EVT 536

ENVIRONMENTAL SAMPLING

OVERVIEW OF AIR SAMPLING & MONITORING

DESIGN

Rusdin Laiman
ENVIRONMENTAL SAMPLING

Air sampling sessions would cover on three topics;

1. Overview of air sampling and monitoring design.

2. Ambient Air sampling: Equipment and devices.
3. Emission monitoring and sampling.
CONTENT OF LECTURE FOR THIS WEEK
Sampling
Sampling Definition
Types of samples
Sampling plan
Quality samples
Adequacy of samples for the analysis requested
Subsampling
Selecting the method
Source of the method
Method validation
INTRODUCTION
Quality data important when performing
analysis.
Quality of data is a critical factor for a
meaningful, useful and reliable data
interpretation.
The data quality not only dependent on the
laboratory work but also starting from field
sampling .
Sampling can be both field sampling and
sampling for laboratory test.
4
SAMPLING
For the environmental studies – sampling can be two parts:

Sampling operations – field sampling

Analytical sampling.
SAMPLING OPERATIONS

6
ANALYTICAL OPERATIONS
SAMPLING DEFINITIONS
Sampling is the process of selecting a portion of
material, in some manner, to represent or provide
information about a larger body of material.

“A defined procedure whereby a part of a substance, a

representative sample of the whole or as required by the
appropriate specifications for which the substance,
matrix, material or product is to be tested”
NAMAS

NAMAS = National Measurement Accreditation Service

8
Types of Sample
There are four types of samples

Representative sample
Selective
Random
Composite sample
Types of Sample
Representative Sample
This is a sample that is typical of the parent material for
the characteristics under inspection. We have to be
careful in the way we defined the characteristics of
interest because a sample may be adequate and
representative if the concentration of the analysis is at a
5% mass/mass level (i.e. 5 parts per hundred) but it
may not be if at 5 ppm level.
To obtain an adequate representative sample we must
take account of the state of the parent material we are to
examine.
There are four types (next Slide)

10
Representative Sample
There are four types of representative sample:

i. Homogenous – Materials homogenously distributed within the matrix -

Example a vegetable oil at 40 0C or filtered aqueous solution.
ii. Heterogeneous – Materials not distributed evenly throughout the
sample location or container – Example palm oil at 15 0C or a sample of
breakfast cereal.
iii. Static (contained ) system – There are many situations of this type, the
composition of the parent material is permanent with respect to position
in space and stable in time. Example a sample of oil in a drum or a
warehouse stock of food.
iv. Dynamic Condition – The parent material is changing in respect to
time. Removal of a portion at any instant represents only a snapshot of
that moment in time and in that particular location. The fact that it can
never be reproduced presents difficulties in applying statistical control
and consequently cannot be the subject of conventional statistical
sampling plans. Example estuarine water or unsaturated and saturated
oils being continuously blended
Types of Sample
Selective sample
This is a sample which is deliberately chosen by using a
sampling plan that screen out materials with certain
characteristics and/or selects only materials with other
relevant characteristics. This may be called directed or
focused sampling.
Can you think of instances where this type of sample
should be taken???????
Example – Rodent of flour by hair or urine, or toxic
gases in a factory atmosphere where the total level may
be acceptable but a localized sample may contain a
lethal concentration.
12
Types of sample
Random Sample
A sample selected by a random process to eliminate
questions of bias in selection and/or to provide a basis
for statistical interpretation of measurement data. The
sample is selected so that any portion of the material
has an equal (or known ) chance of being chosen.

There are three types of random sample;

 Sample random sampling.
 Stratified random sampling.
 Systematic sampling.

13
Random Sample
The three types of random sample

i. Simple random sampling – any sample has an equal

chance of selection.
ii. Stratified random sampling – The lot is
subdivided/stratified and a simple random sample
selected from each stratum.
iii. Systematic sampling – The first sample is selected
at random, then the subsequent samples are taken
according to a previously arranged interval.
Example every 5th, 10th or whatever is appropriate.

