DETERMINATION OF COPPER AT WIDE RANGE CONCENTRATIONS USING INSTRUMENTAL FEATURES OF HIGH-RESOLUTION CONTINUUM SOURCE FLAME ATOMIC ABSORPTION SPECTROMETRY

This work describes a method to determine Cu at wide range concentrations in a single run without need of further dilutions employing high-resolution continuum source flame atomic absorption spectrometry. Different atomic lines for Cu at 324.754 nm, 327.396 nm, 222.570 nm, 249.215 nm and 224.426 nm were evaluated and main figures of merit established. Absorbance measurements at 324.754 nm, 249.215 nm and -1 -1 -1 224.426 nm allows the determination of Cu in the 0.07 – 5.0 mg L , 5.0 – 100 mg L and 100 – 800 mg L concentration intervals respectively with linear correlation coefficients better than 0.998. Limits of detection -1 -1 -1 were 21 μg L , 310 μg L and 1400 μg L for 324.754 nm, 249.215 nm and 224.426 nm, respectively and relative standard deviations (n = 12) were £ 2.7%. The proposed method was applied to water samples spiked with Cu and the results were in agreement at a 95% of confidence level (paired t-test) with those obtained by line-source flame atomic absorption spectrometry.

Abstract: This work describes a method to determine Cu at wide range concentrations in a single run without need of further dilutions employing high-resolution continuum source flame atomic absorption spectrometry.Different atomic lines for Cu at 324.754 nm, 327.396 nm, 222.570 nm, 249.215 nm and 224.426 nm were evaluated and main figures of merit established.Absorbance measurements at 324.754 nm, 249.215 nm and -1 -1 -1 224.426 nm allows the determination of Cu in the 0.07 -5.0 mg L , 5.0 -100 mg L and 100 -800 mg L concentration intervals respectively with linear correlation coefficients better than 0.998.Limits of detection

Introduction
Copper is usually present at different concentration levels in a variety of workable samples [1].The elemental determination by atomic absorption spectrometry [2,3] is frequently performed by line source spectrometers.Although the line-source flame atomic absorption spectrometry (LS FAAS) [4] is a worldwide, robust and well-established technique, the single-element analysis and narrow range calibration may be considered the main drawbacks of the technique when several elements are required: the changing and conditioning of hollow cathode lamps (or electrodeless discharge lamps) and the need for adjusting analyte absorbance within the linear working range of calibrating plots are time consuming and lead to increased analytical cost in large scale routine analyses [5].However, these cumbersome may be circumvented using a high-resolution continuum source flame atomic absorption spectrometry (HR-CS FAAS).With this new concept, a high-resolution double-Echelle monochromator and a charge-couple device detector with a xenon arc-short lamp with a continuum source [6] makes feasible the application of the atomic absorption spectrometry for fast-sequential and multi-element analyses [7] using one or more atomic lines of the same element sequentially, wavelengthintegrated absorbance over the line core to enhance sensitivity, or at least measuring parts of the wings (side pixel registration) to extend the linear working range calibration [8][9][10].The fast-sequential analysis is a particularly advantageous for routine laboratories involved in large scale analyses, because time and analytical costs are significantly reduced.Little attention has been given to the use secondary and alternating lines in flame AAS to reduce sensitivity and increase the dynamic working range to determine low, intermediate and high levels of an element in a single run without need of further dilutions.
In this work the main (324.754nm), secondary (327.396nm) and alternate (222.570nm, 249.215 nm and 224.426 nm) lines of Cu were evaluated in order to develop a method for wide range determination using instrumental facilities of the HR-CS FAAS.

Instrumentation
The measurements were carried out using an Analytik Jena ContrAA 300 high-resolution continuum source flame atomic absorption spectrometer equipped with a xenon short-arc lamp XBO 301 (GLE, Berlin, Germany) [11] with a nominal power of 300 W operating in a hot-spot mode as a continuum radiation source.Main (324.754nm), secondary (327.396nm) and alternate (222.570nm, 249.215 nm and 224.426 nm) atomic lines for Cu were evaluated to establish the figures of merit of all lines.High-purity acetylene (99.7%,Air Liquid, Sertãozinho, Brazil) was used as fuel gas, and an oxidizing air-acetylene flame was used for analyte atomization.A Perkin Elmer AAnalyst 100 line source spectrometer (Shelton, CT, USA) was also used as comparative technique, and the determination of Cu was carried out in an air-acetylene flame using a Perkin-Elmer Lumina™ hollow cathode lamp.

