Collecting and Preparing

Samples
Module 5
Systematic Errors
Errors affecting accuracy:
1. Sampling errors
2. Method errors
3. Measurement errors
4. Personal errors
Systematic Errors
1. Sampling Errors
• Introduced if improper or non-representative sample are
collected and analyzed
• Non-homogenous sample leads to inaccurate results
• Extrapolating results from non-homogenous sample to
population introduces determinate error
• These can be mitigated by collecting true representative
samples
 Systematic Errors
1. Sampling Errors

Trana et al. (2015). J. of Experimental & Marine Biology and ecology, 462, pp: 8-13
Collecting and Preparing Samples
• Not uncommon when we use an analytical method for the first
time that results are inaccurate and imprecise.
• Assuming that sources of indeterminate and determinate errors
were minimized why might a carefully designed/used method
give such poor results?
• One possibility is that errors associated with the sample were
not carefully considered.
Collecting and Preparing Samples
Section Goals
• Consider how collecting samples and preparing them for
analysis can affect the accuracy and precision of our results
• Selecting a sample inevitably introduces a source of
determinate error
• If a sample does not accurately represent the population from
which it is drawn, then the analysis that is otherwise carefully
conducted will yield inaccurate results
• To minimize sampling errors, we must endeavor to collect the
right/representative sample
Designing a Sampling Plan
• A plan that ensures that a representative sample is collected

Five elements to consider when designing a sampling plan:

1. Where within the target population should samples be collected?
2. What type of sample should be collected?
3. What is the minimum amount of sample needed for the analysis?
4. How many samples should be analyzed?
5. How can the overall variance be minimized?
Designing a Sampling Plan
• A plan that ensures that a representative sample is collected

Five elements to consider when designing a sampling plan:

1. Where within the target population should samples be collected?
2. What type of sample should be collected?
3. What is the minimum amount of sample needed for the analysis?
4. How many samples should be analyzed?
5. How can the overall variance be minimized?
Where to sample the target population
• Sampling errors occur when a sample composition is not
identical to that of the population from which it is drawn

Two types of samples

1. Homogenous
2. Heterogenous
Where to sample the target population
1. Homogenous
• Analyte(s) of interest evenly distributed in matrix.
• Individual sample(s) can be taken without regard to sampling
errors
• Determinate sampling errors are insignificant

2. Heterogenous
• Analyte(s) of interest unevenly distributed in matrix
• Analyte(s) can vary in time (temporal) or space (spatial)
• In this scenario determinate sampling errors can be significant
Where to sample the target population
Approaches to sampling

1. Random sampling
2. Judgmental sampling
3. Systematic sampling
4. Convenience sampling
Where to sample the target population
Random Sampling
• Samples collected at random from the target population
• Makes no assumption about the target population making it the
least biased approach to sampling
• Random samples require more time and expense than other
sampling methods because a greater number of samples are
needed to characterize or be representative of the target
population
Whereto sample the target population
Judgmental Sampling
• These are samples collected from the target population using
available information about the analyte(s) distribution within the
population
• This is basically the opposite of random sampling because
judgmental sampling is selective
• Because assumptions about the target population are included
in the sampling plan, judgmental samples are more biased than
random samples
Where to sample the target population
Systematic Sampling
• Random and judgmental sampling represent extremes in bias
and the number of samples needed to accurately characterize
the target population.
• Systematic sampling falls in between these extremes.
• Samples collected from the target population at regular intervals
in time (temporal) and space (spatial).
Where to sample the target population
Systematic Sampling
Where to sample the target population
Convenience Sampling
• Samples collected from the target population because they are
easily obtained
• Cost, expedience and accessibility are the primary factors
Designing a Sampling Plan
• A plan that ensures that a representative sample is collected

Five elements to consider when designing a sampling plan:

1. Where within the target population should samples be collected?
2. What type of sample should be collected?
3. What is the minimum amount of sample needed for the analysis?
4. How many samples should be analyzed?
5. How can the overall variance be minimized?
What type of sample to collect
• After determining where/when to collect a sample, the next step
is designing a sampling plan to decide what type of sample to
collect.