14
Types of Sample
Composite Sample
Composite sampling is a way of reducing the cost of
analyzing large number of samples. A composite samples
consists of two or more portions of material (collected at
the same time) selected so as to represent the material
being investigated.
The ratio of components taken to make up the composite
can be in terms of bulk, time or flow. The components of
the composite sample are taken in proportion to the
amount of the material they represent.
This type of sample may be appropriate when carrying out
food surveys. The samples may for example be bulked in
proportion to the amount normally consumed.
15
SAMPLING PLAN
 Sampling is always done for a specific purpose
and this purpose will determine, to certain
extent, the sampling procedure.
Examples:
 Packaged food has to be examined for both
mass and content.
 Canned food is examined for leakage from the
can, uniformity of contents and contamination.
 Crops need to be inspected during the growing
season for level of pesticide.
16
SAMPLING PLAN
 Legal and statutory Requirements
 There are regulations governing sampling schemes for a whole
range of materials.
 US EPA reference method for the sampling and analysis of
ambient air pollution PM, SO2, NOx, CO, O3 and etc.
 US EPA Method for sampling and analysis of emission from
stationery sources – example Method 5 and Method 17 for TSP
emission from stationary sources.
 EC directives which cover sampling. Examples are sampling fruits
and vegetables for examination for pesticides residues and for
trace elements in fertilizers.
SAMPLING PLAN
Sampling Scheme

There are three types of sampling scheme;

i. Probability sampling
ii. Non-probability sampling
iii. Bulk Sampling
iv. Acceptance sampling

18
Probability Sampling
This is used when a representative sample
is required and there are three approaches
which give rise to the three types of
random sample; namely:
• Sample random sampling,
• Stratified random sampling and
• systematic sampling.

19
Non-Probability Sampling

This is when a representative sample cannot be collected;

it is the appropriate method to produce a selective sample.
 Bulk Sampling

This type of sampling involves the taking of

a sample from material which does not
consist of discrete, identifiable or constant
unit. The bulk material may be gaseous,
liquid or solid.

21
Acceptance Sampling
Acceptance sampling involves the
application of a predetermined plan to decide
whether a batch of goods meets the defined
criteria for acceptance.
The main aim of any acceptance sampling
must be to see that the customer gets the
quality required, while remembering that
financial resources are not unlimited and that
the cost of the article must reflect the cost of
inspection as well as the cost of production.
22
QUALITY SAMPLE
Sample need to ensure of their quality.
Quality of sample could be ensured
through the followings performance:
Preparedness prior to sampling.
Sample CoC (chain-of-custody).
Sample container.
Sample integrity.
Suspended material in water sample.
Visual observation.

23
Preparedness prior to sampling
In situ-measurement – In in-situ measurements, the data are
recorded directly in logbooks or Field Data Cards with
identifying information, field observations, and remarks.
Examples of in-situ measurements are pH, temperature,
D.O., conductivity, and flow measurement.
Samples - Samples other than in-situ measurements, are
identified by a sample tag or label. These samples are
removed from the sample location and transported to a
laboratory for analyses. Before removal, however, a sample
is often separated into portions depending upon the analyses
to be performed. Each portion is preserved in accordance
with applicable procedures and each sample container is
identified by a sample tag / label.

24
Preparedness prior to sampling
Sample tags / labels - labels shall be completed for each sample,
using permanent marker. The information recorded on the sample
tag/label includes:

1. Station Number - a number assigned by the project coordinator.

2. Date - a six digit number indicating the year, month, day of collection.
3. Time - a four digit number indicating time of collection. e.g. 0954.
4. Station Location - sampling station description.
5. Samplers - each sampler is identified.
6. Sample Number - a unique sample # established from the Field Data Card
for each set of samples collected at one time and place.
7. Parameter/pres. - the analysis to be conducted for the sample /sample
preservation
8. Remarks - the samplers record pertinent observations affecting analyses,
if any

25
Preparedness Prior to Sampling
A Field Data Card - which assigns a unique number to each set of
samples, must be completed for each sampling event. The field data card
used for samples contains an appropriate place for designating the sample
as a grab or a composite, and identifying the type of sample collected for
analyses.

