What is sampling?
“In analytical result is never better than the sample on which it is based”. The following statement emphasizes the importance of proper sampling.
It is because every chemical analysis starts with the following steps
(1) Preparation of sample, (2) preservation of the sample & (3) Its conversion into a proper format before subjecting to analysis hence above steps give importance of good sample or proper sampling technique, therefore a sample drawn from the bulk should be a true representative of bulk in all aspects otherwise analysis reported for sample may not correspond to the bulk which may make analysis meaningless.
The term sampling refers to all steps that are carried out to obtain a sample from bulk material.
Definitions of Terms involved in sampling
- The Population or The universe:- The bulk quantity of a material from which a sample is taken is called the universe. The universe may be finite or infinite.
- Sample: – A small portion of material taken from the universe and selected in such a way that it possesses the essential characteristics of the bulk is called a sample.
- Sampling:- Sampling is the process of obtaining a reasonable amount of material that has all the essential properties of the bulk material. Or sampling is the science of extracting from a larger quantity of material, a small portion that is truly representative of the bulk material. Sampling techniques are of 2 types:- i) Random sampling & ii) Systematic Sampling.
- Sampling Unit:- The minimum sized package in the consignment that the sample may represent is known as the sampling unit. E.g.: If one sample is picked from ten bags of ore then every single bag is a sampling unit and a total of ten bags in the universe.
- Increment:- A stated amount of the material that is withdrawn from the sampling unit is defined as an increment. E.g.: – Amount of ore picked from each bag of ore is called a single increment. Increment should be related to the size of the sampling unit.
- Gross Sample:- The total sample obtained by mixing or blending different increments is known as the gross sample.
- Sub Sample:- A smaller sized sample, produced by sub-division of gross sample and which possesses all the essential characteristics of gross sample is known as Sub-Sample
- Analysis Sample:- An accurately weighed amount of the Sub-Sample taken for the analysis is called an analysis sample. E.g.: If tablet sampling is done from one million tablets, the tablet is a sample, and one million tablets are the universe. If all million tablets may be packed in packets of 10 tablets each, then each packet is called a sampling unit. One or two tablets taken at a time from each of the packets of 10 tablets become the increment.
Theory or Laws of Sampling:-
The theory of sampling is based on certain statistical laws which are not exact but are universally true. The laws governing sampling are of two types.
- Laws of Statistical regularity, 2) Law of inertia of large numbers.
Laws of Statistical regulating:- According to this law, a moderately large number of samples chosen at random from the universe and mixed uniformly to give an average sample can be termed a representative sample and is expected to represent the composition of the universe with a high probability
Laws of Inertia of Large numbers:- Random sampling from the universe does not mean the selection is done in a haphazard or careless manner but it only means that sampling is done without any bias or prejudice, whereby a portion of the universe has an equal chance of getting included in to sample.
Laws of Inertia of Large numbers:- According to this law, abnormalities will occur in large samples, but abnormalities will occur in large samples, but abnormalities will compensate, leaving the average un alternated.
If follows from this law that, if large numbers of samples are drawn from different parts of the bulk material then the probability that the gross sample will represent the bulk material is quite high. This is in agreement with the observation that the experimental mean (x) approaches the true mean (m) when a hypothetical infinite number of measurements are made. The reliability of the sample is proportional to the square root of the sample size.
Factors Involved in Sampling
- Nature of bulk material & its homogeneity :-
If the bulk material is homogenous in nature then sampling is easy and even a small amount of sample will give accuracy in observation & readings. While sampling of heterogeneous bulk will involve a greater number of steps to obtain a sample that will be a true representative of bulk.
- The expected accuracy of result
Depending upon the end-use of bulk, the accuracy of bulk sample result is expected, and depending upon it the type or method of sampling is selected.
E.g.: a sampling of drugs will require high accuracy hence will require a costly & elaborate method, while that of raw material used for manufacturing chalk sticks need not be equally elaborate or careful.
- Cost and time of analysis:-
The time required for sampling should not be very long and the cost of sampling should not be more than the cost of the product.
Purpose of Sampling:-
- Judging acceptability :-
A sampling of bulk many times is done to conclude the material from which sample is done to conclude whether the material from which sample is drawn meets the requirements such as purchase or sales specifications. If it meets requirements it can be accepted otherwise rejected.
