Low energy PIXE revisited: is it still alive?

Anuncio
REVISTA MEXICANA DE FÍSICA S 56 (1) 82–84
FEBRERO 2010
Low energy PIXE revisited: is it still alive?
J. Miranda
Instituto de Fı́sica, Universidad Nacional Autónoma de México,
Apartado Postal 20-364 México, D.F. 01000, México.
Recibido el 11 de marzo de 2009; aceptado el 11 de agosto de 2009
Particle Induced X-ray Emission (PIXE) using low energy protons (below 1 MeV) has not been extensively used in the past, mainly because
of the low emission of X-rays, a limited knowledge of theoretical ionization cross sections in this energy range, and the possible damage
to the samples due to the energy transferred by the ion beam. Nevertheless, there are several small laboratories around the world that still
use this variation of PIXE, because of the advantages it presents, such as a lower background radiation in the X-ray range corresponding to
light elements and the smaller corrections due to secondary fluorescence. This results in improved detection limits for light elements (from
Mg to Cl), as compared to the traditional PIXE (with 2 MeV to 3 MeV protons). In the past decade, further efforts in the measurement and
understanding of the X-ray production cross sections, as well as new applications, have been developed. In the present work, a brief summary
of these advances is presented, together with suggestions about how its performance might be enhanced with the use of complementary
techniques, like X-ray fluorescence (XRF) and Rutherford Backscattering (RBS) spectrometry.
Keywords: PIXE; low energy protons; X-rays.
La emisión de rayos X inducida por partı́culas (PIXE) usando protones con bajas energı́as (menores que 1 MeV) no se ha usado extensamente
en el pasado, principalmente a causa de la baja emisión de rayos X, un conocimiento limitado de las secciones de ionización teóricas en este
intervalo de energı́as y el posible daño producido en las muestras por la energı́a transferida a ellas por el haz de iones. Sin embargo, hay varios
laboratorios alrededor del mundo que aún utilizan esta variante de PIXE, gracias a las ventajas que presenta, tales como una menor radiación
de fondo en el intervalo de energı́as de rayos X correspondiente a elementos ligeros y correcciones más pequeñas por la fluorescencia
secundaria. Todo esto resulta en mejores lı́mites de detección para elementos ligeros (entre Mg y Cl), comparados con los que hay en PIXE
tradicional (usando protones con energı́as entre 2 MeV y 3 MeV). En los diez años recientes se han hecho mayores esfuerzos para conocer
bien las secciones de ionización, al igual que nuevas aplicaciones. En este trabajo se presenta un breve resumen de los avances, junto con
sugerencias acerca de cómo se podrı́an mejorar los resultados usando técnicas complementarias, como la fluorescencia de rayos X (XRF) y
la retrodispersión de Rutherford (RBS).
Descriptores: PIXE; Protones de baja energı́a; Rayos-X.
PACS: 32.80.Hd; 33.50.Hv; 34.50.Dy
1. Introduction
Particle Induced X-ray Emission (PIXE) is an important analytical method in many scientific or technological areas [1,2].
Most of the times, PIXE is used with an ion beam having
energies in the range 1.5 MeV to 3 MeV, resulting in limits
of detection as low as 1 mg kg−1 , or even better, according to the sample type and/or preparation. Nevertheless, as
explained in previous papers [3,4], there may also be interest in using proton beams with energies below 1 MeV, because of the advantages PIXE has in this energy range. This
is mostly due to the behavior of the physical quantities involved in PIXE, such as stopping and ionization cross sections, which are different to those found at energies above
1 MeV. As discussed in those papers [3,4], the main benefits
of low energy PIXE (LE-PIXE) are mainly the depth resolution (due to the strongly decreasing ionization cross section and growth of the stopping power), the lower secondary
fluorescence emission, as well as the low background radiation, which improves sensitivity for light elements (between
Mg and Cl). In spite of this, there are effects that are not
considered in “traditional” PIXE, which may affect the data
obtained if LE-PIXE is used, limiting its applications. Unfortunately, the past decade saw a decrease in the number of lab-
oratories dedicated to LE-PIXE, while the published papers
decreased significantly. In a simple search through abstracts
databases, around 2,400 papers using PIXE were published
since 1996; however, in the same period, no more than 25
publications about LE-PIXE appeared in the scientific literature. The advantages of LE-PIXE seem to be highly overlooked by most PIXE research groups.
