µm crater the 206Pb concentration is 21 % lower than the literature value whereas Zr and U are by 4% and 9% too low, respectively (internal standard: Si). The comparison of the Pb intensity between zircon 91500 and NIST 610 (figure 4) illustrate that this loss is crater diameter independent (almost constant with respect to the total ablated mass). The particle size distribution (PSD) measurements show that particles produced during the ablation of both materials (NIST 610 and zircon 91500) are sufficiently small to ensure complete vaporisation in the ICP. Therefore, ICP-induced changes are likely to be negligible for geochronology data using ns-laser ablation. However, PSD measurements on zircon 91500 indicated also that the particles are much smaller than typically measured when ablating NIST 610. Considering the particle transport efficiencies calculated are dependent on particle size21 it might well be that a considerable loss of nm sized particles or vapour occurs. Pb, as the most volatile element, is likely to be more abundant in the smallest particle size fraction or in the vapour phase and therefore lost by diffusion and condensation.
This journal is © The Royal Society of Chemistry 2010 |