Determination of Sulfur Content in Steel by Laser-Produced Plasma Atomic Emission Spectroscopy ANA GONZALEZ, MONTSERRAT ORTIZ, and JOSE CAMPOS* Unidad de Ffsica At6mica y Laseres, Instituto de InvestigaciOn Bt~sica, CIEMAT, Avda Complutense 22, 28040 Madrid, Spain (A.G., M.O.); and Cdtedra de Ffsica At6mica Experimental, Facultad de Ciencias Ffsicas, Universidad Complutense de Madrid, 28040 Madrid, Spain (J.C.) Sulfur content in steel samples has been determined by laser-produced plasma atomic emission spectroscopy with the use of a Q-switch Nd: YAG laser. With the use of time-resolved spectroscopy employing an OMA III (EG&G) as detector, a detection limit of 70 ppm and a precision of 7% have been obtained. Calibration curves are linear, and no notice- able matrix effects have been observed. Index Headings: Laser microprobe; Sulfur analysis; Emission spectros- copy. INTRODUCTION Most conventional methods to determine sulfur in steels are based on gravimetric, spectrometric, and gas-analysis measurements. The most common method is based on SO2 determination. The manufacture of steels with very low sulfur concentration requires improvement of the existing methods and development of new ones in order to obtain reliable and accurate results minimizing the time necessary for the analysis. For these reasons, other methods such as the laser-based infrared absorption technique I and cathodic stripping voltammetry after con- version of the sulfur into hydrogen sulfide2 have been developed. Laser microprobe atomic emission spectroscopy is a simple and fast technique to carry out direct elemental analysis. It is based on laser ablation of the sample and optical emission spectrometry of the microplasma pro- duced. Since the pioneering works of Brech and Cross 3 and Brech and Schuch, 4 numerous studies have been car- fled out in the last decades in the field of laser ablation for elemental analysis? -9 The development ofphotodiode array systems as detectors 1°,11 has improved this tech- nique. This method can be used in solid and liquid 12 samples. Recently, laser microprobe mass spectrometry has been applied to in situ sulfur isotope analysis of geo- logical materials.~3.14 Application of laser-produced plas- ma emission spectrometry to sulfur content in rubber mixtures has been described in Ref. 15. Nevertheless, to our knowledge, there are no published studies of sulfur content measurement in steel by the last method. The aim of the present work is to show the applicability of this technique to sulfur determination in steel samples. In previous works of this g r o u p , 16,17 this method has been applied to carbon determination in steel and molten steel samples. The data of Kaufman and Martin TM have been used to identify the sulfur spectroscopic lines. To provide spectral Received 10 February 1995; accepted 2 August 1995. * Author to whom correspondence should be sent. line wavelengths of other elements, we have used the tables of Reader et al. 19 and Striganov and Sventitskii? ° EXPERIMENTAL The experimental arrangement is similar to the one used in previous works, 16,17 and so it is only briefly de- scribed here. A Q-switch Nd:YAG laser generates 200- mJ pulses of 7-ns duration at 20-Hz frequency and 1065- nm wavelength. The laser beam was focused onto the target with a lens of 12.5-cm focal length. The sample was inside a chamber filled with N2 at 1000 mbars. The diameter of the laser-focusing region on the sample was about 0.5 mm, corresponding to an irradiance of 1.3 101° W/cm 2. The light emission perpendicular to the laser light from the plasma was directed to the entrance of a 1-m Czerny-Tumer monochromator (Spex 1704) provided with a 2400-groves/mm holographic grating. The spectral range was 178-700 nm with 0.03-nm spectral resolution. A time-resolved optical multichannel analyzer (OMA III, EG&G) allowed the detection of spectrum sections at preset times from the laser pulse and during a selected time length. The spectral range of every spectrum section was 10 nm. To measure spectra below 190 nm, it was necessary to fill the monochromator with an inert gas to avoid light absorption by the 02 Schumann band system. In our experiment, the monochromator was filled with N2 at 1000 mbars. The chamber was placed close to the mono- chromator entrance slit to collect the maximum of the emission intensity. RESULTS AND DISCUSSION Calibration curves have been obtained with the use of certified samples whose compositions are given in Table I. The calibration curves, which relate the S(I)/Fe(II) emission intensity ratio to the S/Fe concentration, are linear over the range 0.008 to 0.28% of sulfur concentra- tion. First, it is necessary to select the optimum spectral region. According to the existing references, ~8,21the most intense lines of S(II) lie in the 550-nm spectral region (543.23-, 545.38-, and 547.36-nm lines), and the 400- nm region shows intense S(I) spectral lines. Also in the UV region atomic sulfur emits intense lines at 180.73, 182.03, and 182.62 nm. The line selection study was carried out with a FeSz sample. Spectral interference due to overlap with iron lines was found in the 550- and 400-nm spectral regions. Therefore, the 180.0-nm region was chosen for this ex- 1632 Volume 49, Number 11, 1995 0003-7028/95/4911-163252.00/0 APPLIED SPECTROSCOPY © 1995Society for Applied Spectroscopy