Clean Chemistry

A practical guide to . . .

Clean Sample Preparation for Trace Metals Analysis

by Dr. Robert Richter

Techniques for the Modern Laboratory

A practical guide to . . .

Clean Sample Preparation for Trace Metal Analysis

by Dr. Robert Richter

Techniques for the Modern Laboratory

CLEANHEMISTRY

© 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 Press25 Controls Dr..Shelton, CT 06484

CONTENTS

Section 1:Controlling Contamination in Your Laboratory

Chapter 1:The Analytical Blank

Chapter 2:Laboratory Environment

Chapter 3:Materials for Trace Analysis

Chapter 4:Trace Analysis Reagents

Chapter 5:The Analyst: A Source of Contamination

Section 2:Advanced Sample Preparation Techniques

Chapter 6:Closed-Vessel Microwave Digestion

Chapter 7:Microwave Evaporation

Chapter 8:Improving Method Detection Limits

Section 3:Preparing Your Laboratory for Trace Analysis

Chapter 9:Laboratory Housekeeping Techniques

Chapter 10:Equipment & Supplies for Trace Metal Analysis

Appendix A:References

13

15

21

33

37

43

57

65

77

81

91

5

Clean Chemistry: Techniques for the Modern Laboratory

6

Introduction


8

Introduction

O ver the past several decades, the need for trace and ultra-

trace 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 aware-

ness 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 under-

stand 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

control. The second part deals

with the preparation of the ho-

mogeneous 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 spe-

cial 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 analyti-

cal blank has little or no effect on

the accuracy of the result when

the analyte concentration of the

sample is several orders of magni-

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.

Chapter 1The Analytical

Blank

��

��

��

��

13

On the other hand, as the analyte concentration ap-

proaches 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 undis-

turbed, 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-

nates can also be introduced

to the laboratory from external

sources, through the ventila-

tion 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 Manhat-

tan Project to remove fissionable particles from the air.7

HEPA filters have an efficiency of 99.97% for particles 0.3

Chapter 2Laboratory Environment

15

µm and larger. They effec-

tively remove bacteria, pol-

len, fly ash, and dust. Some

common contaminates, such

as tobacco smoke, are not

removed because their parti-

cles are smaller than 0.3 µm.2

Particulate contamination in

a work area is also minimized

by using a laminar air flow

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 contamina-

tion for trace metal analysis is to construct a clean labora-

tory, 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 operat-

ed clean room, sub-parts-per-trillion measurements can be

16


HEPA air filter

made. More information

on clean rooms and clean

room design can be found

in references 8–11.

A more practical ap-

proach 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

standard units are good

for drying containers and apparatus after cleaning, pro-

tecting 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 plexi-

glass box that is purged with filtered nitrogen or air to

remove air particles. The exhausted air is vented to a con-

17

Ch. 2: Laboratory Equipment

Laminar flow hood

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 deposi-

tion 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 equiva-

lents 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 con-

tent 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

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, espe-

cially 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 contamina-

tion 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.

Silic

on

Ru

bb

er

Teflo

nSi

lico

nV

inyl

Poly

ure

than

ePo

lyst

yren

eM

ylar

Ray

on

Go

ld, P

lati

nu

m

Bra

ss, S

ilver

Har

d R

ub

ber

Wo

od

Stee

lPa

per

Alu

min

um

Silk

Wo

ol

Nyl

on

Hu

man

Hai

rM

ica

Gla

ssH

um

an S

kin

Air

Most Negative (-) Most Positive (+)

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

Ch. 2: Laboratory Equipment

19

20


Chapter 3Materials for

Trace Analysis

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 cop-

per wire mesh, which will con-

tribute Cu, Fe, Co, Cr, and Ni. The

ashless filter paper often used to

filter samples contains trace met-

al contamination on the order of

1ppm, and the filtration assem-

blies can also impart contamina-

tion to the sample. For trace

metal analysis it is best to avoid

these practices when possible. Solid samples should be homog-

enized 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 GlassThe most common material used for laboratory contain-

ers 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 bo-

rosilicate glass, because they will gradually solubilize the

glass according to the following equation4:

2X NaOH + (SiO2)x ]XNa2SiO

3 + XH

20

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+

QuartzAn 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 labora-

tory 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 oxida-

tion and electrical fusion (Type IV) of SiCl4. Both of these

methods produce quartz with low trace metal contamina-

tion.3,4,15 The typical trace metal impurity levels for the

various types of quartz are shown in Table 1.