- Detection of contaminations :-
The second purpose of sampling is to assure that the material under consideration is free from contamination or any unwanted material. E.g.: Urea in drinking milk.
- Identifying material :-
The third purpose of sampling is to identify an unknown material. A carefully drawn sample can adequately serve to establish the identity of bulk material under study.
- Estimation of material : –
Sampling is sometimes done to make an estimate that a particular element or material is present or absent in a given material.
Types of Sampling
For a sampling technique to be reliable and accurate, the method should confirm to following conditions.
- The sample mean should provide an unbiased estimate of the population mean.
- Sampling procedure should lead us to an accurate estimate of the central tendency and dispersion of bulk material for given time & money.
- Test of significance should be applicable on the sample to estimate populations variance.
The sampling procedure can be broadly classified into two types
A) Random Sampling, B) Non-random or Systematic sampling.
A) Random Sampling:-
Random sampling is the selection of samples without bias and in a way that gives full freedom to the operation of chance factors
- In this method every individual item has an equal chance of being selected; if sample size is large enough then sample produced is most likely represent the bulk.
- This type of sampling requires minimum knowledge of the bulk material in advance of being selected.
- When bulk is of homogenous nature, random sampling is comparatively easy. However, when sample is heterogeneous then different procedure followed.
- Heterogeneous material is first divided into relatively homogenous group i.e. material is first divided into groups possessing similar characteristics. Then from each group samples are drawn at random it is called as stratified sampling.
E.g.: ores exist as lumps of various sizes; hence lumps of different sizes are grouped on basis of their sizes then from each group of sizes samples are drawn at random at mixed.
B) Non-random or Systematic Sampling:-
- This type of sampling appears to be more scientific method than random sampling, though not necessary that it will give better sample.
- In systematic sampling, sample units are drawn in a definite sequence at equal intervals from one another.
E.g.: every tenth tablet is selected from the tableting machine or the twelfth soft drink bottle selected from the bottling plant.
- This method has an element of bias or prejudice in favour of tenth tablet or twelth bottle in above example.
- Due to cyclic fluctuations in tabletting machine or bottling plant exactly the tenth tablet or the twelth bottle may be defective or may be perfect. In either case, sample does not truly represent universe.
- Non- random sampling requires a prior list of the items in population & it gives satisfactory result if items of such a list are arranged in a random manner and samples are drawn at definite intervals.
Problems or Difficulties associated with sampling
a. Lack of prior information:- There may be no information on the nature of the distribution of the desired property in the bulk material. The examination of previous data may supply approximate but adequate data required.
b. Excessive Cost:- Sampling is immaterial, as long as the sample removed is truly representative of the bulk material and to obtain a representative sample from bulk material, a large number of units may have to be selected, which may involve direct or indirect expenditure.
c. Physical difficulties:- The sample has to be removed in such a way as not to disturb the composition of the material at the point at which the sample is removed. But because of the physical nature, condition, or location of material to be sampled, the same cannot be achieved hence required randomness in sampling is not possible. And in the case of heterogeneous material, it is still difficult.
E.g. Steel ingots or cast iron samples are collected by drilling, milling, or sawing the ingots. Drilling causes graphite flakes to become loose thereby increasing the carbon content of the sample.
Sample collected from the interior as well as from the surface of the ingot can only be capable of becoming a true representative. Similarly, the liquid is stored in the drum. Should be sampled from the top, bottom & middle of the drum.
In the case of material forming large pile follows a fixed pattern, not a random pattern, in which larger particles roll down and collect at the outer edge of the pile while smaller particles accumulate at the center of the pile.
Gas sampling is a very important concern. A sampling of gases is more difficult than those of liquid or solids. The method used to draw samples of one gas may not be applicable to another gas. Even though gases mix freely by diffusion and become homogenous, there are practical difficulties in drawing a sample of gas. The major difficulty is in the prevention of contamination of gas by atmospheric air. Successive sampling may show the stratifications of the gas samples.
Types of gas sampling
There are two types for sampling for gases and they are,
i) Ambient sampling and ii) Stock sampling
i) Ambient Sampling
Ambient sampling means sampling of air, gases that exist in the free state in air. It is also called atmospheric sampling which is rather difficult than stack sampling.
