The procedure for the determination of nitric acid and ammonia is based on the work of Allegrini et al., (1987, 1989, 1992) with some simplifications. It is suitable for the determination of nitric acid and ammonia in the concentration ranges 050 µg/m3 HNO3-N and 05 µg/m3 NH3-N, respectively. If higher ammonia concentration levels are expected, the sampling procedure must be modified.
Since the denuders give a possibility to determine the individual concentrations of HNO3(g) and NO3-(particle); and NH3(g) and NH4+(particle) it is a recommended method to use in the EMEP network. A detailed discussion of using denuders contra filter pack is found in chapter 3.2.1.
The air is drawn through a series of annular denuders, and filters. The two first denuders are internally coated with sodium carbonate (Na2CO3) and glycerol for the collection of nitric acid and sulphur dioxide, the third is coated with citric acid, oxalic acid or phosphorous acid for the collection of ammonia. Evaporation of the coating layer can be a problem. A new study done on the efficiency of different coating layers to determine ammonia has shown that phosphorous acid is the most suitable denuder coating reagent (Perrino and Gherardi, 1999).
The coated denuders are then followed by a three-filter pack system. The first filter is a membrane filter with high collection efficiency for submicron particles, followed by a filter impregnated with potassium hydroxide for collection of nitric acid which may have evaporated from the particle filter, and a filter impregnated with oxalic acid for the collection of ammonia which may also have evaporated.
Nitrous acid (HNO2) is also absorbed in the alkaline denuders, but will normally not cause a significant interference as it is usually detected as nitrite in the sample extracts. Glycerol prevents oxidation of nitrite to nitrate by ozone. Nitrogen dioxide and PAN is partially absorbed as nitrite, this interference in the eventual determination of nitrous acid may be corrected for from the distribution of nitrite between the first and the second denuder.
A schematic description of the sampling equipment is given in Figure 3.4.1. It consists of a small insulated box, with a fan for internal air circulation and provisions for heating to a temperature ca. 2ºC above the ambient temperature, and with a rack for mounting the denuder sampling trains and filter packs, electromagnetic valves connecting the sampling trains to a manifold, a leak-proof membrane pump which gives a sampling rate of 15 l/min, and a gas meter for recording of the sample volume. The electromagnetic valves activating each of the sampling trains are operated by means of an electronic timer. Denuder and filter pack sampling trains can also be activated and changed manually.
The sampling train consists of an air intake, 2 annular
denuders of length 242 mm, and one denuder of length 120 mm. The denuders
are connected by special threaded connectors, and capped with threaded caps
when not in use. A special connect is used to connect the denuder train to a
3-stage filter pack containing an aerosol filter, a KOH-impregnated filter and
a citric- or oxalic acid-impregnated filter. The pump and sample volume
recording instrumentation is identical to the equipment described under 3.2.4.
Figure 3.4.1: Sampling arrangement for denuder sampling. The number of sampling
lines may be extended up to eight.
Denuders:
242 mm URG-2000-30 BWH#30, annular denuder, heavy wall glass
120 mm URG-2000-30 BWH#30 as above
University Research Glassware, 118 E. Main St., P.O. Box 368 Carrboro, North Carolina 27510, USA.
See section 3.2.4.1
Coating of denuders and preparation of filter packs
The preparation and coating of the denuders should take place in the laboratory that will carry out the chemical analyses of the collected samples. A solution of 1% Na2CO3 and 1% glycerol is prepared by dissolving 1 g of the analytical grade reagents in 50 ml of deionized water, and diluting to 100 ml with methanol. Similarly, 1 g of citric acid (or oxalic- or phosphorous acid) is dissolved in a few drops of deionized water and diluted to 100 ml with ethanol.
The denuders have to be thoroughly cleaned, preferably by immersion in strong detergent for several hours, and then rinsed in deionized water.
For the coating procedure, a source of clean, dry air is needed. Compressed air must be filtered, and may have to be scrubbed. Passing the air through 2 tubes filled with cotton or filter paper impregnated with, respectively, sodium carbonate and citric acid will be safeguards against contamination. It is also possible to use clean and dry nitrogen from a steel cylinder.
Put on a screw-thread cap on one end of the denuder, add 23 ml of the coating solution, and rotate the denuder so that all internal surfaces are wetted. Pour off excess liquid, take of the screw-cap and put on another cap with connection to the dry air source. Pass dry air through the denuder, while rotating the denuder slowly. Observe the evaporation of the liquid film, and continue for a few seconds more to make sure that the evaporation is complete. Close the denuder with screw-caps.