22

Ch. 3: Materials for Trace Analysis

23

Elem

ent

Bo

rosi

licat

e G

lass

Qu

artz

(Ty

pe

I)Q

uar

tz (

Typ

e II

)Q

uar

tz (

Typ

e II

I)

Al

Maj

or

7468

< 0

.25

BM

ajo

r4

0.3

0.1

Ca

1,00

016

0.4

< 0

.1

Cr

0.1

ND

b0.

03

Cu

11

< 1

Fe3,

000

71.

5<

0.2

K3,

000

6<

10.

1

Li7

1N

D

Mg

600

4N

DN

D

Mn

1,00

01

0.2

< 0

.02

Na

Maj

or

95

< 0

.1

Sb2.

90.

30.

10.

1

Tab

le 1

: Tra

ce e

lem

ent

con

cen

trat

ion

s (µ

g/g

) in

bo

rosi

licat

e g

lass

an

d v

ario

us

typ

es o

f qu

artz

.a

a Ad

apte

d fr

om

refe

ren

ces

3 an

d 1

5.b N

D =

No

t D

etec

ted

.

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 manu-

facturing 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.

PolyethyleneThere 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

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).

PolypropylenePolypropylene 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

Table 2:Trace element concentrations (µg/g) in some polymersa

Element LDPE HDPE PP PFA FEP PTFE

Al 0.5 30 55

Ca 800

K > 5,000 > 600 0.2 0.23

Na 1.3 1.5 4.8 0.1 0.4 0.16

Sb 0.005 0.2 0.6

Ti 5 60

Mn 0.01 0.2 0.6

Zn 520

a Adapted from reference 17.

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-fluro-

carbon (PFA), fluorinated ethylene propylene (FEP).

Cleaning MethodsPicking a material with low trace metal impurities

doesn’t guarantee low blank levels. Trace metal impuri-

ties 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 behav-

ior 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 poly-

ethylene, a 48-hour soaking with 10% nitric acid is effective

for initial and routine cleaning.17, 18


26

27


Table 3:Impurities (ng/cm2 of surface) leached from plastic containers in one week with 1:1 HNO

3/Water. HDPE and LDPE were leached at

room temperature while FEP was heated to 80ºC.a

Element LDPE HDPE FEP

Pb 0.7 2 2

Tl 1 < 1 < 1

Ba 2 < 0.2 4

Te < 0.5 0.2 0.6

Sn < 0.8 1 1

Cd 0.2 0.2 0.4

Ag NDb 0.2 < 8

Sr 0.2 1 0.2

Se 3 0.4 0.2

Zn 2 8 4

Cu 2 0.4 2

Ni 0.5 1.6 2

Fe 3 3 20

Cr 0.8 0.2 0.8

Ca 10 0.6 80

K 2 2 2

Mg 0.7 0.6 8

Al 1 1 6

Na 8 10 6

Total 38 50 148

a Adapted from reference 17.b ND = Not Detected.

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 LDPE HDPE FEP

Pb 18 0.6 2

Tl 3 < 0.6 < 1

Ba 0.3 1 2

Te 0.7 NDb 2

Sn < 0.8 < 1 1

Cd 0.2 0.2 0.6

Ag ND ND < 6

Sr 0.2 0.2 < 1

Se < 0.3 0.4 0.8

Zn 1.0 9 4

Cu 0.7 1 6

Ni 0.3 0.8 0.8

Fe 1.0 1 16

Cr 0.3 0.8 4

Ca 0.8 60 2

K 0.7 1 1.6

Mg 0.7 0.4 1.0

Al 10 4 4

Na 42 6 2

Total 38 50 148

a Adapted from reference 17.b ND = Not Detected.


28

Steam cleaning with nitric or hydrochloric acid is also

a very effective cleaning method for containers and appa-

ratus (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 sur-

face 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 re-

agent 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 contamina-

tion is removed.

Figure 2. Acid steam cleaning system.


