Simultaneous determination of metals at trace level in a multicomponent system. Application to real samples

Ekrrochimica Acta, Vol. 41, No. 13, pp. 2011-2017, 19% Copyright 0 1996 ElsevierScienceLtd. Printedin Great Britain.AU rights reserved 0013~4686/96 $15.00 + 0.00

Pergamoo

SIMULTANEOUS DETERMINATION OF METALS AT TRACE LEVEL IN A MULTICOMPONENT SYSTEM. APPLICATION TO REAL SAMPLES-f C. LOCATELLI*

and G. TORSI

Department of Chemistry “G. Ciamician”, University of Bologna, Via F. Selmi 2, I-40126 Bologna, Italy (Receioed 27 September 1995; in reoisedform 14 December 1995) Abstract-Voltammetry is a very suitable, versatile and rapid method for the simultaneous metal determination in complex matrices. The present work, regarding the determination of Cu, Fe, Cr, Sn, Ti, MO and Mn, is a very interesting example of the possibility to determine single elements in real samples. 0.1 M ammonium citrate (pH 6.1 and 8.5) was employed as the supporting electrolytes. Differential pulse voltammetric (DPV) and alternating current voltammetric (ACV) measurements were carried out using, as working electrode, a stationary mercury electrode, as auxiliary a platinum electrode and as reference an AgjAgCl, KCl sat. electrode. The analytical procedure was verified by the analysis of standard reference materials such as Stainless Steel (AISI 321) NBS-SRM 121dl and Highly Alloyed Steel Eurostandard 281-la. Precision and accuracy, expressed as relative standard deviation and relative error, respectively, were of the order of 3-5%, while the detection limit for each element was around 10e9 M. The standard addition technique extends the usefulness of the voltammetric method to very high element concentration ratios. Copyright 0 1996 Elsevier Science Ltd. Key words: voltammetry, trace determination, interferences, real samples, alloys.

INTRODUCTION

Multi-element analysis is a goal which must be pexsued in all analytical methods. In the analysis of metals in real matrices a lot of determinations have been carried out by means of spectroscopic techniques, which unfortunately need several very expensive instruments for simultaneous determination of different metals[l]. It is felt that voltammetric methods can be a valid and effective option in the multicomponent analysis of metals, since a single potential scan in an appropriate supporting electrolyte can give a qualitative and quantitative analysis with good selectivity, employing a simple inexpensive instrumentation. In particular the high sensitivity of the voltammetric method[2, 33 can be combined with the considerable selectivity, especially if the second current technique is harmonic alternating employed[4-61. The sensitivity and selectivity have further been improved by the introduction of new types of electrode[7-91 and new quantitative methods based after multiple standard extrapolation Gditions[lO, 111. In previous communications[9121, sensitive and selective voltammetric methods were reported for the simultaneous determination of several elements in real matrices. The present work is the continuation of this line of research. Here, the simultaneous determination of seven elements, Cu, * Author to whom correspondence should be addressed. t Work partially presented at the 46th ISE Meeting, Xiamen, China, August 27-September 1,1995.

Fe, Cr, Sn, Ti, MO and Mn, in the alloy matrices stainless steel (AISI 321) NBS-SRM 121dl and highly alloyed steel Eurostandard 281-la, is presented. The above mentioned elements, in matrices of this type, frequently interfere with one another[13]. The method proposed reduces the interferences without compromising the precision, accuracy and speed. These results have been obtained by combining an appropriate supporting electrolyte (0.1 M ammonium citrate pH 6.1 and 8.5) with the standard addition method, applied to the case of very high element concentration ratios. EXPERIMENTAL

*pparatus Voltammetric measurements were carried out with an AMEL (Milan, Italy) Model 471 Multipolarograph in conjunction with an AMEL Model 430 polarographic stand. The working electrode was the AMEL polarographic stand operated as a hanging drop mercury electrode, while an Ag/AgCl, KC1 sat. electrode and a platinum wire were used as the reference and the auxiliary electrode, respectively. The voltammetric cell was kept at 25.0 f 0S”C. The solutions were deaerated with pure nitrogen for 15min prior to the measurements, while a nitrogen blanket was maintained above the solution during the analysis. Standard additions were made with Gilson micropipettes with disposable plastic tips. The solutions were deaerated 2 min. after each standard addition.

