Odours can be measured in two ways:
- using chemical or physical-chemical analyses that will provide us with a quantification by product (H2S, CH3SH, NH3…);
- using olfactometric analyses based on human nose reaction and that can be used to quantify a perceived olfactory nuisance. In conjunction with dispersion models, these measurements enable us to map nuisances.
The aim consists in at least analysing the products from the two most common groups as found in foul air: reduced sulphurs and nitrogen compounds (it should be noted that volatile fatty acids, aldehydes and ketones can also be analysed using methods similar to those we shall describe).
Sampling is extremely important if we are to obtain meaningful results.
In order to carry out these analysis under optimum conditions, we need to:
- undertake in situ gas phase chromatography for reduced sulphur;
- capture nitrogen compounds on a solid or liquid medium (ditto for VFAs, aldehydes and ketones) followed by desorption and analysis using HPLC or GCMS (see section analysis).
In the case of in situ chromatography, after having calibrated the equipment with standard gases, the sampling tube can be inserted direct into the gas stream in order to feed the chromatography column. The results, expressed in mg·m–3 or in mg·m–3, constitute instantaneous analysis points.
For solid or liquid media capture, the tube is inserted direct into the stream of gas to be analysed, the gas drawn up is discharged into the capture system (liquid bubblers or activated carbon or activated alumina column) by means of a pump and a flowmeter. These traps are then sent to the analysis laboratory where they are desorbed and analysed.
The results constitute an average analysis applicable to the sampling period (therefore, no allowance is made for any flow rate peaks).
Finally, as far as reduced sulphurs are concerned, in the absence of any portable chromatograph, samples can be taken using a special bag if necessary (made of Tedlar, a material that will not absorb these compounds). These samples are then taken to the laboratory and the pollutants analysed using gas chromatography. Fairly strict rules have to be followed. Details of these rules can be found in the NF EN 13725 and NF X 43-300 standards.
A qualitative and quantitative knowledge of what makes up the atmosphere will not be enough to ascertain the odour properties of the mixture (human nose reaction) created because of the "synergy" or, on the contrary, of the "masking" effects. Furthermore, some odour-emitting products are perceptible at such low concentrations that even the most efficient analysers are often incapable of detecting them.
Olfactometry is based on the only odour sensors available: the human olfactory mucous membranes. Olfactometry has two types of measurement:
- the measurement of the "odour dilution factor at the perception threshold"; in France, this is a constant that has no dimension and that is called K50. In Anglo-Saxon countries, this same constant is expressed as OU·m–3, (odour unit per cubic metre of air).
- the measurement of an "atmosphere’s odour intensity" expressed in relation to levels within a given range on a scale of reference scale.
These measurements are carried out by statistically processing the verbal responses of a panel of experts. They are presented with varying dilutions of the odorant atmosphere to be analysed.
The higher the number of persons on the panel, the more repeatable and reliable the measurement. Therefore, we need to find a compromise between result validity and costs. We recommend using at least:
- 16 subjects for a good level of accuracy;
- 8 subjects in most cases.
The subjects must be qualified on the basis of their individual perception thresholds to 5 pure products representing the major categories of odorous compounds: for each of these products, the threshold for each subject must fall within a range that is 0.1 to 10 times the average threshold provided as a reference.
Always a crucial matter, the NF EN 13725 standard describes the method used to take samples of odorous atmospheres.
This standard applies to both perception threshold dilution factors and to odour intensity measurements.
perception threshold dilution factor
A threshold concentration can be established for each pure element or odorous mixture.
At this concentration, 50% of individuals making up the panel of experts will perceive the odour. The remaining 50% will not. By definition, the odorous mixture concentration will then be equal to the perception threshold (K50).
The NF EN 13725 standard describes this principle.
After having been diluted by an appropriate odourless gas, the odorous mixture is presented to each of the panel members who each individually states whether or not he has perceived the odour of the mixture. For each of the subjects and on the basis of a series of trials, we can then establish the dilution level at which the odour perception probability will be equal to 50%.