Sample CoC - Due to the evidentiary nature of samples collected during

enforcement investigations, possession must be traceable from the time
the samples are collected until they are introduced as evidence in legal
proceedings. To maintain and document sample possession, chain of
custody procedures are followed. A sample is under custody if:
1. It is in your possession, or
2. It is in your view, after being in your possession, or
3. It was in your possession and then you then locked it up to prevent tampering, or
4. It is in a designated secure area.

26
Preparedness Prior to Sampling
Blank – Analyte-free water is collected into a sample
container, of the same lot as the containers used for the
environmental samples. This evaluates contamination
introduced from the sample containers from a common lots.
Equipment/Rinse Blanks: A sample that is collected by
pouring over or running analyte-free water through the
sample collection equipment after decontamination and
before sample collection. The sample is collected in the
appropriate sample container with the proper preservative,
identical to the samples. This represents background
contamination resulting from the field equipment, sampling
procedure, sample container, preservative, and shipment.

27
Preparedness Prior to Sampling
 Field Blank - In the field, analyte-free water is
collected into a sample container with preservatives.
The sample containers are the same lot used for the
environmental samples. This evaluates contamination
introduced from the sample container(s) with
applicable preservatives. Field blanks are not used for
volatile samples.
 Field Replicates/Duplicates - Two or more samples
collected at the same sampling location. Field
replicates should be samples collected side by side or
by collecting one sample and immediately collecting
the second sample. Field replicates represent the
precision of the whole method, site heterogeneity, field
sampling and the laboratory analysis.
Preparedness Prior to Sampling
Field Split Samples - Two or more representative
subsamples taken from one environmental sample in
the field. Prior to splitting, the environmental sample
is homogenized to correct for sample heterogeneity
that would adversely impact data comparability.
Field split samples are usually analyzed by different
laboratories (inter laboratory comparison) or by the
same laboratory (intra laboratory comparison). Field
splits are used to assess sample handling procedures
from field to laboratory and laboratory’s
comparability.

29
 Preparedness Prior to Sampling

• Filter Blank - In the field, analyze-free water is passed

through a filter and collected into in the appropriate
sample container. The filter blank is then preserved. This
procedure is identical to the sample collection.
Sample CoC
There is always a chain of events from the
process of taking samples to the analysis.
We should recognize the weakest part of
the chain and strengthen it so the part
would not have any effect on the final
results.

Example of CoC form

31
Sample CoC

32
Sample CoC

33
Sample CoC

34
Sample Container
Sample container can be glass, polyethylene, low
molecular weight polyethylene.
Glass – glass container may adsorb or desorb elements. Sodium
can desorb from soft glass (glass which melt at relatively low
temperature normally made of soda lime) and also from hard
glass of borosilicate. But soft glass is a more serious problem
when analyzing trace levels of inorganic materials. Glass
containers are often cleaned using phosphate detergents and even
after washing with acid and several rinses with water, high
phosphorus levels are recorded. So, for many trace analysis,
glass may not be suitable.
Polyethylene – suitable for most solids and aqueous samples.
When used for aqueous samples, unlike glass, there will be no
leaching of elements like Na, K, B, and Si.

35
Sample Integrity
Some samples may change due to standing.
The cream separates out from milk samples and the buttery lump has to be
broken before analysis.
In certain cases, the analytes may be in suspension rather than in solution in the
test samples, e.g. metals in engine oils.
There will be also cases where the suspended materials may not be the analytes
but the presence of suspended materials can effect the analyte.
The sample inhomogeneity also can effect the final test results. Take test samples
of different sizes and carry out duplicate measurement on each of the selected
sizes. If the difference between duplicates increase as the size of the sample
decrease then it can be assumed that the materials for testing is inhomogeneous.
ADEQUACY OF SAMPLES FOR THE
ANALYSIS REQUESTED
Most chemical tests are destructive so one cannot
tests all the material. There may be a problem in
taking a representative sample from bulk material
known to be heterogeneous. The sampling plan must
be such that the degree of homogeneity can be tested.
If the validated method requires 1 g of material but
only 100 gm is available, we must find out if the
method is sufficiently robust to stand this amount of
scaling down.
What one aims is the smallest sample capable of
giving the necessary information.
37
Adequacy of Samples

 The first step in deciding the size of the sample is to

decide how large an error, i.e. difference between our
measured value and the true value, can be tolerated in
the estimate. Once the uncertainty has been fixed, the
next step is the level of confidence.

