Clean Chemistry

Techniques for the Modern Laboratory

by Dr. Robert Richter

A practical guide to . . .

Clean Sample Preparation for Trace Metals Analysis

C

LEAN HEMISTRY Techniques for the Modern Laboratory

by Dr. Robert Richter

A practical guide to . . .

Clean Sample Preparation for Trace Metal Analysis

© 2003 Milestone Inc.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written consent of the publisher.

Printed in the U.S.A. Edition 1

Teflon®, Dacron®, and Tyvek® are registered trademarks of E.I. DuPont Corporation.

Milestone Press 25 Controls Dr.. Shelton, CT 06484

CONTENTS Section 1: Controlling Contamination in Your Laboratory Chapter 1: The Analytical Blank

13

Chapter 2: Laboratory Environment

15

Chapter 3: Materials for Trace Analysis

21

Chapter 4: Trace Analysis Reagents

33

Chapter 5: The Analyst: A Source of Contamination

37

Section 2: Advanced Sample Preparation Techniques Chapter 6: Closed-Vessel Microwave Digestion

43

Chapter 7: Microwave Evaporation

57

Chapter 8: Improving Method Detection Limits

65

Section 3: Preparing Your Laboratory for Trace Analysis Chapter 9: Laboratory Housekeeping Techniques

77

Chapter 10: Equipment & Supplies for Trace Metal Analysis

81

Appendix A: References

91 5

Clean Chemistry: Techniques for the Modern Laboratory

6

Introduction


8

O

ver the past several decades, the need for trace and ultratrace elemental analysis has increased. Current drinking water regulations require determinations in the low parts

per billion (ppb) range, while semiconductor applications require parts per trillion (ppt) determinations. There is a growing awareness that improved sample handling and preparation techniques are required to meet these new standards, and the instruments and experimental techniques used to achieve them. This book focuses on helping the analytical chemist understand and adapt to the latest trends in trace metal analysis. The book is divided into three parts. The first part of the book focuses on the importance of the analytical blank, along with techniques for its reduction and

Introduction

control. The second part deals with the preparation of the homogeneous sample solutions for trace metal analysis. Part three talks about preparing your lab for trace metal analysis—what you need and where to get it.

9


10

l

SECTION ONE

l

Controlling Contamination in Your Laboratory


12

T

race elemental methods require the analyst to carry a special sample, known as an analytical blank, through all the steps of the analysis. This analytical blank is a measure of

all external sources of elemental contamination, and is used to make a correction to the measured sample concentration. The variability of the analytical blank, rather than its absolute value, is what determines the accuracy of trace metals analysis. The overall uncertainty for an analytical measurement is calculated using the following formula1:

��

��

The variability of the analytical blank has little or no effect on the accuracy of the result when the analyte concentration of the sample is several orders of magni-

Chapter 1 The Analytical Blank

tude higher than the blank result. For example, given a sample with a measured lead concentration of 500 ± 25 ng and a blank concentration of 10 ± 5 ng, the total uncertainty would be:

�� The reported analytical result would be 490 ± 25 ng of lead. In this example, the blank result has no effect on the accuracy of the result. 13


On the other hand, as the analyte concentration approaches the blank level, the accuracy of the result can be compromised by the uncertainty of the analytical blank measurement. For example, given a sample with a measured lead concentration of 50 ± 2 ng and a blank concentration of 10 ± 5 ng, the total uncertainty would be:

�� The reported analytical result would be 50 ± 5 ng of lead. In this example, all of the uncertainty in the analysis is due to the analytical blank. In order to improve the precision and accuracy of trace metal determinations, steps must be taken to control and reduce the analytical blank. The contributions to the analytical blank come from four major sources: • Laboratory environment • Apparatus and containers used for analysis • Reagents • Analyst The rest of Section One discusses each of these sources in detail, and offers some strategies for controlling these contamination sources in your lab.

14

A

irborne contamination is a major contributor to the analytical blank. Dust particles are created in the laboratory from the slow weathering of solid

objects, or from chemical processes upon materials. The main components of these dust particles are Ca, Si, Al, Fe, Na, Mg, K, Tl, Cu, and Mn.2-4 If left isolated and undisturbed, dust particles will have little effect on trace metal analysis. Unfortunately, this does not usually happen, and the dust particles become airborne. Airborne particulate concentrations can go from 0.2 x 106 in the morning to 1.5 x 106 by noon.5 Once airborne, the heavier particles quickly settle out, while the lighter ones are suspended and carried by the air currents throughout the laboratory. Airborne contami-

Chapter 2 Laboratory Environment

nates can also be introduced to the laboratory from external sources, through the ventilation system. In Dortmund, Germany, analysts discovered that their high iron blanks were caused because nearby industrial plants dispersed 20 tons of ferric oxide, which contaminated their ventilation system.6 Particulate contamination is minimized by filtering the laboratory air with a high efficiency particulate air filter (HEPA). This filter was developed for the Manhattan Project to remove fissionable particles from the air.7 HEPA filters have an efficiency of 99.97% for particles 0.3 15


µm and larger.

They effec-

tively remove bacteria, pollen, fly ash, and dust. Some common contaminates, such as tobacco smoke, are not removed because their particles are smaller than 0.3 µm.2 Particulate contamination in a work area is also minimized by using a laminar air flow HEPA air filter

system to provide directional air flow. Directional air flow

from one end of the work area to the other essentially creates an air curtain, preventing airborne contamination from entering, or settling out. The best approach to controlling airborne contamination for trace metal analysis is to construct a clean laboratory, or “clean room.” This approach is expensive, and is used in the production of computer, pharmaceutical, and aerospace products, and in the production of standard reference materials.

Clean rooms minimize metal-bear-

ing dust by filtering the air that enters the room through HEPA filters, using specialized construction materials, and operating under positive pressure so that the flow of air is always out of the room. Clean rooms also utilize antechambers to isolate them from the general laboratory and provide an area for analysts to change into specialized clean room attire. In a properly maintained and operated clean room, sub-parts-per-trillion measurements can be 16

Ch. 2: Laboratory Equipment

made.

More information

on clean rooms and clean room design can be found in references 8–11. A more practical approach for trace metal laboratories is the use of laminar flow hoods. These units provide areas of HEPA filtered air for sample handling, analysis, and reagent storage. The

Laminar flow hood

standard units are good

for drying containers and apparatus after cleaning, protecting samples in autosampler trays during analysis, and preparation of calibration and quality control standards. A standard unit should not be used to handle or store toxic or hazardous materials, because all the fumes that are generated are blown out at the analyst. Exhausted laminar hoods are available for the handling of hazardous and corrosive substances. These units are constructed from nonmetallic components and the HEPA filtered air flows vertically and is drawn though a perforated work area into the laboratory ventilation system. Clean conditions can also be obtained with a filtered air enclosure. These enclosures consist of a plexiglass box that is purged with filtered nitrogen or air to remove air particles. The exhausted air is vented to a con17


ventional hood. These units are an inexpensive alternative to laminar flow hoods, and are good for drying and storage of containers and apparatus after cleaning, and storage of samples and reagents. Airborne contamination can also be controlled by minimizing the generation, transportation, and deposition of atmospheric particles. The major source for trace metal particulates in the laboratory is the degradation of metals, paints, cements, plastics, and other construction materials. Unnecessary shelving, partitions, and furniture should be removed from the laboratory, because dust and debris can accumulate on them. Metal furniture should be replaced with wood. Stainless steel door handles, hinges, and plumbing should be replaced with plastic equivalents or coated with pigment-free epoxy paint. Bench tops should be coated with epoxy paint, and for added protection, covered with Teflon, or polyethylene sheeting (contact paper). Ceiling panels should be replaced with ones that have a plastic laminate on each side to prevent particle formation. Low fiber emitting and dissolvable content tissues should be used for wiping operations. Floors, benches, and apparatus should be wiped down with D.I. water regularly.

