CAN SPE, SPME, AND SBSE BE AUTOMATED?

272



Trace Environmental Quantitative Analysis, Second Edition



82. HOW DOES THE RAPID TRACE WORK?

The conventional SPE vacuum manifold apparatus requires only a source of mild

vacuum to enable liquids to pass through SPE sorbent beds by creating a partial

vacuum and hence a pressure drop across the SPE barrel type cartridge. This pressure

drop is enough to facilitate a steady flow rate of sample through the sorbent bed.

The Rapid Trace delivers sample to the sorbent bed using positive pressure combined

with a series of valves, mixing vessels, tubing, and pumps to achieve the important

aspects of automated SPE. A schematic of one Rapid Trace module is shown below

along with a description for each important part:122



1

2

3

4



5



8



7



6

9



10

11



12

13

Trace

Rapid

tation

Works

SPE



14



© 2006 by Taylor & Francis Group, LLC



Sample Preparation Techniques



Item



Part



1



Gas valve



2

3



Mixing

vessel

Syringe drive

and liquid

sensor



4



12-port valve



5



Thumb wheel

switch

Power light

Run light

Error light

Column

plunger



6

7

8

9



10



Start/stop

switch



11



Cannula



12



SPE column

turret

Service panel

Standard rack



13

14



273



Function

Activates gas drying of the column bed if the valve is plumbed into the unit

and the dry column step is used in software

Creates a mixed reagent when the add reagents to mixing vessel and mix

reagents in mixing vessel steps are used in the software

Performs all dispense and aspirate functions; the syringe draws up either

sample or reagent through the 12-port valve, as specified in the procedure,

and dispenses it back through the 12-port valve to the column plunger or

cannula as directed by the procedure

Directs the liquid flow from the syringe to one of the following ports as

specified in the procedure:

• Reagents 1–8 (8 different ports)

• Vent

• Cannula

• Column plunger

• Mixing vessel

Selects a module address of 0–9

Lights when the power is on

Lights when the module is running

Lights when an error occurs

Delivers the sample or reagent to the SPE column; the column plunger will

move into the column and place the reagent or sample directly onto the

column bed

Starts the procedure for a magnetically encoded rack, and stops a procedure

while it is running; pressing the start/stop switch while the procedure is

running will cause the procedure to stop; when using magnetically encoded

racks, modules must be connected to the controller

Access the sample test tube to add reagent to the sample, mix the sample, or

to draw the sample into the syringe and load it onto the SPE column, as

written in the procedure

Holds the 1- and 3-mL syringe barrel columns; holds up to 10 syringe barrel

SPE columns

Allows access to shuttle to change internal tubing

Holds up to ten 13 × 100 mm sample tubes and ten 12 × 75 mm collection

tubes; holds the magnets for magnetically encoded racks



Laboratories in general obtain the Rapid Trace Workstation in sets, with a set

containing either 5 or 10 modules. One 10-module set enables up to 10 samples to

be processed sequentially per module (rack), while 10 SPEs can be processed

simultaneously. Let us assume that it takes 30 min to process one SPE cartridge; 10

test tubes containing samples, located at position 1 on each rack, across all 10 racks,

could be simultaneously processed during the first 30 min. To process all 10 racks

that are full of test tubes means that 100 SPEs can be accomplished in 300 min, or

5 h. This is a significant gain in sample prep productivity.



© 2006 by Taylor & Francis Group, LLC



274



Trace Environmental Quantitative Analysis, Second Edition



83. HOW IS THE RAPID TRACE WORKSTATION

PROGRAMMED?

It depends on what you want to do. However, the order of conditioning the sorbent

bed, loading the sample, purging the cannula, rinsing the sorbent, drying the sorbent

to remove surface water, and, finally, eluting the sorbed analytes (collecting) is

critical.123 The table below serves to illustrate RP-SPE and is typical of how the

Rapid Trace is programmed:

Step No.



Step



Source



Output



mL



mL/min



Liq Sense



1

2

3

4

5

6

7

8

9

10



Condition

Condition

Load

Rinse

Purge-cannula

Rinse

Dry

Purge-cannula

Collect

Purge-cannula



MeOH

H2O

Sample

H2O

H2O

Vent

—

MeOH

Rin2

H2O



Org W

Aq W

Bio W

Bio W

Cannula

Aq W

Time =

Cannula

Fract 1

Cannula



2

2

1

2

6

0.1

0.5

6

2

6



12

12

2

8

30

8

Min

30

2

30



No

No

No

No

No

No

No

No

No

No



Note: W = waste; Org = organic; Aq = aqueous; Bio = biological; Fract = fraction.