In this work, an update is presented about how LE-PIXE
was employed in the last 13 years, first discussing the advances in the knowledge of the basic physical process (Xray emission), and then showing some new applications and
proposing improvements to extend them.
2.
Calculation and measurement of X-ray production cross sections
The main physical process involved in PIXE is the ionization
of atomic inner shells by the collision of an energetic positive
ion with a target atom, followed by the de-excitation of this
atom by the decay of an upper-shell electron to fill the vacancy, with the subsequent emission of an X-ray photon, to
eliminate the excess energy, including other processes, like
the Auger electron emission or the radiative Auger process
LOW ENERGY PIXE REVISITED: IS IT STILL ALIVE?
(simultaneous emission of a photon and an electron). An
in-depth discussion of the process can be found elsewhere
[1,5,6]. From these references it is possible to understand
the importance of an accurate knowledge of the X-ray production cross sections, which have been studied both theoretically [7,8] and experimentally, although there is not a full
description of the process at low proton energies. In particular, several papers have been devoted to the measurement of
the cross sections in this energy range [9-17], for K, L, and M
lines, although must of them deal only with the comparison
against the ECPSSR theory of Brandt and Lapicki [18], frequently proving the limitations of the model at these energies.
Some works have been focused to find a better description of
the cross sections for analytical purposes, such as the study of
the effect of the atomic parameters involved in the emission
of L X-rays [19], which was followed by the improvement
in the corresponding databases by Campbell [20,21]. Also,
there was a discussion about the role of multiple ionizations
(MI) in total L X-ray production cross sections in the low energy range [22], showing that to obtain good predictions MI
must be considered in the calculations. Recent global evaluations for the accurate prediction of the cross sections were
published by Kahoul et al. [23] for the K shell; Nekab and
Kahoul [24], Miranda et al. [25], and Smit and Tancek [26]
for the L-shell, while Pajek et al. [27] presented an evaluation
for the M-shell, specifically for PIXE applications. Some of
these works, however, are limited in their proton energy range
and, therefore, are not applicable to LE-PIXE.
3.
New applications of LE-PIXE
As mentioned above, since the most recent review about LEPIXE, only a few applications have been published, the majority of them in materials science. However, experimental
devices and methods have also been described. For example,
Ishikawa et al. [28] built a laboratory based on a low energy
cyclotron, with the advantage of operating it through the internet. The main applications of this laboratory are ion beam
modification and analysis of materials. For modification,
they use ion implantation, while the group applies low energy ion beams (such as 100 keV protons) to carry out PIXE,
RBS or NRA studies. Additionally, Miranda and Rodrı́guezFernández [29] established a method to calibrate the energy
of a small accelerator using PIXE and ion backscattering simultaneously. Their results showed that this method provides
good results, in agreement with the more accepted method of
nuclear resonances.
Regarding the use of LE-PIXE in materials characterization, there is an important advantage of the better depth
resolution and/or sensitivity as compared to that of “normal”
PIXE. For instance, Zahraman et al. [30] used 0.75 MeV
protons to analyze ultra-thin (0.5 nm < t <20 nm) chromium
layers deposited onto large surface quartz substrate, for uniformity control of the film thickness. They demonstrated that
there was a complete agreement with RBS measurements,
and that the signal-to-noise ratios as well as the limits of de-
83
tection for Cr were highly improved with the low energy protons, as compared to beams with energies between 1.5 MeV
and 3 MeV. Furthermore, the collection time for PIXE spectra were around 5 min to obtain a 3 %-4 % uncertainty in
the measurements, a shorter time than those required for ion
backscattering experiments.
