de-ionized water, reagent-grade chemicals, âultrapureâ nitric acid, specified ... sL). All chemicals used were analytical-reagent grade. (J. T. Baker. Chemical. Co.,.
CLIN. CHEM. 27/1, 73-77 (1981)
Zinc Determined in 1O-/2L Serum or Urine Samples by Flameless Atomic Absorption Spectrometry Nancy E. VieIra and James W. Hansen
We describea preciseflameless atomicabsorption and a Model 056 strip-chart spectrometric method requiring only 10 tL ofsample, Corp., Norwalk, CT 06856). which thus permits repeated measurements in the neonate. Standard curves covering the range of 0 to 30 mg of zinc per liter, with standards in various matrixes (albumin, glycerol, serum, or nitric acid) hadslopesranging from 0.49 to 1.17, relative tothatforaqueousstandards. We prepared low-zinc matrixes, which had slopes similar to that of serum, by dialyzing serum or a 50 gIL solution of albumin vs a buffered zinc-free dialysis fluid containing appropriate inorganic constituents. Use of thoroughly de-ionized water, reagent-grade chemicals, “ultrapure”
nitric acid, specified disposable plastic ware, and appropriate pipet tip-rinsing techniques minimized extraneous contamination with zinc. Concentrations of zinc in serum calculated froma “rational-method” calibration algorithm fit tothestandard curveagreedwellwith independent determinations by flame atomic absorption spectrom-
etry. AddItIonal Keyphrases: acid matrixes compared source of . nutritional
albumin, glycerol, serum, and nitric neonatal chemsstiy variation, status trace elements ref.
.
.
erence intervals Information concerning the concentration of zinc in body fluids has become increasingly important because the role of zinc in both normal and abnormal growth and development is better recognized. In addition, oral and parenteral dietary supplements that include zinc are being used more frequently in sick and undernourished patients, including premature infants. In the case of newborns, the current sample volumes required for analysis may produce hematological problems for the patient, which discourages measurements at appropriately frequent intervals to monitor mineral status and nutritional adequacy. The high sensitivity and specificity of flameless atomic absorption spectrometry permitted us to develop a convenient and reliable technique for measuring zinc in 10 sL of sample.
Materials and Methods
Apparatus All determinations were made with a Model 5000 atomic absorption spectrometer equipped with an HGA 2100 graphite furnace with ramp (gradual temperature increase) accessory Neonatal
and
Pediatric
Medicine
Branch,
National
Child Health and Human Development, National Health, Bethesda, MD 20205. Received May 9, 1980; accepted Sept. 10, 1980.
Institute
of
Institutes
of
recorder (all from Perkin-Elmer A Perkin-Elmer hollow-cathode zinc lamp was used as the source, at a current of 15 mA. The spectrometer was operated at 213.9 nm, in the peak height mode, and with a 0.7-nm low slit width. Graphite furnace conditions, established according to Fernandez and lannarone (1), were: dry for 60s with 10-s ramp to 95 #{176}C, char for 30s with 15-s ramp to 450 #{176}C, and atomize for 6 s at 2400
#{176}C. Argon was used as the purge gas flooding the graphite furnace. We adjusted the flow rate to 60 flow-meter divisions (corresponding to 110 mL/min) during atomization, to reduce zinc sensitivity threefold. This avoided excessive dilution of the sample, which can result in significant extraneous zinc contamination. The final sensitivity for a typical zinc assay at these settings is 37 pg/0.1 absorbance unit (A).
Reagents All water used in these studies was processed through deionizers manufactured and maintained by Hydroservice and Supplies, Inc., Durham, NC 27705; it then contained albumin > glycerol. For zinc in plasma or serum, analysis of variance showed within-assay variation to be 3.3% and between-assay variation to be 16% with dialyzed serum or albumin as the matrix used in preparing the standard curve. Zinc values obtained by our technique were compared with those obtained by conventional flame atomic absorption spectrometry. The means of the groups of samples (n = 36) did not differ significantly: 0.77 and 0.81 mg/L, respectively. The standard deviation for zinc values in the samples measured in the graphite furnace (0.14 mg/L) was slightly less than that for those measured by flame (0.18 mg/L), resulting in a narrower reference interval. However, for individual paired samples, the values determined by the graphite furnace were consistently lower and thus the mean difference was 5% lower (p