Let us explain how to interpret this table.

Condition — Steps 1 and 2: Prepare the sorbent to effectively partition or

adsorb the hydrophobic analyte of interest from an aqueous sample matrix.

Always direct waste to the proper outputs, separating organic (Org W),

aqueous (Aq W), and biological fluids (Bio W). Condition at a recommended flow rate of 12 mL/min or 0.2 mL/sec.

Load — Step 3: To remove the entire liquid content from the test tube that

contains sample, use a volume that is 0.2 mL greater than the actual sample

volume. Load sample at a recommended flow rate of 2 mL/min or

0.03 mL/sec.

Rinse — Step 4: Rinse the column with water or alternate aqueous solution

and output this rinse water to the biological waste output. Most methods

require more than one reagent rinse step. Rinse recommended flow rate is

8 mL/min or 0.15 mL/sec.

Purge-Cannula — Step 5: Clean the cannula with water or an aqueous

reagent after the first rinse step. This will remove any remaining sample

matrix from the cannula. Do not use organic solvent when the sample matrix

is on the cannula.

Rinse — Step 6: Most methods require more than one reagent rinse step at

this point.

Dry — Step 7 (optional): To ensure that the dry step is effective, set the gas

pressure at the tank to 45 psi. Recommended dry time is 0.5 min. Set it

© 2006 by Taylor & Francis Group, LLC



Sample Preparation Techniques



275



longer as necessary. If your manual SPE method has a dry column step,

you must precede the dry step with a rinse step. If it does not have one,

use vent as the source. Preceding the dry step with the rinse step positions

the column into a waste station.

Purge-Cannula — Step 8: Clean the cannula with a reagent (usually MeOH

or the elution solvent) to strip remaining analyte from the cannula.

Collect — Step 9: Collect the fraction using a slow flow rate. For vacuum

manifold users, this is an important elution step. Recommended collect

flow rate is 2 mL/min or 0.03 mL/sec.

A schematic of the inner workings of the plunger/cannula is shown below:



84. ARE THERE EXAMPLES OF AUTOMATED SPE

OUT THERE?

Yes there are, but not as many published reports as you might think. Of envirohealth interest is a recent paper from the Centers for Disease Control and Prevention

(CDC) on incorporation of the Rapid Trace SPE Workstation as part of a faster

sample prep approach to isolating and recovering persistent organic pollutants

(POPs) from archived plasma samples. The method consisted of up-front RP-SPE

of selected OCs, followed by NP-SPE cleanup using silica gel, with subsequent

injection into an analytical HPLC column incorporating an analytical gel permeation

column. A fraction is obtained from the GPC column, which is subsequently injected

into a GC or GC-MS. Mean recoveries of the 13C-labeled internal quantification

standards ranged from 64 to 123% for the 11 monitored OCs.124 A semiautomatic

high-throughput extraction and cleanup method developed around the use of the

Rapid Trace has been recently reported and is a subsequent extension of the work

just cited. This paper shows how automated spiking of samples and automated

© 2006 by Taylor & Francis Group, LLC



276



Trace Environmental Quantitative Analysis, Second Edition



RP-SPE and NP-SPE can be coupled together while extending the method to polybrominated diphenyl ethers (PBDEs), PBBs, and PCBs in human serum.125



85. HOW ARE THE METHODS CATEGORIZED

FOR TRACE INORGANICS ANALYSIS?

A recent compilation of EPA methods organizes the numerous analytical methods

for inorganics analysis according to the following six major categories:126

1. Trace metals identified by flame (FlAA) and by graphite furnace (GFAA)

atomic adsorption spectrophotometry

2. Trace metals identified by inductively coupled plasma-atomic emission

spectrophotometry (ICP-AES)

3. Trace metals identified by inductively coupled plasma-mass spectrometry

(ICP-MS)

4. Mercury identified by cold vapor atomic absorption spectrometry

5. Cyanide (total and amenable)

6. Inorganic carbon (total carbon less organic carbon)

Not listed in these categories are analytical methods for the principal inorganic

anions derived from strong acids that are prevalent in groundwater: chloride, bromide, nitrite, nitrate, phosphate, and sulfate. These analytes are currently measured

routinely by the application of either ion chromatography (IC) or specific colorimetric procedures following the conversion of the anion to a colored complex.

Oxyhalides such as the bromate ion have been found in chlorinated drinking water.