2011

for

C. LOCATELLI and G. Tow

2012

Table 1. Experimental peak potentials (-E, V/Ag, AgCI, KCI sat., *O.OOS).Experimental conditions: see Table 2 Voltammetric technique pH = 6.1 DPV

ACV

DPV

ACV

cu Fe Cr Sn Ti MO Mn

0.165 0.225 0.375 0.605 0.905 1.010 1.615

0.170 0.225 0.370 0.610 0.915 1.005 1.615

0.340 0.475 0.190 0.755 1.265 1.630

0.345 0.485 0.185 0.750 1.255 1.620

Reagents and reference solutions

All solutions were prepared with deionized water (Millipore, Milli Q), and all reagents were suprapure grade. Aqueous stock solutions of copper, iron, chromium, tin, titanium, molybdenum and manganese were prepared by dilution of the respective standard 1000mg/dm3 solutions (BDH, England). The Teflon voltammetric cell was rinsed every day with supraTable 2. Experimental conditions for the determination of the elements by differential pulse (DPV) and fundamental harmonic alternating current voltammetry (ACV)

V

nitric acid to minimize potential

Sample preparation

pH = 8.5

Element

E, dE/dt / AE 0 5

pure concentrated contamination.

DPV

ACV

- 0.050 10 50 0.065 0.250

- 0.050 10 100 10 270 + 87 -

E, : initial potential (V/Ag, AgCl, KC1 sat.); dE/dt: potential scan rate (mV/s);f: frequency (Hz): AE: amplitude of alternating current voltage (ACV) and pulse (DPV) superposed (mV); @: demodulation phase angle (degrees); r: pulse duration (s); v: pulse repetition (s).

l

Stainless steel (AISI 321) NBS-SRM 121dl and highly alloyed steel Eurostandard 281-la were prepared according to Thomerson and Price[14]. Approximately 0.2-0.3 g of sample, accurately weighed in a platinum crucible, was dissolved by adding 1 cm3 of 37% (m/m) hydrochloric acid and, subsequently, 1 cm3 of 69% (m/m) nitric acid. After the initial reaction had subsided, 2 cm3 of 60% (m/m) perchloric acid was added, and the solution was evaporated until the sample was fully oxidized and fumes of perchloric acid appeared. The solution so obtained was kept at the same temperature for about 5min. After cooling, the soluble salts were dissolved in 25cm3 of 0.1 M ammonium citrate pH 6.1. Successively, in order to obtain the pH value equal to 8.5, concentrated NH,OH was carefully added. RESULTS

AND DISCUSSION

The metals determined are the main components of the two alloys stainless steel (AISI 321) NBS-SRM 121dl and highly alloyed steel Eurostandard 281-la, that is Cu, Fe, Cr, Sn, Ti, MO and Mn. In the same matrices Co and Ni are also present. The analytical procedure for their determinations has been reported in a previous paper[15]. It must be considered also the fact that Co and Ni do not give reversible electrodic process in 0.1 M ammonium citrate at pH 6.1 and 8.5, and, for this reason, they do not interfere in the determination of the metals above mentioned. Aqueous standard reference solutions

Before the analysis of the standard reference materials, a preliminary study was carried out for the determination of each element alone and in the presence of a large excess of interferents. The voltammetric techniques employed were differential pulse

Table 3. Analytical calibration functions of elements in 0.1 M ammonium citrate (pH 6.1) as supporting electrolyte (aqueous reference solutions) Element

Differential pulse voltammetry

Alternating current voltammetry

cu

i, = (0.02 + 0.02) + (3.85 + 0.03) x lo6 c r = 0.9991 s, = 3.4% DL = 5.19 x 10egMb

i, = (0.01 + 0.02) + (2.69 f 0.06) x lo6 c r = 0.9990 s, = 3.8% DL = 7.43 x 10-gM

Fe

is = (0.01 f 0.02) + (2.99 f 0.05) x lo6 c r = 0.9990 s, = 4.2% DL = 6.69 x lo-‘M

i, = (0.01 f 0.01) + (1.43 + 0.05) x lo6 c r = 0.9993 s, = 2.6% DL = 1.40 x lo-sM

