JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1993 VOL. 0 307 lntracavity Laser Spectroscopic Method for Determining Trace Amounts of Iodine and Barium in Water and Biological Samples" V. S. Burakov A. V. Isaevich P. Ya. Misakov P. A. Naumenkov and S. N. Raikov Institute of Molecular and Atomic Physics Academy of Sciences of Belarus Minsk 220072 Belarus An intracavity laser spectroscopic method and the necessary instrumentation are described for the direct determination of trace amounts of iodine in water and in biological media. Minimal sample preparation is required for the laser probing of vapours over the surface of the heated liquid in a closed cell. This laser method is also applied to measurements of ultra-low barium contents in water by sample evaporation in a standard graphite furnace electrothermal atomizer.Detection limits of 0.015 mg I-' and 0.2 ng I-' were obtained for iodine and barium respectively which are 2-3 orders of magnitude lower than those obtained using a pulsed dye laser. The precision of the method was 10% for the lowest concentrations measured. The procedure is comparatively simple quick and inexpensive. Keywords Laser spectroscopy; atomic and molecular absorption; electrothermal probe atomization; urine; iodine; barium The determination of low levels of biologically significant elements such as iodine in human tissues water and food is an important problem for the practical care of public health. The prevalence of goitre is endemic in some regions for example in Belarus i.e. the population lives under conditions of natural iodine insufficiency.Along with the determination of iodine in water an important indication of the iodine supply of an organism is the study of the iodine excreted in urine. This permits an assessment of the extent to which goitre is endemic in a particular region to be made as well as providing the possibility of establishing scientifically substan- tiated dosing of iodine-containing preparations used for treat- ing the disease and its prophylaxis. It is also considered necessary to develop new techniques for determining trace amounts of toxic elements among them barium in water and other environmental samples because conventional analytical methods have almost reached their practical limits. It is known that even the smallest amounts of such elements or their compounds have a toxic effects on humans and other organisms The determination of iodine in a water sample by the conventional spectrophotometric methods presents no unsur- mountable problems.For example indicator reactions have been proposed for determining iodine in a water sample using non-coloured organic reagents that form intensively coloured products on reaction with iodine ions in an acidic medium in the presence of hydrogen peroxide.' With spectrophotometry in the visible range an iodine detection limit of 1 pg 1-1 has been achieved for such solutions. However the above method has an essential disadvantage the determination is impeded by the presence in the solution of iron which is widely distributed as well as bromide molybdenum silver and mercury. In real biological media for example in human secretions the spectrometric determination of small amounts of iodine presents a fairly difficult problem because of the intensive light absorption by organic and inorganic products. The physico- chemical and thermal treatment and the preparation of samples to be analysed is not effective moreover it can lead to uncontrollable changes in the concentration of the iodine in the samples especially at low levels.Therefore a colorimetric method is usually used to determine iodine in urine at levels higher than 1 mg 1-'. The application of ionomers with iodide ion-sensitive elec- trodes enables a detection limit for iodine of 0.03 mg 1-1 to be * Presented at the XXVIII Colloquium Spectroscopicum Internationale (CSI) York UK June 29-July 4 1993.attained.2 In real samples however in addition to iodine a number of other ions that impede measurements by ion- selective electrode can be present. In practice there are also chemical methods for measuring iodine in biological media with the same detection limit.3 The disadvantages of these methods are the length of time they take and the necessity of using highly toxic compounds (arsenic brucine etc.). For barium the lowest detection limits attainable are reached by using atomic absorption spectrometry with electrothermal atomization or by inductively coupled plasma mass spec- trometry (ICP-MS). There are a series of modern spectrometers for which detection limits for barium of 0.002-0.04 pg 1- ' can be obtained.However further development of these methods now depends mainly on improvement of service and software which will not lead to noticeable increases in sensitivity. Thus the need for ultra-sensitive laser methods for determi- nations at trace levels is becoming increasingly desirable as the required detection limits for elements decrease. In this paper a laser spectrometric technique is proposed for (i) direct measurements of trace amounts of iodine in water and urine with minimal sample preparation by using the method of intracavity laser spectroscopy (ICLS) with conditions for pro- bing the vapours over the surface of a heated liquid in a closed cell; and (ii) intracavity laser spectral analysis of trace amounts of barium in water by sample evaporation in a standard graphite furnace electrothermal atomizer. The ICLS method is based on the strong dependance of laser radiation intensity on the cavity losses resulting from the introduction into the laser cavity of a layer of a substance having absorption lines (bands) within the active medium gain contour of the laser.