 A level of confidence of 95% means that we accept

that 5% of samples may have a value which lies
outside our chosen uncertainty limit.
Adequacy of Samples
Ifthe sample mean is , then, assuming normal distribution the
limits of the mean are:

where; ss is the sample standard deviation; ns is the number

of observation; t is the value tabulated in the student’s t
distribution tables for (ns – 1) degrees of freedom.

If we assume a normal distribution, for a sample mean of ,

and a value of n ≥ 30, then at 95% confidence level this
approximates to
±
(this is because for this situation, t∞ = 1.96 = 2)

39
 Adequacy of Sample (cont’d)
 

40
Adequacy of Sample (contd)
 

41
SUBSAMPLING
 A subsample is a split of a sample; prepared in such a way that
there is some confidence that it has the same concentration of
analyte as that in the original sample. Because if inhomogeneity,
differences may occur between samples but there should not be
any between subsamples.
 The minimum size of a subsample can be determined using the
concept of sampling constant. The sampling constant Ks has unit
of mass. It is the subsample weight necessary to ensure a relative
subsampling error of 1% (68% confidence level) in a single
determination.
 The value of is numerically equal to the coefficient of variation
for results obtained on 1 g subsamples in a procedure free from
analytical error.
SUBSAMPLING

 
SUBSAMPLING
 

44
SUBSAMPLING
Calculate CV and hence determine Ks. Then
use this value of Ks with the target value of
CV to determine the subsample size required.
Once Ks has been evaluated for an analyte in
a particular sample type the relative SD for
the same analyte in a future subsample of
weight wf is then estimated by:

(CV)F =

45
SUBSAMPLING (contd)
SUBSAMPLING PROCEDURE

 The most effective way of improving the subsampling characteristics

of a laboratory sample is by fine grinding. However, this can produce
problems because of the possibility of contamination.
 It is not always possible to grind sample, e.g. food stuff analysis. The
best general procedure is to reduce the size of the laboratory sample
by a suitable means, for example quartering.
 There may be legislation governing the subsampling protocol for the
analysis to be undertaken. In such a case we need to refer to
appropriate protocol.
SELECTING THE METHOD
 In this section we should be able acquire knowledge to enable to:
 Identify factors which have to be considered when choosing
a method.
 Know where to look for suitable methods
 Decide when and how to modify standard methods for a
particular requirement (analysis).
 Understand how to validate analytical methods.
PURPOSE OF ANALYSIS
 Before starting work on sample, it is vital to enquire why the work
is being done, what will happen to the results and to find out what
decisions will be taken depending on the numerals values obtained.

 Purpose of analysis – your suggestions???

 Possible purpose!!
1. Preparation of a databank of figures to establish trends, e.g. changes in
pesticide residues in foods with season, or from year to year.
2. Acceptance /rejection of a chemical /products before use in a manufacturing
operation…….

48
SOURCES OF METHODS
 There are basically two types of analytical methods; namely

 Qualitative

 Quantitative

49
SOURCES OF METHODS
 Qualitative
 To determine whether a particular analyte is present or not –
certainly not how much to any degree of accuracy.
 If a negative results is required ( i.e. confirmation of absence
from the product) then one only has to worry about the
sensitivity ( or limit of detection) of the test used.
 Many tests to confirm the absence of impurities in
pharmaceutical products fall into this category.
 Rapid test for positive confirmation are often made on unknown
substance.