Cr

i, = (0.02 + r = 0.9993 i, = (0.01 + r = 0.9991

i, = (0.01 f r = 0.9989 i, = (0.02 f r = 0.9990

Sn Ti MO Mn

0.02) + (1.87 + 0.06) x lo6 c s, = 2.7% DL = 1.07 x lo-* M 0.01) + (2.48 k 0.04) x lo6 c s, = 3.9% DL = 8.06 x lo-‘M

i, = (0.01 f 0.02) + (6.52 f 0.07) x lo* c r = 0.9995 s, = 2.3% DL = 3.07 x lo-sM i, = (0.01 & 0.01) + (2.89 f 0.05) x lo* c I = 0.9992 s, = 3.0% DL = 6.92 x lo-sM i, = (0.02 f 0.02) + (2.06 f 0.05) x lo* c r = 0.9989 s, = 4.8% DL = 9.71 x lo-* M

0.02) + (0.87 f 0.06) x lo6 c s, = 3.9% DL = 2.30 x lo-*M 0.02) + (1.02 f 0.07) x 10” c s, = 2.8% DL = 1.96 x lo-sM

i, = (0.02 k 0.03) + (3.06 f 0.08) x lo* c r = 0.9988 s, = 4.7% DL = 6.54 x lo-sM i, = (0.01 f r = 0.9990 i, = (0.02 + r = 0.9989

0.02) + (1.61 f 0.08) x lo* c s, = 3.6% DL = 1.24 x lo-‘M 0.02) + (1.05 * 0.09) x lo* c s, = 4.3% DL = 1.90 x lo-‘M

The errors correspond to a probability of 95%; i, = peak current (PA); c = concentration of the electroactive species (M). b Limit of detection (DL) is expressed according to IUPAC[16] and corresponds to a probability of 99%.

l

Simultaneous determination of metals at trace level in a multicomponent system

2013

Table 4. Analytical calibration functions of the elements in 0.1 M ammonium citrate (pH 8.5) as supporting electrolyte (aqueous reference solution@ Element

Differential pulse voltammetry

Alternating current voltammetry

cu

i, = (0.01 f 0.02) + (4.69 f 0.07) x 10” c r = 0.9989 s, = 4.6% DL = 4.26 x lo-‘Mb

i, = (0.02 + 0.03) + (3.01 + 0.03) x 10” c r = 0.9989 s, = 5.2% DL = 6.64 x lO+M

Fe

i, = (0.03 f 0.04) + (2.69 f 0.04) x lo6 c r = 0.9990 s, = 3.8% DL = 7.43 x 1O-9 M

i, = (0.02 f 0.03) + (1.23 f 0.07) x lo6 c

Cr

i, = (0.02 f 0.03) + (0.96 f 0.08) x lo6 c r = 0.9988 s, = 5.1% DL = 2.08 x lo- * M

i, = (0.02 rf: 0.02) + (5.63 f 0.08) x 10’ c

Sn

i, = (0.01 f 0.02) + (1.07 + 0.06) x lo6 c

r = 0.9991 s, = 4.0%

r = 0.9990 s, = 3.5% r = 0.9992 s, = 3.4%

DL = 3.55 x 1O-8 M

i, = (0.01 + 0.02) + (1.01 f 0.05) x lo6 c r = 0.9990 s, = 3.9% DL = 1.98 x 10-8M -

DL = 1.87 x lo-‘M -

Ti

DL = 1.63 x 1O-8 M

MO

i, = (0.03 f 0.04) + (4.32 f 0.08) x lo5 c r = 0.9994 s, = 2.1% DL = 4.63 x lo-‘M

i, = (0.02 & 0.03) + (2.77 f 0.06) x lo5 c r = 0.9987 s, = 5.3% DL = 7.22 x lo-‘M

Mn

i, = (0.02 f 0.02) + (3.77 * 0.06) x lo5 c r = 0.9991 s, = 3.2% DL = 5.31 x 10-8M

i, = (0.03 + 0.04) + (2.18 f 0.06) x 10’ c r = 0.9989 s, = 5.1% DL = 9.17 x 10e8M

