Simultaneous Multielement Analysis of Biologically Related Samples

2 Simultaneous Multielement Analysis of Biologically Related Samples with RF-ICP

1

Downloaded by AUBURN UNIV on September 11, 2017 | http://pubs.acs.org Publication Date: February 1, 1979 | doi: 10.1021/ba-1979-0172.ch002

F R A N K N . A B E R C R O M B I E , M . D . SILVESTER, and R O M A N A B. C R U Z

2

Barringer Research Limited, 304 Carlingview Drive, Rexdale, Ontario, Canada M 9 W 5G2

This chapter discusses the advantages and limitations of the multielement analysis of biologically related samples using induction-coupled plasma optical emission. The sample categories covered include grains, feeds, fish, bovine liver, orchard leaves, and human kidney stones. These materials have been simultaneously analyzed for copper, nickel, vanadium, chromium, phosphorus, cobalt, lead, potassium, zinc, manganese, iron, strontium, sodium, aluminum, calcium, magnesium, silicon, boron, and beryllium, often with limited amounts of sample.

he development and the acceptance of the Induction-Coupled Argon Plasma ( I C A P ) as an analytical tool since the reports from Greenfield ( J ) and Fassel (2) have been monitored by scientists i n many disciplines. Publications describing instrumentation design, performance characteristics, and analytical applications during the early years originated primarily from the Albright and Wilson and the Iowa State U n i versity researchers. More recently, review articles have appeared describing the various aspects of I C A P emission spectroscopy (3,4,5,6, 7). The performance claims inferred that the I C A P would become a basic trace analysis tool for the disciplines of geological, metallurgical, 1

Current address: Montana Bureau of Mines and Geology, Montana Tech., Butte, M T 59701. 2 Current address: CSA Ltd., 55 Holmdene Close, Beckenham, Kent, U.K. 0-8412-0416-0/79/33-172-010$05.00/0 © 1979 American Chemical Society Risby; Ultratrace Metal Analysis in Biological Sciences and Environment Advances in Chemistry; American Chemical Society: Washington, DC, 1979.


2.

ABERCROMBiE

E T

A L .

Multielement Analysis with RF-ICP

11

agricultural, medical, and environmental sciences. The development and acceptance of the technique was accelerated when A p p l i e d Research Laboratories (1974) and Jarrell A s h (1975) offered commercial, directreading emission spectrometers interfaced to I C A P sources. (Prior to this event, modular instrumentation interfacing was necessary if a laboratory intended to use an I C A P either for research or analysis. ) F r o m this current time perspective, it is possible to note that the performance attributes and analytical utility of the I C A P claimed by Greenfield and Fassel have been realized. It is now conservative to suggest that the I C A P can be regarded as an accepted analytical tool. Several commercial instrumental versions are available with more versions i n various stages of planning and manufacturing. M a n y industry groups and most technically oriented government agencies use plasma systems within their laboratories. The first industrial application of an induption-coupled plasma occurred within a custom analytical laboratory. A t the present time, i n Canada alone, there are five commercial laboratories with I C A P systems. W i t h i n our own organization, the two systems were acquired i n 1974 and 1975. A t that time there was a confidence that the I C A P would become a key instrument for the service facility. Since that time, extensive experience which reinforces this view has been obtained i n the routine application of the I C A P to diverse analytical problems and samples. In this chapter, the application of the I C A P to biologically related samples is discussed. Analytical data are presented for several diverse sample types and reference standards which demonstrate the suitability of one I C A P for the analysis of biologically related samples. T w o reports (8,9) have shown potential applications for I C A P analysis i n health and related areas. The data presented i n the extensive report by Dahlquist (10) also supports this viewpoint. W h i l e only one specific example is presented for health-related specimens i n this chapter, an implied general area of potential application is intended. H u m a n lung, brain, heart, hair, and blood specimens have been analyzed i n this laboratory but the nature of this work must remain undisclosed at this time. Other samples which have been successfully analyzed by I C A P but do not bear a detailed discussion in this report include herring gull egg shells, egg contents and whole body homogenates, rat bodies, pet food, mollusks, wheat straw and grain, tree seeds, and various legumes. The analytical results reported here, where possible, are compared with certified values or values obtained by other analytical techniques. I n a few examples, however, the data are presented to demonstrate a trend; i n these instances values are for information only as no reference data are available by alternate techniques.

