a p p l i c at i o n N o t e Inductively Coupled Plasma – Optical Emission Spectroscopy Authors Laura Thompson Zoe Grosser, Ph.D. Paul Krampitz PerkinElmer, Inc. Shelton, CT 06484 USA Increased Sample Throughput for ICP-OES on Drinking Water and Waste Water Abstract The application of an SC-FAST sample introduction system to the analysis of natural and certified water samples is described. The SC-FAST system consists of an autosampler, sample loop, switching valve, high efficiency nebulizer and a glass cyclonic spray chamber to perform analysis by direct nebulization. The potential benefits of this introduction system are numerous and include: increased throughput, reduced memory effects, increased stability, lower reagent consumption and reduced instrument maintenance. Results indicate that sample throughput can be nearly tripled while still meeting the requirements. Sample-to-sample analysis is accomplished in 77 s with significantly improved washout compared to ICP-OES analysis by conventional introduction. Introduction Experimental The analysis of drinking water and wastewater for trace metal contamination is an important step in ensuring human and environmental health. More productive analyses make better use of public dollars and provide laboratories with a better cost of ownership for instrumentation. One way to improve productivity for metals analysis is by using a more sophisticated and automated sample introduction system to maximize the time spent on measurements and minimize the time spent on wash-in and wash-out of the sample. This work describes the coupling of the Optima™ 7300 DV with the ESI SC-FAST sample introduction system. Instrument The SC-FAST sample introduction system consists of an autosampler, sample loop, vacuum pump, 6-port switching valve, a sea spray nebulizer and utilizes flow injection to perform analysis by direct nebulization. The FAST system provides a number of advantages over conventional ICP-OES introduction systems, the most significant of which is higher sample throughput and reduced memory effects. The sample loop is in close proximity to the nebulizer which reduces the time required for sample uptake. Furthermore, the sample loop prevents samples from contacting the peristaltic pump tubing, which greatly improves sample washout. In addition to higher throughput and reduced memory effects, the FAST system allows for the online addition of internal standards which simplifies sample preparation and helps minimize errors and contamination. An Optima 7300 DV (PerkinElmer®, Shelton, CT) was used for the analysis of all samples described in this work. An SC-FAST (Elemental Scientific Inc., Omaha, NE) was coupled to the ICP-OES for sample introduction. The FAST system is controlled through the Optima WinLab32™ software and is shown schematically in Figure 1. The wavelengths monitored and the viewing modes used for the analytes are listed in Table 2. Instrument conditions for the Optima ICP-OES and FAST, as well as experimental parameters used throughout this work, are listed in Table 3. Standards All solutions were prepared using ASTM Type I (>18 MΩ-cm) water and double-distilled nitric acid. All acid concentrations reported in this document are described as a relative (v/v) percentage. Reference materials for this work were obtained from High Purity Standards (Charleston, SC) and from NIST®, (Gaithersburg, MD). Summary of Method The procedure followed in this work is summarized in Table 1. Table 1. Summary of Method. Establish Initial Performance Data 1. Linear Range 2. Perform IDLs and MDLs 3. Analyze Quality Control Samples with acceptable performance Daily Analysis 1. Light