Bottles, containers, samples, reagents,

and equipment should be kept isolated from laboratory air using laminar flow hoods, plastic snap-top Tupperware boxes, or polyethylene bags.3,4,8-13 When an electrostatic charge is present, charged atmospheric particles are attracted to oppositely charged 18

Sili c Tefl on R u Sili on bber con Vin y Po l lyu Po ret h ly My styre ane l ne Ra ar yon Go l Bra d, Pla t s Ha s, Silvinum rd Wo Ruber o Ste d ber el Pap Alu er m Silk inu m Wo ol Ny lo Hu n m Mi an H ca air G la s Hu s m Air an S k in

Ch. 2: Laboratory Equipment

Most Negative (-)

Most Positive (+)

Figure 1. Materials’ tendency to generate a static charge.

surfaces. Electrostatic charges are usually generated by friction between, and/or separation of, two dissimilar materials, at least one of which is a nonconductor or a poor conductor of electricity. The accumulated charge (static) resides on the surface of, rather than within, the charged nonconductive object. The highest static charges accumulate under low humidity, on insulating surfaces that have a low moisture content.4 Teflon has a tendency to generate a negative charge, while the components of airborne contamination become positively charged, especially at low humidity (Figure 1). The opening of a Teflon bottle produces a negative charge on the neck of the bottle, so any positively charged airborne contamination can immediately become attached to the cap or neck of the bottle.12 Electrostatic charges on airborne contamination can be minimized by keeping humidity levels above 50%. The surface charges can be removed by wiping with a lint-free cloth, lightly wetted with high-purity ethanol or water. Commercially manufactured static eliminators are also available. 19


20

B

efore a sample is analyzed for trace metals, it has been collected, stored, processed, and prepared. During this sequence of events, the sample comes in contact with

many different laboratory tools, containers, and apparatus, which can deposit trace metal contamination into the sample. Standard laboratory mortars are made of alumina or glass, and will therefore contaminate the sample with Al, Si, and Fe. Mills and blenders are made of stainless steel, and have tungsten carbide blades, which will contaminate the sample with W, Ni, Cr, Co, and Fe. Sieves are usually made of stainless steel with copper wire mesh, which will contribute Cu, Fe, Co, Cr, and Ni. The ashless filter paper often used to filter samples contains trace metal contamination on the order of 1ppm, and the filtration assem-

Chapter 3 Materials for Trace Analysis

blies can also impart contamination to the sample.

For trace

metal analysis it is best to avoid these practices when possible. Solid samples should be homogenized by digesting several samples and combing the digestates, or by digesting a large sample and aliquoting. Aqueous samples should be centrifuged in plastic tubes (see below) to remove particulate material.

Borosilicate Glass The most common material used for laboratory containers and apparatus is borosilicate glass. It is resistant to most 21


acids, but should not be used with HF or boiling H3PO4. Alkaline solutions should not be heated or stored in borosilicate glass, because they will gradually solubilize the glass according to the following equation4: 2X NaOH + (SiO2)x

]XNa SiO 2

3

+ XH20

Borosilicate glass is not a good material for trace analysis, because it contains high levels of trace metals (Table 1), and has the potential to absorb analytes from the sample according to the following equation4,14: GlassSiOH + M+

] GlassSiOM + H

+

Quartz An alternative to borosilicate glass is quartz.

Like

borosilicate glass, it is resistant to most acids, but should not be used with HF, boiling H3PO4, or alkaline solutions. Quartz is composed almost entirely of SiO2, and its trace metal concentration depends on the type of quartz and the method of production. Naturally occurring quartz laboratory components are made by electric (Type I) or flame melting (Type II). Type II quartz has a lower trace metal concentration because some of the metals are volatilized in the flame. Synthetic quartz laboratory components are made by the vapor phase hydrolysis (Type III) or oxidation and electrical fusion (Type IV) of SiCl4. Both of these methods produce quartz with low trace metal contamination.3,4,15 The typical trace metal impurity levels for the various types of quartz are shown in Table 1. 22

23 b

a

600 1,000 Major 2.9

3,000 3,000

Borosilicate Glass Major Major 1,000

Adapted from references 3 and 15. ND = Not Detected.

Element Al B Ca Cr Cu Fe K Li Mg Mn Na Sb

Quartz (Type I) 74 4 16 0.1 1 7 6 7 4 1 9 0.3

Quartz (Type II) 68 0.3 0.4 NDb 1 1.5 <1 1 ND 0.2 5 0.1

Quartz (Type III) < 0.25 0.1 < 0.1 0.03 <1 < 0.2 0.1 ND ND < 0.02 < 0.1 0.1

Table 1: Trace element concentrations (µg/g) in borosilicate glass and various types of quartz.a

Ch. 3: Materials for Trace Analysis


Synthetic Polymers The low levels of trace metal contamination make quartz an ideal material for trace metal analysis, but the cost and availability of common laboratory containers and apparatus in quartz limit its use in trace metal analysis. Synthetic polymeric materials are now being employed as materials for containers and apparatus for trace metal analysis. The trace metal impurities of these materials are comparable to those of quartz, but will vary based on the manufacturing environment, type of molding, molding components, and polymerization process. A dirty manufacturing environment will lead to the incorporation of airborne particles. Molded components can contain high levels of the metals (Ni, Al, Mn, Cu, Fe) used to make the mold.12,16 The most common materials used for trace metal analysis are polyethylene, polypropylene, and fluorinated polymers.

Polyethylene There are two types of polyethylene used in trace metal analysis, conventional (low density) and linear (high density). Low density polyethylene (LDPE) is produced by high pressure polymerization of ethylene. High density polyethylene is produced at low pressures, catalyzed by transition metal oxides ([Al]R3, TiCl4, ZrCl3, VCl3, CrCl3). Polyethylene is resistant toward concentrated HCl and HF, but is oxidized by dilute HNO3 and aqua regia. Prolonged storage of dilute solutions of HNO3 causes the material to turn brown or yellow. The maximum temperature for LDPE 24


Table 2: Trace element concentrations (µg/g) in some polymersa Element Al Ca K Na Sb Ti Mn Zn a

LDPE 0.5 > 5,000 1.3 0.005

HDPE 30 800 > 600 1.5 0.2 5 0.01 520

PP 55

4.8 0.6 60 0.2

PFA

FEP

PTFE

0.1

0.2 0.4

0.23 0.16

0.6

Adapted from reference 17.

is 80°C and 110°C for HDPE. The use of LDPE is preferable to HDPE because it has less trace metal contamination. (Table 2).

Polypropylene Polypropylene is produced catalytically (Al, Ti) from propylene, and, like HDPE, has elevated levels of some trace metal contaminates (Table 2). Polypropylene is less resistant to concentrated HCL and becomes yellow or brown with prolonged exposure. Its resistance to dilute HNO3 and aqua regia is similar to that of polypropylene. Polypropylene is harder and more rigid than polyethylene and is stable up to 135°C. Polypropylene is well suited for open vessel digestion containers, and applications that require sterilization. 25


Fluorinated Polymers (Teflon®) Fluorinated polymers (Teflon®) are more expensive than polyethylene or polypropylene, but have lower trace metal impurities (Table 2), and greater chemical resistance. Fluorinated polymers are attacked only by molten alkali metals and fluorinated organic compounds at elevated temperatures. The greater resistance is due to the high-energy C-F bonds and the protection of the carbon backbone by the fluorine atoms. The three most common fluorinated polymers used in trace metal analysis are poly(tetrafluroethylene)(PTFE), perfluoroalkoxy-flurocarbon (PFA), fluorinated ethylene propylene (FEP).

Cleaning Methods Picking a material with low trace metal impurities doesn’t guarantee low blank levels. Trace metal impurities can be leached from the container and apparatus, by the sample and reagents used for trace metal analysis. Comprehensive cleaning procedures must be adopted to ensure the lowest blank levels. There have been a variety of cleaning methods reported in the analytical literature, and the method chosen will vary with the chemical behavior of the element of interest. A good general procedure for cleaning all types of containers is sequential leaching with hydrochloric and nitric acids (Tables 3 and 4). For polyethylene, a 48-hour soaking with 10% nitric acid is effective for initial and routine cleaning.17, 18

26


Table 3: Impurities (ng/cm2 of surface) leached from plastic containers in one week with 1:1 HNO3/Water. HDPE and LDPE were leached at room temperature while FEP was heated to 80ºC.a Element Pb Tl Ba Te Sn Cd Ag Sr Se Zn Cu Ni Fe Cr Ca K Mg Al Na Total a b

LDPE 0.7 1 2 < 0.5 < 0.8 0.2 NDb 0.2 3 2 2 0.5 3 0.8 10 2 0.7 1 8 38

HDPE 2 <1 < 0.2 0.2 1 0.2 0.2 1 0.4 8 0.4 1.6 3 0.2 0.6 2 0.6 1 10 50

Adapted from reference 17. ND = Not Detected.

27

FEP 2 <1 4 0.6 1 0.4 <8 0.2 0.2 4 2 2 20 0.8 80 2 8 6 6 148


Table 4: Impurities (ng/cm2 of surface) leached from plastic containers in one week with 1:1 HCl/Water. HDPE and LDPE were leached at room temperature while FEP was heated to 80ºC.a Element Pb Tl Ba Te Sn Cd Ag Sr Se Zn Cu Ni Fe Cr Ca K Mg Al Na Total a b

LDPE 18 3 0.3 0.7 < 0.8 0.2 ND 0.2 < 0.3 1.0 0.7 0.3 1.0 0.3 0.8 0.7 0.7 10 42 38

HDPE 0.6 < 0.6 1 NDb <1 0.2 ND 0.2 0.4 9 1 0.8 1 0.8 60 1 0.4 4 6 50

Adapted from reference 17. ND = Not Detected.