Risby; Ultratrace Metal Analysis in Biological Sciences and Environment Advances in Chemistry; American Chemical Society: Washington, DC, 1979.

12

U L T R A T R A C E

M E T A L

ANALYSIS

I N SCIENCE

A N D E N V I R O N M E N T


Experimental Digestions. After the initial sample preparation, such as surface cleaning, size reduction, homogenizing, etc. (where required), a wetashing acid digestion is used. T o one gram of sample i n a treated borosilicate beaker, 8 m L Ultrex nitric acid (J. T. Baker, Canlab, Toronto, Ontario) and 2 m L Aristar perchloric acid (British D r u g Houses Chemi cals, Toronto, Ontario) is added. (The borosilicate glassware is treated by soaking overnight at room temperature i n 1:1 concentrated n i t r i c sulfuric acid. ) T h e beaker is covered with a watch glass; the mixture is allowed to stand for one hour, and then is heated over medium heat on a hot plate. T h e temperature should be adjusted to maintain a steady boiling and refluxing for 1-2 hr. A t the end of this period there should be a clear solution. The watch glass is removed to evaporate a l l excess acid. (If brownish coloration appears toward the end, indicating incom plete breakdown of the sample, more nitric acid should be added w i t h refluxing repeated. ) T h e solution is evaporated to dryness. T h e residue is then taken up i n 5 m L of 0.5N H C 1 (Ultrex) and diluted to a suitable final volume, usually 10 m L . A greater dilution is used to decrease the calcium and magnesium concentration to within the dynamic range of the instrument or, for the case of some samples, to completely dissolve the potassium perchlorate present. Instrumentation. A 32-element I C P - O E S system was used through out. A description is provided i n Table I. Throughout the course of the work the spectrometer was modified as recommended by the manufac turer. The primary modification relative to this work was the installation of interference filters and photomultiplier masking assemblies. The data presented i n Tables II, III, V , V I , and V I I I were obtained after the interference filter and photomultiplier modification. The main Barringer Research modification was the addition of a monochromator to provide a dedicated channel for potassium at the 766.49-nm line which is not within the wavelength region of the QA-137 and 1920 L / m n grating combination. Spectral Lines Used (nm) Ag Al Β Ca Cd Co Cu

328.07 308.21 249.77 315.89 226.50 345.35 324.75

Fe Κ Mg Mn Mo Na Ni

259.94 766.49 279.55 257.61 386.41 330.23 231.60

Ρ Pb Sr Ti V Zn

214.91 220.35 407.77 334.95 292.40 202.55

Calibration and Solution Analysis. Calibration is accomplished b y recording the emission signals from a series of 15 solutions containing varied concentrations of the elements of interest. T h e combination of elements i n each of the individual calibration solutions is selected to include only elements that do not cause chemical or spectral interferences w i t h the other elements within the same solution. ( F o r example, lead and zinc calibration solutions do not contain calcium and magnesium salts to avoid the spectral interference from calcium and magnesium.


2.

ABERCROMBiE

E T

A L .

Multielement Analysis with RF-ICP


Table I.