Three common methods to obtain samples

1. Grab sampling
2. Composite sampling
3. In situ sampling
What type of sample to collect
Grab sample
• A single sample removed from the target population at a given
time and space
• Represents a snapshot of target population
• Easily adapted to any of the sampling schemes described
• If target population is homogenous and uniform in time and
space a set of grab samples collected at random can be used
• A systematic sampling plan using grab samples can be used to
characterize a target population whose composition varies over
time and space.
What type of sample to collect
Composite Sample
• A set of grab samples that are combined to form a single
sample
• After thoroughly mixing, the composite sample can be analyzed
• Analyzing a single composite sample instead of many individual
grab samples provides an appreciable saving in time and cost
• Neither grab/composite samples can be used to continuously
monitor a time-dependent change in target population
What type of sample to collect
In-situ Sample
• A sample taken within the population without physically
removing the sample
• In-situ sampling in which an analytical sensor is placed directly
in the target population
• Allows for continuous monitoring without removing individual
grab samples
How much sample to collect
• To minimize sampling errors, a randomly collected grab sample
must be of an appropriate size
• If too small, composition of the sample may differ from the
target population resulting in significant sampling error
• If too large, composite may require more time and money to
collect and analyze with little improvement in sampling error
How many samples to collect
• It is obviously impractical to analyze the entire population
• How many samples are required to achieve a desired maximum
sampling error
• If samples drawn from the target population are normally
distributed, then the following equation describes the
confidence interval for the sampling error

𝑡𝑠𝑠 𝑡 2 𝑠𝑠2 n = # of samples

𝜇 = 𝑥ҧ ± 𝑛=
𝑛 (𝜇 − 𝑥)ҧ 2
t = statistical factor
Ss = SD for sampling
Minimizing overall variance
• Overall variance of any analysis is a combination of the
variance due to the method and the variance due to sampling
• We can improve the variance due to sampling by collecting
more samples of proper size.
• Increasing the number of times we analyze each sample
improves the variance due to the method
Implementing the Sample Plan
Involves three steps:
1. Physically removing the samples from its target population
2. Preserving the sample
3. Preparing the sample for analysis

• Once a sample is withdrawn from a target population, there is a

danger that it may undergo a chemical and/or physical change
• Problem if this happens is that sample no longer represents the
target population
Implementing the Sampling Plan
• Samples are therefore preserved in some way once in the
laboratory
• Samples then need to be processed prior to analysis
Preparing the samples for analysis
• In most real-world situations, we must separate our target
analyte(s) from more than one component in a complex matrix
• How then can we remove our target analyte(s) from other
interfering components?
• Interfering component (interference/interferent): any chemical
that contributes to the measured signal of the analyte(s)
Separating Analyte from Interferents
• Let’s revisit equation for a calibration curve:

y = kx + c
• Where,
y = sample signal
k = analyte sensitivity
x = analyte concentration
c = background signal
Separating Analyte from Interferents
• For argument purposes, let’s assume that we do not have any
interferences and that our background signal is negligible
y=k×x
Ssamp= kA × CA

• Let’s now say we have an interferent (I) with a determined

signal
kI × C I
Separating Analyte from Interferents
Ssamp= kA × CA

• Then this relationship now becomes:

Ssamp = (kA × CA) + (kI × CI)

• Selectivity: If the method responds primarily to the analyte of

interest and is little affected by other substances

• A method’s selectivity is determined by the relative difference in

its sensitivity toward the analyte and interferent
Separating Analyte from Interferents
• Selectivity coefficient, KA,I, characterizes the selectivity of the
method:

Ssamp = (kA × CA) + (kI × CI)

= (kA × CA) + (KA,I × kA × CI)
= kA × (CA + KA,I × CI)
Separating Analyte from Interferents
Ssamp = kA × (CA + KA,I × CI)

• An interferent will not pose a problem as long as:

CA >> KA,I × CI

• When an interferent contribution is significant, any accurate

analysis must begin by separating the analyte from the
interferent.
Last Class – October 15
• Module 5!
• Samples, Samples, Samples!
• How many?
• How big?
• From where?
• Preservation
• Preparation – selectivity and separation

• Module 4 HW – Achieve – Due October 25th at 11:59 PM

How many samples to collect
The standard deviation for sampling is +/- 1.5% for an analysis.
How many samples are required to achieve a percent relative
sampling error of +/- 0.90%? How does
this work??