IC methods have been developed for various oxyhalide ions in recent years. Water

that is free of dissolved organics and heavy metal can be directly injected into ion

chromatographs or filtered if particulates are present. Aqueous samples that contained dissolved biomatter or heavy metal ions pose severe challenges to IC because

the columns employed in the technique are susceptible to column fouling.

In this chapter, we will discuss the basis of sample preparation for five of the

six categories listed above and focus on the determinative steps for all six in Chapter

4. Total organic carbon (TOC) is a combustion technology in which aqueous samples

can be injected directly without the need for sample preparation; therefore, we will

not discuss it any further in this chapter. Let us start with a discussion of the principles

of sample prep with respect to trace metals.



86. HOW DO YOU PREPARE AN ENVIRONMENTAL

SAMPLE TO MEASURE TRACE METALS?

Sample preparation for the determination of trace concentration levels of the many

priority pollutant metals is strongly connected to the nature of the determinative

technique. Historically, FlAA was first used to measure metals. The more sensitive

GFAA technique followed. Along about the same time as GFAA was being developed,



© 2006 by Taylor & Francis Group, LLC



Sample Preparation Techniques



277



ICP-AES came along. ICP-AES afforded the opportunity to measure more than one

metal in a sample at a time, the so-called multielement approach. In recent years,

the development of ICP-MS has carried trace metal analysis to significantly lower

IDLs and introduced the opportunity to identify and quantitate the various elemental

isotopes.

The metals of greatest interest to TEQA are listed in Table 3.15 along with the

author’s comments on the chemical and toxicological nature of each. The design of

instrumentation for either FlAA, GFAA, ICP-AES, or ICP-MS requires that the

sample be introduced as a liquid. These systems easily accommodate aqueous

solutions in contrast to gas chromatographs, which require, for most of the polysiloxane capillary columns, the injection of a nonaqueous liquid such as an organic

solvent. Aqueous samples that contain inorganics, in contrast to aqueous samples

that contain organics, can be introduced into an FlAA, GFAA, or ICP without

removal of the analyte from its sample matrix. Chemically, metals might exist in

the environment as ions in one or more oxidation states or partially or wholly chelated

to ligands of various sorts. They may be bound or complexed to soil/sediment

particulates and therefore cannot be easily released via a liquid–solid extraction or

leaching. Some metals form the structural composition of solid matrices derived

from the environment, such as aluminosilicates in clay. In these cases, decomposition

of the sample removes the organic portion of the matrix. Decomposition methods

include the following:

1. Combustion with oxygen with and without fluxes

2. Digestion with acids

Both methods for decomposing the sample matrix require heat, and both have

benefited from replacement of resistive heating (e.g., laboratory hot-plate equipment)

with microwave heating.127

The following procedure has been used by this author to prepare a sample of

fly ash for determination of priority pollutant metals.128 Assume that the ash has

been obtained from a previous combustion procedure.

1. Weigh 0.2 g of sample into a 100-mL beaker. Record the weight to the

nearest 0.001 g using an analytical balance.

2. Add 5 mL of concentrated nitric acid (HNO3) and 5 mL of concentrated

hydrochloric acid (HCl).

3. Place a watch glass over the beaker and digest at medium heat for 60 min.

4. Evaporate to dryness.

5. Add 5 mL of concentrated HNO3 and evaporate to dryness.

6. Add 1 mL of concentrated HNO3 and warm.

7. Add 1 mL of distilled deionized water and warm. Filter into a 25-mL

volumetric flask.

8. Cool and dilute to the mark using 1% HNO3. This gives exactly 25 mL

of an aqueous sample containing the solubilized metal ion.

9. Aspirate into a previously calibrated FlAA and record the absorbance.



© 2006 by Taylor & Francis Group, LLC



278



Trace Environmental Quantitative Analysis, Second Edition



TABLE 3.15

Toxicity, Common Oxidation States, and Chemical Forms of the Most

Environmentally Significant Chemical Elements

Toxicity

Element

Mg

Al

Cr

Mn

Fe

Co

Ni

Cu

Zn

Ga

As

Se

Mo

Ag

Cd

In

Sn

Sb

Te

Ba

Pt

Au

Hg

Tl

Pb

Bi



A



B



C



D



X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X



No. of Oxidation States



Common Chemical Forms



1

1

9

11

9

7

7

6

2

3

3

6

8

4

2

3

2

3

7

1

5

6

2

2

2

4



Mg(2+)

Al(3+)

Cr(3+), Cr(VI)

Mn(IV), Mn(VII)

Fe(2+), Fe(3+)

Co(2+)

Ni(2+)