plasma, allow 15 minutes for warm-up 2. Record Instrument Sensitivity a. Use 1 ppm Mn for axial view, 10 ppm Mn for radial view b. Record the counts for each and watch for significant changes as compared to signals obtained in previous days/weeks 3. Calibrate using blank and standards 4. Examine data and adjust background and measured wavelengths as needed 5. Screen new samples for relative levels and natural presence of internal std elements 6. Run instrument performance QCS 7. Run analytical QCS 8. Run samples 9. Review results of quality control samples for PASS/FAIL criteria 2 Figure 1. Schematic of a FAST introduction system. Table 3. FAST-Optima 7300 DV Instrumental Conditions and Experimental Parameters. Table 2. Wavelengths Monitored and Viewing Modes. Analyte Symbol Wavelength Monitored (nm) View Optima 7300 DV Parameters RF Power 1450 watts Axial Plasma Gas Flow 15 L min-1 188.979 Axial Auxiliary Gas Flow 0.2 L min-1 Ba 233.527 Axial Nebulizer Gas Flow 0.6 L min-1 Beryllium Be 313.042 Radial Peristaltic Pump Speed 0.85 mL min-1 Boron B 249.677 Radial Nebulizer/Spray Chamber Sea Spray Cadmium Cd 226.502 Axial Torch Cassette Position -3 Calcium Ca 315.887 Radial Purge Normal Cerium Ce 413.765 Radial Chromium Cr 267.716 Axial Resolution Normal Cobalt Co 228.616 Axial Integration Time 2 s min/5 s max Copper Cu 327.393 Axial Read Delay 14 s Iron Fe 238.204 Radial Wash Time 1s Lead Pb 220.353 Axial Number of Replicates 3 Lithium Li 670.784 Radial Magnesium Mg 285.213 Radial Manganese Mn 257.610 Axial Mercury Hg 194.168 Axial Molybdenum Mo 202.035 Nickel Ni Phosphorus P Aluminum Al 308.215 Radial Antimony Sb 206.836 Arsenic As Barium FAST Parameters Sample Loop Volume 2 mL Sample Loop Fill Rate 27 mL min-1 Carrier Pump Tubing Black/Black (0.76 mm i.d.) Axial Sample Load Time 7s 231.604 Axial Rinse 5s 213.617 Axial Analysis Time (total) 77 s (sample-to-sample) Experimental Parameters Potassium K 766.490 Radial Selenium Se 196.026 Axial Carrier Solution 1% HNO3 Silicon Si 251.611 Axial Rinse Solution 1% HNO3 Silver Ag 328.068 Axial Acidity of Stds/Samples 1% HNO3 Sodium Na 589.592 Radial Strontium Sr 407.771 Radial Thallium Tl 190.801 Axial Tin Sn 189.927 Axial Titanium Ti 334.940 Axial Vanadium V 292.402 Axial Zinc Zn 206.200 Axial Internal Standards Yttrium Y 371.029 Radial/Axial Tellurium Te 214.281 Radial/Axial Ce 413.764 Radial Interference Check Cerium Sensitivity Check Solutions containing 1 ppm Mn and 10 ppm Mn were analyzed periodically to monitor the sensitivity of the instrument. The Mn solutions were analyzed weekly and after the initial installation of the introduction system. The 1 ppm and 10 ppm Mn solutions were prepared by diluting 50 μL and 500 μL into 50 mL of 1% HNO3, respectively. Internal Standards All solutions were spiked with 1.5 ppm Y and 2.5 ppm of Te. The spiking solution was made from single element stock solutions. Calibration Since this method can cover drinking water, waste water and soils, each element was calibrated to levels typically encountered in those samples. The concentrations used in the calibration standards are listed in Table 4. Each standard contained all elements listed in Table 4. 