'/ United States Environmental Protection Agency Water Engineering Research Laboratory Cincinnati OH 45268 Research and Development EPA/600/S2-85/063-Aug. 1985 vxEPA Project Summary Microprocessor-Controlled Anodic Stripping Vo I tarn meter for Trace Metal Analysis in Tap Water R. G. Clem, F. W. Park, F. A. Kirsten, S. L Phillips, and E. P. Binnall This report discusses the construc- tion and use of a portable, micro- processor-controlled anodic stripping voltammeter for onsite, simultaneous metal analysis of copper (Cu), lead (Pb), and cadmium (Cd) in tap water. The in- strumental system consists of a pro- grammable controller, permits keying in analytical parameters such as sparge time and plating time, a rotating cell for efficient oxygen removal and amalgam formation, and a magnetic tape that can be used for data storage. Analysis time can be as short as 10 to 15 min. The stripping analysis is based on a pre- measurement step during which the metals from a water sample are con- centrated into a thin mercury film by deposition from an acetate solution of pH 4.5. The concentrated metals are then electrochemically dissolved from the film by applying a linearly increas- ing anodic potential. Typical peak- shaped curves are obtained. The heights of these curves are related to the concentration of metals in the water by calibration data. Results of tap water analysis showed 3 ± 1 (ig/L Pb, 22 ± 0.3 fig/L Cu, and less than 0.2 fig/L Cd for a Berkeley, California, tap water. For 10 samples of Seattle, Washington, tap water analyses showed 1 to 1 |j.g/L Cu and 1 to 2 |tg/L Pb. Recommenda- tions are given for a next generation instrument system. This Project Summary was devel- oped by EPA's Water Engineering Re- search Laboratory, Cincinnati, OH, to announce key findings of the research project that is fully documented in a separate report of the same title (see Project Report ordering information at back). Introduction Trace metals in public drinking water supplies are of special concern because of their potential long-term harmful ef- fects when ingested by humans. Water supply resources for municipalities con- tain various types and concentrations of dissolved metals. In untreated water, these concentrations are generally sig- nificantly lower than the EPA Interim Primary Drinking Water Standards. But chemicals used in water treatment may contain traces of metals as impurities and may introduce them to water at the treatment plant. Another potential source of metals is corrosion and disso- lution of the metallic piping of tank inte- riors through which the treated water passes or is stored before ingestion. Thus a need exists for measuring the concentration of dissolved metals at the user's tap, where the water is actually ingested. Although a number of methods are available for measuring trace metals in drinking water, anodic stripping voltam- metry (ASV) was selected for this study because it offers a number of advan- tages, including (a) simultaneous analy- ses of more than one metal, (b) sensitiv- ity to concentrations of less than 1 M-g/L, (c) portable instrumentation, (d) metal speciation at ambient concentrations, (e) easy automation, and (f) rapid analy- ses at a relatively low cost. ------- This report describes a portable, microprocessor-controlled ASV system that was constructed for the onsite anal- ysis of Cu, Pb, and Cd in tap water (Fig- ure 1). The instrumental parameters such as a sparge and plating time were keyed into the computer, and analyses were then done automatically. The ASV system was applied to the analysis of one sample of Berkeley, California, tap water and 10 samples of tap water from Seattle, Washington. The report dis- cusses calibration of the instrument with known metal concentrations and analyzes the results of the tap water analyses. Selected information is given on (1) the anodic stripping electro- chemistry of Cd, Pb, and Cu; (2) the ASV method using mercury thin films plated onto graphite electrodes; and (3) the in- strumentation electronics. Anodic Stripping Voltammetry at Mercury Film Electrodes Anodic stripping is applied to the analysis of ^g/L quantities of mercury- soluble metals such as Cd and Pb using thin films (4 to 100 jim) formed by the deposition of mercury (Hg) onto