’ Se-efootnote in Table 3. b See footnote in Table 3. voltammetry (DPV) and fundamental harmonic alternating current voltammetry (ACV). Such techniques can be independently used. The experimental peak potentials and analysis conditions are reported in Tables 1 and 2, respectively. In Tables 3 and 4 the analytical calibration functions obtained with the differential pulse and fundamental harmonic alternating current voltammetry in aqueous solutions of 0.1 M ammonium citrate at pH 6.1 and 8.5 are shown. In Table 4 the analytical calibration function relevant to Ti is not reported since the electrodic process of this element at pH 8.5 is not reversible. The correlation coefficients were found satisfactory in all cases, while the precision of the techniques, expressed as relative standard deviation (s,

I

I

1

I

1

I

I

3

I

smaller than 5%) was considered good. The corresponding detection limit at the confidence level of 99% for each analytical regression function is also reported. The analytical sensitivities show that, in the experimental conditions reported above, the differential pulse voltammetry is slightly more sensitive than the fundamental harmonic alternating current technique. The simultaneous determination of all the elements was studied in a wide range of concentration ratios in order to establish the concentration ratio intervals within which no interferences were verified. As expected, in the case of a large excess of one component, an increasing overlapping of signals was observed in the measurement of the element present at lower concentration.

I

5

I

I

1

b

I

CC”(MxlOS)

Fig. 1. Differential pulse voltammetric determination of copper in 0.1 M ammonium citrate pH 6.1 by standard addition method in the presence of iron excess. Concentrations: cFc = 2.28 x 10w3M; cc” = 1.82 x IO-‘M; cFe : c,-”= 125.3. Experimental conditions are given in Table 2.

C. LOCATELLI and G. TORSI

2014

Table 5. Half-peak width (W,,,) of the elements in alternating current voltammetry Element CU Fe Cr Sn Ti MO Mn

pH 6.1

pH 8.5

47 51 63 52 82 69 53

56 51 47 59 70 55

To evaluate the interference degree, the peak current of each element was measured in a wide range of concentration ratios. The data were compared with those determined by using the analytical calibration function of the individual element (see Tables 3 and 4) and the relative errors were calculated for the species present at the lower concentration. The determination of the metals is particularly critical in the case of Cu, Fe, Cr, Ti and MO at pH 6.1 and of Cu, Fe, Cr and Sn at pH 8.5. The concen-

tration ratios, within which a maximum experimental error of 5% is present are: 51 : 1 > k,, : cFc> 1141; 371 : 1 > cc, : CFc> 1 : 359; 116:l > cTi : cyo > 1 : 90 when 0.1 M ammonium citrate pH 6.1 was employed and 160: 1 > ccr : ecu > 1 : 137; 169 : 1 > cc” : cFe> 1 : 143; 403 : 1 > Csn: CFe> 1 : 389 when the pH was 8.5. The above mentioned concentration ratios have also been confirmed by mono- and bivariate analysis of the data. However, also in the case of strong interference between two neighbouring elements, that is outside the above reported concentration ratio intervals, it is possible the determination of the single metal employing the standard addition method. This method consists in suitable additions of the element present at low concentrations. The analysis of the i, vs concentration curve shows a non-linear behaviour, but linearity is attained when the element concentration ratio is again inside the notinterference concentration interval (see interval concentration ratios above reported). An example is shown in Fig. 1. Extrapolation of the linear section of the i, vs concentration analytical

Table 6. Analytical calibration functions of the elements in stainless steel (AISI 321) NBS-SRM 121dl (a) pH 6.1” Differential pulse voltammetry

Alternating current voltammetry

cu *

i, = (69.69 + 0.02) + (3.84 k 0.05) x lo6 c r = 0.9990 s, = 2.3% DL = 5.21 x 10T9 Mb

i, = (48.69 f 0.05) + (2.73 f 0.04) x lo6 c r = 0.9992 s, = 3.6% DL = 7.33 x 10m9M

Fe

i, = (1.92 + 0.06) x lo4 + (3.15 + 0.02) x lo6 c r = 0.9988 s, = 4.8% DL = 6.35 x 10e9 M

i, = (7.82 f 0.08) x lo3 + (1.28 + 0.09) x lo6 c r = 0.9990 s, = 2.8% DL = 1.56 x 10-s M

Cr

iP = (3.39 + 0.07) x lo3 + (2.03 + 0.08) x lo6 c r = 0.9987 s, = 5.3% DL = 9.85 x 1O-9 M

i, = (1.15 f 0.06) x 10” + (0.69 f 0.05) x lo6 c r = 0.9990 s, = 4.3% DL = 2.90 x IO-sM