- The main advantages of this method are its ultra-high sensitivity to small concentrations of absorb- ing species and high spectral resolution given by the mode structure of the laser radiation.The ICLS method differs favourably from such widely used laser methods as laser- induced fluorescence and from laser-enhanced ionization because of the possibility of simultaneously recording a large number of spectral line shapes in a single experiment. The ICLS system used is similar to that employed in continuum source atomic absorption spe~trometry,~ but the latter was reported to be a few orders of magnitude less sensitive.A detailed description of the proposed ICLS method the basic equations and analytical procedure have been reported Experimental Measurements were carried out using the intracavity laser spectrometer that has been described in detail previ~usly.~~'~308 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1993 VOL. 0 It is shown schematically in Fig. 1 and consists of a tunable flash-lamp pumped pulsed (5 ps) dye laser as the primary light source radiating a smooth broadband (10 nm) spectrum in the range 430-700nm; a special cell for liquid samples or a standard graphite furnace atomization source both located in the laser cavity; and a 0.001 nm resolution kchelle spectrograph with an optical multichannel analyser.An hermetic quartz (or glass) cell 270 mm long and 25 mm inner diameter coaxially arranged inside an ohmic heater providing heating of the vapours of up to 100°C and above is used for liquid samples. The temperature inside the cell is controlled to within 1 "C and 10 ml doses of the liquids to be analysed (aqueous solutions and urine) are introduced into the cell. The laser beam passes over the liquid surface along the geometrical axis of the cell. The use of a 5 ps laser makes it possible to provide an effective path length of about 1 km for the laser radiation through the vapours. For the determination of iodine the laser radiated a spectrum in the 580-590 range falling within the region where the visible system of the iodine molecular absorption bands is situated.Iodine in urine is present almost exclusively in the form of inorganic salts mainly potassium iodide. Therefore to prepare standards for calibrating the spectromer aqueous solutions of dried chemically pure potassium iodide were used. A stock solution (1000 mg 1-l) remains unchanged for a long period of time in the absence of light. Working solutions were prepared daily by appropriate dilution of the stock solution with de-ionized triply distilled water. Just before measurements were made the solutions and urine were subjected to the minimum of chemical pre-treatment to decompose the iodide salts using a conventional analytical method a few drops of sulfuric acid and hydrogen peroxide were added directly to the cell.'' A standard electrothermal atomizer' was used in the intra- cavity laser spectrometer for probe atomization.Aqueous samples of BaCl (20 pl) were dried ashed and atomized using a pyrolytic graphite coated graphite tube (28 mm long) with the following programme (i) dry 100 "C 40 s; (ii) ash 300 "C 10 s; (iii) atomize 2600"C 6 s; and (iv) clean 2800"C 3 s. Atomic absorption signals were measured at the barium wave- length of 553.548 nm. The laser radiated a spectrum in the 550-560nm range. The effective path length of the laser radiation through the furnace was about 100 m. Results and Discussion By means of preliminary experiments the most intensive electronic absorption band of molecular iodine with a peak wavelength at about 585 nm was chosen for the measurements. To increase the iodine yield over the surface of the prepared solution the walls and windows of the cell were heated Back mirror output mirror Prism laser I I High uniformly.As the vapour temperature increased there was a noticeable increase in the iodine absorption signal. Strong heating of the cell however led to excessive evaporation of the water the wide absorption bands of which could be superimposed on and hence mask the iodine bands. Therefore the experimentally established optimum vapour temperature for the analysis was 80-90 "C. The dependence of the sensitivity of the determination of iodine in two absorption spectra recording regimes were initially investigated. The first procedure chosen was low resolution (0.04 nm) of the spectrograph when the whole laser spectrum is recorded with wide electronic absorption bands of the evaporated iodine molecules.The second procedure was high spectral resolution (0.001 nm) when a narrow range of the laser spectrum (585.0-585.5 nm) at the peak of the elec- tronic absorption band of iodine is recorded. In the latter the vibrational-rotational line structure of the band is completely resolved [Fig. 2(a)]. With complete resolution of the line absorption spectrum of iodine some gain in sensitivity and accuracy in the determination of trace amounts of iodine is observed. In the first recording regime the observable width of the electronic band is comparable to the width of the laser spectrum. Therefore a change in the iodine content in the laser cavity causes a noticeable transformation of the shape of the laser spectrum which strongly impedes quantitative measurements especially measurement of the background laser intensity beyond the absorption band.In the second method of recording measurements of relative absorption at the narrow vibrational-rotational iodine line centre and beyond it gives much more stable results. Conventionally the relative absorp- tion is expressed as (E -E)/E where Eo = (Eol + E,,)/2 as shown in Fig. 2(b). A calibration graph for iodine was plotted using the high spectral resolution (Fig. 3). The precision (maximum value of the relative standard deviation) of the measurement obtained for ten replicate determinations of iodine in a standard solution was approximately 10% for the lowest measured (0.05 mg 1-l) iodine content.A detection limit derived from 0.1 x the relative I C 585.4 585.2 Q .? 