carbon electrodes. The Hg is deposited along with metals (Cd, Pb, Cu) being analyzed. The first step involves electroplating, during which the metals being deter- mined are concentrated on the order of 100-fold in 5 min within the Hg film. This deposition step is not exhaustive (non- stoichiometric); only 2% to 3% of the total amount in the water sample is de- posited. The applied potential is stepped to a sufficiently negative value so that the electro-deposition rate is lim- ited only by the rate of mass transfer of the dissolved metals from the bulk of the solution to the electrode surface. The solution is stirred by rotating the cell to enhance mass transfer. The plated metals establish a concentration gradient within the Hg film whereby the concentration of metal at the electrode- solution interface is larger than that to- ward the interior of the electrode. After a set time (such as 5 to 10 min), stirring is stopped and the solution be- comes quiet. During this quiescent pe- riod, the concentration gradient within the Hg film is essentially eliminated be- cause deposition is almost negligible, and the plated metals diffuse through the film from regions of high concentra- tions to lower concentrations. A quies- cent time of 60 sec was used here. The third step is anodic stripping of the metals from the amalgam by appli- Figure 1. ASV instrument constructed by the Lawrence Berkeley Laboratory. 2 ------- cation of a linearly increasing positive potential. At potentials near the Nerns- tian value for a particular metal, the metal begins to dissolve (strip), giving rise to typically peak-shaped curves at more positive potentials. A peak is ob- tained for each metal, and thus a charge-voltage curve can consist of more than one peak. Usually this peak height is measured and related to the metal concentration in the solution by means of calibration data. Peak areas have also been measured. Instrument System Components The ASV system constructed here was designed to automate settings for the analytical procedure, to acquire data relating peak height to metal concentra- tion in water samples, and to store data from the analysis of tap water for subse- quent calculations. The system consists mainly of the following items: a rotating cell containing the water sample, elec- trodes, and inlets for the sparge gas; a digital potentiostat for automating the procedure; and a pen recorder or mag- netic tape for data storage and calcula- tion. The functions of these four major components are described in the fol- lowing sections and in Table 1. Rotating Cell A cylindrical rotating cell machined from Lucite* was used to provide both for quick oxygen removal during the sparging step and for a high rate of metal preconcentration during the metal deposition. Only 60 sec is re- quired for effective oxygen removal, thereby reducing the time required for analysis. The solution volume used was generally 15 ml. During oxygen sparging, the cell con- taining the 15 mL of solution is rotated automatically in a clockwise direction of 0.9 sec. A counterclockwise pulse is then imparted for a duration of 0.1 sec. Under these conditions, the dissolved oxygen concentration is lowered suffi- ciently so that the background current resulting from oxygen reduction does not interfere with the analysis proce- dure. Nitrogen or other inert gas such as carbon dioxide is introduced through the bottom of the electrode holder at a rate of ~5 L/min to effect sparging. At the conclusion of the 60-sec sparging Table 1. Major Functions of the ASV System Components Component Functions Rotating cell Digital potentiostat Microcomputer controller Recorder Contains samples and electrodes, sparges, oxygen, preconcentrates trace metals, and performs anodic stripping of amalgams. Applies and maintains plating potential, applies anodic stripping ramp potential, and measures stripping peak heights. Starts and stops the following: cell rotation for sparg- ing, cell rotation for plating, sparge and plating times, anodic stripping ramp; also transmits peak height data to cassette and strip chart recorders. The strip chart records a background trace and records peak height traces. The cassette provides fast-rate data storage for later calculation. 