Another application of LE-PIXE was the determination
of Mg and Mn contents in Bax Sr1−x TiO3 films by Dedyk et
al. [31]. The authors employed proton beams with energies
from 150 keV to 300 keV, reducing the background by more
than four orders of magnitude. Although the ionization cross
sections decrease also, the sensitivity to the impurities of the
elements with Z < 18 increased for the aforementioned energy range in comparison with MeV ion incident energies.
Furthermore, in this work a 230 keV neutral H beam was
also used to eliminate charging effects on the sample.
A further successful work where LE-PIXE gave better results was the determination of depth profiles in thin films and
ion-implanted samples when the X-ray detection angle was
varied. Miranda et al. [32] found that when lower energy
beams induced the X-ray emission it was possible to measure the profiles with a much better spatial resolution. Nevertheless, in their method it was necessary to assume a-priori
the shape of the distribution of the foreign atoms in a singleelement substrate.
A subject where LE-PIXE had never been used is the
analysis of atmospheric aerosols [1,2], one of the main applications of PIXE. The main reason is the need for a high sensitivity for most of the elements in samples that frequently have
no more than a total mass of 100 µg. However, very often the
results for the determination of light elements is obscured by
the high background in the X-ray energy range between 1
keV and 3 keV, namely, the corresponding to K lines of elements between Mg and Cl. Additionally, besides the low Xray yield, for these low energy photons it is necessary to make
important sample matrix corrections, due to proton stopping
and X-ray attenuation in the sample. In spite of all this, Espinosa et al. [33] obtained a significant upgrading in the limits of detection for light elements in aerosol samples taken in
Mexico City when irradiating the filters with 0.7 MeV protons, as compared to the results with 2.5 MeV protons. This
divergence is displayed in Fig. 1. It was not possible to measure Mg and Al concentrations with the 2.5 MeV beam, while
LE-PIXE gave good results for those elements. It is necessary
to stress the fact that no reliably measurements were obtained
for elements heavier than Ca employing the 0.7 MeV proton
beam.
Finally, a few words may be added to suggest improvements to the overall elemental sensitivity of LE-PIXE. As
explained in the past [4], its main limitation is the low X-ray
yield for elements heavier than Cl. A possible complementary technique may be X-ray fluorescence (XRF), which is
more sensitive for elements with atomic number above 20 if
the primary X-ray energy is higher than the absorption edge
of the analyzed element (such as that produced by an X-ray
tube with a Cu, Mo, or Rh anode). XRF, however, is not very
Rev. Mex. Fı́s. S 56 (1) (2010) 82–84
84
J. MIRANDA
good for lighter elements, such as those detected with LEPIXE. Thus, a simultaneous combination of LE-PIXE and
XRF applied to thin or homogenous thick targets might offer
a more uniform sensitivity curve for many elements. It must
be pointed out, however, that the physical process involved
in the emission of X-rays by an atom suffering irradiation simultaneously with protons and X-ray photons have not been
widely studied. This opens, thus, a new field of study in basic
atomic physics.
4.
Conclusions
F IGURE 1. Limits of detection (µg cm−2 ) for PM2.5 and PM10
aerosol samples collected in downtown Mexico City, and deposited
onto polycarbonate filters. The PIXE analyses were carried out
with proton beams of energies 0.7 MeV and 2.5 MeV, using a Si(Li)
detector and an in-vacuum Si-PIN detector, respectively.
The use of LE-PIXE, in spite of its benefits, has not seen important advances in the past fifteen years, except for the studies on X-ray production by proton beams with energies below
1 MeV. The most relevant applications were developed in materials science. The possible aid of XRF to analyze heavy
elements may open new utilizations of LE-PIXE. Nevertheless, LE-PIXE, although useful, does not seem to see a strong
advancement in the near future.
1. S.A.E. Johansson and J.L. Campbell, PIXE: A Novel Technique
for Materials Analysis (John Wiley, Chichester, 1988).
17. S. Szegedi, P. Raics, and M. Fayez-Hassan, J. Radioanal. Nucl.
Chem. 260 (2004) 429.
2. S.A.E. Johansson, J.L. Campbell, and K.G. Malmqvist, eds.
Particle Induced X-Ray Emission Spectrometry (John Wiley &
Sons, New York, 1995).