50
SOURCES OF METHODS
Quantitative
Quantitative methods are used in a variety of situations an a
variety of different methods can be used. What should be
remember is that the method used be fit for the purposes.
Suitable methods fall into a number of categories and there are
many sources where methods may be found.
i.In house methods developed by one laboratory for their own
special needs.
ii.Methods published in the open scientific literature, e.g. Analyst,
Journal AOAC International, Journal of the Association of
Public Analyst, Journal of Chromatography and etc.
iii.Methods supplied by trade organizations .
SOURCES OF METHODS
• Quantitative (contd)

 Method employed for the analysis procedure has to be

appropriately validated.
 What validation means is that the method has been subjected to
a study which shows that, as applied in the user’s laboratory, it
provides results which are fit for their intended purpose.
 In the field of trace analysis where analyst are attempting to
determine very low levels of analytes (ppm, ppb, or mg/kg,
µg/kg) in a very complex matrix, it is often necessary to
examine large number of samples using methods that might
take anything from a few minutes to a whole week to complete.
SOURCES OF METHODS
FACTORS TO CONSIDER IN CHOOSING A METHOD.

Factors to consider include;

i. Limit of Detection (LOD)
ii.Accuracy
iii.Precision
iv.Speed
v.Equipment required
vi.Sample size
vii.
Cost
viii.
Safety
ix.Specificity
LIMIT OF DETECTION
 Limit of detection is especially important in trace analysis, when one
has to decide whether a contaminant is present below or above the legal
limit.
 Ideally the limit of detection of the method selected should be at least
one-tenth of the concentration to be measured.
 For example, if the legal limit for lead in tap water is 0.05 mg/ dm3
(5ppb).
 In some cases a limit of quantification may need to be considered
where it is necessary not only to detect the presence of an analyte but
also to determine the amount present with a reasonable statistical
certainty.
LIMIT OF DETECTION
 Note that limit of quantitation is equivalent to 10 x LOD.
 It will then be necessary to take into account the accuracy and
repeatability of the determination as well as the standard deviation
of determinations made on blank samples which do not contain the
analyte.

55
ACCURACY
 Accuracy is defined as the closeness of the agreement between the
results of a measurement and the true value of the quantity being
measured.
 Very often a high degree of accuracy is not important for trace
analysis where the concentration of the contaminant is well below
the permitted level.
 For example, the permitted maximum residue level of fluorine in
complete animals feeding stuffs is 150 mg/kg. If a sample is
analysed and found to contain 50 mg/kg, it does not matter if the
analysis is in error by even 100% as the level of contamination is
still well below the permitted maximum.
 Where the concentration of a contaminant, or permitted additive, is
close to the maximum allowed, accuracy becomes more important.
PRECISION
 Precision is the closeness of a series of replicate measurement to
each other.
 The same arguments as accuracy apply to precision.
 A high degree of precision may not necessary unless the result
obtained lies close to the margin.

57
SPEED OF ANALYSIS
• If a large number of samples have to be analyzed, a method
which is rapid is to be preferred so that data can be acquired
quickly and with the minimum effort and cost.
• In the initial survey of methods we might be able to decide
the followings:
There is no problem and, therefore, no further work
is required, or
There is evidence for a need to make additional
determination. This might involve the use of the
same method on additional samples. Or an
alternative method which takes lnger to carry out
but can be targeted on more selective areas.

58
EQUIPMENT REQUIRED
 While a method using a mass spectrometer may be ideal
for the study, if no such equipment is available, the job
will have to be contracted out to another laboratory, or
another approach agreed with the customer.
 Neutron Activation analysis or radiochemical
measurement require special equipment and dedicated
laboratory facilities and safety procedure.
 Such techniques are often not generally available and
are better left to specialist laboratories.