Cu(1), Cu(2+)

Zn(2+)

Ga(III)

As(III), As(V)

Se(IV), Se(VI)

Mo(VI)

Ag(+)

Cd(2+)

In(III)

Sn(II), Sn(IV)

Sb(III), Sb(IV)

Te(IV)

Ba(2+)

Pt(IV)

Au(III)

Hg(+), Hg(2+)

Tl(+)

Pb(2+), Pb(IV)

Bi(III)



Note: Toxicity categories are as follows:

A = major toxic metals with multiple effects

B = essential metals with potential for toxicity

C = metals with toxicity related to medical therapy

D = minor toxic metals

Source: Data from Klasser, C.D., Ed., Casarett & Doull’s Toxicology: The Basic Source of

Poisons, 5th ed., McGraw-Hill, New York, 1996; Elmsely, J., The Elements, Oxford University

Press, Oxford, 1989.



© 2006 by Taylor & Francis Group, LLC



Sample Preparation Techniques



279



Notes:

1. Be sure to choose the appropriate wavelength and oxidizer gas for each

metal of interest. The aqueous solutions used as calibration standards

should be prepared in 1% HNO3.

2. For every batch of samples that has been digested, one blank, one matrix

spike, and one matrix spike duplicate should also be prepared. For the

matrix spike and matrix duplicate, designate one sample from the batch

for these.

Sample matrices from a biological origin such as blood, urine, and serum can

be merely diluted with water, dilute nitric acid, or a dilute surfactant such as Triton

X-100 and aspirated directly into the flame for FlAA or placed directly into the

graphite tube for GFAA. With respect to GFAA, a plethora of matrix modifiers have

been developed over the years to deal with spectral and chemical types of interference.129 Spectral interferences arise when the absorption of emission of an interfering

species either overlaps or lies so close to the analyte absorption or emission that

resolution by the monochromator becomes impossible. Chemical interferences result

from various chemical processes occurring during atomization that alter the absorption

characteristics of the analyte.130 Spectral interferences refer to the presence of concomitants that affect the quantity of the source light that reaches the detection system,

whereas chemical interferences reduce the analyte absorbance signal by interactions

with one or more concomitants compared to standards without concomitants.131



87. WHAT IS MATRIX MODIFICATION IN GFAA?

Because matrix modification is one aspect of sample preparation, even though it is

most often accomplished automatically in GFAA, we will discuss it here. For

example, National Institute for Occupational Safety and Health (NIOSH) Method

7105 recommends a matrix modifier that consists of a mixture of ammonium dihydrogen phosphate, magnesium nitrate, and nitric acid for the determination of airborne Pb. Because the graphite furnace can be viewed as a chemical reactor whereby

the sample with its matrix is placed on a graphite platform (the L’vov platform,

discussed in Chapter 4) and heated to a very high temperature, reactions can take

place that involve both the metal analyte of interest and sample matrix components.

The concept of matrix modification, from the sample prep perspective, is to add

to the sample a chemical reagent that will cause a desirable chemical reaction or

inhibit an undesirable reaction.132 For metals that tend to volatilize, one can add a

modifier that reduces analyte volatility by increasing the volatility of the matrix.

Consider the determination of Pb in highly salted aqueous samples such as seawater.

Seawater contains appreciably elevated levels of chloride salts. Adding an ammonium ion to the seawater, followed by heating the sample to a high temperature,

causes the following reaction to occur:

Cl(−aq ) + NH +aq )

4

© 2006 by Taylor & Francis Group, LLC



heat NH 4 Cl(g )

→



280



Trace Environmental Quantitative Analysis, Second Edition



This reaction removes chloride ions from the sample while minimizing the loss of

Pb as the more volatile PbCl2.

Harris133 discusses the findings of Styris and Redfield,134 who studied the effect

of magnesium nitrate on the determination of Al. At high temperature, MgO is

formed and steadily evaporates. This maintains a steady vapor pressure of MgO in the

GFAA tube. The presence of MgO serves to keep Al as the oxide by establishing

the following equilibrium:

3 MgO + 2 Al(S)





→

← 3 Mg ( g ) + Al2 O 3(s )





When most of the MgO has evaporated, the equilibrium begins to shift to the left

as Al2O3 is converted to elemental Al. This reaction serves to delay the Al from

evaporating until a higher temperature is reached.

Butcher and Sneddon135 describe the work of Schlemmer and Welz,136 who

investigated the use of Pd and Mg nitrates as matrix modifiers in the determination

of nine metallic elements. They showed that higher pyrolysis temperatures could be

used, compared to no modifier or using other common modifiers. Thus, we have

shown how additions to the sample matrix led to improved performance in GFAA.