30 a c - Q 10 I I + I J 553.6 553.5 Wavelengthln m Fig. 1 Schematic representation of the intracavity laser spectrometer Fig. 2 Partial laser spectra showing absorption lines of (a) iodine and (b) bariumJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1993 VOL. 0 309 0.9 a 0.7 + 2 0.5 a .- + 0.3 K 0.1 I I I I 0.01 0.1 1 10 [Iyrng I-' Fig. 3 Calibration graph for molecular absorption of iodine 5.0 1.5 0) c m fl 8 1.0 -Q m a .- c.' - a 0.5 cc I I I 0.01 1 100 [Bal/pg I-' Fig. 4 Calibration graph for atomic absorption of barium absorbance signal of 0.015 mg 1-l for iodine in solution was achieved. The detection limit for iodine attained in this work can easily be lowered if necessary by 2-3 orders of magnitude by simply replacing the pulsed dye laser of microsecond duration by a continuous wave dye laser (where the effective path length exceeds 1 x lo3 km proportionally increasing the laser pulse duration') all other modules of the intracavity spectrometer remaining unchanged.In this case the detection limit for iodine achieved with the intracavity spectrometer can be compared with the detection limit for iodine using ICP-MS of 0.008 mg 1-'.13 The ICLS method was applied to the measurement of iodine in the urine of patients who had not knowingly taken any iodic compounds in the form of food or medicine. No signifi- cant spectral interferences from the urine matrix were encoun- tered.The organic and mineral components in a urine sample have no effect on the absorption of the laser radiation over the liquid surface by the iodine molecules. In the urine being analysed the iodine contents varied over the range 0.1-1.0 mg I-'. Fig. 2(b) shows the part of the laser radiation spectrum that is transmitted through a furnace in the laser cavity near the barium spectral line. There are no absorption signals in the blank furnace. The calibration graph for barium formed by plotting the relative absorbance versus the concentration in solution is shown on Fig. 4. Fot a relative absorbance not only is the relative magnitude of the absorption signal in the centre of the transmission profile used (by analogy with measurements of the iodine contents) but also the relative width of the atomic absorption line for large barium contents under conditions of total absorption when E=O (points on the graph).The precision of the measurements obtained for ten replicate analyses of barium standard solutions was approximately 10% for the lowest measured (0.7 ng I-') barium content. The accuracy of the ICLS method was determined mainly by the stability of the spectral distribution of the laser radiation intensity. A 20 p1 volume of de-ionized triply distilled water gave an absorption signal at the barium line at a level of 0.5 ng 1-'. The detection limit derived from 0.1 x the relative absorbance signal of 0.2ngl-1 of barium in solution (or 0.004 pg for a 20 pl probe) was achieved. Similar results for intracavity laser spectrometry have been obtained previously for other element^.^^^^ Conclusion In this work the process of intracavity laser spectral determi- nation of trace amounts of iodine in water and urine and barium in water has been optimized.The detection limits for iodine and barium are lower as compared with conventional methods and low iodine contents in real biological samples have been measured directly with minimum preparation of the samples being probed. The proposed method and equipment can be used to validate routine laboratory measurements of low iodine contents in urine and other biological media as well as to determine ultra-low amounts of iodine barium and other elements in real samples without preconcentration. 1 2 3 4 5 6 7 8 9 10 11 12 13 References Kreingold S.U. Sosenkova I. I. Panteleimonova A. A. and Lavrelashvili L. V. Sov. J. Anal. Chem. 1978 33 2168. Econics Ion-Selective Electrodes Econics Moscow Russia. Kienya A. I. Rosental I. A. and Kulchitskaya S. N. Lab. Delo. 1985 9 534. Pachomycheva L. A. Sviridenkov E. A. Suchkov A. F. Titova L. V. and Churilov S . S. Pis'ma Zh. Eksp. Teor. Fiz. 1970 12,60. Peterson N. C. Kurylo M. J. Braun W. Bass A. M. and Keller R. A. J. Opt. SOC. Am. 1971 61 746. Trash R. J. von Weyssenhof H. and Shirk J. S. J. Chem. Phys. 1971 55 4559. Jones B. T. Smith B. W and Winefordner J. D. Anal. Chem. 1989 61 1670. Baev V. N. Belikova T. P. Sviridenkov E. A. and Suchkov A. F. Sov. JETP 1978 74 43. Burakov V. S. Voitovich A. P. Mashko V. V. and Raikov S. N. Lasers-Physics and Applications World Scientific Singapore New Jersey London Hong Kong 1989 p. 39. Burakov V. S. Gvozdev A. A. Kovalev A. Ya. Misakov P. Ya. and Raikov S. N. J. Appl. Spectrosc. (Engl. Transl.) 1989,51 1123. Altgausen A. Ya. Laboratory Clinical investigations Medicine Moscow 1964.. Burakov V. S. Esilevski V. A. Misakov P. Ya. Naumenkov P. A. Pelieva L. A. Pimanov Yu. P. Uzunbadjakov A. S. and Sharnopolski A. I. J. Appl. Spectrosc. (Engl. Transl.) 1985 42 247. Atomic Spectroscopy Detection Limits Perkin-Elmer. Paper 3/03905 F Received July 16 1993 Accepted September 22 1993