28

FEP 2 <1 2 2 1 0.6 <6 <1 0.8 4 6 0.8 16 4 2 1.6 1.0 4 2 148


Steam cleaning with nitric or hydrochloric acid is also a very effective cleaning method for containers and apparatus (Table 5).19,20 In this method the container is placed over a PTFE-coated glass rod. Acid in a lower reservoir is heated, and purified acid vapor travels up through the glass rod and condenses on the container, removing surface contamination (Figure 2). This method of cleaning is a popular alternative to the traditional soaking methods for the following reasons: 1. The trace metal contamination found in the reagent grade acid remains in the lower reservoir and does not come in contact with the component to be cleaned. 2. The clean component does not remain in contact with the cleaning acid after the surface contamination is removed. ��

��

��

��

��

29

Figure 2. Acid steam cleaning system.


3. The critical surfaces of the clean component are dry when the cleaning process is complete.

This

eliminates the need for rinsing and air drying. 4. The cleaning process takes place in a sealed container, which minimizes airborne contamination and provides a clean environment for the components to be stored until they are needed.

A Closer Look A machine that performs fully automated acid steam cleaning, as described in the text above. Purified acid vapor condenses on the containers placed in the machine, and the sealed environment protects them until they are needed. (See page 82 for more information.)

30

a

≤ 85

≤ 55 ≤ 56

144 ± 39 ≤ 85 ≤ 72 ≤ 261 ≤ 57

995 ± 80

Ni Co Cu Cr Cd Tl Pb Zn


117 ± 12

≤ 474

Fe

≤ 876

≤ 57

≤ 261

≤ 72

≤ 56

≤ 55

≤ 474

≤ 121

≤ 121

Element Al Mg Na

TFM Teflon Vessel Acid Leachingb Steam Cleaningb 287 ± 46 258 ± 24 289 ± 22 232 ± 15

b

1,005 ± 124

≤ 57

≤ 261

≤ 72

≤ 85

109 ± 9

≤ 56

≤ 55

≤ 474

Error expressed as one standard deviation (n = 3).

1,640 ± 1,000

≤ 57

≤ 261

≤ 72

170 ± 15 176 ± 57

≤ 56

≤ 55

≤ 474

Quartz Vessel Acid Leaching Steam Cleaningb 398 ± 28 327 ± 18 441 ± 56 347 ± 26 1190 ± 350 608 ± 67 b

Table 5: Comparison of high-temperature acid leaching cleaning vs. acid steam cleaning. Trace metal contamination (pg/g) in 5% HNO3 blanks prepared after cleaning are listed below. The acid leaching was performed at 180ºC with mixture of HCL and HNO3. The steam cleaning performed with HNO3 only.a



32

T

he instrumentation used for trace analysis (ICP/OES or ICP/MS) requires homogeneous solutions for calibration and analysis. Calibration solutions are prepared by dilu-

tion with water, and samples are treated with mineral acids. The purity of the reagents is important because the amount of reagent used is usually several orders of magnitude larger than the original sample size. Trace metal contamination in the reagents must be low enough to measure accurately the analyte concentration in the sample. For example, to measure 10 ng of lead in a sample that was prepared with 25 mL of reagents, the lead contamination in the reagents must be less than 0.2 ng. No single purification method is capable of removing all impurities from reagents.

Sub-boiling

distillation has been shown to

Chapter 4 Trace Analysis Reagents

be the method of choice for acid purification.

This method uses

infrared heaters to vaporize the surface liquid. The vaporized liquid is collected on an inclined water-cooled condenser and drips into the collection container (Figure 3). Vaporization without boiling is the key element of this purification process because it prevents aerosolized particles from depositing on the surface of the condenser and being carried over to the purified acid. High-purity acids produced in this way have essentially the same concentration as the acids used to produce them.12,21-23 Table 6 shows the quality of acids that can be obtained from sub-boiling distillation. 33


Figure 3. Sub-boiling acid distillation system.

A Closer Look A fully automated sub-boiling distillation system allows chemists to produce their own high-purity acids, and even re-purify contaminated acids. (See page 81 for more information.)

A good water supply is also essential for trace metal analysis. The most common methods for achieving this are sub-boiling distillation (previously discussed), reverse osmosis, and ion-exchange. Reverse osmosis separates dissolved material from the water by forcing contami34

Ch. 4: Trace Analysis Reagents

Table 6: Trace metal contamination (pg/g) in nitric acid produced by sub-boiling distillation in a quartz still. Element Mg Al Ca Ti V Cr Mn Fe Co Ni Cu Zn Se Sr Ag Cd Sn Ba Tl Pb B

Source 1a 42 147 157 8 11 5 2 210 1 23 21 49 1 1 2 2 9 4 <1 3 ND

Source 2b 90 700 110 ND ND 60 ND 350 ND 80 50 60 60 20 6 20 20 20 60 30 ND

Source 3c 400 900 400 800 <3 100 7 800 < 10 30 200 80 ND ND ND < 30 ND 20 ND 40 200

Reference 24 Microwave evaporation with ICP-MS analysis. Reference 22 Hot plate evaporation in Class 100 hood with ID-SSMS analysis. c Reference 23 Inverted Pyrex with filtered air evaporation with SSMS analysis. a

b

35


nated water through a membrane against osmotic pressure. The membrane preferentially allows water to pass, rejecting 90-99% of dissolved ions and particulates. The reverse osmosis system is generally used as a pre-treatment technique for water before it is further purified by either sub-boiled distillation (previously described) or ion-exchange. In the ion-exchange method the contaminated water passes through a column of resin. The resin is composed styrene-divinylbenzene copolymers engineered to have an affinity for either cations or anions. The resins exchange hydrogen and hydroxyl ions for the charged metal contaminates. This results in an exchange of the trace metal contamination for clean water.

Sub-boiling

distillation and ion exchange both produce water with low levels of trace metal contamination (Table 7).

Table 7: Trace metal contamination (pg/g) in high-purity water. Element Mg Al Ca Ti V Cr Mn a

Source 1a 42 147 157 8 11 5 2

Source 2b 90 700 110 ND ND 60 ND

Adapted from reference 23. 36

N

early all trace metals analysis procedures require some intervention by the analyst. Serious sample contamination can occur as a result of careless manipulation of the

sample. Touching or handling of equipment with bare hands can cause serious contamination. Dried skin contains 6 µg/g of Zn and 0.7 µg/g of copper, and sweat contains Na, K, Pb, Ca, and Mg. Hand lotions and creams contain Al, Zn, Ti, and Mg oxides, and other trace metal contaminates.

Traces of iron, copper,

gold, silver, platinum, and chromium can be deposited from watches, rings, and bracelets. Human hair contains on average 100 µg/g of zinc, 20 µg/g of copper and iron and 10 µg/g of lead. The use of cosmetics by the analyst can be a significant source of contamination, because of the metal oxides and other materials which are added to such products for

Chapter 5 The Analyst: A Source of Contamination

color and texture (Table 8). Some hair dyes and shampoos contain selenium and lead. Analysts should avoid the use of these products when carrying out analyses. In order to control analyst contamination the analyst must first be isolated from the samples.

The most fundamental

precaution is to wear gloves when performing procedures that require manipulation of the sample. The gloves must be impervious to skin oils and perspiration and must be powder-free. Clear polyvinyl chloride or polyethylene gloves are the best for routine handling of the samples. When working with concentrated acids, 37


Table 8: Trace metal contamination in cosmetics. Cosmetic Lipstick Eye Shadow Blush Mascara Foundation Face Powder

Trace Metals Present Bi, Zn, Fe, Mg, Ti, Mn Bi, Si, Fe, Mn, Ti, Al, Cr, Mg Si, Fe, Mg, Ti, Ca Na, Fe, Mg, Ti, Cr, Al Ti, Al, Zn, Fe Ti, Si, Bi, Fe, Zn, Mg, Ca

nitrile gloves are a good compromise between chemical resistance and cleanliness, and are often worn in conjunction with long-cuff vinyl gloves. One should remember that gloves are only as clean as the last thing touched, and must be changed on a routine basis to avoid the gloves’ becoming a contamination source. The best results are achieved when gloves are worn in conjunction with a head cover, shoe covers, and a laboratory coat. Protective garments (lab coats, head and shoe covers) worn during trace metal analysis should not be cotton or linen. These materials should be avoided because they produce a considerable amount of lint, which can contaminate the sample. The best materials for laboratory garments are nylon, Dacron® polyester, and Tyvek®. Garments made from these materials do not shed fibers, are lightweight, and are resistant to acids and other reagents. Garments used in trace metal analysis should also have all cut edges enclosed, and should be made without com38

Ch. 5: The Analyst—A Source of Contamination

ponents that are susceptible to corrosion, such as metal buttons and zippers. The analyst can also be responsible for cross-contamination from other samples or work being performed in the laboratory. If the analyst is not aware of the history of the laboratory containers and apparatus used for the analysis, contamination can occur as a result of using a vessel or container that was exposed to high concentrations of the analytes of interest. The use of paper towels in another part of the laboratory can introduce large quantities of airborne contamination into the atmosphere.