With respect to FlAA and ICP, the addition of so-called matrix modifiers developed for GFAA serves no useful purpose. FlAA and ICP use nebulization into an

oxidizing-reducing high-temperature source to introduce a liquid into the flame and

plasma, respectively, and because of this, both techniques require that samples have

a low dissolved solids content so as to prevent clogging. ICP is essentially free from

most spectral and chemical interferences due to the extremely high temperature of

the plasma (8000 to 10,000˚C), whereas these interferences are prevalent in FlAA

and serve to influence the IDL for a given metal.



88. HOW DO I PREPARE A SOLID WASTE, SLUDGE,

SEDIMENT, BIOLOGICAL TISSUE, OR SOIL SAMPLE?

EPA Method 3050 from the SW-846 series of methods involves solubilizing a solid

sample with acids and peroxide and removing the insoluble residue by filtration.

EPA Method 3051 is a microwave-assisted acid digestion procedure. EPA Method

200.3 is applicable to the preparation of biological tissue samples prior to using

atomic spectrometry for quantifications of Al, Sb, As, Ba, Be, Cd, Ca, Cr, Co, Cu,

Fe, Pb, Li, Mg, Mn, Hg, Mo, Ni, P, K, Se, Ag, Na, Sr, Tl, Th, U, V, and Zn, and an

outline of the sample prep procedure is as follows:137

•



•



Place up to a 5-g subsample of frozen tissue into a 125-mL Erlenmeyer

flask. Any sample-spiking solutions should be added at this time and

allowed to be in contact with the sample prior to the addition of acid.

Add 10 mL of concentrated nitric acid and warm on a hot plate until the

tissue is solubilized. Gentle swirling of the sample will aid in this process.



© 2006 by Taylor & Francis Group, LLC



Sample Preparation Techniques



•



•



•

•

•



281



Increase the temperature to near boiling until the solution begins to turn

brown. Cool sample, add an additional 5 mL of concentrated nitric acid,

and return to the hot plate until the solution once again begins to turn brown.

Cool sample, add an additional 2 mL of concentrated nitric acid, return

to the hot plate, and reduce the volume to 5 to 10 mL. Cool sample, add

2 mL of 30% hydrogen peroxide, return sample to the hot plate, and reduce

the volume to 5 to 10 mL.

Repeat the previous step until the solution is clear or until a total of 10 mL

of peroxide has been added.

Cool the sample, add 2 mL of concentrated hydrochloric acid, return to

the hot plate, and reduce the volume to 5 mL.

Allow the sample to cool and quantitatively transfer to a 100-mL volumetric flask. Dilute with DDI, mix, and allow any insoluble material to

separate. The sample is now ready for either ICP-AES, ICP-MS, or GFAA.



89. WHAT ARE EPA’S MICROWAVE DIGESTION METHODS?

EPA Methods 3015A (applicable to an aqueous sample such as groundwater) and

3051A (applicable to soils, sediments, sludges, and oils) utilize advances in microwave heating technology. Microwave heating significantly reduces the more labor

intensive hot-plate techniques described earlier in this chapter.138 These recently

developed methods enable environmental samples to be digested so that a quantitative determination of up to 26 metals can be made. These metals are listed as their

chemical symbols as follows:



Al

Sb

As

Ba

Be



B

Cd

Ca

Cr

Co



Cu

Fe

Pb

Mg

Mn



Hg

Mo

Ni

K

Se



Ag

Na

Sr

Tl

V



Zn



Scheme 3.9, adapted for Method 3051A, is a flowchart that outlines the procedure and logic to prepare soil, sediments, sludges, and oils for trace metals analysis.

To summarize Method 3052A:138

A representative sample of up to 0.5 g is extracted and/or dissolved in 10 mL of

concentrated nitric acid or 9 mL of concentrated nitric acid and 3 mL of concentrated

hydrochloric acid for 10 min using microwave heating with a suitable laboratory unit.

The sample and acid(s) are placed in a fluorocarbon polymer or quartz vessel or vessel

liner. The vessel is sealed and heated in a microwave unit. After cooling, the vessel

contents are filtered, centrifuged or allowed to settle and then diluted to volume and

analyzed by the appropriate determinative method.



Safety considerations are paramount in digestion involving microwave heating

technology. It cannot be overemphasized enough that a suitable laboratory microwave

© 2006 by Taylor & Francis Group, LLC