Factors such as wind (direction, velocity), temperature, height, rain, geographical condition, etc. Should be considered during ambient sampling, because this variable can change the composition of gases from time to time and from place to place.
The method also depends on the chemical and physical properties of the substances which are present in the atmosphere and which are under study.
In collecting air samples, the following points must be satisfied.
a) An accurate airflow device must be used to trap the contamination in the air. Certain
filters or absorbing solutions are used. In general fine filters or specific trapping, solutions are used in and filtration action is controlled by the porosity of the filter and in solutions, certain chemicals are used to trap unwanted components.
b) A widely used technique is to collect air samples in plastic bags of various sizes at definite points at different times. Then all results are computed to give a general pattern.
ii) Stock sampling:-
A sampling of gases from closed systems such as tanks, cylinders, or flowing through pipes is called stack samplings. The industrial gases are generally sampled. Continuously as they flow through pipes by this method. Only care should be taken that the sample collected represents a constant fraction of the total flow and all portions of the stream are sampled.
Apparatus used for sampling of gases
Generally, the apparatus used for air sampling consists of
(a) sample probe, (b) Sample container, (c) Delivery line, (d) Mercury trap.
- Sample Probe : Sample probe is that part of the apparatus which is attached to or which is extended into vessel or a pipe containing the gas. The probe should extend into vessel to about one-sixth diameter of the gas container.
- Sample container : – Sample container is the vessel in which the gas sample is collected and this may vary in size from 250 cm3 to several cubic centimetres depending upon material to be sampled. The containers are made of glass, steel or iron.
- Delivery Line :- Delivery line carries the sample.
- Mercury trap: – It is an arrangement connected to the sample container which consists of tube dipped in a vessel containing mercury. It helps in releasing excess of pressure developed.
All the tubes are provided with stopcocks or valves to control gas flow. The stopcocks should be cleaned & well greased with high vacuum grease before a sample is taken; the gas in question is allowed to come in contact with lubricant to establish equilibrium.
Methods used for removal of gas samples
A sampling of relatively pure gas can be done in three following methods.
i) Expansion method, ii) Displacement method, iii) Flushing method
i) Expansion method: In this method, the sample container is evacuated by using a vacuum pump attached to the container. The container is also warmed to remove gases absorbed on the container wall. The sample probe is attached to the gas container (and stopcock D is opened) the gas flows into the sample container by natural expansion.
Since the perfect vacuum is never obtained some residual impurities may still stay on, hence the process of evacuation & filling of the sample gas is repeated several lines. This method is generally used for Ambient sampling.
ii) Displacement Method:- In this method, a liquid generally mercury, water, or saturated salt solution is used to displace all the air from the sample container, sample probe & delivery line. This method is generally used to draw samples from cylinders filled with gas at high pressure i.e. in other words for stack sampling.
This method however not suitable for accurate sampling because some gas samples may get dissolved in the liquid used for displacing the air. Another possibility is that sample gas may be saturated with vapors of liquid used for displacing.
iii) Flushing method:- In this unwanted gas in the sample container is completely removed or flushed out by the gas being sampled. This may be done by the apparatus by the gas being sampled. This may be done by the apparatus shown in the figure. Initially, the stopcock ‘B’ is closed. The probe is inserted into the gas container and stopcock ‘A’, ‘C’, and ‘D’ are opened. The gas flows into the sample container and displaces air from pt. When it is determined that all air has been completely flushed out. ‘A’ is closed and sample gas is allowed to completely fill the sample container, then stopcock ‘C’ and ‘D’ are closed.
Sampling of Liquids
A sampling of liquids involves three different cases viz. Homogenous liquid, heterogeneous liquids and flowing liquids. In each case, a different technique or procedure has to be employed for sampling.
For a sampling of homogenous & heterogeneous liquid sample thief (figure given below) can be used while multiple tube sampler (fig shown below) can be used for sampling of flowing liquids.
I) Sampling of Homogenous (Static) Liquid:-
A homogenous liquid has the same composition throughout. Hence sample can be withdrawn at random from the top, middle, and bottom layers. For this purpose, sample thief is used.
