Interference Effects of Excess Ca, Ba and Sr on Mg Absorbance During Flame Atomic Absorption Spectrometry: Characterization in Terms of a Simplified Collisional Rate Model

Group II elements constitute an environmentally important group of metals. Calcium ranks 5th in relative abundance in nature. It occurs in limestone, dolomite, gypsum and gypsiferous shale, from which it can leach into underground and surface waters. Calcium content of natural waters can range from zero to several hundred milligrams per liter, depending on the source and treatment of the water. Magnesium occurs in nature in close association with calcium. It ranks 8th in abundance among the elements, and is a common constituent of natural waters. As with calcium, concentrations of magnesium in natural water may vary from zero to several hundred milligrams per litre, depending on source and treatment of the water. Calcium and magnesium salts contribute to water hardness. Concentrations of Mg above 125 mg/L can have cathartic and diuretic effects on the water (APHA, 1992). Barium ranks 16th in relative abundance in nature, and occurs in trace amounts in natural waters. Strontium resembles calcium, and interferes in the determination of calcium by gravimetric and titrimetric methods. Although most portable water supplies contain little strontium, levels as high as 39 mg/L have been detected in well water (APHA, 1992). FAAS and ICP-AES are the preferred methods for determining Gp II elements including Mg. Signal enhancement and/or depression were reported previous when Gp II elements were determined by atomic absorption spectrometry in the presence of other Gp II elements as interferents by several authors (Zadgorska and Krasnobaeva, 1977; Czobik and Matousek, 1978; Kos’cielniak and Parczewski, 1982; Smith and Browner, 1984; Zaranyika and Chirenje, 1999). Our approach to the study of interelement effects in atomic spectrometry involves a technique of probing changes in the number densities of the excited and ground states. Experimental analyte line emission intensity (I) and line absorbance (A) signal ratios, I’/I and A’/A, respectively, where the prime denotes readings taken in the presence of the interferent, are determined and compared to theoretical values derived assuming steady state kinetics. The method was used to follow collisional processes on the excitation and


Introduction
Group II elements constitute an environmentally important group of metals. Calcium ranks 5 th in relative abundance in nature. It occurs in limestone, dolomite, gypsum and gypsiferous shale, from which it can leach into underground and surface waters. Calcium content of natural waters can range from zero to several hundred milligrams per liter, depending on the source and treatment of the water. Magnesium occurs in nature in close association with calcium. It ranks 8 th in abundance among the elements, and is a common constituent of natural waters. As with calcium, concentrations of magnesium in natural water may vary from zero to several hundred milligrams per litre, depending on source and treatment of the water. Calcium and magnesium salts contribute to water hardness. Concentrations of Mg above 125 mg/L can have cathartic and diuretic effects on the water (APHA, 1992). Barium ranks 16 th in relative abundance in nature, and occurs in trace amounts in natural waters. Strontium resembles calcium, and interferes in the determination of calcium by gravimetric and titrimetric methods. Although most portable water supplies contain little strontium, levels as high as 39 mg/L have been detected in well water (APHA, 1992). FAAS and ICP-AES are the preferred methods for determining Gp II elements including Mg. Signal enhancement and/or depression were reported previous when Gp II elements were determined by atomic absorption spectrometry in the presence of other Gp II elements as interferents by several authors (Zadgorska and Krasnobaeva, 1977;Czobik and Matousek, 1978;Kos'cielniak and Parczewski, 1982;Smith and Browner, 1984;Zaranyika and Chirenje, 1999). Our approach to the study of interelement effects in atomic spectrometry involves a technique of probing changes in the number densities of the excited and ground states. Experimental analyte line emission intensity (I) and line absorbance (A) signal ratios, I'/I and A'/A, respectively, where the prime denotes readings taken in the presence of the interferent, are determined and compared to theoretical values derived assuming steady state kinetics. The method was used to follow collisional processes on the excitation and ionization of K resulting from the presence excess Na as interferent (Zaranyika et al., 1991). The approach assumes no change in the rate of introduction of analyte atoms into the excitation source, and no change in the temperature of the torch or flame, upon the simultaneous introduction of an easily ionized interferent element. Our argument is that according to the local thermodynamic equilibrium (LTE) approach, atomic line absorption and atomic line emission intensities are directly proportional to the population of the ground and excited states, respectively, i.e., AN 0 and IN j whereas N j and N 0 are related by the Boltzmann equation (Boumans, 1966): where g j and g 0 are the statistical weights of the excited and ground states, respectively, k is the Boltzmann constant, T is the absolute temperature and E is the difference in the energies of the two electronic states involved in the transition. If the rate of introduction of the analyte atoms into the plasma is kept constant, and we assume no change in the plasma temperature on simultaneous introduction of interfering metal atoms with the analyte, we may write: where the primes denote the actual populations of analyte ground state and excited atoms in the presence of the interferent. Combining equations 1 and 2, we have where n u is number density of the excited state. Equation 4 suggests that the effects of collisional processes on the excitation, and line emission of the analyte atoms resulting from the presence of interferent atoms may be followed by measuring the absorption or emission intensities of a given concentration of analyte atoms in the absence and presence of the interferent, and comparing the I'/I and A'/A ratios plotted versus analyte concentration. Applying the approach to absorption spectrometry, the following situations may be identified: i. No collisional effects, and therefore no change in the populations of the ground and excited states of analyte atoms: A'/A=1 ii. Increase of ground state, e.g. suppression of ionization: A'/A>1 iii. Depopulation of ground state, e.g. charge transfer reactions: A'/A<1 iv. Increase in excited state population: A'/A>1 v. Depopulation of excited state: A'/A<1 The aims of the present work were to investigate and characterize, in terms of a simplified collisional rate model, the interference effects observed when Mg is determined by airacetylene FAAS in the presence of excess Ca and Sr as interferents.