3 Table 4. Calibration Standard Concentrations. Analytes Standard Concentration μg L-1 Al, As, Ag, B, Ba, Be, Ca, Cd, Ce, Co, Cr, Cu, Fe, Hg, K, Li, Mg, Mn, Mo, Na, Ni, P, Pb, Sb, Si, Se, Sn, Sr, Ti, Tl, V, Zn 1000 The linear range results should be viewed with the understanding that a combination of elements in the presence of a complicated matrix can cause interference effects and reduce the linear range for a number of elements. For results that more accurately reflect an individual experiment, the linear range should be established using standards in a matrix that replicates the sample matrix as closely as possible. Monitored Wavelengths Table 5. Optima 7300 DV IDLs, MDLs and Linear Ranges. As mentioned earlier, the wavelengths monitored, along with the viewing mode used for each analyte are listed in Table 2. All analyses were performed using the instrument’s auto-integration feature with a minimum of 2 seconds and a maximum of 5 seconds. IDL MDL Analyte Wavelength (ppb) (ppb) MDL Spike Linear Level Range Ag 328.068 0.5 1.1 5 100 Al 308.215 1.5 6.6 20 2000 Initial Performance Demonstration As 188.979 1.8 1.2 20 100 IDLs B 249.677 2.0 (0.4) 1.9 (0.2) 5 2000 Ba 233.527 0.2 0.2 2 25 Be 313.107 0.2 0.5 2 50 Ca 317.933 1.3 (0.3) 1.0 (0.1) 5 900 Cd 226.502 0.4 0.4 2 100 Ce 413.764 7.6 (0.9) 12.8 (1.4) 20 100 Co 228.616 0.3 0.5 2 250 MDLs Cr 267.716 0.4 0.6 2 50 Method detection limits (MDLs) were based upon seven replicate measurements of a series of spiked calibration blanks. Each blank solution was spiked with analytes at concentrations between 2 and 10 times the calculated IDLs. The MDL was calculated by multiplying the standard deviation of the seven replicate measurements by the appropriate Student’s t test value according to: Cu 327.393 0.4 0.3 2 300 Fe 259.939 1.0 (0.5) 0.7 (0.2) 5 400 K 766.490 41.1 (2.1) 28.1 (3.6) 100 2000 Mg 279.077 2.5 (0.4) 4.6 (1.0) 20 700 Mn 257.610 0.1 0.5 2 40 Mo 202.031 0.4 0.5 5 125 Na 589.592 7.6 (0.2) 4.7 (0.5) 20 900 Ni 231.604 0.7 0.5 2 125 Pb 220.353 0.6 1.1 20 100 P 213.617 2.9 5.6 20 3000 Li 670.784 0.5 (0.01) 0.5 (0.1) 5 200 Hg 253.652 1.6 9.6 20 100 Sb 206.836 2.6 2.1 20 100 Se 196.026 1.1 1.8 20 100 Si 251.611 2.6 (1.3) 13.3 (4.7) 5 2500 Sr 421.552 1.6 (0.1) 1.6 (0.5) 2 50 Sn 189.927 5.6 (0.5) 6.7 (1.9) 20 2000 Ti 334.940 0.1 0.5 2 50 Tl 190.801 1.1 1.8 20 100 V 292.402 0.2 0.6 2 50 Zn 206.200 0.2 0.4 2 100 Instrument detection limits (IDLs) were estimated using multiple replicate measurements of the calibration blank (1% nitric acid). The IDL was calculated to be the concentration equal to three times the standard deviation of those replicate measurements; results are shown in Table 5. The IDL calculation was followed according to the procedure outlined. MDL = (S) x (t) Note that the Student's t-value is based on a 99% confidence level. Both the Student’s t-value and the standard deviation are based on n-1 degrees of freedom (t = 3.14 for six degrees of freedom). Linear Range A linear calibration range was established for each element listed. No special detector optimization was done prior to conducting this procedure. The linear dynamic range for each analyte was calculated to be the highest concentration for which the recovery was within ±10% of the true (i.e., known) value of the standard. The results from this study are based upon multi-element standards in a 1% nitric acid matrix and are given in Table 5. ( ) = axial 4 Table 6. Precision and Recovery Data for High Purity “Trace Metals in Drinking Water” (CRM). Analyte Avg Meas. Conc. μg L-1 Std. Dev. μg L-1 % RSD μg L-1 Certified Value μg L-1 Recovery of Certified Value (%) Spike Level μg L-1 Avg. Spike Recovery (%) Std. Dev. of Spike Rec % RSD Ag --- --- --- 2 --- 100 99.4 0.8 0.6 As 69.9 0.7 1.0 80 87.4 100 102 3.0 2.9 B --- --- --- NA --- 100 99.7 1.1 1.1 Be 19.8 0.1 0.5 20 99.0 100 102 0.3 0.3 Ca 33400 18 0.1 35000 95.4 100 --- --- --- Cd 9.66 0.2 1.7 10 96.6 100 101 1.0 1.0 Co 23.3 0.2 0.8 25 93.2 100 100 0.2 0.2 Cu 19.8 0.7 3.4 20 99.0 100 102 1.0 1.0 Cr 19.4 0.3 1.5 20 97.0 100 102 0.7 0.7 Fe 93.2 1.9 2.0 100 93.2 100 102 5.0 4.9 K 2380 31 1.3 2500 95.2 100 --- --- --- Li 19.3 0.8 4.3 