Sample thief enables the collection of the liquid at different depths without disturbing it. Thus a composite sample can be prepared. Sample thief consists of a bottle shown in the figure. The bottle is fitted with two tubes t1 and t2. The tube t2 is much longer than the height of the liquid in the container, so when the thief is dipped into the container, the tube t2 projects well outside the container, the tube t1 is always open. The thief is lowered to different depths in the container with stopcock ‘A’ closed. When the thief is at an appropriate depth, ‘A’ is opened. The liquid enters the bottle through ‘t2’ by displacing the air through ‘t2’ and in this way sample is collected.
II) Sampling of Heterogeneous (Static) or Immiscible liquids:-
Heterogeneities of a liquid may arise due to complete immiscibility of components of liquids mixed together or, due to formation of emulsion or suspension or due to volatility of liquid, leaving behind partially crystallized solid. The different procedure has to be used in all different cases.
a) Heterogeneous liquids that are immiscible:-
Accurate sampling of immiscible liquids which forms layers is done by determining the volume of each immiscible layer by using the cross-section of the container and the height of the layer samples are then taken from each immiscible layer by using sample thief, care is being taken to see that the volumes of sample collection are in the same ratio as the ratio of the volume of the layers in the container originally.
The individual samples are then mixed to get a gross sample. If the liquid container is non-uniform or odd-shaped, then the liquid has to be transformed into a cylindrical or rectangular container in order to determine the ratio of different liquid layers.
b) Emulsion or unstable suspensions:-
The best way to sample an unstable suspension is to separate the two (solid & liquid) phases by filtration. The two phases are weighed and sampled separately. If this is not possible then the entire suspension is rapidly and efficiently stirred to get the uniform mixture and then the resultant liquid is rapidly sampled reproducibility of this sampling technique should be checked.
c) Liquid containing partially crystallized solids:-
Semi solidified liquids or liquid containing crystallized solids are first heated, before subjecting to sampling. Heating is continued till solid dissolves in the liquid or melts, and then the entire mars is stirred efficiently and then rapidly sampled.
III) Sampling of Flowing (not-static) Liquids :
The sampling technique of flowing liquids has to be modified since the composition of flowing liquid changes with time and position. It is, therefore, necessary to draw samples at different positions and at different time intervals, this leads to the elimination of bias introduced due to variation of composition with time.
Generally, the composition of the liquid at the center of flow is not the same as that at other points of flow. Hence to get a satisfactory sample under this condition a multiple tube sampler is used. This tube consists of tubes of varying lengths ending up into one single long tube. The overall appearance of this device is similar to the palm of a hand, with tubes of varying lengths appearing like fingers.
Multiple Tube Sampler
When such a tube is placed in the path of the flow of a liquid, the tube extends to different points in the liquid stream. Samples collected at random or at regular intervals are mixed to give a gross sample.
General precaution during sampling of liquids.
To obtain a good representative sample of liquid certain requirements are cleanliness of apparatus and containers used preservation of sample composition and the scrupulous care of sampling apparatus. Precaution is also needed during handling, transportation, and storage.
Cleanliness denotes the exclusion of foreign materials from apparatus & containers before and during sampling. Sampling connection, sample container should be rinsed & drained before it is actually used.
The sample composition should be preserved. The volatile sample must be protected against evaporation. Liquid to be sampled if contain solid or immiscible droplets, care should be taken so as to transfer it quantitatively into the sample. Dissolved gases should not be allowed to escape if they are present.
To allow contraction & expansion, the container should be only 80% filled. Light-sensitive samples should be collected in an opaque or amber-coloured container.
Sampling of Solids
A sampling of solid is more tedious and difficult than a sampling of gases and liquids. This is because of the heterogeneity in which the solids are obtained. A solid exists either in compact form or a particular form. Even if materials appear homogenous they turn out heterogeneous due to localized concentrations of impurities. The sampling technique adopted to obtain a true representative of bulk will depend mainly on the form in which solids exist.
The conversion of a gross sample suitable for analysis in the lab requires reduction of particle size along with reduction of mass, which depends on three factors.
- Heterogeneity of bulk material, 2. Particle size of bulk material and 3. Degree of accuracy of results.
Sample size in solid sampling:-
The gross sample should have the same particle size distribution as the bulk. The size of the sample depends upon various factors such as the particle size, the bulk size, the degree of precision required, and the actual amount of sample used for testing. In this concept the bulk ratio is important.
Another factor used for determining the number and size of increments that are to be brought from bulk to form gross samples is the size to weight ratio.
