Application Update 55 Determination of Cations and Amines in Hydrogen Peroxide by Ion Chromatography Using a RFIC (Reagent-Free System Introduction Hydrogen peroxide is an essential chemical in the fabrication of integrated circuit and microcircuit devices. Maximum allowable contaminate levels for semiconductor grade hydrogen peroxide can be as low as <00 ppt (ng/l per individual inorganic cation. A number of semiconductor manufacturers include specifications for maximum allowable levels of < ppb (µg/l for trimethylamine and related amines. It has been reported that an IonPac CS0 column with a 40 mm methanesulfonic acid mobile phase and suppressed conductivity detection successfully determined inorganic cations in hydrogen peroxide. In addition, simple amines and alkanolamines were determined. While these separations did not use an organic solvent modifier in the eluent, the addition of organic solvent will improve amine peak shape on this column. This application uses an IonPac CS column to determine trace cations and amines in hydrogen peroxide with a large-loop injection. The CS column separates amines without the organic solvent eluent modifier needed for separating amines when using older cation-exchange ion-chromatography (IC columns. Using the RFIC system, this application easily determines sub-µg/l concentrations of cations and amines. Equipment Dionex ICS-500 * consisting of: GP50 Gradient pump with vacuum degas option EG50 Eluent Generator with CR-CTC and EluGen EGC-MSA cartridge ED50A Electrochemical Detector with conductivity cell LC5 Chromatography Oven Chromeleon 6.6 Chromatography Workstation * This application can be performed on any Dionex RFIC system. An autosampler was not used for the work presented in this AU, but either an AS40 or AS autosampler can be used as long as the proper precautions are taken to produce a clean blank. For these precautions see the Sample Loading and the Precautions sections of Dionex Application Note 5.. Reagents and Standards Deionized water (DI, Type I reagent grade, 8 MΩ-cm resistivity. Lithium hydroxide (LiOH Sodium chloride (NaCl Ammonium acetate (CH COONH 4 Potassium chloride (KCl Magnesium sulfate (MgSO 4 Calcium chloride (CaCl Methylamine (CHNH Ethylamine (C H 5 NH Dimethylamine (CH NH Trimethylamine (CH N Application Update 55
Conditions Column: IonPac CS 4 x 50 mm (P/N06055 Guard: IonPac CG 4 x 50 mm (P/N 060560 Eluent: Flow Rate: EGC-II MSA, 0.8 mm isocratic to 8.5 min, gradient to 4 mm at min, gradient to 8 mm at 50 min ml/min Temperature: 0 C Detection: Suppressed conductivity, CAES, Inj. Volume: recycle mode ml Background: < Preparation of Solutions and Reagents Eluent The MSA eluent is prepared automatically by pumping DI water through the EG50 equipped with an EGC-MSA cartridge. Standard Solutions Stock Standards Prepare 000 mg/l standards for each of the cations and amines in DI water. Standards should be prepared from the highest purity compounds available. Table provides the amounts needed to prepare 00 ml of each standard. Concentrated standards are stable for at least one month when stored at 4 C. Table. Mass or Volume of Compounds Used to Prepare 00 ml of,000 mg/l Cation Standards Composite Standard Solutions Appropriate mixed standards are prepared from the 000 mg/l stock standards. First prepare mg/l standards and then use these secondary standards to prepare the composite working standards at low µg/l concentrations. For this application we prepared the composite standards listed in Table. Secondary standards should be prepared fresh weekly and working standards prepared fresh daily. If an accurate measurement of ammonium is required, run a separate ammonium standard because degradation of amine standards will produce ammonium. Table. Standard Concentrations for Method Calibration