Sn *

i, = (7.51 f 0.09) + (2.58 + 0.04) x lo6 c r = 0.9991 s, = 3.8% DL = 7.75 x 10m9M

i, = (2.82 f 0.06) + (0.97 + 0.06) x lo6 c r = 0.9990 s, = 3.2% DL = 2.06 x lO_sM

Ti

i, = (23.63+ 0.12) + (6.62 + 0.06) x 10’ c 0.9988 s, = 5.2% DL = 3.02 x lo-‘M i, = (2.34 f 0.07) + (2.72 f 0.06) x 10’ c r = 0.9988 s, = 4.2% DL = 7.35 x lo-* M i, = (34.60+ 0.18) + (2.11 k 0.08) x lo5 c r = 0.9994 s, = 2.0% DL = 9.48 x lo-“M

i, = (11.42f 0.11) + (3.20 + 0.09) x lo5 c r = 0.9992 s, = 3.8% DL = 6.25 x 10eBM i, = (1.45 + 0.06) + (1.69 f 0.09) x lo5 c

Element

r =

MO Mn

r = 0.9987

s, = 5.0%

DL = 1.18 x lo-’ M

i, =

(19.52f 0.21) + (1.19 f 0.03) x lo5 c r = 0.9990 s, = 3.6% DL = 1.68 x lo-‘M

(b) pH 8.5’ Element

Differential pulse voltammetry

Alternating current voltammetry

cu l

i, = (88.33f 0.23) + (4.88 f 0.12) x lo6 c r = 0.9990 s, = 4.2% DL = 4.10 x 10V9Mb

i, = (52.31+ 0.11)+ (2.89 + 0.18) x lo6 c r = 0.9987 s, = 5.2% DL = 6.92 x 10w9M

Fe

i, = (1.69 + 0.23) x lo4 + (2.77 + 0.13) x lo6 c r = 0.9993 s, = 3.9% DL = 7.22 x 10e9M

i, = (6.78 f 0.16) x 10’ + (1.11 f 0.04) x lo6 c r = 0.9991 s, = 3.8% DL = 1.80 x lo-sM

Cr

i, = (1.72 + 0.03) + (1.03 f 0.04) x lo6 c r = 0.9987 s, = 4.8% DL = 1.94 x lO_sM

i, = (9.13 f 0.19) x lo* + (5.47 f 0.22) x 10’ c

Sn

i, = (2.79 k 0.15) + (0.96 + 0.17) x lo6 c

-

r = 0.9985 s, = 5.6% DL = 3.66 x lO_sM i, = (3.20 k 0.11) + (1.10 f 0.10) x 10” c r = 0.9987 s, = 5.0% DL = 1.82 x IO-‘M -

i, = (3.88 f 0.19) + (4.51 f 0.23) x 10’ c r = 0.9993 s, = 2.9% DL = 4.43 x lO_sM

i, = (2.31 f 0.14) + (2.69 f 0.18) x lo5 c r = 0.9994 s, = 2.1% DL = 7.43 x lo-sM

i, = (60.52 k 0.17) + (3.69 f 0.19) x 10s c s, = 4.3% DL = 5.42 x lo-* M

i, = (37.23 + 0.23) + (2.27 f 0.15) x lo5 c r=0.9991 s,=2.9% DL=8.81 x lo-*M

r = 0.9990 s, = 3.5% Ti MO Mn

DL = 2.08 x lo-*M

r = 0.9989

’ See footnote in Table 3. b See footnote in Table 3. l Extrapolation of the linear section (see text).