Protective garb—including gloves, head cover, shoe covers, and lab coat—is essential for best results.

Careless handling of the reagents can lead to contamination of the reagents, the sample, and the calibration standards. Contamination from these sources is very unpredictable. As an analyst, one must be constantly aware of one’s actions, and think about how those actions will affect the blank. Most importantly, the analyst must avoid those actions that tend to increase the blank, or whose effects on the blank are unknown.

39


40

l

SECTION TWO

l

Advanced Sample Preparation Techniques


42

T

he reliability of most methods of analysis depends on quantitative conversion of solids to homogeneous solutions. Conventional wet-sample preparation methods for

the decomposition of solid samples are usually carried out in vessels containing the sample and a large volume of decomposition reagent(s), typically 15 to 100 mL. This mixture is heated for several hours using hot plates, heating mantles, or ovens. Heating is terminated when the analyst decides that the decomposition of the sample is sufficiently complete.

This type of

open-vessel digestion has many drawbacks, which include the use of large volumes (and multiple additions) of reagents, potential for contamination of the sample by materials and laboratory environment, and the exposure of the analyst and the laboratory to corrosive fumes. The

high-pressure

Chapter 6 Closed-Vessel Microwave Digestion

closed-

vessel wet-ashing technique, originally described by Carius, is a more efficient way to decompose samples for analysis. The increased pressures allow temperatures beyond the atmospheric boiling point of the reagent to be reached. While Carius’s method improves the efficiency of decomposition, it also has several drawbacks. Analytes can be lost during the opening of the tube when the contents, under elevated pressure, are released suddenly. The analyst is frequently exposed to corrosive reagents, as well as flying pieces of glass, during the opening of the tube. 43


A Closer Look A closed-vessel microwave sample digestion machine is the most efficient means of converting solids to homogeneous solutions. (See pp. 81-82 for more information.)

Steel-jacketed Teflon lined bombs are now available to perform similar high-pressure and temperature reactions in thermal ovens. While the higher pressures and temperatures inside the closed vessels increase reaction rates, digestions may still require several hours due to the inefficient, “outside-in” heating mechanism. In addition, the high pressures involved with both of these conventional closed vessel methods tend to increase the safety risk of applying these techniques. Closed-vessel microwave decomposition, on the other hand, uses significantly different technology and fundamentally unique principles to accomplish sample decomposition. Heating by microwave energy is a “cold” in situ process, producing heat only when there is absorption or coupling of the microwave energy to the solution or microwave-absorbing objects. The two primary mechanisms for the absorption of microwave energy by a solution are dipole rotation and ionic conductance. In the dipole rotation mechanism, molecular dipoles align with the applied 44

Ch. 6: Closed-Vessel Microwave Digestion

electric field. Oscillation of the electric field results in forced molecular movement of the dipole molecules with the resulting friction heating the solution. At 2.45 GHz, the frequency of most laboratory microwave ovens, the dipoles align, then randomize 5 billion times a second. In the ionic conduction mechanism, the ionic species present in solution migrate in one direction or the other according to the polarity of the electromagnetic field. The accelerated ions meet resistance to their flow, and heating is a natural consequence.1

These two unique heating mechanisms

result in rapid heating of solutions, in comparison with conduction and convection. The heating is so fast that, in open vessels, vaporization alone can dissipate the excess energy. This results in solutions’ being able to sustain superheating above their normal boiling points by as much as 50C for water to 260C for acetonitrile.2,3 To use closed-vessel microwave decomposition effectively, one must understand the unique temperature and pressure relationships involved.

Gas pressures

inside microwave-closed vessels are not what would be predicted from the temperature of the liquid phase. The pressure inside a microwave vessel is dependent upon the volume of the vessel, and the temperature and composition of the gas phase. For example, when water is placed in a high-pressure steel-jacketed Teflon bomb and heated in a convection oven, an equilibrium vapor pressure is established.

This vapor pressure is dependent upon

the rate of evaporation and condensation of the water vapor.

When the temperature is increased, there is a 45


corresponding increase in the evaporation rate and a decrease in the condensation rate, because the vessel walls heat both the solution and gas phase. The decrease in condensation rate leaves more water in the vapor phase, increasing the internal pressure. In contrast, when water is heated to the same temperature in a microwave closedvessel, the internal pressure is significantly less than its steel-jacketed counterpart. This phenomenon is a direct result of the microwave heating mechanism used, and the materials of which the microwave vessel is composed. The microwave-closed vessel’s liner and outer casing are microwave transparent and have no insulating capacity. Thus, they remain relatively cool during the heating process. The cooler the vessel walls, the more efficient they will be at removing water molecules from the vapor phase.

The

increased condensation rate results in lower internal

pressures

at

higher

temperatures.

This microwave reflux action is illustrated in Figure 1. A more complex example of this pheFigure 1. Reflux conditions inside a microwave closed vessel. Adapted from reference 4. 46


nomenon is the closed-vessel microwave heating of nitric acid. Nitric acid is a polar and partially ionized mixture which heats rapidly in a microwave field. When heated, nitric acid partially decomposes into NOx gas. The gas phase inside a closed vessel becomes a mixture of NOx gas, nitric acid and water vapor. The pressure of nitric acid during closed-vessel microwave heating is still lower than predicted, even after taking into account the partial pressure of the NOx gas (Figure 2). The decrease in pressure results from the previously described reflux action and from the loss of the ionic conductance mechanism of microwave heating in the gas phase. The loss of the 70

60

Thermal Equilibrium Microwave Vessel

Pressure (atm)

50

40

30

20

10

0 121

130

133

150

165

170

190

192

Temperature (Celcius)

Figure 2. Internal pressure of nitric acid at different temperatures. 47

210

219


ionic conductance mechanism in the gas phase means the NOx gas does not convert microwave energy into heat efficiently, keeping the pressure increase associated with the heating of the gas to a minimum. This unique temperature and pressure relationship, found only in closed-vessel microwave heating, becomes more complex and unpredictable as additional reagents and samples are added to the solution. Also, the condensation rate varies with the microwave vessel materials and geometry, the liquid volume, the duration of heating, and the system’s ability to dissipate the excess energy. At the present time there is no conventional method of predicting the decrease in internal pressure associated with closedvessel microwave heating. This phenomenon is one of the reasons that pressure control is not applicable for standardizing microwave sample preparation methods. The use of closed vessels in microwave decomposition allows the reagents to be heated above their atmospheric boiling points. The higher temperatures achieved in the closed system give microwave decomposition a kinetic advantage over hot plate digestion, as described by the Arrhenius Equation: dlnk dT

=

Ea RT 2

Integration of this equation gives:

ln

(

k2 Ea 1 1 = – k1 2.303R T1 T2 48

)


In this expression k1 and k2 are rate constants for the reaction of interest at T1 and T2 respectively, Ea is the activation energy, and R is the ideal gas constant. These equations show that the reaction rate increases exponentially with increasing temperature. This translates into approximately a 100-fold decrease in the time required to carry out a digestion at 175 °C when compared to 95 °C digestion.4,6,7 In addition, because the mineral acid converts the microwave energy into heat almost instantaneously, rapid heating of the sample is achieved, further decreasing the reaction times. The effectiveness and safety of the microwave digestion procedure depends on the choice of digestion reagents. The most common reagents used for microwave digestion are nitric, hydrochloric, and hydrofluoric acids. Nitric acid is a strong oxidizing agent and liberates trace elements as highly soluble nitrate salts. The oxidizing properties of nitric acid are retained only in concentrated form and are lost when the acid is diluted below 2M. Nitric acid is used for the dissolution of metals and organic/biological materials. Some metals form insoluble oxide films (Al, Cr, Ti, Nb, Ta) and require the addition of a complex forming species in order to obtain complete dissolution. The nitrate ion forms weak complexes with tin, tungsten, and antimony, and these elements can become hydrolyzed and precipitate from solution. Hydrochloric and hydrofluoric acids are complexing acids. Many metal carbonates, peroxides and alkai hy49


droxides are dissolved by hydrochloric acid. Hydrofluoric acid is the only acid which will readily dissolve silica-based materials, and creates strong complexes with elements that form stable refractory oxides. Alkaline earths, lanthanide and actinide elements form insoluble or sparingly soluble complexes with hydrofluoric acid. These acids are often used in conjunction with nitric acid to dissolve metal alloys and noble metals.