'Mention of trade names or commercial products does not constitute endorsement or recommenda- tion for use. period, the microprocessor stops cell rotation, and the solution drains com- pletely to the floor of the cell, because a Lucite surface is not wetted by water. The the agitation (stirring) period be- gins, during which amalgam is accumu- lated in the preconcentration step. A low flow rate of nitrogen (e.g., 1 to 2 L/min) is maintained during the plat- ing step to prevent infusion of air into the cell and, thereby, to maintain a low background current. Efficient stirring is important for short analysis times such as those used here. Stirring is produced in the rotated cell during plating by controlled reversals of the direction of cell rotation every sec- ond. The graphite working electrode is placed on a radius of the cell and is po- sitioned to face the cell wall. Turning the electrode through 180° so that it faces the center of the cell results in a marked decrease in the rate of amalgam accu- mulation because of a substantially lowered stirring rate. Thus the elec- trode, once positioned and locked into place with the set screw, must not be moved until the analysis is completed. Digital Potentiostat The digital potentiostat controls the potential between the working elec- trode and the reference electrode by in- jecting and extracting pulses of charge. In this manner, the feedback current is digitized. Advantages of digitized cur- rent include the capability for direct readout on a scalar, direct computer compatibility, and the elimination of a need for analog to digital converters. The current or integral of charge is mea- sured by counting the number of charge pulses over a specified period of time with the scalar. A schematic of the digi- tal potentiostat appears in Figure 2. The circuit diagram in Figure 2 con- sists of the electrochemical cell, opera- tional amplifier, digitalizing logics, and current pumps. The arrangement shown is used to maintain a control po- tential and to measure the current change digitally during electroplating and stripping. Two pumps exist in series—one a 0.1 milliampere (ma) neg- ative constant current supply, and the other a positive, 0.6- to 1.0-ma ad- justable and controllable constant cur- rent supply. The cell (which is con- nected between the two current supplies to ground) provides an alter- nate current path. The maximum posi- tive or negative voltage drop across the cell depends on the cell impedance and the current being provided by the posi- tive and negative source. All voltages between these extremes are regulated by the feedback loop, which consists of the reference electrode, the voltage comparator, the J-K flip-flop, the cur- rent switch transistors Q3 and Q4, and diode D2. When a pulse from the cur- rent switch forward-biases D2, the neg- ative 0.1 current supply is pulling the cell voltage down. Since the positive current supply can provide an average of 0.3 to 0.5 ma, it will pull the cell volt- age up when being pulsed. The positive source actually is a 0.6- to 1.0-ma sup- ply, which is normally sunk through diode D2. Only when Q4 is turned off will this supply provide current to the cell. The PULSE ON pulses, which come from the J-K flip-flop, are also output through Q of the same J-K to a pulse counter whose counting period is program-selected. Voltage is maintained across the cell by the use of one fixed 0.1-ma supply and one controllable 1.0-ma supply. ------- both connected to the cell auxiliary elec- trode so that the sum of the currents across the cell impedance will generate the voltage called for by the systems program. The feedback loop consists of the reference electrode and an LM 310 (IC-4) unity gain buffer leading to a dif- ferential amplifier whose output goes to an LM 311 voltage comparator. When the reference electrode is lower than the voltage demanded by the program, a current switch (2 each 2N 2369) back-biases diode D3 (1N 4148) so that current from the 1-ma constant current source (IC-1 and 2N 2608) is directed to the cell auxiliary electrode. To facilitate relative current measurement, pulses are injected into the cell. A 1-MHz clock is input to a 7470 J-K flip-flop; when the J-K is enabled by a high signal from the LM 311 voltage comparator, its output will