29

��

��

��

��

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 con-

tainer, 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


TFM

Tefl

on

Ves

sel

Q

uar

tz V

esse

l

Elem

ent

Aci

d L

each

ing

bSt

eam

Cle

anin

gb

Aci

d L

each

ing

bSt

eam

Cle

anin

gb

Al

287

± 4

625

8 ±

24

398

± 2

832

7 ±

18

Mg

289

± 2

223

2 ±

15

441

± 5

634

7 ±

26

Na

≤12

1≤

121

1190

± 3

5060

8 ±

67

Fe≤

474

≤47

4≤

474

≤47

4

Ni

≤55

≤55

≤55

≤55

Co

≤56

≤56

≤56

≤56

Cu

144

± 3

911

7 ±

12

170

± 1

510

9 ±

9

Cr

≤85

≤85

176

± 5

7≤

85

Cd

≤72

≤72

≤72

≤72

Tl≤

261

≤26

1≤

261

≤26

1

Pb≤

57≤

57≤

57≤

57

Zn

995

± 8

0≤

876

1,64

0 ±

1,0

001,

005

± 1

24

Tab

le 5

: Co

mp

aris

on

of h

igh

-tem

per

atu

re a

cid

leac

hin

g c

lean

ing

vs.

acid

ste

am c

lean

ing.

Tra

ce m

etal

co

nta

min

atio

n (p

g/g

) in

5%

HN

O3 b

lan

ks p

rep

ared

aft

er c

lean

ing

are

list

ed b

elo

w. T

he

acid

leac

hin

g w

as

per

form

ed a

t 18

0ºC

wit

h m

ixtu

re o

f HC

L an

d H

NO

3. Th

e st

eam

cle

anin

g p

erfo

rmed

wit

h H

NO

3 on

ly.a

a Ad

apte

d fr

om

refe

ren

ce 2

0.b E

rro

r exp

ress

ed a

s o

ne

stan

dar

d d

evia

tio

n (n

= 3

).


32


Chapter 4Trace Analysis

Reagents

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 re-

agent 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 con-

centration 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 impuri-

ties from reagents. Sub-boiling

distillation has been shown to

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 car-

ried 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

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 contami-

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.)


34

Table 6:Trace metal contamination (pg/g) in nitric acid produced by

sub-boiling distillation in a quartz still.

Element Source 1a Source 2b Source 3c

Mg 42 90 400

Al 147 700 900

Ca 157 110 400

Ti 8 ND 800

V 11 ND < 3

Cr 5 60 100

Mn 2 ND 7

Fe 210 350 800

Co 1 ND < 10

Ni 23 80 30

Cu 21 50 200

Zn 49 60 80

Se 1 60 ND

Sr 1 20 ND

Ag 2 6 ND

Cd 2 20 < 30

Sn 9 20 ND

Ba 4 20 20

Tl < 1 60 ND

Pb 3 30 40

B ND ND 200

a Reference 24 Microwave evaporation with ICP-MS analysis.b 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.

Ch. 4: Trace Analysis Reagents

35

nated water through a membrane against osmotic pres-

sure. 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-treat-

ment technique for water before it is further purified by

either sub-boiled distillation (previously described) or

ion-exchange. In the ion-exchange method the contami-

nated 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 Source 1a Source 2b

Mg 42 90

Al 147 700

Ca 157 110

Ti 8 ND

V 11 ND

Cr 5 60

Mn 2 ND



36

Chapter 5The Analyst: A Source of

Contamination

N early all trace metals analysis procedures require some

intervention by the analyst. Serious sample contamina-

tion 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 con-

tamination, because of the metal

oxides and other materials which

are added to such products for

color and texture (Table 8). Some

hair dyes and shampoos con-

tain 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 impervi-

ous 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

nitrile gloves are a good compromise between chemical

resistance and cleanliness, and are often worn in conjunc-

tion 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 con-

taminate the sample. The best materials for laboratory

garments are nylon, Dacron® polyester, and Tyvek®. Gar-

ments 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 com-

Cosmetic Trace Metals Present

Lipstick Bi, Zn, Fe, Mg, Ti, Mn

Eye Shadow Bi, Si, Fe, Mn, Ti, Al, Cr, Mg

Blush Si, Fe, Mg, Ti, Ca

Mascara Na, Fe, Mg, Ti, Cr, Al

Foundation Ti, Al, Zn, Fe

Face Powder Ti, Si, Bi, Fe, Zn, Mg, Ca


38

Table 8:Trace metal contamination in cosmetics.

ponents that are susceptible to

corrosion, such as metal buttons

and zippers.

The analyst can also be re-

sponsible for cross-contamina-

tion from other samples or work

being performed in the labora-

tory. 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 ex-

posed 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 con-

tamination into the atmosphere.

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.

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

best results.