Equipment
An AA-6401 Shimadzu Spectrophotometer with an aberration-corrected Czerney-Turner mounted monochromator, automatic two step gain adjustment beam balance, automatic baseline drift correction using electrical double beam signal processing in peak height and area modes, was used in conjunction with an air-cooled 100-mm slot burner with a stainless steel head with a Pt-Ir capillary nebulizer with a Teflon orifice, glass impact bead and polypropylene chamber, and an air-acetylene flame. The air was supplied by a Toshiba Toscon compressor at 0.35 MPa input pressure, while the fuel gas was supplied from a pressuresed tank (BOC Zimbabwe (Pvt) Ltd, Harare), at 2.0 mL/min. Under these conditions the temperature of the flame was approximately 2573K [3][4][5], confirmed by a personal communication obtained from Shimadzu Inc., Japan. The spectrophotometer is equipped with automatic fuel gas flow rate optimization for each element. Experiments were carried out using Hamamatsu Photonics hollow cathode lamps (Hamamatsu, Japan) as source. The lamps was operated at the recommended minimum current of 8 mA for Mg respectively. Measurements were made using the 285.2 nm Mg line. The spectrophotometer employs a high speed 2-wavelength simultaneous measurement Deuterium lamp to correct for background signal. The nebulization chamber of the spectrophotometer was cleaned with triply distilled water to remove any deposited solids after each set of runs. The instrument was optimized for absorbance measurements, and care was taken not to change the instrumental settings/conditions until all measurements involving a particular interferent were completed. The average absorbance reading was recorded on the instrument computer monitor 5 seconds after aspiration. Instrumental parameters employed were as follows: Spectral band pass, 0.5 nm; Burner height, 7 mm; Burner angle, 0 o ; Acetylene fuel flow rate, 2.0 L/min.; Air input pressure, 0.35 MPa. Minimum Hollow cathode lamp current 8 mA. A mean aspiration rate of 3.000.06 mL/min. (n = 8), and mean nebulization efficiency of 6.31.7% (n = 8) were obtained.