Simultaneous determination of metals at trace level in a multicompcnent system calibration curve permitted the evaluation of the metal content (Cu in the present case) in the solution with acceptable accuracy. Reversibility of the electrode process

The reversibility of each electrode process has been evaluated by measuring the half peak width values Wr,* in the fundamental harmonic alternating current voltammetry. It is well-known[3] that for a totally reversible process the Wiiz value is 90/n mV, where n is the number of electrons involved in the electrodic process. Table 5 reports the II’,,, values of the studied elements in both supporting electrolytes. The reversibility influences the analytical sensitivity of the analytical calibration curves. A comparison with the Tables 3 and 4 clearly shows, as expected, that the reversibility degree is linked to the analytical sensitivity. Real samples The preparation of the real samples is reported in the experimental section, while the measurement experimental conditions are listed in Table 2. Tables 6(a), (b) and 7(a), (b) show the analytical calibration functions measured in the real matrix solutions for Table

2015

each element. The comparison among the relevant analytical sensitivities determined in aqueous reference (Tables 3 and 4) and in real matrix solutions (Tables 6a, b and 7a, b) shows that there are no matrix interferences, at least at 95% confidence interval. In the analysis of stainless steel (AISI 321) NBS-SRM 121dl alloy (Tables 6a, b), it is possible to determine simultaneously all the elements either at pH 6.1 or 8.5, using the respective analytical calibration functions, directly or by the standard addition method in the case of very strong interferences. In the Tables 6a, b, the analytical calibration functions of Cu and Sn (0.1 M ammonium citrate pH 6.1) and of Cu (0.1 M ammonium citrate pH 8.5), for both differential pulse and fundamental harmonic alternating current voltammetry were calculated by extrapolation of the respective linear section. In fact at pH 6.1 Fe and Cr interfere strongly in the determination of Cu and Sn, respectively (A,??,, FeCu = 60mV and AE, Cr-Sn = 230mV with molar concentration ratios crc : ecu = 337.6 and cc, : cs,,= 573.9). At pH 8.5 both Fe and Cr interfere with Cu (AE, Fe-Cu = 135mV and AE, Cr-Cu = 150mV

7. Analytical calibration functions of the elements in highly alloyed steel Eurostandard 281-la

(a) pH 6.1’ Differential pulse voltammetry

Alternating current voltammetry

N.D.

N.D.

r = 0.9989

iP = (2.02 f 0.08) x 104 + (3.21 f 0.12) x lo6 c s, = 4.9% DL = 6.23 x 10m9Mb

i, = (1.03 f 0.11) x lo4 + (1.63 f 0.18) x lo6 c r = 0.9991 s, = 3.4% DL = 1.23 x 1Oes M

Cr

i, = (3.48 f 0.23) x 10” + (1.99 f 0.15) x 10” c r = 0.9994 s, = 1.8% DL = 1.01 x lo-*M

i, = (1.86 f 0.35) x 10” + (1.06 f 0.12) x lo6 c r = 0.9992 s, = 3.6% DL = 1.89 x 10e9 M

Sn (*)

i, = (0.81 f 0.08) + (2.15 f 0.31) x lo6 c r = 0.9995 s, = 1.7% DL = 9.30 x 10-9M

i, = (0.34 f 0.09) + (0.91 f 0.07) x lo6 c r=0.9994 s,=1.7% DL=2.20x 10e8M

Ti

i, = (14.28 f 0.37) + (6.32 f 0.12) x 10’ c r = 0.9987 s, = 5.3% DL = 3.16 x 10e8M

i, = (7.41 f 0.08) + (3.28 f 0.11) x 10’ c r = 0.9989 s, = 4.7% DL = 6.10 x lo-* M

Mo (*)

i, = (3.11 f 0.05) x lo-* + (2.99 f 0.28) x lo5 c r = 0.9992 s, = 3.1% DL = 6.69 x 10-8M

i, = (2.01 f 0.09) x lo-’ + (1.93 +_ 0.16) x lo5 c r = 0.9993 s, = 3.0% DL = 1.04 x lo-‘M

Mn

i, = (13.37 f 0.21) + (1.87 f 0.13) x 10’ c r = 0.9985 s, = 5.3% DL = 1.07 x lo-‘M

i, = (8.58 * 0.12) + (1.20 + 0.12) x 10’ c r=0.9992 s,=3.4% DL=1.67x lo-‘M

Element

cu Fe

(b) pH 8.5’ Element

Differential pulse voltammetry

Alternating current voltammetry

r = 0.9990

i, = (0.38 f 0.06) + (4.80 f 0.23) x lo6 c s, = 5.1% DL = 4.17 x 10e9Mb

i, = (0.25 f 0.06) + (3.20 f 0.07) x lo6 c r = 0.9987 s, = 6.3% DL = 6.25 x 10m9M