𝑡𝑠𝑠 𝑡 2 𝑠𝑠2 n = # of samples

𝜇 = 𝑥ҧ ± 𝑛=
𝑛 (𝜇 − 𝑥)ҧ 2
t = statistical factor
Ss = SD for sampling

Can use %
relative error
Degrees of Confidence Confidence Confidence Confidence Confidence Confidence Confidence
freedom level (%): level (%): level (%): level (%): level (%): level (%): level (%):
50 90 95 98 99 99.5 99.9
1 1.000 6.314 12.706 31.821 63.656 127.321 636.578
2 0.816 2.920 4.303 6.965 9.925 14.089 31.598
3 0.765 2.353 3.182 4.511 5.841 7.453 12.924
4 0.741 2.132 2.776 3.747 4.604 5.598 8.610
5 0.727 2.015 2.571 3.365 4.032 4.773 6.869
6 0.718 1.943 2.447 3.143 3.707 4.317 5.959
7 0.711 1.895 2.365 2.998 3.500 4.029 5.408
8 0.706 1.860 2.306 2.896 3.355 3.832 5.041
9 0.703 1.833 2.262 2.821 3.250 3.690 4.781
10 0.700 1.812 2.228 2.764 3.169 3.581 4.587
15 0.691 1.753 2.131 2.602 2.947 3.252 4.073
20 0.687 1.725 2.086 2.528 2.845 3.153 3.850
25 0.684 1.708 2.060 2.485 2.787 3.078 3.725
30 0.683 1.697 2.042 2.457 2.750 3.030 3.646
40 0.681 1.684 2.021 2.423 2.704 2.971 3.551
60 0.679 1.671 2.000 2.390 2.660 2.915 3.460
120 0.677 1.658 1.980 2.358 2.617 2.860 3.373
∞ 0.674 1.645 1.960 2.326 2.576 2.807 3.291
Need to use iteration to solve!
• We cannot get the correct t value without knowing how many
samples!
• Begin by assuming infinite samples.

𝑡 2 𝑠𝑠2 1.960 2 (1.5)2

𝑛= 2
= 2
= 7.52 ≈ 8
(𝜇 − 𝑥)ҧ (0.90)

𝑡 2 𝑠𝑠2 2.365 2 (1.5)2

𝑛= 2
= 2
= 15.54 ≈ 16
(𝜇 − 𝑥)ҧ (0.90)
Need to keep going!!!!
Finding the number of samples
𝑡 2 𝑠𝑠2 2.131 2 (1.5)2
𝑛= 2
= 2
= 12.61 ≈ 13
(𝜇 − 𝑥)ҧ (0.90)

𝑡 2 𝑠𝑠2 2.179 2 (1.5)2

𝑛= 2
= 2
= 13.18 ≈ 13
(𝜇 − 𝑥)ҧ (0.90)

• Once two successive calculations give the same number of

trials, the problem is solved.
Theory of Separation Efficiency
• The goal of any analytical separation is to remove either the
analyte or the interferent(s) from the sample matrix.

• To bring about a separation, there must be a significant

difference in either the physical or chemical property of the
analyte and interferent.
Theory of Separation Efficiency
• A separation efficiency, also referred to as recovery, is:
𝐶𝐴
𝑅𝐴 =
(𝐶𝐴 )𝑜

where RA = recovery (typically represented as %)

CA = amount of analyte after separation
(CA)o = amount of analyte initially present
Theory of Separation Efficiency
Q1. 5 ng of DDT is added to water and extracted by liquid-liquid
extraction using hexane as the extracting solvent. After
extraction, the amount of DDT left behind in water is 1 ng. What
is the separation efficiency of the method?