Analyte Level Level Level Level 4 Level 5 (µg/l (µg/l (µg/l (µg/l (µg/l Lithium 0.5.50.5.00.5 Sodium.00 6.00 9.00.00 5.00 Ammonium.50.00 4.50 6.00.50 Methylamine.50 45.00 6.50 90.00.50 Potassium 4.50 9.00.50 8.00.50 Ethylamine.50 5.00.50 0.00.50 Dimethylamine 45.00 90.00 5.00 80.00 5.00 Trimethylamine 5.00 0.00 45.00 60.00 5.00 Magnesium.50.00 4.50 6.00.50 Calcium.50.00 4.50 6.00.50 Sample Preparation Hydrogen peroxide samples were treated with platinum to eliminate the hydrogen peroxide as described in reference. Cation Compound Amount (g Lithium Lithium hydroxide 0.4 Sodium Sodium choloride 0.45 Ammonium Ammonium acetate 0.4 Potassium Potassium chloride 0.9 Magnesium Magnesium sulfate 0.56 Calcium Calcium chloride 0. Amine Amount (ml Methylamine (d=0. g/ml 0.4 Ethylamine (d= 0.8 g/ml 0. Dimethylamine (d= 0.68 g/ml 0.4 Trimethylamine (d= 0.6 g/ml 0.58 Determination of Cations and Amines in Hydrogen Peroxide by Ion Chromatography Using the RFIC (Reagent-Free System
Results and Discussion Chromatography Separating inorganic cations and amines in the same sample has traditionally been a challenging application and has required the addition of organic solvents to achieve adequate peak shapes for the amines. Unfortunately, addition of an organic solvent to the eluent reduces detection sensitivity. Using advanced resin synthesis techniques, Dionex produced a cation-exchange column, the IonPac CS, that delivers good peak shapes for amines without the addition of an organic solvent to the eluent. Using the CS, a separation of the common inorganic cations, ammonium, and the four amines, mono-, di-, and trimethyl amine, and ethylamine was developed (Figure. This separation starts with a weak (0.8 mm MSA eluent followed by two shallow gradients. Normally, such a separation could suffer from reproducibility problems due to eluent preparation errors, but the RFIC system eliminates this potential problem. The eluent generator accurately produces the required eluent with high precision. Although potassium and methylamine are not baseline resolved, accurate quantification of these ions is possible using integration tools available in Chromeleon. The unknown peaks at 0.60 and 9.50 min are likely a result of the gradient changes near those times in the eluent program. Overall, the baseline drift due to eluent concentration change is very low, which is a consequence of the high purity of the electrolytically generated MSA eluent. Linearity and Calibration Prior to sample analysis we performed a method calibration of our RFIC system using the five mixed standards listed in Table. The ranges for each cation reflect the expected range for these cations in the hydrogen peroxide sample. Table shows linear response for each cation in its chosen range. 0.5 Peaks:. Lithium. µg/l. Sodium 9. Ammonium 4.5 4. Methylamine 68 5. Potassium 4 5 4 6 0.0 0 5 0 5 0 5 0 5 40 45 50 6. Ethylamine. Dimethylamine 5 8. Trimethylamine 45 9. Magnesium 4.5 0. Calcium 4.5 8 9 0 Table. Method Calibration Results as Reported by Chromeleon Analyte % Corr. Coeff. r Offset Slope Lithium 99.6549 99. 0.050 0.05 Sodium 99.05 99.406 0.090 0.0085 Ammonium 99.804 99.60 0.008 0.0095 Methylamine 99.4 99.449 0.0 0.000 Potassium 99.6 99.5 0.09 0.006 Ethylamine 99.64 99.85 0.044 0.004 Dimethylamine 99.89 99.49 0.040 0.000 Trimethylamine 99.64 99.6 0.0 0.004 Magnesium 99.98 99.858 0.00 0.008 Calcium 99.6450 99.9 0.009 0.008 Figure. Separation of mixed cation and amine standard Application Update 55