Microwave Decomposition and Clean Chemistry Closed-vessel microwave decomposition has several unique characteristics that have led analysts to reduce the sources of error and contributions to the analytical blank that were previously obscured by lengthy sample preparation procedures. As described in Chapter 2, the laboratory air that comes in contact with the sample can deposit airborne contaminants into the sample. Open-vessel methods cannot prevent this type of contamination because the samples are continuously exposed to laboratory air currents. The use of closed vessels for microwave decomposition isolates the sample from laboratory air currents during the decomposition process, reducing airborne contamination. By preparing the samples for closed-vessel microwave dissolution under clean air conditions, we can isolate the sample, so that it will not come in contact with laboratory air. Thus, the possibility for contamination by airborne 50


particles in the laboratory air is eliminated. Similarly, by performing post sample processing, such as rinsing and dilution in clean environments, airborne contamination can be further reduced. Reagent contamination contributes to the concentration of the analytes present in the sample. The amount contributed is a function of the total quantity of reagent used. For example, using 50 mL of a reagent that contains a contaminant at the 100 ppb level will contribute 5 µg of that species as contamination. Unlike open-vessel methods that require large quantities of reagent due to evaporation losses, closed-vessel microwave methods continually reflux the reagents and do not allow the reagent vapors to escape. The reflux conditions inside a microwave-transparent vessel allow much smaller volumes of the reagent to be used, thus reducing the contamination from the decomposition reagents. The materials used for microwave vessels (fluorinated polymers and quartz) are chemically inert to most dissolution reagents, and provide a non-contaminating environment for sample preparation. TFM is preferred over PFA for the construction of microwave vessels because of its superior chemical and thermal resistance and lower blank levels (Tables 1 and 2). The skill of the analyst is perhaps the most difficult factor to evaluate. For open-vessel methods, the analyst must monitor the progress of the dissolution in several 51


Table 1: Trace element concentrations (µg/L) of acid blanks prepared from identical reagents in PFA and TFM vesselsa Element Al B Ba Bi Cd Co Cr Mo Pb Sb W Zn Zr

PFA 2.7 7.5 0.9 0.5 4.7 0.8 0.8 0.6 6.7 0.6 5.4 0.6 1.0

TFM ND 1.8 DL DL 0.1 DL 0.1 DL 0.03 DL DL 0.2 DL

Adapted from reference 8. ND = Not determined. DL = Concentration below instrument detection limit. a

vessels simultaneously. Compounding the problem, the temperature of each vessel varies with its position on the hot plate. The analyst must then subjectively decide whether dissolution is sufficiently complete, and remove the sample from the heat source. In microwave methods, the efficiency of the heating, coupled with the direct monitoring of the conditions inside the vessel, removes the subjective judgment of the analyst from the digestion pro52


Table 2: Analytical blank values for TFM microwave vessels after fifty digestions of environmental samplessa Element As B Ba Be Cd Ce Co Cu Ga a

µg/L 0.24 0.40 0.04 0.03 0.01 0.19 0.02 0.03 0.02

Element Li Mo Pb Sb Sc Se Sr Ta Zn

µg/L 0.09 0.26 0.15 0.01 0.04 0.01 0.05 0.01 0.31


cess. The automation and standardization of the sample preparation method through the use of closed-vessel microwave decomposition serves to reduce the impact of the analyst’s judgment upon much of the sample preparation process. The combined effect of closed-vessel microwave decomposition and clean chemistry techniques is shown in Tables 3 and 4, and Figure 3. The blank results show significant reduction in the contributions of the outside environment, reagents, and materials to the analytical blank level and overall measurement uncertainty. From Table 4, we can see that this combination of techniques allows trace metal analysis to be completed with increased 53


accuracy and precision. Figure 3 (on page 56) shows that closed-vessel microwave decomposition, coupled with clean chemistry techniques, is accurate and precise enough to prepare instrument calibration standards.

Table 3: Comparison of analytical blank results obtained from hot plate and microwave digestion of a certified reference soil.a Concentration expressed as µg/g. Analyte As Cd Cr Cu Pb Hg Ni Se Tl V Zn a b

Hot Plateb 0.204 ± 0.106 0.318 ± 0.122 3.35 ± 2.85 0.060 ± 0.020 0.171 ± 0.076 0.037 ± 0.004 0.375 ± 0.069 0.548 ± 0.264 0.028 ± 0.020 2.35 ± 0.45 2.92 ± 1.43

Microwaveb 0.074 ± 0.013 0.029 ± 0.019 0.104 ± 0.059 0.030 ± 0.017 0.040 ± 0.019 0.017 ± 0.007 0.060 ± 0.063 0.172 ± 0.022 0.028 ± 0.020 1.20 ± 0.51 1.66 ± 0.93

From reference 10. Error expressed as 95% confidence interval (n=4). 54


Table 4: Comparison of analytical blank results obtained from hot plate and microwave digestion of a certified reference soil.a Concentration expressed as µg/g. Analyte As Cd Cr Cu Pb Hg Ni Se Tl V Zn

Hot Plateb 12.3 ± 2.27 0.31 ± 0.09 68.8 ± 7.5 23.8 ± 2.7 10.8 ± 1.3 0.97 ± 0.14 63.4 ± 3.9 1.61 ± 0.34 0.29 ± 0.05 65.6 ± 8.8 113 ± 13.5

Microwaveb 17.6 ± 0.9 0.41 ± 0.06 123 ± 3 33.5 ± 1.2 17.5 ± 1.1 1.42 ± 0.10 83.2 ± 3.0 1.54 ± 0.33 0.63 ± 0.02 119 ± 6 102 ± 6.1

Certified Valuec 17.7 ± 0.8 0.38 ± 0.01 130 ± 4 34.6 ± 0.7 18.9 ± 0.5 1.40 ± 0.08 88 ± 5 1.57 ± 0.08 0.74 ± 0.05 112 ± 5 106 ± 3

From reference 10. Error expressed as 95% confidence interval (n=4). c 95% confidence interval as reported on certificate. a

b

55


Figure 3. ICP-MS calibration using microwave digested certified reference materials. The three calibration points represent separate digestions. Adapted from reference 10.

56

S

olutions submitted for trace elemental analysis often need to be evaporated prior to analysis. The most common reason for evaporation is to concentrate the sample, because

initial analyte levels are below the instrument detection limits. The other reason for evaporation is that the sample solution contains matrix elements that will present problems for analysis. Traces of hydrofluoric acid will etch the glass components of ICP and ICP-MS systems, releasing trace element contamination. Chlorine and fluorine form polyatomic ions that interfere with the ICP-MS analysis of many common elements. For example, 40 Ar35Cl+ interferes with 75As, and 35

Cl16O+ interferes with 51V. Obtaining

solutions

for

trace elemental analysis is often complicated due to the formation of volatile species during

Chapter 7 Microwave Evaporation

the evaporation process. As the number of solvent molecules decreases, the ions begin to recombine, and at dryness the residue will consist of a mixture of recombined salts. These salts will have an associated vapor pressure and boiling point that will vary with the oxidization state and counter anion (Table 5). The use of traditional heating methods to perform evaporations can lead to the loss of volatile analytes, because as the evaporation proceeds, the temperature of the evaporation vessel approaches the temperature of the heat source. At dryness these temperatures can be in excess of 150°C. 57


Table 5: Potentially volatile salts from solution. Element Arsenic Antimony Selenium Tin Vanadium Chromium

Volatile Salts AsCl3 AsF3 SbF5 SbCl5 SeCl4 SeF4 SnCl4 VCl4 VF5 CrF5

Boiling Point (ºC) 130.2 -63 150 79 170–196a 107.8 115 152 111.2 117

SeCl4 sublimes. Data taken from the following references: a

Dean, J.A., ed., Lange’s Handbook of Chemistry, 12th ed., New York: McGraw-Hill, 1979. David, R. Linde, ed., CRC Handbook of Chemistry and Physics, 71st ed., Cleveland: CRC Press, 1990.