59
SAMPLE SIZE
In many industrial areas as well as food and agriculture, the
amount of sample available to the analyst is not normally a
limiting factor.
However, in clinical chemistry the opposite applies as no patient
is willing to donate large volumes of blood for analysis.
Similarly in forensic work the sample material may also be
limited in size. Sample size is linked to the limit of detection.
Improved detection levels can sometimes be achieved by taking
a larger weight of samples.
Where a large weight of sample is essential, destruction of
organic matter is preferably carried out by dry ashing in a
muffle furnace.
COST OF ANALYSIS
 Most analytical chemists and their customers have to be concerned
with the cost of an analysis.
 Whilst the major factors are human resources and the cost of
running and maintaining a laboratory, the choice of method may
have small bearing on the total cost of the job.
 Analysis of a single sample will always be charged at a higher rate
than a batch of six.
 Analysis requiring techniques such as mass spectrometry or
nuclear magnetic resonance spectroscopy will be more expensive
than classical techniques because of the capital cost of the
equipment used and the high grade staff required to interpret the
data produced by such techniques.
SAFETY
 Analytical procedures involving ionizing radiation and radioactive
material require special safety facilities. Other safety factors may
also influence our choice of method.
 For example we may wish to avoid the use of methods which
require toxic solvents such as benzene, certain chlorinated
hydrocarbons (e.g. tetrachloromethane, trichloromethane) or
reagents such as potassium cyanide if alternative procedures are
available.
 Where statutory methods have to be used there may be no
alternatives. In such cases it is essential that staff are fully aware of
the hazards involved and are properly supervised.

62
SPECIFICITY
 The degree of discriminations between the analyte and other
substances present in, or extracted from, the matrix must be
carefully considered.
 Attention will have to be paid to the clean up procedures used
and the discriminating power of the detection system.
 Testing using likely interfering compounds may be necessary.

63
METHOD VALIDATION
• To ensure the competency of the applied analytical
method as well as the competency of the staff performing
the analysis, method validation has to be performed on the
employed method.
• The method validation will involve the following
performance:
i. Replication
ii. Recovery test
iii. Blank
iv. Alternative methods
v. Alternative detection
vi. Reference material
vii. Collaborative studies

64
REPLICATION
Performing the analysis a number of times (as opposed to
once only) ensures two things;
The method is working properly, that is it is capable of
giving repeatable results and that the analyst has
understood the methods and is carrying out the
operations correctly.
There is no large sampling error, this is provided that the
replicate analysis are made on separate portions of the
sample. And not replicate measurements on the aliquot
portions of the final solution.
It is perhaps the single most important step in validation
where no certified reference material is available.

65
RECOVERY TEST
 Where the nature and composition of the analyte is clearly defined
recovery test should be performed in which a known amount of
analyte is added to the sample and the analysis is then performed
before and after addition so that the amount recovered can be
calculated.
 Alternatively, if a satisfactory blank material is available(i.e.
sample known to contain no analyte) this can be used instead.
BLANK
There are two types of blank test:
1. Firstly Analysis on the reagent used through the test procedure exactly
as described in the method. This serves as a check on contamination in
the reagents; and this is particularly important in trace analysis.
2. Secondly , it serves to check for the presence of contamination in the
general laboratory atmosphere and equipment, including glassware.
ALTERNATIVE METHOD

 When an analysis has been completed and a result obtained, one

should consider whether another method can be used to determine
the same analyte.
 Preferably this should be a method based on a completely different
physico-chemical principle.

68
ALTERNATIVE DETECTION
AND REFERENCE
MATERIAL
 Apart from alternative method it should be wise if one could
check the analysis using alternative detection.
 Using a reference material containing the analyte at a known ,
certified concentration is the ultimate test of a method and of the
analyst.
 The reference material should be as close as possible in chemical
composition to that of the sample and should also contain the
analyte at about the same concentration as is present in the
sample.
CONCLUDING REMARKS

Systematic sampling technique in addition to appropriate

application of analytical technique will ensure not only the
accuracy and precision of the analytical results but also the fitness
for the purpose of the analysis performance.
 Equipment for Air Sampling

Common air sampling

equipment includes:
a) High volume, total
suspended particle (TSP)
samplers
b) PM-10 samplers
c) High volume PS-1 samplers
d) Personal sampling pumps
e) Canister samplers

High Volume Sampler Video 1

High Volume Sampler Video I
I

71
 Common Air Sampling Collection Media/Device

a) Rigid whole air coll

ection canister (SU
MMA canister)
b) Filter cassette for p
articulate sampling
c) Impinger
d) Sorbent tube/cartrid
ge
e) Polyurethane foam
(PUF)
.

72