Minimum Detection Limit (MDL The MDL of each analyte was determined using seven injections of a low level standard (reported in Table 5. The amount determined, the RSD of that amount, and calculated MDL for each cation are shown in Table 4. These MDLs show that the method has the required sensitivity for this application. Sample Analysis Three platinum-treated hydrogen peroxide samples were analyzed with the RFIC method (Figures 4. Of these samples, only samples and (Figures and 4 had detectable amines, with a small amount of trimethylamine. Sample contained alkali and alkaline earth cations and ammonium and Sample contained ammonium and alkaline earth cations. Table 4. Determination of MDLs for Cations and Amines Analyte Average Amount RSD (% MDL a (µg/l (µg/l (µg/l Lithium 0.69 0. 0.005 Sodium.50 0.4 0.05 Ammonium.5.8 0.066 Methylamine 8.00 0.84 0.05 Potassium.5 0.66 0.068 Ethylamine 6.6 0.549 0.08 Dimethylamine 6.040 0.9 0.0 Trimethylamine.80 0.4 0.8 Magnesium 0.5.55 0.059 Calcium 0.85 5.58 0.48 Reproducibility The short-term reproducibility was measured by injecting seven replicates of a low-level standard. The results reported in Table 5 show the excellent retention time reproducibility expected with the RFIC system, as well as good peak area reproducibility for a low-level standard. Table 5. Method Reproducibility.0 0. Peaks:. Sodium.6 µg/l. Ammonium 6.. Potassium 0.5 0 5 0 5 0 5 0 5 40 45 50 5. Unknown 6. Magnesium 0.5. Calcium 5.80 4 5 6 Figure. RFIC Analysis of Hydrogen Peroxide. 0.5 Peaks:. Sodium 0.8 µg/l. Ammonium. Potassium 0. 5. Unknown 6. Trimethylamine 0.6. Magnesium 0.5 8. Calcium.0 Analyte Concentration (µg/l RSD (% Retention Amount Area time Lithium 0.6 0.86 0. 5.05 Sodium.5 0.60 0.4.00 Ammonium. 0..8.94 Methylamine 8 0.0 0.84.06 Potassium.5 0.5 0.66 5.654 Ethylamine 6. 0.9 0.549 9.44 Dimethylamine 6 0.94 0.9 4.6 Trimethylamine.8 0. 0.4 6.65 Magnesium 0. 0.056.55 5.60 Calcium 0.85 0.0 5.58 8.044 8 5 6 4 0. 0 5 0 5 0 5 0 5 40 45 50 Figure. RFIC Analysis of Hydrogen Peroxide. Pass Determination of Cations and Amines in Hydrogen Peroxide by Ion Chromatography Using the RFIC (Reagent-Free System
0.5 Peaks:. Ammonium.6 µg/l. Unknown. Unknown 5. Trimethylamine 0.4 6. Magnesium 0.5. Calcium 5.80 Summary This application uses an IonPac CS column and a RFIC system to separate cations and amines in a hydrogen peroxide sample. These cations are determined at sub-µg/l concentrations using a large-loop direct injection. 0. 45 6 0 5 0 5 0 5 0 5 40 45 50 4 References:. Kerth, J., Rattmann, C., Jensen D. GIT-Labor-Fachz. 000, 44(, 4.. Monitoring for Trace Anion Contamination in the Extracts of Electronic Components. Application Note 5, Dionex Corporation, Sunnyvale, CA.. SEMI C0-06999 Specifications and Guidelines for Hydrogen Peroxide. Figure 4. RFIC Analysis of Hydrogen Peroxide. Method Accuracy To evaluate the method accuracy, a standard of ethylamine, dimethylamine, and trimethylamine was spiked into the sample and the recovery determined. Results of the spike recovery are shown in Table 6. These results show good recovery at the low concentrations of each of these analytes. Table 6. Recovery of Amines from Hydrogen Peroxide (0 % Samples Ethylamine Trimethylamine ion. Power. Productivity. Dimethylamine Amount (µg/l 0.0000 0.5000 0.60 0. Amount (µg/l 0.0000 0.9904 99.04 0.0000 0.9.9 Amount (µg/l 0.68 6.8 0.59.5.459. 0.96.5 0.5 0.5 Reagent-Free and RFIC are trademarks and CAES, Chromeleon, Elugen and IonPac are registered trademarks of Dionex Corporation. Passion. Power. Productivity. Dionex Corporation 8 Titan Way P.O. Box 60 Sunnyvale, CA 94088-60 (408-000 North America Sunnyvale, CA (408-85 Westmont, IL (60 89-660 Houston, TX (8 84-565 Atlanta, GA (0 4-800 Marlton, NJ (856 596-0600 Canada (905 844-9650 www.dionex.com Europe Austria (4 66 5 5 Belgium ( 5 494 Denmark (45 6 6 90 90 France ( 9 0 0 0 Germany (49 66 99 0 Italy (9 06 66 5 505 The Netherlands ( 6 4 4 0 Switzerland (4 6 05 99 66 United Kingdom (44 6 69 Asia Pacific Australia (6 940 5 China (85 48 8 India (9 64 5 Japan (8 6 6885 Korea (8 65 580 Application Update 55 5 LPN 8 PDF 08/06 006 Dionex Corporation