In contrast to traditional evaporation methods, the unique heating mechanism exclusive to microwave-assisted heating allows for the retention of volatile analytes during evaporation. As the evaporation proceeds and the solvent is removed from the system, the matrix volume will be reduced. As the mass of the sample solution decreases, the amount of microwave energy absorbed decreases according to the following equation: 58

Ch. 7: Microwave Evaporation

A Closer Look Microwave-assisted heating methods allow for the retention of volatile analytes during evaporation. (See pp. 81-82 for more information.)

KCp∆Tm t In this equation P is the apparent absorbed power in watts, K is the conversion factor for calorie/s to watts, Cp is the specific heat, ∆T is change in temperature, m is the total mass of sample in the microwave, and t is irradiation time. This unique relationship between sample mass and energy absorption, along with the microwave vessels being microwave transparent, leads to a decrease in temperature as the sample approaches dryness. Lower temperatures at dryness decrease the potential for loss of volatile species, resulting in more complete recoveries for volatile analytes11,12 (Figures 4 and 5). Pabs=

Another factor that plays a significant role in evaporation losses is the oxidization state of the analyte of interest. It has been reported that even mild heating of an HF solution results in a 20% loss of Se(IV) and a 45% loss of As(III), and a losses of 65%-100% upon dryness. Losses of antimony during evaporation are due to 59


Figure 4. Evaporation recoveries of select elements from a 9:3 solution of HNO3/HCl. Initial solution concentration was 500 ppb.

the formation of poorly soluble compounds, and mercury is due to its being present in its elemental form. These losses can be prevented by converting these elements to higher oxidation states i.e. Se(VI), As(V), Sb(V), Hg(II). The use of traditional decomposition methods can not ensure a uniform oxidation state, due to matrix interferences and reaction rate limitations.14 This can be overcome by coupling closed-vessel microwave decomposition with microwave evaporation. The elevated temperatures and pressures decompose the matrix interferences and form stable complex ions. 60


Figure 5. Percent recovery of 2.5 ng spikes from 10ml of HCl. Uncertainties are expressed as 95% confidence intervals with n≥4. Adapted from references 4 and 13.

For example: Sb2O2(s) + HCL + HNO3 AsF3(aq) + HF + HNO3

SbCl6-(aq) AsF6-(aq)

The formation of these stable complex anions leads to complete retention of these traditionally volatile elements. (Figures 6 and 7).

61

Concentration (µg/g)


Element

Figure 6. Concentration of analytes in SRM 1566A (Oyster Tissue) following Microwave-Assisted Evaporation of the digestate compared with the certified total concentrations. Uncertainties are expressed as 95% confidence intervals with n≥3. Adapted from references 4 and 11.

62

Concentration (µg/g)


Element

Figure 7. Concentration of analytes in SRM 2710 (Montana Soil) following Microwave-Assisted Evaporation of the digestate compared with the certified total concentrations (* = noncertified concentration). Uncertainties are expressed as 95% confidence intervals with n≥3. Adapted from references 4 and 11.

63


64

A

nalytical chemists are being required to measure lower and lower levels of trace metals in samples. This frequently requires the analyst to make measurements near

the method detection limit, which usually results in decreased accuracy and precision. The advantages of using closed-vessel microwave decomposition in combination with clean chemistry techniques for lowering method detection limits have been discussed in Chapter 6 of this book, but sometimes these methods are not enough, and another approach is required to achieve the desired levels. Closed-vessel microwave digestion techniques require a minimum volume of 10 mL to achieve accurate temperature monitoring of the reaction conditions. Modern spectroscopic techniques

require

samples

submitted for analysis to have

Chapter 8 Lowering Your Method Detection Limits

acid concentration of 10% (v/v) or less. This requires samples that were digested with 10 mL of acid to be diluted by a factor of 100 or more. This problem can be overcome by increasing the sample size. This works well for samples that do not contain a large amount of organic material. For samples with high organic content this is usually not an option, because the secondary gases (CO2 and NOx) produced during the digestion can cause the vessel to vent when larger sample sizes (greater than 0.5 grams) are used. Recent advances in microwave chemistry have led to the development of vessel-inside-vessel technol65


ogy as a means to lower method detection limits for highly organic samples. Vessel-inside-vessel technology was developed through by the collective efforts of Milestone Inc, srl, and GmbH in the late 1990s. Vessel-inside-vessel technology uses a smaller secondary vessel inside the primary microwave vessel. The secondary vessel contains the sample and digestion reagents, and the primary vessel contains the 10 mL of solution required to achieve accurate temperature monitoring (Figure 8). This configuration reduces the amount of acid required for digestion to near stoimetric quantities, which reduces the dilution factor and increases the detection limit. The use of vessel-inside-vessel technology is also used for the processing of larger organic sample sizes. This is accomplished by controlling the reaction kinetics and lowering the pressure inside the microwave vessel. Controlling reaction kinetics is especially important when trying to digest large quantities (0.5 to 1.0 g) of organic material, because the potential for auto-catalytic decomposition increases. When the sample size is small (0.25 g), the heat released by the oxidization of the organic material does not cause a significant change in the temperature of the reaction mixture. As the sample size is increased, the heat released from the oxidation can cause the reaction mixture to heat faster that the programmed rate. The rise in temperature promotes further decomposition, which results in the microwave vessel’s venting (sometimes at 66

Ch. 8: Lowering Your Method Detection Limits

Figure 8 (a and b). Photo and schematic of vessel-inside-vessel technology.

67


pressures lower than its rating) due to the sudden increase in pressure resulting from the self-sustaining auto-catalytic decomposition (runaway reaction) of the sample (Figure 9). The use of vessel-inside-vessel technology helps to control these self-sustaining auto-catalytic reactions by providing a heat sink for the energy liberated during oxidization. This is accomplished by placing water in the outer microwave vessel. The water draws the heat away from the reaction mixture, slowing down the reaction kinetics and preventing a runaway reaction. (Fig 10). The amount of sample that can be safely digested is limited by the amount of pressure generated during the decomposition process. Current microwave vessel tech-

Temperature (ºC)

nology limits the internal pressure to 100 atm (1450 psi).

Figure 9. Example of runaway reaction during a closed-vessel microwave digestion. 68

Temperature (ºC)


Temperature (ºC)

Figure 10a. Digestion of 5 grams of fresh liver using conventional microwave decomposition. Note the runaway reaction.

Figure 10b. Digestion of 5 grams of fresh liver using vessel-inside-vessel technology. There is no runaway reaction. 69


For most organic samples, this limits the sample size to 0.5 to 0.7 grams. In order to digest organic samples larger than 0.7 grams, secondary reaction chemistry must be employed to lower the pressure during microwave digestion. This is accomplished with vessel-inside-vessel technology by adding H2O2 to the outer microwave vessel to convert NOx and CO2 into HNO3 and HCO3 respectively. Primary decomposition reaction (CH2)x + 2xHNO3

xCO2 + 2xNO + 2xH2O

Secondary reactions

2H2O2

∆

2H2O + O2

2NO + O2

2NO2

2NO2 + H2O

HNO2

∆

CO2 +H2O

HNO3 + HNO2

HNO3 + 2NO + H2O

H2CO3

The quantitative effect of this technique is shown in Figure 11. As you can see, this technique effectively doubles the amount of organic sample that can be digested using closed-vessel microwave technology. There is also no transfer of analytes from the inner vessel to the outer vessel. (Tables 6-8.)

70


Red line = Temperature

Blue line = Pressure

Figure 11. Quantitative effect of vessel-inside-vessel technology. These are overlay plots for the microwave digestion of polypropylene pellets. Line A represents 0.35 grams digested using conventional microwave digestion. Line B represents 0.35 grams digested with vessel-inside-vessel technology. Line C represents 0.60 grams digested with vessel-inside-vessel technology.

71


Table 6: Microwave digestion of cell culture media, using vessel-insidevessel technology followed by ICP-AES analysis. Results of six replicate samples. Element Ca Zn Fe Mo Mg K P Na

Average (µg/g) 454 2.22 9.44 7.03 428 19,552 7,527 79,461

% RSD 5.04 8.24 2.52 5.49 4.13 5.66 3.69 2.88

Table 7: Iron spike recoveries for microwave digestion of cell culture media, using vessel-inside-vessel technology followed by ICP-AES analysis. Spike Amount 10 µg 25 µg 50 µg

% Recovery 95.2 103 98.8

The average spike recovery is 99.2% with an RSD of 4.3%.

72


Table 8: Microwave digestion of SRM 1577A (Bovine Liver), using vesselinside-vessel technology followed by ICP-MS analysis. Element Cd Co Mo Pb V a

Measureda (ng/g) 427 ± 50 208 ± 1.9 3,540 ± 46 135 ± 14 104 ± 7

Certifieda (ng/g) 440 ± 60 210 ± 50 3,500 ± 500 135 ± 15 98.7 ± 1.6

Error expressed as 95% confidence interval.