Reagents, analytical solutions and samples
High-purity de-ionized water obtained using ® a Millipore Rios 5 reverse osmosis and a Millipore ® Milli-Q Academic deionizer system (resistivity 18.2 MÙ cm, Millipore, Bedford, MA, USA), and nitric ® acid (Suprapur , Merck, Germany) were used throughout to prepare all solutions. -1 A 5000 mg L Cu standard stock solution was prepared by dissolving 1.0 g of metallic copper (99.9% purity) in a minimum volume of concentrated nitric acid.The final solution was transferred to a 200 mL volumetric flask and the volume completed with 1.0% (v/v) HNO solution.-600 mg L were used to check the accuracy and the effectiveness of the proposed method.All solutions and samples were stored in highdensity polypropylene bottles (Nalgene®, Rochester, USA).Plastic bottles and glassware materials were cleaned by soaking in 10% (v/v) HNO at 3 least 24 h and rinsed abundantly in de-ionized water before use.

Measurement procedure
Absorbance measurements were carried out at 324.754 nm, 327.396 nm, 222.570 nm, 249.215 nm, and 224.426 nm.The equipment was optimized to provide the best sensitivity.For wavelengthintegrated absorbance equivalent to 3 pixels, 5.0 mL -1 min of sample flow-rate, the absorbance for blanks, working standard solution and spiked samples solutions were measured at the selected atomic lines for Cu to obtain the calibration curves for all wavelengths.All measurements were carried out in six replicates using an injection module (SFS 6) enabling the computer controlled aspiration of blanks, working standard solutions and samples.
All samples were also analyzed by LS FAAS at the main line of Cu at 324.754 nm.Because the instrumental limitations of technique, spiked samples were diluted to adjust the analyte absorbance within the linear working range.All measurements were carried out in triplicate.
The limit of detection (LOD) and limit of quantification (LOQ) for Cu were calculated according to the IUPAC recommendation [12].

Results and discussion
In HR-CS FAAS the continuum radiation source provides a better SNR than hollow cathode lamps usually found in LS FAAS.In this context, the use of secondary and alternate atomic lines of a given element is a good strategy to reduce sensitivity and extend the working range calibration.

Optimization of the best condition for different wavelengths for Cu
The performance of atomic lines for Cu at 324. 754  After burner height and flame composition optimization, the linear working range was evaluated by plotting curves of absorbance versus analyte -1 concentration within the 0.05 -1000 mg L intervals (Figure 1).Analysis of figure reveals that calibration  determination of Cu within the 0.07 -800 mg L concentration range using only 3 wavelengths: -1 324.754 nm (0.07 -5.0 mg L ), 249.215 nm (5.0 --1 -1 100 mg L ), and 224.426 nm (100 -800 mg L ).

Analyses of real samples
After method development, three analytical  2 shown results found by HR-CS FAAS using the proposed method.Results described in the table are in agreement at 95% of confidence level (paired ttest) with those obtained by LS-FAAS.Accuracy studies were carried out by recoveries tests and the results obtained by HR-CS FAAS varied within the 98 -103% and for LS FAAS within the 97 -104% intervals.Relative standard deviations for the measurements (n = 12) varied from 1.2% to 2.0% (324.754nm), from 1.6% to 2.6% (249.215nm) and from 1.5% to 1.8% (224.426nm) by HR-CS FAAS and from 2.2% to 3.9% by LS FAAS.Limits of

Conclusions
The use of different atomic lines of a given element in HR-CS FAAS is feasible to elemental determination in a wide range concentration without need of further dilutions of samples, reducing the errors associated to the excessive sample handling.The example here is the determination of low, medium and high concentrations of Cu in waters, but can be extended to other element and matrix.

3 1 the
Copper working standard solutions in the -1 0.05 -1000 mg L interval were prepared by appropriate dilution of the stock standard solutions and acidified to 1.0% (v/v) with HNO .For LS FAAS 3 concentration interval was 0.5 -5.0 mg L .Tap water samples spiked with Cu within 1.0 -1
The main figures of merit of studied atomic lines by HR-CS FAAS and LS-FAAS are shown in Table1.Analysis of this table reveals that the highest

Table 1 .
Figures of merit of main (324.754nm), secondary (327.396nm) and alternate (222.570nm, 249.215 nm and 224.426 nm) atomic lines for Cu by HR-CS FAAS and for 324.754 nm by LS-FAAS.
a Linear correlation coefficient