Fe

i, = (1.50 f 0.38) x 104 + (2.39 f 0.18) x lo6 c r = 0.9988 s, = 3.9% DL = 8.37 x 10e9M

r = 0.9985

Cr

i, = (1.42 f 0.16) x lo3 + (0.81 f 0.09) x lo6 c r = 0.9992 s, = 4.1% DL = 2.47 x lo-‘M

i, = (9.47 f 0.18) x 10’ + (5.41 + 0.10) x 10s c r = 0.9989 s = 5.0% DL = 3.70 x lo-* Iid

i,, = (0.46 f 0.06) + (1.21 f 0.11) x lo6 c s, = 3.7% DL = 1.65 x IO-‘M -

iP= (0.44 f 0103) + (1.15 + 0.06) x lo6 c r = 0.9989 s, = 4.4% DL = 1.74 x lo-‘M -

MO

i, = (4.79 f 0.05) x lo-’ + (4.61 f 0.23) x 10’ c r = 0.9986 s, = 5.3% DL = 4.34 x lo-*M

i, = (3.12 f 0.04) x 1O-2 + (3.00 f 0.31) x lo5 c r = 0.9989 s, = 5.2% DL = 6.67 x lo-* M

Mn

i, = (28.67 f 0.26) + (4.01 f 0.28) x lo5 c r = 0.9990 s, = 4.6% DL = 4.99 x lOTaM

i, = (16.37 f 0.14) + (2.29 * 0.18) x 10’ c

cu *

Sn Ti

r = 0.9991

’ Se-efootnote in Table 3. b See footnote in Table 3. * Extrapolation of the linear section (see text).

i, = (8.87 f 0.35) x lo3 + (1.41 f 0.15) x lo6 c s, = 5.8% DL = 1.42 x lo-* M

r = 0.9993

s, = 3.0%

DL = 8.73 x 10e8M

C.

2016

LLICATELLIand

with molar concentration ratios cre : c,-- = 337.6 and ccI : cc- = 92.3). In these cases, as above reported, the standard addition method is necessary. Therefore, in the case of stainless steel (AISI 321) NBS-SRM 121d1, the change of the 0.1 M ammonium citrate pH from 6.1 to 8.5 by appropriately adding concentrated NH40H is not indispensable. In the determination of the same elements in highly alloyed steel Eurostandard 281-la, such pH change is absolutely necessary, owing to the high interferences. In fact Cu can be determined only at pH 8.5. For this reason in Table 7(a) (pH 6.1) the Cu analytical calibration function is not reported, since the interferences from Fe and Cr are very strong (AEp Fe-Cu = 60mV, with a molar concentration ratio crc : ecu = 79 924, and AE, Cr-Cu = 210mV, with a molar concentration ratio ccI : c,-” = 22 236). Always in 0.1 M ammonium citrate pH 6.1, Ti and Cr interfere, respectively, in the determination of MO (AEp Ti-Mo = 105mV with molar concentration ratio cn : cMo= 217.3) and of Sn (AE, CrSn = 230mV, with molar concentration ratio Ccr : CS”= 4617). MO and Sn are so evaluated by the

G. Tonal

standard addition method, extrapolating the linear section of the respective analytical calibration function. At pH 8.5 (Table 7b), it is possible, by the standard addition method, the determination of Cu in a very suitable way (precision, expressed as relative standard deviation equal to 5.1% and 6.3% in DPV and ACV, respectively). The analytical results, expressed as w/w per cent, at the confidence level of 95% are in agreement with the standard reference material certified values, and are reported in Table 8. As it appears from the same Table 8, the values of precision and accuracy, expressed as relative standard deviation (s,%) and relative error (e%), are in all cases less than 5%, so confirming the validity of the proposed analytical procedure. CONCLUSION From the data above reported it can be concluded that it is possible, by choosing the supporting electrolyte, and using the standard addition method, to carry out a multielement voltammetric analysis with

Table 8. Analytical results Stainless steel (AISI 321) NBS-SRM 121dl Element

Certified concentration (%)

Determined concentration (%)

Error (e), (%)

Rel. std. dev. (s,), (%)