Ch. 5: The Analyst—A Source of Contamination

39

40


l SECTION TWO l

Advanced Sample Preparation

Techniques


42

Chapter 6Closed-Vessel

Microwave Digestion

T he reliability of most methods of analysis depends on

quantitative conversion of solids to homogeneous solu-

tions. 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 decomposi-

tion 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 decom-

position 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 en-

vironment, and the exposure of

the analyst and the laboratory to

corrosive fumes.

The high-pressure 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 sud-

denly. The analyst is frequently exposed to corrosive reagents, as

well as flying pieces of glass, during the opening of the tube.

43

Steel-jacketed Teflon lined bombs are now available

to perform similar high-pressure and temperature reac-

tions 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 conven-

tional closed vessel methods tend to increase the safety

risk of applying these techniques. Closed-vessel micro-

wave 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 mi-

crowave-absorbing objects. The two primary mechanisms

for the absorption of microwave energy by a solution are

dipole rotation and ionic conductance. In the dipole rota-

tion mechanism, molecular dipoles align with the applied

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.)

44


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 di-

poles 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 su-

perheating 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 composi-

tion 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

Ch. 6: Closed-Vessel Microwave Digestion

corresponding increase in the evaporation rate and a de-

crease 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 closed-

vessel, 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 capac-

ity. 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 phe-

Figure 1. Reflux conditions in-side a microwave closed vessel. Adapted from reference 4.


46

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

121 130 133 150 165 170 190 192 210 219

70

60

50

40

30

20

10

0

Thermal Equilibrium

Microwave Vessel

Temperature (Celcius)

Pre

ssu

re (a

tm)

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 pres-

sure 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


47

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 condensa-

tion 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 closed-

vessel 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 decomposi-

tion allows the reagents to be heated above their atmo-

spheric 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:

Integration of this equation gives:

dlnk Ea

dT RT2=

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 exponen-

tially 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 di-

gestion.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/bio-

logical 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 complex-

ing acids. Many metal carbonates, peroxides and alkai hy-


49

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, lan-

thanide 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 can-

not prevent this type of contamination because the sam-

ples 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 concen-

tration 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 meth-

ods that require large quantities of reagent due to evapora-

tion losses, closed-vessel microwave methods continually

reflux the reagents and do not allow the reagent vapors to

escape. The reflux conditions inside a microwave-trans-

parent 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 (fluori-

nated polymers and quartz) are chemically inert to most

dissolution reagents, and provide a non-contaminating en-

vironment 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

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 moni-

toring of the conditions inside the vessel, removes the

subjective judgment of the analyst from the digestion pro-


52

Table 1:Trace element concentrations (µg/L) of acid blanks prepared

from identical reagents in PFA and TFM vesselsa

Element PFA TFM

Al 2.7 ND

B 7.5 1.8

Ba 0.9 DL

Bi 0.5 DL

Cd 4.7 0.1

Co 0.8 DL

Cr 0.8 0.1

Mo 0.6 DL

Pb 6.7 0.03

Sb 0.6 DL

W 5.4 DL

Zn 0.6 0.2

Zr 1.0 DL

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

cess. The automation and standardization of the sample

preparation method through the use of closed-vessel mi-

crowave 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

Table 2:Analytical blank values for TFM microwave vessels after fifty

digestions of environmental samplessa

Element µg/L Element µg/L

As 0.24 Li 0.09

B 0.40 Mo 0.26

Ba 0.04 Pb 0.15

Be 0.03 Sb 0.01

Cd 0.01 Sc 0.04

Ce 0.19 Se 0.01

Co 0.02 Sr 0.05

Cu 0.03 Ta 0.01

Ga 0.02 Zn 0.31


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.


54

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 Hot Plateb Microwaveb

As 0.204 ± 0.106 0.074 ± 0.013

Cd 0.318 ± 0.122 0.029 ± 0.019

Cr 3.35 ± 2.85 0.104 ± 0.059

Cu 0.060 ± 0.020 0.030 ± 0.017

Pb 0.171 ± 0.076 0.040 ± 0.019

Hg 0.037 ± 0.004 0.017 ± 0.007

Ni 0.375 ± 0.069 0.060 ± 0.063

Se 0.548 ± 0.264 0.172 ± 0.022

Tl 0.028 ± 0.020 0.028 ± 0.020

V 2.35 ± 0.45 1.20 ± 0.51

Zn 2.92 ± 1.43 1.66 ± 0.93

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

55


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 Hot Plateb Microwaveb Certified Valuec

