LABORATORY INFORMATION BULLETIN
DFS/ORA No. 3994
Medical Devices
November 1995 Page 1 of 10
Further Investigation of Chromium Determination in Surgical Stainless Steel:
Interferences Noted for High-Carbon Steels
John B. Crowe, Barbara S. Barnes, Douglas T. Heitkemper,
S. Frank Platek, and Karen A. Wolnik
U.S. Food and Drug Administration
Forensic Chemistry Center
Cincinnati, OH
INTRODUCTION
Since 1989, FDA has been involved in the determination of chromium in imported stainless steel surgical instruments by flame atomic absorption spectrophotometry (FAAS)1. ASTM Designation F 899-84 "Standard Specification for Stainless Steel Billet, Bar, and Wire for Surgical Instruments" classifies stainless steel material for surgical instruments into one of 4 classes; Austenitic, Martensitic, Precipitation Hardening and Ferritic2. Furthermore, within each "class" a number of different "types" are specified which have a set of composition requirements associated with them, including the range for allowable chromium (Cr) content. For example, the specification for Martensitic steel of types 410 and 410X specifies allowable limits for Cr of 11.50% - 13.50% while the specification for chromium in Martensitic steel of type 440A is 16.00% - 18.00%. However, the purpose of LIB 3750 was to screen for a minimum chromium content in surgical stainless steel rather than to classify steel by type. The minimum chromium level is 11.50 % or greater for all classes and types of sugical stainless steel with the exception of Precipitation Hardening steel of type XM-16 which can be as low as 11.00 % chromium.
Note:The Laboratory Information Bulletin is a tool for the rapid dissemination of laboratory methods (or information) which appear to work. It may not report completed scientific work. The user must assure him/herself by appropriate validation procedures that LIB methods and techniques are reliable and accurate for his/her intended use. Reference to any commercial materials, equipment, or process does not in any way constitute approval, endorsement, or recommendation by the Food and Drug Administration.
In addition to chromium, limits are given for minor and trace concentrations of other elements. In Martensitic steel of types 410 and 410X these specifications include Ni (1.00 % max.), Mn (1.00 % max.), S (0.03 % max.), Si (1.00 % max.), P (0.04 % max.), and C (0.09 - 0.21 %). Inductively coupled plasma atomic emission spectroscopy (ICP-AES) has become more commonly available in FDA district and regional laboratories making it advantageous to develop multielement methods for the analysis of stainless steel by ICP-AES3. One purpose of this communication is to demonstrate the usefulness of the previously described hot-plate digestion of stainless steel1 for multielement determinations by ICP-AES. However, the primary pupose is to alert the reader to potential errors in the determination of chromium in steel classes containing high carbon levels.
A few laboratories have reported to us on the formation of a black precipitate when digesting some stainless steel samples. Preliminary studies in our laboratories have suggested that the precipitate forms in steels containing higher carbon content. In this work a series of 10 reference steels containing variable amounts of carbon are digested and analyzed for Cr, Mn, and Ni by ICP-AES. The results of this analysis are presented and discussed in terms of the interference and its effect on the analysis of imported surgical stainless steel.
EXPERIMENTAL
Reagents & Standards
Methanol, reagent grade.
Concentrated HNO3, Baker Instra-Analyzed or equivalent.
5% (v/v) HNO3.
Concentrated HCl, Baker Instra-Analyzed, or equivalent.
Distilled deionized H2O, metered resistance 18 Mm/cm.
Elemental standard solution, 1000 µg/mL, atomic spectral grade.
Fe standard solution, 10,000 µg/mL, atomic spectral grade.
SRM 73c, National Institute of Standards & Technology, Gaithersburg, MD.
SRM C1289, National Bureau of Standards, Washington, D.C.
Reference Standard 226-1, Analysen-Kontrollprobe, Berlin, Dortmund, Dusseldorf, Germany.
Reference Standards C13X and C13X PH70, MBH Analytical LTD., Barnet, England.