The dilution factor is provided by the following formula:
where Q1 is the flow rate of odourless gas used to dilute the flow rate Q2 of odorous gas.
The measurement requires an olfactometer. The olfactometer is a device that allows to control the dilution of the odorous mixture with the odourless gas and to present the diluted mixture to the panel member. Only dynamic olfactometers are used. It must be possible to rapidly vary the dilution level between 10 and 10,000. This requires a device comprising three interchangeable channels (two for the odourless gas and one for the gas to be tested). The recommended flow rate is approximately 2 Nm3·h–1. Therefore, the olfactometer is not a measurement apparatus; this measuring function is carried out by the human nose.
measuring the odorous intensity of an atmosphere
The odorous intensity of an environment is the intensity of the sensation produced. The odorous intensity depends on the concentration of the odorous mixture.
Odorous intensity is measured using arbitrary units and psycho-physical measurements carried out by a panel of experts whose responses are processed statistically.
measurement method: please refer to the Afnor NF X43-103 standard
The measurement method uses a range of reference intensities produced by means of several samples containing either 1-butanol diluted solutions, or pyridine diluted solutions.
The measurement consists in comparing the intensity of the odour perceived against the intensity of the samples taken from the reference range. The measurement can be carried out direct in the environment concerned. Responses given by panel members are confidential. Results are processed statistically.
By definition, the log of the mean value of the odorous intensity of an atmosphere is the arithmetical mean of the N values of their log.
The confidence interval calculation is based on this figure. The probability level selected is equal to 0.95, i.e. the actual value of the intensity has a 95% chance of falling within the interval of confidence.
application to odours created by water treatment plants
measuring emissions (odour flow rate)
By definition, the odour emitted by a source is the dilution air flow for which the odorous gases concentration in the effluent diluted in this manner are equal to the perception threshold concentration.
Odour output (Nm3·s–1) = K50 × discharged air flow (Nm3·s–1).
If we know the odour flow rate for a source, using atmospheric dispersion models, we can establish the area in which the odour will be detected continuously, within the threshold distance (distance at which the odour is only perceived in 50% of cases).
If the odour flow rate is characteristic of the extent of the polluted area, the source’s intensity provides the pollution level existing in its vicinity.
There are two possible cases:
- effluent is collected in an extraction duct before being discharged into the environment, we can measure K50 and the odour intensity direct at source;
- effluent is released into the ambient environment from a source with a surface area such as a settling tank. Studies have demonstrated that emissions are proportional to the tank’s free surface area. Consequently, a given volume of odour-emitting effluent is taken as a sample and introduced into a mock-up in order to establish the threshold dilution factor (K50).
measurements taken in the environment
These are intended for use as part of an evaluation that is as quantitative as possible of the odour level in a given area. This level is measured using the odorous intensities estimated by a group of trained observers.
Any intensity recorded by the panel must be supplemented with information on:
- the precise location of this perception:
- its time;
- the duration of this perception (continuous or in waves);
- a qualifier (industrial, urban, agricultural, etc. odour).
Results are usually produced as colour charts (one for each set of weather conditions) into which are entered the mean figures for the intensities perceived.
These measurements describe the state of the site before and after installation of a plant or of a wastewater treatment plant, for example. In order to evaluate the level of inconvenience experienced by populations close to a site, we can use volunteers selected from the local population.
Based on the physiological reactions of a group, these olfactory methods are undoubtedly the most representative of sensations and, where applicable, of discomfort felt by operators or by those living in the vicinity of a water treatment plant.
Nevertheless, these are expensive methods due to the number of "expert noses" that have to be gathered and of the meticulously reproducible nature of the results to be obtained. Also, incontrovertible chemical methods are also of interest, in particular when establishing and verifying odour control efficiency.