5 ng = 100%
4 ng = x%
x = 80%
Separation efficiency = 80%
Theory of Separation Efficiency
• In the same manner, the recovery of the interferent, RI is
defined as:
𝐶𝐼
𝑅𝐼 =
(𝐶𝐼 )𝑜

• The degree of separation given by the separation factor, S I,A is a

measure of the effectiveness of a separation at separating an
analyte from an interferent.
Theory of Separation Efficiency

𝐶𝐴 𝐶𝐼
𝑅𝐴 = 𝑅𝐼 =
(𝐶𝐴 )𝑜 (𝐶𝐼 )𝑜

• So, separation factor, SI,A: 𝑅𝐼 𝐶𝐼 × (𝐶𝐴 )𝑜

𝑆𝐼,𝐴 = =
𝑅𝐴 (𝐶𝐼 )𝑜 × 𝐶𝐴

• Ideal separation, RA= 1; RI= 0, SI,A = 0

 Theory of Separation Efficiency
Question: An analysis to determine the concentration of Cu in an
industrial plating bath uses a procedure for which Zn is an interferent.
When a sample containing 128.6 ng/L Cu is carried through a separation
to remove Zn, the concentration of Cu remaining is 127.2 ng/L. When a
134.9 ng/L solution of Zn is carried through the separation, a
concentration of 4.3 ng/L remains. Calculate the recoveries for Cu and Zn
and the separation factor.

𝐶𝐴 𝐶𝐼 𝑅𝐼 𝐶𝐼 × (𝐶𝐴 )𝑜
𝑅𝐴 = 𝑅𝐼 = 𝑆𝐼,𝐴 = =
(𝐶𝐴 )𝑜 (𝐶𝐼 )𝑜 𝑅𝐴 (𝐶𝐼 )𝑜 × 𝐶𝐴
Classifying Separation Techniques
• An analyte and interferent can be separated if there is a
significant difference in at least one of their chemical or physical
properties.
 Classifying Separation Techniques

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

Classifying Separation Techniques
Separation based on size
• The simplest physical property that can be exploited in a
separation is molecular size.
• Done by using a “porous medium” through which only the target
analyte or interferent can pass.
 Classifying Separation Techniques
Separations based on size – Dialysis
• A semi-permeable membrane is used to separate target analyte
from interferent.
• Usually constructed from cellulose with pore size between 1-5 nm.
• Liquid sample is placed inside cellulose and then immersed in a
solution whose composition is different to the sample.
• If concentration of target analyte is not the same on the 2 sides of
the membrane the resulting concentration gradient will be the
driving force for diffusion across the membrane.
• Small particles that can fit through pores pass through (providing
there is a concentration gradient) while the larger ones remain
inside the membrane.
 Classifying Separation Techniques
Separations based on size - Dialysis

http://www.spectrumlabs.com/dialysis/Fund.html
 Classifying Separation Techniques

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

Classifying Separation Techniques
Separations based on size – Size Exclusion Chromatography
• Also referred to as molecular exclusion chromatography, gel
filtration or gel permeation chromatography
• Word ‘chromatography’ was coined by the Russian botanist,
Tswett in 1901, by combining the Greek words ‘color’ and ‘to
write’.
• Chromatography is a physical method of separation that
distributes components to be separated between two phases,
one stationary (stationary phase), the other (mobile phase)
moving in a definite direction.
Classifying Separation Techniques
Separations based on size – Size Exclusion Chromatography
• In SEC, a column is packed with a stationary phase of small
porous particles of cross-linked dextrin or polyacrylamide.
 Classifying Separation Techniques
Separations based on size – Size Exclusion Chromatography
• The pore-size (~1-10 m) of the stationary phase is controlled by
degree of cross-linking
• Greater cross-linking = smaller pore size
• Sample is added to the head of the column and allowed to pass
either by gravity or using a pump at a fixed flow rate.
• Chemicals with a molecular size too large to enter the pores flow
through the column unrestricted.
• Molecules small enough to enter the pores of the stationary phase
will do so and take longer to pass the length of the column.
Classifying Separation Techniques
Separations based on size – Size Exclusion Chromatography
• Smaller molecules, which penetrate more deeply into the pore
structure, take the longest time to pass through the column
• Used heavily in biochemistry in analysis of proteins.
Classifying Separation Techniques
 Classifying Separation Techniques