Experimental Conditions

Generator output power employed frequency

1600 W 27.12 M H z

Gas argon coolant plasma aerosol

vapor from bulk liquid 10 L / m i n 1 L/min 1 L/min

Sample uptake rate mode Spectrometer configuration grating reciprocal linear dispension detectors entrance slit exit slits plasma observation height Monochromator optics

monochromator slits grating photomultiplier

13

2-2.5 m L / m i n cross-flow pneumatic, with Scott (11) chamber Model QA-137 (Applied Research Laboratories, Sunland, C A ) 1-m Paschen-Runge mount 1920 rulings/mm, blazed at 350 nm 0.48-0.52 nm/mm, first order R300 (Hamamatsu Corp., Middlesex, N J ) 12 /xm 50 jum 16 mm above load coil, 4-mm vertical section 50-mm focal length, 25.4 mm quart, 1:1 image focused on a 500-mm fiber bundle, placed immediately in front of the entrance slit y M Ebert ( J a r r e l - A s h Division, Fisher Scientific Co., Boston, M A ) 50 /nm ruling information unknown, near-infrared blaze R787 (Hamamatsu Corp., Middlesex, N J ) 2

L e a d and silver calibration solutions do not contain other metals as the chloride salts. Also, sodium and potassium calibration solutions do not contain molybdenum, phosphorus, silicon, and chromium, w h i c h are used as the potassium and sodium salts). The acid concentration of the samples is adjusted to within ± 1 0 % of the concentration of the calibration solutions. This procedure effectively minimizes solution uptake rate and nebulization efficiency effects which have been reported (JO, 12,13). The emission spectrometer is interfaced to a programmable calculator ( M o d e l 9830, Thermal Printer 9866, Hewlett-Packard, Mississanga, Ontario). During calibration raw millivolt data of the matrix acid blank are subtracted from the standard data. Slopes and intercepts are obtained through a second-order polynomial regression. D u r i n g analysis all data listed on the thermal printer and stored on the % - i n . magnetic tape cassettes is i n solution concentration units with appropriate sample


14

U L T R A T R A C E

M E T A L

ANALYSIS

I N SCIENCE

A N D

E N V I R O N M E N T

identification and dilution information. T h e data is relisted, reporting the sample concentration for the elements of interest i n an appropriate tabular format.


Results and Discussion Current commercial I C A P systems are designed for relatively nonviscous liquid samples a n d standards. Thus, solid samples must be dissolved or leached via chemical procedures. The procedures w h i c h are commonly used include: digestion-oxidation i n a fluorocarbon polymer bomb containing an oxidizing acid; low-temperature plasma ashing fol lowed by acid leaching of the residue; controlled temperature dry ashing in a mufHe furnace (approximately 4 5 0 ° C ) followed by acid leaching of the residue; and a wet digestion-oxidation b y a strong oxidizing acid reflux. The digestion used throughout this chapter is a controlled reflux in H N O 3 / H C I O 4 . ( W a r d and Marciello (14) have recently compared several digestion procedures and H F / H N O 3 / H C I reflux seems to be a superior digestion for the analysis of samples containing silicon a n d elements which tend to form silicates. ) W h i l e the H N O 3 / H C I O 4 diges tion had been successfully used i n the past at Barringer for several ele ments b y classical atomic absorption sequential analysis, no work had demonstrated the applicability of the digestion or the capability of the staff to carry out simultaneous trace multielement analyses on biological samples. T h e average number of elements determined per sample i n creased from five for atomic absorption spectroscopy to approximately 20-25 for the I C A P approach. Standard reference materials were the first samples analyzed to evaluate the preparation chemistry and instrumental capability. Evalua tion of the initial analytical results demonstrated that a greater degree of analytical expertise was required to ensure precise and accurate multi element analysis. I n many instances, lack of agreement w i t h the certified value was traced to a chemical incompatibility i n the digestion procedure Table II. Ni (ppm)

Comparison of Barringer (BRL) Ρ (%)

Να (ppm)

Orchard leaves N B S 1571 BRL

1.3 ± 0.2 1.8 ± 0.2

0.21 ± 0.01 0.20 ± 0.05

82 ± 6 140 ± 12

Bovine liver N B S 1577 BRL

Simultaneous Multielement Analysis of Biologically Related Samples

Recommend Documents