be a 500-kHz square wave. Microcomputer Controller In addition to the applied potential, control is also provided for the follow- ing functions: time for each step, se- quencing of each step, storage of the ASV data, output of the data, and a vis- ual indication of the stripping process using a strip chart recorder. This control is provided by a microcomputer based on an Intel 8080A 8-bit microprocessor central processing unit. Changes in test parameter variables are entered by the operator from a 16-button keyboard on the front panel of the instrument. A flow chart detailing the various controlled functions appears in Figure 3. The POWER ON or SYSTEM RESET buttons return all parameters to the ini- tial conditions programmed for the plat- ing and stripping parameters. When ei- ther of these buttons is pressed, preprogrammed values from PROM memories are set into the microproces- sor. These are the following input com- mands: Sparge: 60 sec Start: 1000 mV Step: 5 mV Plate: 420 sec Delay: 1 msec Count: 400 msec Day: 0 Test No.: 0 Spike A: 0 Spike B: 0 The MONITOR light will come ON. Pressing any key will show the value programmed. For example, to see "SPIKE", press SPIKE and A, or SPIKE and B. To enter a value of 25 for SPIKE B, press SPIKE, press B, press ENTER, press 2, press 5, press ENTER and the change is made. The MONITOR light will go ON. Data Handling The output signal of analytical inter- est from the ASV is the stripping peak height as a function of applied potential. For this instrument, the stripping peak heights were mainly measured from the tracing on either a strip chart recorder or an X-Y recorder. Output to the strip chart recorder is the digital-to-analog (DAC) conversion times a scale factor of the counts accumulated during each 5-mV step (or other selected step mag- nitude) during the stripping cycle. The scale factor (sensitivity) of the strip chart recording can be chosen by the operator as appropriate for a particular + 72 1 ma J Q >'- 2 K Q 500KHZ Bursts _nnn_nnn_ Aux Sirrent A/itch -12 Voltage Comparator 0.1 ma -12 Volt DAC Figure 2. Schematic for Anodic Stripping Voltammeter. 4 ------- (initialize'} Power On • Set Run No. To 1 • . * , , Aux. Relay Open . Cell Drive MTR 'Off * Ref DAC To EndMV I Set Test Parameters PS T Close >- Open Relay Day XXXX Test XXXX Ref order LSE X Spike A XXX 1 Sparge 060 Sec Start -1000MV Step 5 MV Fixed End +350 MV Program Plate 350 Sec Values Delay IMS Count 400 MS Install Cell With Conditioned Sample r (Push Test^) Freon Valve To High Flow , • Spin Cell 0.9 Sec On 0. 1 Sec Off * For 60 Sec * Freon Valve To Low Flow ® 1 Oscillate Cell » Ref DAC To Start MV » Aux Electrode Relay 'On' » Plate For Selected Time » Stop Cell Oscillation Strip Chart Recorder 'On Wait 5 Sec For Cell Stability Manually Turn On Freon , X 7 1 DAC Steps By Selected MV Reset & Start CNT r-= — 7— -i Cnunter «*„_«,.. DAC To Holds For Count rlVo*"^ ^H'MVs' Stop DAC Ramp t Open Aux Electrode Relay * Cell Drive MTR 'Off » MAG Record I// Parameters And Data » Advance to Next Run [ Wait [ Enter Spike 'ype & Amour * Push Test » Stan TNT l"""<5 rKr "AM -.. .,.—.-.. stop c/v/ ^ Sfef) 1 | '"" . 1 Recorder > Rerun Data ToS/C Recorder tP.S.) Operator adds spike of A or B solution. Type and amount depending on peak observed on t * chart recorder. Attempt is made to double peak. Repeat Of Run No. 1 Except Run 2 & 3 Plate Time is Run 1 Time - 2 _, I < End Of Run 3 Completes Tes Of A Sample 1 Advance Test Number » Wait For Next Test t Figure 3. Flow chart for Anodic Stripping Voltammeter. ------- measurement. Data are also recorded on a cartridge digital tape recorder, where they are available for further off- line processing. Each water analysis is identified by both test number and run numbers. A complete analysis must have three se- quential runs. These three runs are per- formed on: (1) the water sample; (2) the water sample with spike A added; (3) the water sample with spikes A and B added. Runs 2 and 3 establish the cali- bration of the instrument for that run. A sample of tap water from the Lawrence Berkeley Laboratory was ob- tained by running the water for 1 to 2 min then collecting 500 ml in a polyethylene bottle that had been cleaned and rinsed with the tap water. Hydrochloric acid was added to pH 