British Chemical Standards No.'s 332, 339, 462, 466/1, and 472, Bureau of Analysed Samples, LTD., Middlebrough, England.
Obtaining the Analytical Sample
The reference steel standards used in this work were received in the form of shavings or slivers and could be digested directly. One exception was NBS SRM C1289 which was sampled as described previously1.
Sample Preparation
Five replicates of each reference material were prepared using the procedure described in LIB 37501 with the following modifications. Sample weights were 30 mg ± 10 mg and the digest was diluted to 100 mL volumetrically with ddH2O.
Standards Preparation
Standard 1:2% HNO3, 2% HCl.
Standard 2:20 æg/mL Cr, 10 æg/mL Mn, Ni, 2% HNO3, 2% HCl.
Standard 3:200 æg/mL Fe, 2% HNO3, 2% HCl.
Standard solutions were prepared in acid washed volumetric flasks using calibrated transfer pipets. Commercially available 1000 æg/mL stock standards of Cr, Mn and Ni were used. The Fe stock standard concentration was 10,000 æg/mL.
Determination by ICP-AES
The ICP-AES instrument used in this study was a Thermo Jarrell-Ash® PolyScan 61E ICP-AES. The instrument was operated in the simultaneous mode using the parameters as follows:
Power 1150 W
Sample Flow Rate 1.5 mL/min
Torch Gas Flow High
Auxiliary Gas Flow 1.0 L/min
Nebulizer Pressure 31 PSI
Integration Time 5 sec
No. of Integrations 4
Spectrometer PurgeArgon
The instrument was calibrated using a two-point calibration prepared from the standards described above. Standard checks were analyzed after every 5 or 6 samples and reagent blanks (method blanks) were analyzed at 10% frequency. The linear ranges for Mn, Ni, Cr, and Fewere verified previously. Manganese and Ni were linear from the detection limit to 100 æg/mL. Chromium was linear to 200 æg/mL and Fe to 500 æg/mL.
Two-point background correction was used when possible. This corrects for a baseline shift between matrices; however, it does not correct for spectral overlap. The following lists the spectral lines and background correction points used in this work:
Line Background Correction
Cr 267.7 nm +0.07 nm -0.06 nm
Ni 231.6 nm +0.04 nm -0.03 nm
Mn 257.6 nm +0.07 nm -0.06 nm
Fe 259.9 nm 0 -0.06 nm
Line selection for each element is important because of potential interferences from multiple atom and ion lines arising from high levels of Fe in the sample digests. Readers are cautioned to check for Fe interferences, especially if alternative lines are utilized. A scan (expanded to observe the baseline) of each line while nebulizing a solution of Fe at a slightly higher concentration than present in the sample digests is a good way to insure the validity of the sample analysis.
Calculation of Elemental Concentration
Calculate weight % in the steel standards as follows (example Cr):
Wt % Cr= 100 * (æg/mL Cr) (100mL)
1000 æg/mg * sample wt. (mg)
Blank subtraction is used when the method blank solution value exceeds the limit of detection.
Scanning Electron Microscopy/Energy Dispersive X-Ray Analysis
Two of the steel reference sample digests which contained visible black solid particles were filtered through a Nuclepore® polycarbonate filter with a 0.1 æm pore diameter. The filter containing precipitate was transferred to a labeled petri dish and allowed to dry. The dry filter was mounted (precipitate side up) on a 25 mm diameter, vitreous carbon planchette withan even contact-region painting of colloidal graphite and allowed to dry. No sample coating was applied to the filter.