Voltammetric technique

cu

0.230

0.219 k 0.010 0.223 k 0.009

-4.8 - 3.0

3.7 3.1

Fe

68.263

66.122 k 2.542 69.690 f 3.015

-3.1 +2.1

Cr

17.40

16.43 k 0.99 16.62 f 0.95

-5.6 -4.5

2.9 3.8 4.6 3.4

Sn

0.069

0.071 f 0.004 0.072 f 0.003

+ 2.9 +4.3

2.6 3.6

DPV ACV DPV ACV DPV ACV DPV ACV

Ti

0.342

0.360 * 0.019 0.353 f 0.016

+ 5.3 +3.2

4.6 3.7

DPV ACV

MO

0.165

0.169 k 0.007 0.155 + 0.011

+ 2.4 -6.1

3.0 5.1

DPV ACV

Mn

1.80

1.70 * 0.12 1.82 f 0.08

-5.6 +1.1

4.8 2.0

DPV ACV

Highly alloyed steel Eurostandard 281-la Element

cu

Certified concentration (%) 0.0010

Determined concentration (%)

Error (e), (%)

Rel. std. dev. (s,), (%)

Voltammetric technique

0.00095 f 0.000 19 0.00095 f 0.000 12

-5.0 -5.0

4.8 4.5

DPV ACV DPV ACV DPV ACV

Fe

70.307

74.006 f 3.609 69.053 f 2.869

+5.3 -1.8

5.1 2.3

Cr

18.17

19.03 f 0.94 18.91 f 0.78

+4.7 +4.1

3.1 3.9

-5.6 +3.3 +4.2 +1.4 -5.0 - 5.0 -3.3 +4.6

3.9 2.1

Sn

0.0090

Ti

0.216

MO

0.0020

Mn

0.786

0.0085 f 0.0093 + 0.225 + 0.219 + 0.0019 f 0.0019 f 0.760 f 0.822 k

0.0007 0.0005 0.011 0.007 0.0003 0.0005 0.036 0.047

3.0 1.7 3.6 4.4 3.0 4.1

DPV ACV DPV ACV DPV ACV DPV ACV

Simultaneous determination of metals at trace level in a multicomponent system good precision and accuracy. The Authors think that the proposed analytical procedure is valid for all the metal alloy matrices and in particular the stainless steel, even if the choice of the pH value is evidently conditioned by the concentration ratios of the metals present in the same real matrix. Acknowledgements-This work was supported by a grant from Ministry of University and Scientific and Technological Research (Grant 40%).

REFERENCES 1. G. P. Thomas, L. C. Goldstone, Inc. Lab. 24, 13 (1994). 2. J. Wang, Stripping Analysis-Principles, Instrumentation and Applications; Chapters 4 and 5. VCH Publishers Inc, Deerfield Beach, FL (1985). 3. A. M. Bond, Modern Polarographic Methods in Analytical Chemistry, M. Dekker, ed., New York (1980). 4. A. Bobrowski. A. M. Bond. Croatica Chemica Acta 66. 499 (1993).

2017

5. A. M. Bond, R. J. O’Halloran, I. Ruzic, D. E. Smith, Anal. Chem. 48,8 12 (1976). 6. C. Locatelli, F. Fagioli, C. Bighi, T. Garai, Talanta 34, 529 (1987). 7. J. K. Christensen, L. Kryger, N. Pind, Anal. Chim. Acta 141, 131 (1982). 8. E. Wang, W. Sun, Y. Yang, Anal. Chem. 56, 1903

(1984). 9. C. Locatelli, F. Fagioli, T. Garai, C. Bighi, Anal. Chem. 60,2402 (1988). 10. C. Locatelli, F. Fagioli, T. Garai, C. Bighi, R. Vecchietti. Anul. Chim. Acta 204. 189 (1988). 11. F. Fagioli, C. Locatelli, S. Landi, k. Vdcchietti, Electroanalysis 1,449 (1989). 12. C. Locatelli, E. Vasca, C. Bighi, F. Fagioli, T. Garai, Electroanalysis accepted for publication (1995). 13. J. Alary, J. Esclassan, J. Vandaele, Analyst 111, 593 (1986). 14. D. R. Thomerson, W. J. Price, Analyst %, 825 (1971). 15. C. Locatelli, F. Fagioli, T. Garai, Anal. Chem. 63, 1409 (1991). 16. Spectrochim. Acta, Part. B 33B, 219 (1978).