1, and the solution was permitted to equi- librate for about 48 hr. The sample was then discarded, and an additional 500 mL was collected in the same manner. After 48 hr, 15 mL was added to the ro- tating cell, and ammonia-treated sodium acetate was added drop-by- drop to achieve pH 4.7. Then 25 jo-L of 0.0025 M mercury nitrate solution was pipetted in, and background anodic stripping curves were obtained. Known pi concentrations of Cu, Pb, and Cd standards were pipetted in, and strip- ping curves were again obtained. The peak heights were measured on an X - Y recorder, and the concentrations of the three metals in the tap water were determined by the method of standard additions. Any Cd in the water was be- low the limits of detectability of the pro- cedure used. Analysis by atomic ab- sorption detected no Cd either. The results obtained for Cu, Pb, and Cd were 22, 3, and <0.2 ppb, respectively. Re- sults obtained for 10 Seattle water sam- ples are shown in Table 2. Recommendations Based on the experience gained for tap water analysis with the instrument described in this report, a next- generation ASV might include the fol- lowing items: (1) A cell that incorpo- rates self-cleaning capabilities. Carbon electrodes with electro-deposited mer- cury films age with use and give both spurious peaks as well as increased .background current during the stripping step. The result is a loss in sensitivity and possible erroneous analyses be- cause of peaks shown for metals that are otherwise not present in detectable amounts. Currently, cleaning is done Table 2. Analysis of Municipal Water From Seattle, Washington Sample Designation NE 24 105NE 7914 NE 26 9850 Belfair Rd, Standing 9850 Belfair Rd, Running Taylor Creek Well 7/4 Taylor Creek Well 7/13 Jones Res. Running Jones Res. Standing Park Place Standing, 1st run Park Place Standing, 2nd run AA Values Cu Pb 1000 ppb 10 ppb 680 < 10 950 < 10 510 <10 <6 <10 <6 <10 — — — — — ASV Cu — 550 ppb 1100 800 2.6 1.4 12.4 1800 WOO 1000 Values Pb — 1.3 ppb N.D. N.D.* N.D. N.D. 2.3 N.D.* N.D.* N.D.* N.D. = Not detected. N.D.* = Non detection likely due to solid phase amalgam formation. manually by periodically rubbing the surface gently with filter paper. (2) Methods for removing large quantities of dissolved Cu. Plated Cu interferes in two ways: by forming intermetallic compounds with Cd (which reduces the anodic peak height for Cd) and by mask- ing the Pb and Cd peaks. Large concen- trations of dissolved Cu (estimated at >800 |ig/L) interfered in this way with measurement of Pb and Cd in the sam- ples of Seattle water. Cu plates more easily than either Pb or Cd, so one ap- proach to this problem might be a pre- analysis step involving deposition of Cu with a dual-working electrode system. (3) Methods for measuring the total con- centration of dissolved metals. Metals can form more than one species in natu- ral waters; some of these species are not measured by ASV. A pre-ASV step such as ozonolysis might be effective in destroying complexes between metals and electro-inactive species. (4) Pro- gramming of the ASV system so that the readout will be in ^.g/L of the metals being measured. Currently the output signal is either in analog form on a strip chart recorder or in digital form on mag- netic tape. Both must be further pro- cessed. The full report was submitted in fulfill- ment of Contract No. EPA-70-D-X0507 by the University of California under the partial sponsorship of the U.S. Environ- mental Protection Agency. ft US GOVERNMENT PRINTING OFFICE. 1985 559-111/20638 ------- R. G. Clem, F. W. Park, F. A Kirsten, S. L Phillips, and E. P. Binnall are with University of California, Berkeley, CA 94720. Marvin C. Gardels is the EPA Project Officer (see below). The complete report, entitled "Microprocessor-Controlled Anodic Stripping Voltammeter for Trace Metal Analysis in Tap Water," (Order No. DE 85-002 781; Cost: $10.00, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Water Engineering Research Laboratory U.S. Environmental Protection Agency Cincinnati, OH 45268 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 Official Business Penalty for Private Use $300 EPA/600/S2-85/063 0000329 PS U S ENVIR PROTECTION AGENCY REGION 5 LIBRARY 230 S DEARBORN STREET CHICAGO It 60*04 ------- |