As 12.3 ± 2.27 17.6 ± 0.9 17.7 ± 0.8

Cd 0.31 ± 0.09 0.41 ± 0.06 0.38 ± 0.01

Cr 68.8 ± 7.5 123 ± 3 130 ± 4

Cu 23.8 ± 2.7 33.5 ± 1.2 34.6 ± 0.7

Pb 10.8 ± 1.3 17.5 ± 1.1 18.9 ± 0.5

Hg 0.97 ± 0.14 1.42 ± 0.10 1.40 ± 0.08

Ni 63.4 ± 3.9 83.2 ± 3.0 88 ± 5

Se 1.61 ± 0.34 1.54 ± 0.33 1.57 ± 0.08

Tl 0.29 ± 0.05 0.63 ± 0.02 0.74 ± 0.05

V 65.6 ± 8.8 119 ± 6 112 ± 5

Zn 113 ± 13.5 102 ± 6.1 106 ± 3

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

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


56

Chapter 7Microwave Evaporation

S olutions submitted for trace elemental analysis often need

to be evaporated prior to analysis. The most common rea-

son 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 contamina-

tion. Chlorine and fluorine form polyatomic ions that interfere

with the ICP-MS analysis of many

common elements. For example, 40Ar35Cl+ interferes with 75As, and 35Cl16O+ interferes with 51V.

Obtaining solutions for

trace elemental analysis is often

complicated due to the forma-

tion of volatile species during

the evaporation process. As the

number of solvent molecules de-

creases, 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 tempera-

tures can be in excess of 150°C.

57

Table 5:Potentially volatile salts from solution.

Element Volatile Salts Boiling Point (ºC)

Arsenic AsCl3

130.2

AsF3

-63

Antimony SbF5

150

SbCl5

79

Selenium SeCl4

170–196a

SeF4

107.8

Tin SnCl4

115

Vanadium VCl4

152

VF5

111.2

Chromium CrF5

117

a SeCl4 sublimes.

Data taken from the following references:

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-as-

sisted 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


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 ir-

radiation 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).

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 dry-

ness. Losses of antimony during evaporation are due to

Pabs= KCp∆Tmt

A Closer Look

Microwave-assisted heat-ing methods allow for the retention of volatile analytes during evaporation. (See pp. 81-82 for more information.)

59

Ch. 7: Microwave Evaporation

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.

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

60


For example:

Sb2O2(s) + HCL + HNO3 SbCl6-(aq)

AsF3(aq) + HF + HNO3 AsF6-(aq)

The formation of these stable complex anions leads to

complete retention of these traditionally volatile elements.

(Figures 6 and 7).

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.

61


Co

nce

ntr

atio

n (µ

g/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

Element

Co

nce

ntr

atio

n (µ

g/g

)

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 fre-

quently 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 dis-

cussed 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 con-

ditions. Modern spectroscopic

techniques require samples

submitted for analysis to have

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 con-

tain a large amount of organic material. For samples with high

organic content this is usually not an option, because the sec-

ondary 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 technol-

Chapter 8Lowering Your

Method Detection Limits

65

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 micro-

wave vessel. The secondary vessel contains the sample

and digestion reagents, and the primary vessel contains

the 10 mL of solution required to achieve accurate tempera-

ture 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 ma-

terial, because the potential for auto-catalytic decomposi-

tion 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

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

67

Ch. 8: Lowering Your Method Detection Limits

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 oxidi-

zation. 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-

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

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

Tem

per

atu

re (º

C)


68

Tem

per

atu

re (º

C)

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

Tem

per

atu

re (º

C)

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 em-

ployed 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 + 2xHNO

3 xCO

2 + 2xNO + 2xH

2O

Secondary reactions

The quantitative effect of this technique is shown in

Figure 11. As you can see, this technique effectively dou-

bles 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.)

2H2O2 2H2O + O2

2NO + O2 2NO2

2NO2 + H2O HNO3 + HNO2

∆

HNO2 HNO3 + 2NO + H2O∆

CO2 +H2O H2CO3


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

72


Table 6:Microwave digestion of cell culture media, using vessel-inside-vessel technology followed by ICP-AES analysis. Results of six

replicate samples.