20 96.5 100 102 1.4 1.4 Mg 8530 125 1.5 9000 94.8 100 --- --- --- Mn 38.6 0.5 1.3 40 96.5 100 103 1.2 1.1 Mo 101 0.7 0.6 100 101 100 101 1.6 1.6 Na 5580 6.9 0.1 6000 93.0 100 --- --- --- Ni 57.8 0.5 0.9 60 96.3 100 98.6 2.5 2.6 Pb 39.7 3.1 7.7 40 99.2 100 103 2.9 2.8 Sb --- --- --- 10 --- 100 107 2.8 2.7 Se --- --- --- 10 --- 100 110 6.5 5.9 Sn --- --- --- NA --- 100 100 2.2 2.2 Sr 246 0.4 0.2 250 98.4 100 94.3 3.0 3.2 Ti --- --- --- NA --- 100 105 0.4 0.4 Tl --- --- --- 10 --- 100 99.5 4.7 4.7 V 29.8 0.4 1.4 30 99.3 100 103 0.5 0.5 Zn 69.5 0.6 0.9 70 99.3 100 98.8 1.6 1.7 NA = Not Applicable Data Handling All data from the Optima 7300 DV was collected using a desktop computer attached to the instrument. The analytical results presented here were computed using the WinLab32 software and exported as report files. The text and supporting data tables were generated using Microsoft® Word and Excel®. Sample Analysis/Quality Control The accuracy and precision of the above-described method were verified using certified reference materials and a local drinking water sample. Certified reference materials were analyzed without modification to determine the accuracy as compared to the certified values. Recoveries of spiked reference materials and the local drinking water sample were calculated. Results from High Purity Standards “Trace Metals in Drinking Water”, NIST SRM 1643e “Trace Metals in Water” and a local drinking water sample are listed in Tables 6, 7 and 8, respectively. The results for the analysis of High Purity Standards interference check standards “INFCS I + INFCS IV” are listed in Table 9. Note that INFCS I was diluted 10,000 fold and INFCS IV was diluted 1,000 fold before the standards were combined into one sample vial. 5 Table 7. Precision and Recovery Data for NIST SRM 1643e “Trace Elements in Water”. Avg. Meas. Certified Analyte Conc. μg L-1 Std. Dev. μg L-1 % RSD μg L-1 Value μg L-1 Rec. of Certified Spike Avg. Spike Std. Dev. Value (%) Level μg L-1 Recovery (%) of Spike Rec % RSD Ag --- --- --- 1.062 --- 100 90.3 1.7 1.9 Al 133 0.3 0.3 141.8 94.1 100 94.0 1.1 1.2 As 60.2 1.9 3.7 60.45 99.6 100 109 4.3 4.0 B 150 0.3 0.2 157.9 95.0 100 98.5 2.4 2.4 Ba 504 11 2.1 544.2 92.6 100 --- --- --- Be 13.6 0.1 0.5 13.98 97.3 100 104 0.5 0.4 Ca 29600 95 0.3 32300 91.6 100 --- --- --- Cd 6.00 0.2 4.1 6.568 91.4 100 102 1.6 1.6 Co 25.2 0.2 0.6 27.06 93.1 100 103 0.9 0.9 Cr 20.4 0.2 0.8 20.40 100 100 105 1.0 0.9 Cu 23.6 0.3 1.2 22.76 101 100 108 0.2 0.1 Fe 97.6 7.0 7.2 98.1 99.5 100 99.7 4.8 4.9 K 1920 46 2.4 2034 94.4 100 --- --- --- Li 18.3 0.9 5.0 17.4 105 100 104 0.8 0.8 Mg 7600 70 0.9 8037 94.5 100 --- --- --- Mn 38.7 1.6 4.0 38.97 99.4 100 102 1.1 1.1 Mo 125 0.3 0.2 121.4 103 100 105 1.4 1.3 Na 19100 100 0.5 20740 92.1 100 --- --- --- Ni 59.5 0.9 1.4 62.41 95.3 100 102 1.1 1.1 Pb 19.4 0.5 2.6 19.63 98.8 100 107 2.1 2.0 Sb 59.9 3.7 6.1 58.30 103 100 102 0.8 0.7 Se --- --- --- 11.97 --- 100 105 5.3 5.0 Sr 319 0.9 0.3 323.1 98.7 100 101 9.5 9.4 Sn --- --- --- n/a --- 100 104 0.03 0.03 Ti --- --- --- n/a --- 100 108 0.3 0.3 Tl --- --- --- 7.445 --- 100 106 6.4 6.1 V 37.0 0.2 0.5 37.86 97.7 100 107 0.6 0.6 Zn 76.6 0.7 0.9 78.5 97.6 100 101 0.7 0.6 NA = Not Applicable 6 Table 8. Results and Spike Recoveries for Local Drinking Water. Analyte Wavelength (nm) LDW Conc. μg L-1 Low Spike Level μg L-1 Low Spike Low Spike Results μg L-1 % Recovery High Spike Level μgL-1 High Spike Results μg L-1 High Spike % Recovery Al 100 100.3 500 475 95.0 308.215 7.50 100.3 As 188.979 < MDL 100 102.9 102.9 500 487 97.4 B 249.677 11.2 100 100.1 100.1 500 472 94.4 Ba 233.527 69.0 200 192.9 96.5 1000 946 94.6 Be 313.042 0.120 100 103.8 