Analysis was performed in a LEICA® 260 scanning electron microscope equipped with an OXFORD® eXL energy dispersive x-ray spectrometer and a K & E Electronics® backscattered electron detector. Each preparation was analyzed using the scanning electron microscope (SEM) under the following conditions:
Accelerating Voltage 20 kV
Working Distance _28 mm
Detector Window Be
Specimen Analysis Mode Area
Detector Dead Time _8-21%
Count Rate _2000 CPS
EDXA Analysis Time 100 seconds
ZAF Analysis Elemental Quantitation
RESULTS AND DISCUSSION
In this study, chromium, manganese, and nickel were determined. The addition of manganese and nickel to the analysis for chromium is relatively easy because these elements are found in stainless steels at levels which will not require the analysis of more than one dilution of the samples. Tables 1-3 compare the results obtained in this study with the reference values for chromium, manganese and nickel, respectively. The experimental values given represent the mean and standard deviation for five replicate analyses. Note that a few of the reference materials were provided with mean values only. Table 1 also reports the certified value for carbon for each of the reference samples.
Earlier studies conducted in our laboratories suggested that steels with higher carbon levels resulted in low Cr recoveries in the digested solution. The low Cr recovery may be the result of the formation of insoluble carbide(s) of chromium4-6. Analysis of a series of reference materials with varying carbon content has allowed characterization of the resulting error in the determination of Cr. As shown in Table 1, the values obtained for chromium ranged from 101% to 106% of the reference value when the carbon level was less than 0.2 percent. The Cr values ranged from 80% to 93% of the reference value when the carbon was present at levels greater than 0.2 percent. The recovery of chromium as compared to the reference value for the series of reference steels can be seen in Figure 1. As the percentage of carbon increases past 0.2% the recovery of Cr decreases. Approximately 20% of the Cr isunaccounted for by ICP-AES in reference steels with greater than 0.3% carbon. These results suggest that higher carbon levels are responsible for low Cr values.
Black particulate was noted in the digests of reference steels with higher carbon levels. Scanning electron microscopy with energy dispersive x-ray analysis was used to analyze the precipitate found in digests of BCS 339 and NIST SRM 73 C. The SEM EDXA results are shown in Table 4. The results are presented as a ratio of the concentration of individual elements to the total concentration of measured elements. Carbon was not measured in this experiment. The particles contain a relatively high amount of chromium and some iron. Other trace elements were present but not at significant levels. This suggests that the particles are not undissolved steel but some insoluble complex with chromium as a main constituent. Further analysis by SEM EDXA has shown the presence of carbon in the precipitate; however, only a qualitative determination of carbon was made. The complex is believed to be similar to chromium carbide (Cr3C2), a highly insoluble complex which can form when a carbon steel matrix is in acid solution4-6.
The experimental values for manganese and nickel are shown in Tables 2 and 3, respectively. The manganese values ranged from 93% to 103% of the reference values with one exception. For the reference steel NIST SRM 73C which contains 0.31% carbon, 88% of the manganese reference value was obtained. The nickel values ranged from 96% to 106% of the reference values. Higher carbon levels in the steels did not appear to significantly affect the manganese or nickel analyses. This is shown graphically in Figure 2. The nickel recovery remains relatively constant with increasing amounts of carbon in the reference materials. Slightly lower analyte recovery values in materials containing higher carbon levels could be the result of co-precipitation with the chromium complex.
Of the four ASTM designated classes for surgical stainless steel, only Martensitic steels of types 420 or 440 have higher carbon levels, ranging as high as 0.50% and 1.20% carbon for types 420 and 440, respectively. Type 440 steels also have higher levels of chromium (16.00% to 18.00 %). Generally, these steels would still meet the requirement for a minimum Cr content of 11.5 percent. The most significant errors will occur with Martensitic type 420 steels where the low end of the chromium range is 12.00 percent. For Martensitic steels of type 410 or 410X no interference is expected. It should be noted that the class and type of steel being imported should be included with the import submission7.
CONCLUSIONS
Both FAAS and ICP-AES determinations are useful for verifying a minimum chromium content in surgical stainless steels in most cases. However, the readers are cautioned that the determination of Cr in surgical stainless steel by either technique has limitations. In particular, steels of higher carbon content (> 0.30%) may result in a lowchromium determination. In our experience, neither hot plate nor microwave digestions completely solubilized steels containing higher carbon concentrations. The sample's class and type should be noted. Martensitic steels of type 420 are particularly susceptible to problems using these methods.