73


74

l

SECTION THREE

l

Preparing Your Laboratory for Trace Analysis


76

T

he first two parts of this book provide the theory and background for low level trace metal analysis. This chapter outlines simple housekeeping and laboratory techniques

that can have an immediate effect on your trace metals determinations.

Laboratory Environment As previously discussed, the laboratory environment can significantly affect low level trace metal analysis. If clean air facilities are not available for any reason, then the sample preparation and analysis areas must be isolated from the main laboratory. Rooms with access to the outside via doors or windows, with high pedestrian traffic, or with

exposed

metal

surfaces

should be avoided. Any exposed

Chapter 9 Laboratory Housekeeping Techniques

metal surfaces should be painted with epoxy paint to prevent metal contamination. The work surfaces should be sealed with a polyurthane or polyacrylic finish (clear heavy-duty contact paper can also be used) to trap potential contamination. Airborne contamination can be controlled by covering air vents with additional filtration media and using a portable HEPA filter system that can be purchased from a home supply store. Floors, benches, and apparatus should be wiped down with D.I. water regularly with lint free towels to remove any dust accumulation. 77


Reagent Handling Clean reagents are essential to the success of trace metal analysis. All reagents and solutions used in trace metal analysis must be handled with care, in order to prevent contamination from outside sources. Reagents should be dedicated to trace metal analysis, and not be used for other laboratory procedures. Reagents bottles should be stored in a clean environment to prevent the accumulation of dust, which can be transferred to the sample during handling. For calibration solutions, a commercially available desiccator cabinet (without the dessicant) works well. For digestion acid bottles, large polyethylene bags used in layers work well for 100-1,000 mL Teflon bottles. No foreign objects should ever be introduced into the original reagent container, and the original reagent bottle should be opened as little as possible, and for the minimum amount of time. To minimize opening of the original container, a sub-sample should be transferred into a clean container for everyday use. Excess reagent should not be returned to the original bottle under any circumstances.

Preparation Procedures Preparation procedures for standards and samples should be designed to minimize the potential for contamination. Large dilutions of standards or samples should be prepared using a micropipette with disposable tips. This approach minimizes the number of dilutions, and prevents contamination from a glass pipette. When using this approach, the tips should be rinsed first with DI water to 78

Ch. 9: Laboratory Housekeeping Techniques

remove any dust, then with a 10% acid solution to remove any trace metal contamination, and last with two volumes of the final solution before use. Traditional glass volumetric apparatus should also be avoided, because they can contribute trace metal contamination to the solution. Plastic volumetric apparatus are available, but disposable graduated polypropylene centrifuge tubes work best. The graduation marks are accurate to 1% RSD or better (dilutions can be done by weight if greater accuracy is needed). These tubes can be used for preparation of calibration solutions as well as sample dilutions, and, because they are disposable, carry-over and cross-contamination problems are eliminated.

Miscellaneous As an analyst, one must be constantly aware of one’s actions, and think about how those actions will affect the blank. Most importantly, the analyst must avoid those actions that tend to increase the blank, or whose effects on the blank are unknown. For example, if you are wearing gloves to prevent contamination from your hands, but you touch a dusty or metallic surface, your glove will pickup trace metal contamination that can be transferred to the sample. The best approach to trace metal analysis is to have a routine and follow that routine for every analysis. If you have a set routine, it will help when a contamination problem occurs, because you can check each step until the source of the contamination is found. 79


80

T

his chapter lists some of the laboratory equipment and supplies you will need to perform trace metal analysis, and provides contact information for many reputable

suppliers of such equipment.

Instrumentation Acid Purification Milestone has developed a self-contained sub-boiling distillation system called duoPUR, which allows chemists to make their own high-purity acids.

The system consists of

two high-purity quartz distillation units.

Each unit contains

two infrared heating elements that supply a maximum power of 1,250 W, a water cooled condenser, bottle. is

and

a

collection

Chapter 10 Equipment and Supplies for Trace Metal Analysis

The distillation process

microprocessor-controlled,

allowing the user to set distillation times and power level. Distillation rates range from 10 to 200 mL per hour, depending on the power setting and the temperature of the cooling water. An added benefit of the system is that you can re-purify contaminated acids instead of downgrading them. 81


Closed-Vessel Microwave Decomposition Milestone’s Ethos Plus line of microwave digestion systems was designed with the trace metals chemist in mind.

The

system is constructed from corrosion-resistant stainless steel. The inner chassis is coated with five layers of electrosprayed PTFE for added corrosion resistance. The unique software interface allows precise temperature ramping. You simply create a temperature profile, and software modulates the power output to follow the defined heating profile. A variety of digestion rotors and accessories (evaporation, micro-inserts, etc.) are available to accommodate all your digestion needs. Ultra-trace Cleaning Milestone’s traceCLEAN system is a fully automated

acid

steam

cleaning system for trace metal analysis accessories. This self-contained system houses an accessories rack and an acid reservoir.

Once loaded,

the accessories are lowered into a sealed cham82

Ch. 10: Equipment and Supplies for Trace Metal Analysis

ber. Nitric acid is repeatedly evaporated and condensed throughout the chamber, thoroughly cleaning the accessories. The main benefit of the system is that any trace metal impurities that are present in the acid do not come in contact with the cleaned accessories. Clean Benches and Hoods Several companies offer clean benches and hoods for trace metal sample preparation.

Labconco offers the

Purifier

®

Trace Metals Work

Station, designed specifically for the demands of trace metals analysis. This work station is made of non-metallic components and provides a class 100 working environment. This enclosure is well-suited to applications involving corrosive chemicals, such as acid digestions, with its optional exhaust fan and PVC duct work. Several other models are also available. Terra Universal Inc. offers everything from free standing modular clean rooms to portable clean booths. All units offer HEPA filtration, with some units providing class 1-10 working conditions. Terra also offers static control via optional ionizers, and static dissipative materials neutralize any static imbalances that may exist in the work area.

83


High-Purity Water There are three major ers

of

water

manufacturhigh-purity equipment:

Millipore, Barnstead International, and Labconoco. All three manufacturers offer Free-standing modular cleanroom

a variety of systems,

from water softeners and reverse osmosis to high end polishers capable of producing Type 1 18.2 megohm-cm water.

Reagents and Standards Calibration and Quality Control Standards There are three major suppliers of calibration standards for trace metals analysis. They are SPEX CertiPrep Inc., Inorganic Ventures, and High Purity Standards. All three companies offer single, multi-element, and custom calibration standards for AA, ICP, and ICP-MS analysis. SPEX offers proficiency testing samples, and Inorganic Ventures and High Purity Standards offer certified reference materials for quality control procedures. Resource Technologies Corporation (RTC) and the National Instituted of Standards and Technology (NIST) provide “real world” quality control material for trace metals analysis. Samples range from soils and sediments 84


to glasses and metal alloys. Each sample comes with a certificate of analysis identifying major and minor trace element constitutes. Reagents If you are not going to make your own high-purity acid, then you will have to buy it. For general and low trace metal analysis, high-purity acids can be purchased from J.T. Baker, Fisher Scientific, and High Purity Standards. Each company offers double-distilled nitric and hydrochloric acids in high-purity Teflon bottles. For ultra-trace work, TAMA Chemicals offers TAMAPURE AA-100, guaranteed to contain trace metal impurities less than 100 pg/mL, and TAMAPURE AA-10, guaranteed to contain trace metal impurities less than 10 pg/mL. All bottles of high-purity acid come with a certificate of analysis.

Lab Supplies Gloves There are many different types and brands of gloves that are suitable for trace metals analysis. I prefer to use N-Dex Nitrile gloves (available from Fisher Scientific) as my first layer, and Oak powder-free vinyl gloves as my top layer. The Oak gloves are available in standard or long cuff. I use the long cuff because it offers better protection. The Oak gloves are available from Fisher Scientific or Terra Universal.

85


Plasticware As discussed in Section One, plastics are the materials of choice for trace metals work. Nalgene makes a wide variety of plastic bottles and accessories for trace metal analysis. One hard-to-find item that they make is plastic volumetric flasks. Sizes range from 50mL to 1,000mL, and flasks are made from either polypropylene or polymethylpentene. Another item that I have found I can’t live without is the free-standing graduated polypropylene tube. These tubes are available from Fisher Scientific, and are perfect for sample dilutions and storage. Miscellaneous A variety of clean chemistry accessories are available from Terra Universal and Fisher Scientific. Items range from tacky mats to lint free towels.