Procedure
Four sets of standard solutions each containing 1 -30 mg/L Mg were prepared from a freshly prepared stock solution. Three sets were spiked with 1000 mg/L Ca or Sr each www.intechopen.com respectively; while one set each was left unspiked. The concentration of the interferent was kept in excess at 1000 mg/L while that of the analyte was kept at 0 -30 mg/L to minimize changes in the physical properties of the test solution. Any such changes in physical properties would affect the set of solutions to be analyzed to the same extent, and this was compensated for by taking blank readings of the solution containing the interferent only. Absorbance (A) readings were made for the spiked sets of solutions, as well as the unspiked set using distilled water as blank. The readings for the spiked sets of Mg solutions were then adjusted for blank readings of the solution containing the interferent only. A'/A ratios were calculated and plotted against Mg concentration in the test solution in Figure 1. Preliminary experiments were run to determine the aspiration rate and nebulization efficiency for the type of solutions under analysis. A mean aspiration rate (n = 8) of 3.000.06 mL/min and a mean nebulization efficiency (n = 8) of 6.31.7% were obtained.

Theoretical calculations
Ground state number densities were calculated assuming an aspiration rate of 3 mL/min. and 6 % nebulization efficiency as measured above, and a temperature of 2573 K for the airacetylene flame as noted above. Ionic number densities were calculated on the basis of the Saha relationship (Allen, 1955). Data obtained are shown in Table 1, in column headed 'n o (cm -3 s -1 )'. a Based on Saha relationship. M = element;  = degree of ionization assuming 2573 K flame temperature. c = analyte concentration in the test solution. Table 1. Ground state atom, ion and pre-thermal equilibration "hot" electron number densities, n e* .

Results and discussion
The experimental A'/A curves in Fig. 1 show a sharp increase in line absorbance signal enhancement with decrease in the concentration of the analyte in the test solution below about 10 mg/L. Similar results were reported previously in a study of mutual atomization interference effects between Group I elements (Zaranyika and Makuhunga, 1997). Absorbance signal enhancement is attributed to suppression of ionization, which is equivalent to collisional ion-electron radiative recombination. The major processes affecting analyte ground state population in the flame are represented schematically in Fig. 2.   Fig. 2. Proposed kinetic scheme: The subscripts AX2, o, u, and + denote analyte salt, ground state, excited state and ion respectively; M denotes interferent; k D , k  , k  ' denote rate constants for thermal dissociation, excitation from the ground state and excitation the excited state respectively; k CT and k CT' denote rate constants for collisional charge transfer involving interferent atoms and ions respectively; k c(o) and k c(1) denote rate constants for collisional radiative recombination to the ground state and excited state respectively; k hv and k R denote rate constants for radiative relaxation and analyte atom and counter-atom recombination respectively.
Assuming a steady state with respect to the analyte ground state and the excited state: Since k R n X '>>k R n X , n o' /n o will be less than unity, i. e., a depression of absorbency signal is expected in this case, contrary to the experimental results in Fig. 1. Although the present work reports analyte line absorption signal enhancement, analyte line absorption signal depression has been reported by several workers (Herrmann and Alkemade, 1963;Brown et al., 1987). Analyte line absorption signal depression conforming to Eq. 12 was also reported previously at low flame temperature when 0.2 -1.0 mg/L K solutions were determined in the presence of 1000 mg/L Na in the secondary reaction zones of the air-acetylene flame (Zaranyika et al., 1991).
Since n e ' = n + n e = n + + n mo , where n e represents the change in the electron number density in the presence of the interferent, and  is the degree of ionization of the interferent, Eq. 14 becomes  (17) i.e., although the absorbance signal enhancement is still dependent on interferent number density and inversely proportional to analyte number density, the enhancement observed will be reduced by a factor of at least n X /n X '. In addition, if we assume the A'/A = n o '/n o , a plot of A'/A versus n mo /n + should give a non-linear slope dependent on the concentration of interferent metal atom and counter atom, and with an intercept that is less than unity and equal to at least n X /n X '. Table 2 shows the slope, intercept and R 2 values for the regression plots of A'/A versus n mo /n + obtained when the enhancement factor A'/A for Mg is determined in the presence and absence of Ca and Sr as interferents. It is apparent from the data in Table 2 that the intercept values of 1.09 and 1.22 obtained are close to unity, in close agreement with Eq. 11. These data suggest that the signal enhancement obtained when Mg is determined in the presence of excess of Ca and Sr as interferents conforms to LC IIA. Signal enhancement in the presence of easily ionizable elements (EIEs) in atomic absorption spectrometry is commonly attributed to suppression of ionization (Foster Jr. and Hume, 1959;Smit et al., 1951). Supression of ionization is in effect collisional radiative recombination, and assumes that in Eq. 14 () CR o e CT mo kn n k n n    so that Eq. 14 reduces to where n e is the change in electron number density upon the addition of the interferent, i.e., n e = n m+ , E a ' and E a are the activation energies for the electrons from the ionization of the interferent and analyte respectively, and. k c is the collisional rate constant given by (Weston and Schwarz, 1972): where Q 12 is collision cross-section between particles 1 and 2, and  is their reduced mass. If we assume thermal equilibrium conditions, then all the electrons will require the same activation, i.e., E a ' = E a , and Eq. 19 reduces to 1 oe oe nn nn where n e = n m+ and n e = n + . It is apparent that substitution for n e and n e from Table 1 will yield values of n o '/n o up to three orders of magnitude greater than experimental A'/A values at low Mg concentrations. If however we assume pre-thermal equilibrium collisional radiative recombination, then the electrons from the ionization of the interferent would require further activation by an amount of energy equal to the difference between the ionization potentials of the analyte and that of the interferent, i.e., in eq. 19, www.intechopen.com where the subscripts a and m denote analyte and interferent respectively. If we make this assumption, then Eq. 15 becomes Table 3 shows the fraction of such electrons having the appropriate energy, and Table 1 shows the corresponding pre-thermal equilibration number density of "hot" electrons, n e *, arising from the analyte, Mg (n e * = n + ), and the interferents Ca and Sr (n e * = n + exp(-E a '/kT)). * E a ' = IP a -IP m .
A major objective for kinetic modeling of interference effects is to be able to predict the interference observed experimentally. If assume that It is apparent from Fig. 3A that Eq. 24 leads to close agreement between theory and experiment at Mg concentrations below about 5 mg/L when Mg is determined in the presence of excess Ca, while the theoretical curve deviates slightly from the experimental curve above this concentration. In contrast to the case when Mg is determined in the presence of excess Ca, exact agreement between theory and experiment is obtained throughout the range of Mg concentrations studied when determined in the presence of excess Sr. Although the work presented in this paper is rather limited in scope, the remarkable success of the model in predicting the interference of excess Sr on the absorbance signal of Mg, confirms the potential of the model as represented by Eq. 24 in simulating the absorbance signal enhancement interference effects during flame atomic spectrometry. Further work is underway to test the model on other systems.