Element Average (µg/g) % RSD

Ca 454 5.04

Zn 2.22 8.24

Fe 9.44 2.52

Mo 7.03 5.49

Mg 428 4.13

K 19,552 5.66

P 7,527 3.69

Na 79,461 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 % Recovery

10 µg 95.2

25 µg 103

50 µg 98.8

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

73


Table 8:Microwave digestion of SRM 1577A (Bovine Liver), using vessel-

inside-vessel technology followed by ICP-MS analysis.

Element Measureda (ng/g) Certifieda (ng/g)

Cd 427 ± 50 440 ± 60

Co 208 ± 1.9 210 ± 50

Mo 3,540 ± 46 3,500 ± 500

Pb 135 ± 14 135 ± 15

V 104 ± 7 98.7 ± 1.6

a Error expressed as 95% confidence interval.

74


l SECTION THREE l

Preparing Your Laboratory for

Trace Analysis

76


Chapter 9Laboratory

Housekeeping Techniques

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 determi-

nations.

Laboratory EnvironmentAs 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 prepara-

tion and analysis areas must be

isolated from the main labora-

tory. Rooms with access to the

outside via doors or windows,

with high pedestrian traffic, or

with exposed metal surfaces

should be avoided. Any exposed

metal surfaces should be painted

with epoxy paint to prevent metal

contamination. The work surfaces should be sealed with a poly-

urthane or polyacrylic finish (clear heavy-duty contact paper can

also be used) to trap potential contamination. Airborne contami-

nation 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 HandlingClean 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 ac-

cumulation 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 mini-

mum 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 ProceduresPreparation procedures for standards and samples

should be designed to minimize the potential for contami-

nation. 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 ap-

proach, the tips should be rinsed first with DI water to

78


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 contam-

ination to the solution. Plastic volumetric apparatus are

available, but disposable graduated polypropylene centri-

fuge 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 dilu-

tions, and, because they are disposable, carry-over and

cross-contamination problems are eliminated.

MiscellaneousAs 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

Ch. 9: Laboratory Housekeeping Techniques


80

Chapter 10Equipment and

Supplies for Trace Metal Analysis

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.

InstrumentationAcid 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 distilla-

tion units. Each unit contains

two infrared heating elements

that supply a maximum power

of 1,250 W, a water cooled

condenser, and a collection

bottle. The distillation process

is 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 contami-

nated 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 cor-

rosion-resistant stainless steel.

The inner chassis is coated with

five layers of electrosprayed

PTFE for added corrosion re-

sistance. The unique software

interface allows precise tem-

perature 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 trace-

CLEAN system is a fully

automated acid steam

cleaning system for trace

metal analysis accesso-

ries. This self-contained

system houses an acces-

sories rack and an acid

reservoir. Once loaded,

the accessories are low-

ered into a sealed cham-


82

ber. Nitric acid is repeatedly evaporated and condensed

throughout the chamber, thoroughly cleaning the acces-

sories. 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 prepara-

tion. Labconco offers the

Purifier® Trace Metals Work

Station, designed specifically

for the demands of trace met-

als analysis. This work station

is made of non-metallic com-

ponents 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 stand-

ing 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 neutral-

ize any static imbalances that may exist in the work area.

Ch. 10: Equipment and Supplies for Trace Metal Analysis

83

High-Purity Water

There are three

major manufactur-

ers of high-purity

water equipment:

Millipore, Barnstead

International, and

Labconoco. All three

manufacturers offer

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 StandardsCalibration and Quality Control Standards

There are three major suppliers of calibration stan-

dards 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 Ven-

tures 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

Free-standing modular cleanroom

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 met-

al 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 SuppliesGloves

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 protec-

tion. 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 polymethyl-

pentene. Another item that I have found I can’t live with-

out 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 InformationMilestone 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

Appendix AReferences

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: Phi-

losophy and Practice, Wiley, New

York, 1993 Murphy, T.J. “The Role

of the Analytical Blank in Accu-

rate Trace Analysis”, Proceedings

from the Seventh Materials Re-

search Symposium; U.S. Govern-

ment 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. 1358A-

1376A.

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-Clean-

rooms 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 Control-

ling the Extent and Sources of Industrial Lead Contamina-

tion 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 Decomposi-

tion 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 Chemis-

try 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 Stan-

dards 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.

Appendix A: References

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 Decomposi-

tion in Inorganic Analysis, CRC Press, Boca Raton, FL, 1989.


94

95

Appendix A: References

About the Author

Dr. Robert C. Richter is a research professor at Chi-

cago 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, Spec-

troscopy, 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

Documents

Clean Chemistry