103.8 500 490 98.0 Ca 315.887 7120 100 --- --- 500 --- --- Cd 226.502 < MDL 100 98.8 98.8 500 484 96.8 Co 228.616 < MDL 100 102.1 102.1 500 486 97.2 Cr 267.716 < MDL 100 101.6 101.6 500 484 96.8 Cu 327.393 97.6 100 106 106 500 478 95.6 Fe 238.204 83.5 100 94.5 94.5 500 491 98.2 K 766.490 877.7 100 --- --- 500 --- --- Li 670.784 < MDL 100 102 102 500 492 98.4 Mg 285.213 1710 100 --- --- 500 --- --- Mn 257.610 16.4 100 102 102 500 503 101 Mo 202.035 < MDL 100 101 101 500 480 96.0 Na 589.592 12800 100 --- --- 500 --- --- Ni 231.604 0.730 100 99.4 99.4 500 482 96.4 P 213.617 < MDL 100 99.0 99.0 500 476 95.2 Pb 220.353 < MDL 100 99.7 99.7 500 487 97.4 Sb 206.836 < MDL 100 102 102 500 492 98.4 Se 196.026 < MDL 100 105 105 500 525 105 Sn 189.927 2.94 100 97.4 97.4 500 475 95.0 Sr 407.771 52.4 100 101 101 500 488 97.6 Ti 334.940 < MDL 100 103 103 500 503 101 Tl 190.801 < MDL 100 104 104 500 475 95.0 V 292.402 0.220 100 103 103 500 487 97.4 Zn 206.200 47.2 100 100 100 500 478 95.6 7 Table 9. Results for High Purity INFCS I + INFCS IV. Analyte Wavelength (nm) INFCS Run #1 μg L-1 INFCS Run #2 μg L-1 INFCS Avg. Result μg L-1 Duplicate RPD INFCS True Value μg L-1 % Recovery As 188.979 110 111 111 0.8 100 111 Ba 233.527 30.0 28.5 29.3 5.2 30 97.7 Be 313.042 10.9 10.9 10.9 0.1 10 109 Ca 315.887 4950 4910 4930 0.8 5000 98.6 Cd 226.502 32.2 33.2 32.7 2.9 30 109 Co 228.616 31.0 31.5 31.3 1.6 30 104 Cr 267.716 31.5 31.4 31.5 0.4 30 105 Cu 327.393 31.3 31.1 31.2 0.5 30 104 Fe 238.204 5230 4910 5070 6.2 5000 101 Hg 194.168 4.97 5.54 5.30 10.8 5 105 K 766.490 2130 2150 2140 0.7 2000 107 Mg 285.213 5070 5050 5060 0.3 5000 101 Mn 257.610 20.8 19.5 20.1 6.4 20 100 Ni 231.604 30.9 31.5 31.2 2.1 30 104 Pb 220.353 105 108 106 2.9 100 106 Se 196.026 53.9 48.8 51.4 10.1 50 103 Tl 190.801 102 103 103 1.1 100 103 V 292.402 31.1 31.1 31.1 0.2 30 104 Zn 206.200 34.8 32.6 33.7 6.7 30 112 Conclusions Acknowledgement The FAST system coupled with the Optima 7300 DV has been shown to produce results while increasing sample productivity over 200% when compared to analyses with conventional introduction systems. Since the FAST system eliminates virtually all of the rinse and read delay times, most of the time is now spent running samples, therefore increasing productivity. Also, since the FAST reaches a steady state signal much more quickly than conventional sample introduction, instrument and method detection limits are improved almost 2-fold for many analytes. When used in conjunction with the SC autosampler and Optima 7300 DV, the FAST system provides a rugged, automated sample introduction system that can significantly reduce labor costs and improve laboratory productivity. We would like to thank Maura Mahar, a graduate student at University of Massachusetts and summer intern for her excellent work contributing to this application. Special thanks also to Stan Smith, PerkinElmer Product Specialist and to Matthew Knopp of ESI for their time and input. References 1.EPA Method 200.7, “Determination of Metals and Trace Metals in Water and Wastes by Inductively Coupled Plasma-Atomic Emission Spectrometry,” Revision 4.4, 1994, Environmental Monitoring Systems Laboratory, Office of Research and Development, United States Environmental Protection Agency, Cincinnati, OH 45268. PerkinElmer, Inc. 940 Winter Street Waltham, MA 02451 USA P: (800) 762-4000 or (+1) 203-925-4602 www.perkinelmer.com For a complete listing of our global offices, visit www.perkinelmer.com/ContactUs Copyright ©2009, PerkinElmer, Inc. All rights reserved. PerkinElmer® is a registered trademark of PerkinElmer, Inc. All other trademarks are the property of their respective owners. 008710_02