Conclusions
From the foregoing discussion we conclude that the interference effects between Group II elements can be characterized using a simplified rate model that takes into account collisional radiative recombination, charge transfer between analyte and interferent species, and collisional recombination between analyte atom and counter atom. The model predicts that, depending on the specific experimental conditions employed, interference effects in flame atomic spectrometry can manifest themselves as enhancements (Eq. 15) or depressions (Eqs. 12 and 17) of the analyte line absorbance signal. Data relating to the signal enhancement interference effects of excess Ca and Sr on Mg absorbance signal during airacetylene flame absorption spectrometry suggest that the signal enhancement can be simulated on the basis of a simplified rate model that assumes pre-LTE ion-electron radiative recombination.

Acknowledgements
This work was supported by a grant from the Research Board of the University of Zimbabwe (UZ). Atomic Absorption Spectroscopy is an analytical technique used for the qualitative and quantitative determination of the elements present in different samples like food, nanomaterials, biomaterials, forensics, and industrial wastes. The main aim of this book is to cover all major topics which are required to equip scholars with the recent advancement in this field. The book is divided into 12 chapters with an emphasis on specific topics. The first two chapters introduce the reader to the subject, it's history, basic principles, instrumentation and sample preparation. Chapter 3 deals with the elemental profiling, functions, biochemistry and potential toxicity of metals, along with comparative techniques. Chapter 4 discusses the importance of sample preparation techniques with the focus on microextraction techniques. Keeping in view the importance of nanomaterials and refractory materials, chapters 5 and 6 highlight the ways to characterize these materials by using AAS. The interference effects between elements are explained in chapter 7. The characterizations of metals in food and biological samples have been given in chapters 8-11. Chapter 12 examines carbon capture and mineral storage with the analysis of metal contents.