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

Classifying Separation Techniques
Separations based on mass or density
• If there is a difference in mass (or density) of target analyte and
interferent(s) then a separation using centrifugation may be
possible.
• Sample, as suspension, is spun at high angular velocity
(revolutions per minute, rpm)
• Particles experience a greater centrifugal force have faster
sedimentation rates and are preferentially pulled or forced
toward the bottom of the centrifuge tube.
• Used heavily in biochemistry to separate cellular components.
 Classifying Separation Techniques
Separations based on mass or density

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

 Classifying Separation Techniques
Separations based on mass or density
Centrifugal force (g) = and revolutions per minute (RPM)

𝑅𝑃𝑀 2
𝐶𝐹 𝑔 = 1.12 × 𝑅 × ( )
1000

R = diameter in mm

http://clinfield.com/2012/07/how-to-convert-centrifuge-rpm-to-rcf-or-g-force/
 Classifying Separation Techniques

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

Classifying Separation Techniques
Separations based on Complexation Reactions – Masking
• Techniques for preventing an interference from interfering with
the target analyte
• Done by transforming interferent into a form that is not detected
• Typically done by binding the interferent to a masking agent
• Technically not a separation technique because target analyte
and interferent are never physically separated from each other
Classifying Separation Techniques
Separation based on Complexation Reactions - Masking

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

 Classifying Separation Techniques

Separations based on Complexation Reactions – Masking

Other Masking agents

1. Diuretics – causes concentration of chemicals in urine to decrease.

2. Dextran – causes concentration of chemical in blood to decrease.
3. Epitestosterone (E) – checks for testosterone (T) employ T/E ratio
- adding E decreases T/E ratio
 Classifying Separation Techniques

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

Classifying Separation Techniques
Separations based on Partitioning between phases
• Selective partitioning (i.e. dividing) of the target analyte or
interferent between 2 immiscible phases
• When a phase containing a solute S, is in contact with a second
phase, the solute partitions itself between the 2 phases until an
equilibrium is reached

Sphase 1  Sphase 2
Classifying Separation Techniques
Separations based on Partitioning between phases

• The equilibrium constant for this is [𝑆𝑝ℎ𝑎𝑠𝑒2 ]

𝐾𝐷 =
[𝑆𝑝ℎ𝑎𝑠𝑒1 ]

where KD is the partition coefficient

• If KD is large, solute is driven from phase 1 → phase 2

Classifying Separation Techniques
Separations based on Partitioning between phases

• The separation process by which a solute is transferred from

one phase to a new phase is known as extraction
• In a simple extraction, the sample is extracted one or more
times with portions of the second phase.
• Several separation techniques are based on this: liquid-liquid,
liquid-solid, solid-liquid and gas-solid extractions.
Classifying Separation Techniques
Separations based on Partitioning between phases
Q22.7 Solute S has a partition coefficient of 4.0 between water
(phase 1) and CHCl3 (phase 2). (a) Calculate the concentration of
S in CHCl3 if [Saq] is 0.02 M. (b) If the volume of water is 80.0 mL
and volume of CHCl3 is 10.0 mL, find the quotient (mole of S in
CHCl3)/(mol of S in water).
(a) KD = [SCHCl3]/[SH2O] = 4

(b) V x M is the key here!

Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)
• Two immiscible liquids placed in separatory funnel and shaken
to increase the surface area between the phases.
• In most cases, one phase is aqueous while the other is organic
• More dense phase will settle to the bottom
• Target analyte is initially present in 1-phase but after extraction
will be present in both phases
• Extraction efficiency determined by equilibrium constant for the
analytes partitioning between the 2-phases.
Classifying Separation Techniques
Separation based on Partitioning between phases

interface
Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)
KD = [Sorg] / [Saq]
KD = distribution ratio D for LLE
(moles in aq)o = (moles in aq)1 + (moles in org)1
[Saq] = (moles in aq)1 / Vaq
[Sorg] = (moles in org)1 / Vaq