For a sample to be a true representative of populations, the bulk ratio should be as large as possible and size: weight be as small as possible. Even due to sampling is minimized when results from a large number of samples are arranged. Sampling error is found to be inversely proportional to the square root of a number of sample average.
If the impurities are present as a particle of large size and cubic shape then it is estimated that the standard deviation ‘s’ of the percentage impurity due to sampling is given by,
S = (b.p.)1/2
b= Size : weight
p = percentage of impurity
Preparation of Sub-sample: Size Reduction
The Gross sample obtained in a manner is then reduced to a size suitable for analysis. This involves both reduction of particle size as well as bulk. This is done by following different methods, before applying any mentioned methods; the gross sample has to be powdered to coarse size and mixed. This can be done by using a ball mill or grinder. The methods for size reduction are,
a. Coning & Quartering method:- This is one of the oldest techniques for reducing the size of the gross sample. The material is first crushed to suitable fineness and is placed into a conical pile. The pile is then flattened, and then divided into quarters, the two opposite quarters are selected, and the other two are rejected. This procedure of coning & quartering is repeated till a sample of the desired size is obtained.
b. The Long pile and alternate shovel method:- In this method, the material is shoveled in the form of a pile about 10 feet in length and width corresponding to that of the shovel. The alternate shovel is discarded; the sample size thus gets reduced.
c. Tabelling, rolling, and quartering:- In this method sample is uniformly spread on a square polythene sheet placed on the table hence the name tabelling. The tabelled material is mixed carefully by lifting one end of the polythene sheet and then repeated from another end this process makes particles roll over one another, hence name rolling. These steps are repeated to make a homogenous mixture. Then this material is uniformly spread on the table and then is divided into four quarters. Two opposite quarters are selected and two are rejected. This is continued till the sample size of the desired amount is obtained.
d. Method using riffles:- A riffle consists of a through that is divided into an even number of segments. Alternate segments deliver the sample to the opposite side of the trough the starting material is divided into two approximately equal portions. One part may be paused through the rifle repeatedly until the sample of the desired size is obtained.
Different sampling Equipments and sampling methods.
Sampling equipment used and the method of sampling depends upon the types of solids to be sampled. Solids can be of two types a) Compact solids and b) Particulate solids.
a. Compact Solids:- Compact solids consists of various forms of solids such as broken clunk of original solid, resolidified molten material, natural deposits of soil, the material of various hardness. In this case, the samples are obtained by drilling, sawing, filing material sampling of compact solid involves the following equipment.
i) Auger Sampler ii) Split – Barrel Sampler iii) Split–tube thief.
i) Auger Sampler:- It is a small helical screw of about 4 cm minimum diameter with a ‘T’ style handle. It is turned into the material and then pulled straight out. The material is then knocked or scraped off with a spatula.
ii) Split-Barrel Sampler:- The drilling equipment used should provide a clean hole to permit the driving of the sampler to obtain an undisturbed sample. The driving head should be made up of hard steel, is at least 45 cm long. It is detached from the coupler & opened to remove the sample.
iii) Split-tube thief:- It is a metallic tube with a slot running the full length of the tube. It has a T-type handle fixed to one end. The end of the tube is sharp and can cut through a container. To remove a sample, the thief is inserted into the container by rotating the handle until the thief reaches the center of the container. The thief is then carefully withdrawn and the material it has scouped out is knocked or scraped out.
b) Particulate Solids:- In particulate form, solids exist as a particle of different sizes. In this case, the particle is important. Sampling is done by making size to weight ratio as the criterion of sample size i.e. ‘b’ is considered.
Different types of devices used are
Split-tube thief, concentric tube thief & Hand scoups, or shovel.
Concentric tube thief:-
It consists of two concentric tubes which are closely fitted to each other. Both tubes have holes cut into the corresponding positions. Outer tubes have sharp ends & can pierce through containers. During insertion holes are closed, when the thief reaches an appropriate position, the tube is rotated to open holes and material gets deposited into the inner tube. And then holes are closed & the thief is pulled out.
Hand scoups or shovel:- Hand scoup or shovels of suitable size are also used for taking cross-sectional samples of particulate solids while in motion e.g.:- on the conveyer belt. The scoups can also be used to take samples from the surface of drums, bags, barrels & other containers.