If a precipitate is formed in the digest of a sample and the sample fails to meet minimum requirements for chromium, the sample must be reanalyzed for chromium by an alternative method. A number of techniques which utilize solid sampling could be used including arc-spark emission spectroscopy, laser ablation ICP-AES or ICP-MS, and glow discharge ICP-AES or ICP-MS.
REFERENCES
(1) Barnes, B.S., Crowe, J.B., Brueggemeyer, T.W, and Wolnik, K.A, "Determination of Chromium in Stainless Steel by Atomic Absorption Spectrometry" Laboratory Information Bulletin No. 3750, Vol 9, No. 3, March, 1993.
(2) ASTM Designation: F 899-84 Standard Specification for Stainless Steel Billet, Bar, and Wire for Surgical Instruments. pages 287-291
(3) Holak,W, and Yip, W.L, "Determination of Chromium and Other Elements in Stainless Steel Surgical Instruments by ICP After Microwave Acid Digestion" Laboratory Information Bulletin No. 3903, Vol. 10, No.10, October, 1994 #10
(4) Cotton, F.A., and Wilkinson, G., Advanced Inorganic Chemistry A Comprehensive Text, 4th edition, John Wiley and Sons, New York, 1980.
(5) Brackett, B., American Iron and Steel Institute, personal communication, 1994.
(6) Tyler, B., Telemetal Industries, personal communication, 1994.
(7) Shirk, C., US FDA, CDRH, personal communication, October 1995.
Table 1. ICP-AES Cr Results vs. Cr Certified Values (%) Sample C1289 BCS BCS BCS MBH MBH BCS BCS NIST AK 466 332 462 C13 X C13X 472 339 73c 266-1 Cr 12.43 17.91 13.52 12.85 14.52 15.51 14.55 11.48 10.28 10.99 ±.07 ±.14 ±.16 ±.13 ±.19 ±.18 ±.14 ±.20 ±.26 ±.29 Cr* 12.12 17.65 12.80 12.30 14.10 14.90 15.82 12.40 12.82 13.67 ±.01 ±.04 ±.049 ±.093 ±.05 ±.06 C* 0.014 0.062 0.063 0.092 0.16 0.2 0.227 0.29 0.31 0.416 ±.001 ±.002 ±.006 ±.003 ±.007 Table 2. ICP-AES Mn Results vs. Mn. Certified Values (%)
Sample C1289 BCS BCS BCS MBH MBH BCS BCS NIST AK 466 332 462 C13 X C13X 472 339 73c 266-1 Mn 0.33 0.69 0.81 0.74 1.49 0.67 1.01 0.40 0.29 0.40 ±.002 ±.009 ±.008 ±.009 ±.019 ±.005 ±.011 ±.005 ±.004 ±.004 Mn* 0.33 0.70 0.80 0.74 1.44 0.66 1.02 0.41 0.33 0.43 ±.01 ±.008 ±.023 ±.03 ±.013 Table 3. ICP-AES Ni Results vs. Ni Certified Values (%) Sample C1289 BCS BCS BCS MBH MBH BCS BCS NIST AK 466 332 462 C13 X C13X 472 339 73c 266-1 Ni 4.18 8.69 12.89 12.82 5.76 3.19 1.97 0.38 0.24 0.14 ±.018 ±.094 ±.45 ±.16 ±.077 ±.034 ±.022 ±.008 ±.003 ±.004 Ni* 4.13 8.64 12.40 12.50 5.65 3.00 1.95 0.37 0.25 0.14 ±.01 ±.04 ±.07 ±.045 ±.03 ±.014 *denotes certified values; all values in % by weight, results average of 5 weighings. Table 4. Primary Elements Identified by EDXA as a Ratio of the Concentration of Each Element to the Total Concentration of Measured Elements Standard Chromium Iron Silicon Manganese BCS 339 68.5% 27.7% 2.2% 0.97% NIST 73c 73.1% 24.2% 0.56% 0.0%
FIGURES 1 and 2