86


Suppliers’ Contact Information Milestone Inc. 25 Controls Drive Shelton, CT 06484 Telephone: 866-995-5100 Fax: 203-925-4241 Website: www.milestonesci.com Labconco Corporation 8811 Prospect Avenue Kansas City, Missouri 64132-2696 Phone: 816-333-8811 Fax: 816-363-0130 Website: www.labconco.com Terra Universial Inc. 700 N. Harbor Blvd. Anaheim, CA 92801 Phone: 714-526-0100 Fax: 714-992-2179 Website: www.terrauniversal.com Millipore Corporation 80 Ashby Road Bedford, MA 01730 Phone: 1-800-MILLIPORE Website: www.millipore.com

87


Barnstead International 255 Kerper Blvd. Dubuque, IA 52001 Phone: 1-800-446-6060 Fax: 536-589-0516 Website: www.barnsteadthermolyne.com SPEX CertiPrep 203 Norcross Avenue Metuchen, NJ 08840 Phone: 1-800-LAB-SPEX Fax: 732-603-9647 Website: www.spexcsp.com Inorganic Ventures 195 Lehigh Avenue Suite 4 Lakewood, NJ 08701 Phone: 1-800-569-6799 Fax: 732-901-1903 High Purity Standards 4741 Franchise Street Charleston SC 29423 Phone: 843-767-7900 Fax: 843-767-7906 Website: www.hps.net

88


Resource Technologies Corporation 2931 Soldier Springs Rd. Laramie, WY 82070 Phone: 1-800-567-5690 Fax: 307-745-7936 Website: www.RT-Corp.com National Institute of Standards and Technology Standard Reference Materials Program Building 202, Room 204 Gaithersburg, MD 20899 Phone: 301-975-6776 Fax: 301-948-3730 Website: www.nist.gov Mallinckrodt Baker, Inc. (J.T. Baker) 222 Red School Lane Phillipsburg NJ 08865 U.S.A. Phone: 1-800-582-2537 Fax: 908-859-6905 Website: www.JTBaker.com Fisher Scientific Phone: 1-800-766-7000 Website: www.fishersci.com

89


Tama Chemicals Distributed by Moses Lake Industries 8249 Randolph Rd Moses Lake, WA 98837 Phone: 509-762-5336 Fax: 509-762-5981 Nalge Nunc International 75 Panorama Creek Drive Rochester, NY 14625 Phone: 1-800-625-4327 Fax: 585-586-8987 Website: www.nalgenunc.com

90

For Section One 1.

Harris, D.C. Quantitative Chemical Analysis, 2nd Ed., W.H.

Freeman and Company, New York, 1987. p. 38. 2. J. Ruzicka and J. Stary, Substoichiometry in Radiochemical Analysis, Pergamon Press, New York, 1968 p. 54-58. 3. Murphy, T.J. “The Role of the Analytical Blank in Accurate Trace Analysis”, Proceedings from the Seventh Materials Research Symposium; U.S. Government Printing Office, Washington D.C., 1976. p. 509-539. 4. Howard, A.G. and Statham, P.J. Inorganic Trace Analysis: Philosophy and Practice, Wiley, New York, 1993 Murphy, T.J. “The Role of the Analytical Blank in Accurate Trace Analysis”, Proceedings from the Seventh Materials Re-

Appendix A References

search Symposium; U.S. Government Printing Office, Washington D.C., 1976. p. 509-539. 5. P.W. Morrison, ed. Contamination Control in Electronic Manufacturing, Van Nostrand Reinhold, New York, 1973, p. 245. 6. Specker, H.I. Z. Erzbergbau Metallhue 17, 132 (1964). 7. Gilbert, H. and Palmer, J.H. High Efficiency Particulate Air Filter Units, TID-7023 USAEC , Washington D.C., August 1961. 8. Moody, J.R., Analytical Chemistry 52, (1982) p. 1358A1376A. 9. Whyte, W. (Ed.) Cleanroom Design, Wiley, Chichester, 1991. 91


10.

Boutron, C.F., Fresenius Journal of Analytical Chem-

istry, 337 (1990) p. 482-491. 11. Austin, P.R., Encyclopedia of Clean Rooms, Bio-Cleanrooms and Aseptic Areas, 3rd ed., Acorn Industries, 2000. 12. Zief, M. and Mitchell, J.L. Contamination Control in Analytical Chemistry, Wiley, New York, 1976. 13. Patterson, C.C. and Settle, D.M. “The Reduction of Orders of Magnitude Errors in Lead Analysis of Biological Materials and Natural Waters by Evaluating and Controlling the Extent and Sources of Industrial Lead Contamination Introduced During sample collecting, Handling and Analysis”, Proceedings from the Seventh Materials Research Symposium; U.S. Government Printing Office, Washington D.C., 1976. p. 321-351. 14. Sulcek, Z. and P. Povondra. Methods of Decomposition in Inorganic Analysis, CRC Press, Boca Raton, FL, 1989. 15. Hetherington, G., Stephenson, G.W., Witerburn, J.A. Electronic Engineering, May 1969 p. 52 16. Stevens, M.P. Polymer Chemistry: An Introduction, Oxford, New York, 1990 17. Moody, J.R. and Lindstrom, R.M. Analytical Chemistry 49, (1977) p. 2264-2267. 18. Laxen, D.P.H and Harrison, R.M. Analytical Chemistry 53, (1981) p. 345-350. 19. Tschopel, P.,Kotz, L., Schulz, W., Veber, G. and Toelg G. Fresenius’ Z. Anal. Chem. 302, (1980) p. 1 20. Richter, R.C. Spectroscopy, 16(6), (2001) p. 21-24. 21. Kuehner, E.C., Alvarez, R., Paulsen, P.J., and Murphy, T.J. Analytical Chemistry, 44, (1977) p. 2050-2056 92


22. Moody, J.R and Beary E.S.Talanta, 29, (1982) p. 1003-1010. 23.

Dabeka, R.W, Mykytiuk, A., Berman, S.S., and Rus-

sell, D.S. Analytical Chemistry, 48, (1976) p. 1203-1207. 24. Richter, R.C., Link, D., Kingston, H.M., Spectroscopy, 15(1), (2000) p. 38.

For Section Two 1. Mingos, D. M. P.; Baghurst, D. R. Chem. Soc. Rev. 20, (1991) p. 1-47. 2. Hoopes, T.; Neas, E.; Majetich, G. Abstr. Pap. Am. Chem. Soc. 201, (1991) p. 231. 3. Baghurst, D. R.; Mingos, D. M. P. J. Chem. Soc., Chem Commun. (1992) p. 674. 4. Richter, R.C., Link, D.D., Kingston, H.M. Analytical Chemistry (2001) 73(1) p30A 5. Journal of Research of the National Bureau of Standards 30, (1943) p. 110. 6. Kingston, H. M.; Walter, P. J. Spectroscopy 7 (1992) p. 22-27. 7. Kubrakova, I. V.; Formanovskii, A. A.; Kudinova, T. F.; Kuz’min, N. M. J-Anal-Chem (1999) p. 460-465. 8. T. Noltner, Spectroscopy 5(4) 1989. 9. Visini and Rampazzo, Department of Environmental Science, University of Venice, Private Communication 10. Richter, R.C. and H.M. Kingston, Department of Chemistry, Duquesne University Private Communication. 93


11.

Link, D.L. and Kingston, H.M. Analytical Chemistry

(2000) 72(13) p. 2908-2913. 12. Link, D.L. and Kingston, H.M., Havirilla, G.L, Colletti, L.P. Analytical Chemistry (2002) 74(5) p. 1165-1170. 13. Han, Y., Kingston, H.M., Richter, R.C., Pirola, C., Analytical Chemistry (2001) 73(6) p. 1106-1111. 14. Sulcek, Z. and Povondra, P., Methods of Decomposition in Inorganic Analysis, CRC Press, Boca Raton, FL, 1989.

94


95


About the Author Dr. Robert C. Richter is a research professor at Chicago State University, where he develops and coordinates student research projects involving microwave-enhanced chemistry. He has taught analytical chemistry and sample preparation, for various universities and the American Chemical Society, since 1992. He has also served as a consultant on a wide range of corporate and governmental projects—developing microwave-assisted methods for a variety of applications, and conducting seminars on clean chemistry techniques. Dr. Richter is also Senior Applications Chemist for Milestone’s clean chemistry, microwave digestion, and microwave extraction product lines. In this role, he has helped academic and commercial laboratories across North America to adapt themselves for clean chemistry procedures and standards. Articles written by Dr. Richter have been published in Analytical Chemistry, American Laboratory News, Spectroscopy, Chemosphere, and Fresenius’ Journal of Analytical Chemistry, among other publications. He has contributed to papers presented at the Pittsburgh Conference, and at meetings of the American Chemical Society.

96

Clean Chemistry

Recommend Documents