18. W. Brandt and G. Lapicki, Phys. Rev. A 23 (1981) 1717.
3. J. Rickards, A. Oliver, J. Miranda, and E.P. Zironi, Appl. Surf.
Sci. 45 (1990) 155.
4. J. Miranda, Nucl. Instr. and Meth. B 118 (1996) 346.
5. J. Miranda, Particle Induced X-Ray Emission (PIXE), in C.
Vázquez, ed., Topics in X-Ray Spectrometry (Comisión Nacional de Energı́a Atómica, Buenos Aires, 2007) pp. 201-216.
6. W. Maenhaut and K.G. Malmqvist, Particle-Induced X-ray
Emission Analysis, in R.E. Van Grieken and A.A. Markowicz, eds., Handbook of X-Ray Spectrometry, 2nd ed. (Marcel
Dekker, New York, 2002) pp. 719-810.
19. J. Miranda, C.M. Romo-Kröger and M. Lugo-Licona, Nucl. Instr. and Meth. B 189 (2002) 21.
20. J.L. Campbell, At. Data and Nucl. Data Tables 85 (2003) 291.
21. J.L. Campbell, At. Data and Nucl. Data Tables 95 (2009) 115.
22. J. Miranda, O.G. de Lucio, E.B. Téllez, and J.N. Martı́nez, Rad.
Phys. Chem. 69 (2004) 257.
23. A. Kahoul, M. Nekkab, and B. Deghfel, Nucl. Instr. and Meth.
B 266 (2008) 4969.
24. M. Nekab and A. Kahoul, Nucl. Instr. and Meth. B 245 (2006)
395.
7. G. Lapicki, Nucl. Instrum. Meth. B 189 (2002) 8.
25. J. Miranda, L.R de la Vega, J.C. Pineda, and E.B. Téllez,
Avances en Análisis por Técnicas de Rayos X XIII (2007) 305.
8. G. Lapicki, X-ray Spectrom. 34 (2005) 269.
26. Z. Smit and E. Tancek, X-Ray Spectrom. 37 (2008) 188.
9. S.J. Cipolla, AIP Conf. Proc. 392 (1997) 117.
27. M. Pajek et al., Nucl. Instr. and Meth. B 150 (1999) 33.
10. S.J. Cipolla, M.J. Dolezal, and L.O. Casazza, AIP Conf. Proc.
329 (1997) 113.
11. J. Miranda, R. Ledesma, O.G. de Lucio, Appl. Radiat. Isotop.
54 (2001) 455.
12. L.Rodrı́guez-Fernández, J. Miranda, J.L. Ruvalcaba-Sil, E. Segundo, and A.Oliver, Nucl. Instrum. and Meth. B 189 (2002)
27.
13. S.J. Cipolla, AIP Conf. Proc. 680 (2003) 15.
14. H. Mohan and A.K. Jaina, Nucl. Instr. and Meth. B 266 (2008)
1203.
15. H. Mohan, A.K. Jaina and S. Sharma, AIP Conf. Proc., in press.
16. S.J. Cipolla, AIP Conf. Proc., in press.
28. I. Ishikawa et al., J. Nucl. Sci. Technol. 44 (2007) 1039.
29. J. Miranda and L. Rodrı́guez-Fernández, Rev. Mex. Fı́s. 42
(1996) 826.
30. K. Zahraman, B. Nsouli, M. Roumie, J.P. Thomas, and S.
Danel, Nucl. Instr. and Meth. B 249 (2006) 447.
31. A.I. Dedyk et al., Ferroelectrics 286 (2003) 267.
32. J. Miranda, J. Rickards, and R. Trejo-Luna, Nucl. Instr. and
Meth. B 249 (2006) 397.
33. A.A. Espinosa, J. Miranda, and J.C. Pineda, Abstracts of the XI
Latin-American Seminar on Analysis with X-rays, SARX 2008
(Universidade Federal de Rio de Janeiro, Rio de Janeiro, Brazil,
2008).
Rev. Mex. Fı́s. S 56 (1) (2010) 82–84
Descargar