(o = before extraction; 1 = after extraction)

Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)
(moles in aq)o = (moles in aq)1 + (moles in org)1
(moles in org)1 = (moles in aq)o – (moles in aq)1
[Sorg] = (moles aq)o – (moles in aq)1 / Vorg

[(𝑚𝑜𝑙𝑒𝑠 𝑖𝑛 𝑎𝑞)𝑜 − (𝑚𝑜𝑙𝑒𝑠 𝑖𝑛 𝑎𝑞)1 ]/𝑉𝑜𝑟𝑔

𝐾𝐷 =
(𝑚𝑜𝑙𝑒𝑠 𝑖𝑛 𝑎𝑞)1 /𝑉𝑎𝑞

(o = before extraction; 1 = after extraction)

Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)
• Fraction of solute remaining in aqueous phase (Faq) after one
extraction:
(𝑚𝑜𝑙𝑒𝑠 𝑖𝑛 𝑎𝑞)1 𝑉𝑎𝑞
(𝐹𝑎𝑞 )1 = =
(𝑚𝑜𝑙𝑒𝑠 𝑖𝑛 𝑎𝑞)𝑜 𝐷 × 𝑉𝑜𝑟𝑔 + 𝑉𝑎𝑞

(Forg)1 = 1 – (Faq)1

(o = before extraction; 1 = after extraction)

Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)
• What if a second extraction was performed:

(𝑚𝑜𝑙𝑒𝑠 𝑖𝑛 𝑎𝑞)2 𝑉𝑎𝑞

(𝐹𝑎𝑞 )2 = =
(𝑚𝑜𝑙𝑒𝑠 𝑖𝑛 𝑎𝑞)1 𝐷 × 𝑉𝑜𝑟𝑔 + 𝑉𝑎𝑞

• If the volumes used in extraction 1 and 2 are the same

(1 = after extraction 1; 2 = after 2 extractions)

Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)
2
𝑉𝑎𝑞
(𝑄𝑎𝑞 )2 = (𝐹𝑎𝑞 )1 × (𝐹𝑎𝑞 )2 =
𝐷 × 𝑉𝑜𝑟𝑔 + 𝑉𝑎𝑞

• For a series of n identical extractions

n
𝑉𝑎𝑞
(𝑄𝑎𝑞 )𝑛 =
𝐷 × 𝑉𝑜𝑟𝑔 + 𝑉𝑎𝑞
Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)

Example on page 538:

Solute A has a partition coefficient (D) of 3 between toluene and
water, with 3 times as much in the toluene phase. Suppose that
100 mL of a 0.01 M aqueous solution of A is extracted with
toluene. What fraction of A remains in the aqueous phase (a) if
one extraction with 500 mL is performed or (b) if 5 extractions
with 100 mL are performed?
 Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)
D = 3; Vaq = 100 mL; Vorg = 500 mL
𝑉𝑎𝑞
(𝑄𝑎𝑞 )𝑛 =
𝐷 × 𝑉𝑜𝑟𝑔 + 𝑉𝑎𝑞
5
100 100
𝑎 = = 0.0625 𝑏 = = 0.000977
3 × 500 + 100 3 × 100 + 100

Forg= 0.938 or 93.8% Forg= 0.999 or 99.9%

Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)
• In general, extraction efficiencies increases dramatically with
the first few multiple extractions, then quickly decreases as the
number of extractions increase
• In addition, the example highlighted that many small volume
extractions are more efficient than one or two large volume
extractions
 Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-liquid extraction (LLE)

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-solid extraction (LSE)

• Liquid sample is passed through a column containing solid

particulates that serve as an adsorbent material
• Often referred to as a solid-phase extraction (SPE)
• Choice of solid adsorbent used is based on the properties of
target analyte and the matrix it is in
 Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-solid extraction (LSE)

David Harvey, Modern Analytical Chemistry, McGraw Hill, 2000

 Classifying Separation Techniques
Separations based on Partitioning between phases
Liquid-solid extraction (LSE)

http://www.greyhoundchrom.com/topic/31-solid-phase-extraction-columns.aspx