At least 3 of the denuders should be set aside for the determination of blank values for each batch of denuders, which is being prepared in this way.
(Alternatively, sodium chloride may be used as an internal coating for the absorption of nitric acid in D1and D2. This absorbent may partly absorb SO2 and will not absorb nitrous acid or PAN).
Preparation of impregnated filters with potassium hydroxide and with oxalic acid, and determination of filter blank values, is described under Section 3.2.8. Filter packs with impregnated filters should also be prepared and loaded in the laboratory, and sealed for transport to the sampling site. One set of denuders and one filter pack should be reserved for use as field blanks every week. These are to be sent to the field sampling site, and returned to the laboratory without being exposed at the site.
Denuders and filter holders will have been prepared and should be transported to the sampling site in special transport containers. They are to be connected to the sampling equipment according to specific instructions, and should be marked with sampling date and time period (from-to) and denuder number (D1, D2, D3) in the laboratory before transport to the site. Make sure that connections are leak proof. Denuder and filter pack trains should be changed at 0800h in the morning or a timer and magnetic valve arrangement should be used to make an automatic sampling train change at 0800h. The gas meter should be read or recorded every day, together with the temperature in the gas meter.
The temperature of the ambient air and the temperature inside the box holding the denuder trains, should also be checked periodically, to avoid overheating and to check that the system functions properly.
A sampling form should be filled in, with date and identification of denuders and filterpacks, gas meter readings, and notes of observations, which may be of interest in connection with the evaluation of the results.
The denuders and filter packs should be capped after exposure, and put in the container used for transport to the chemical laboratory (together with the field blanks).
The following solutions are recommended for extraction of
the denuders and the filters in the filter pack:
D1 (Na2 CO3-imp) |
10 ml deionized water with H2O2 (0.3%) |
D2 (Na2CO3-imp) |
10 ml deionized water with H2O2 (0.3%) |
D3 (Citric acid) |
10 ml 0.01M HNO3 |
F1 (aerosol filter) |
10 ml deionized water |
F2 (alkaline imp. filter) |
10 ml deionized water with H2O2 (3%) |
F3 (oxalic acid filter) |
10 ml 0.01 M HNO3 |
Unscrew the cap at one end of the denuder to be extracted,
add exactly 10 ml of the extraction solution with a pipette, put on the cap and
shake the denuder, then transfer the extract to a stoppered test tube for
subsequent analysis. It is essential to work quickly in order to minimize
contamination hazards! The filters may be folded and transferred to the same
kind of stoppered test tubes to which 5 ml of extraction solution is added.
Stopper and agitate shortly. If a hydrophobic (e.g. fluoropore) membrane filter
is used as the first filter in the filter pack, immersion in an ultrasonic bath
may be useful. Chemical analyses of the extracts are to be made as follows:
Sample |
Ions to be determined |
Analysis methods |
Reference |
D1 |
NO3--N, SO42--S (Cl-, NO2--N) |
Ion chromatography |
Section 4 |
D2 |
NO3--N, SO42--S (Cl-, NO2--N) |
Ion chromatography |
Section 4 |
D3 |
NH4+-N |
Spectrophotometry or ion chromatography |
Section 4 |
F1 |
NO3--N, SO42--S, NH4+-N (Cl-, NO2--N) |
Ion chromatography and spectrophotometry |
Section 4 |
F2 |
NO3--N, SO42--S (Cl-, NO2--N) |
Ion chromatography |
Section 4 |
F3 |
NH4+-N |
Spectrophotometry or ion chromatography |
Section 4 |
The results from the chemical analyses will be given in µg/ml of the respective ions. After subtraction of blank values, the following algorithm have been proposed by Allegrini et al. (1987) for the subsequent calculation of the concentrations of HNO3-N, SO2-S and NH3-N in air:
HNO3 N (µg/m3) =
[(D1-D2)/0.94]*(10/V),
SO2-S (µg/m3) = [(D1-D2)/0.96]*10/V,
NH3-N (µg/m3) = D3*10/V,
where D1, D2, and D3 stand for the concentrations of the relevant components in the respective denuder extracts and V is the sample air volume in m3. The correction factors are based on theoretically calculated absorption efficiencies and it is assumed that interfering particles and other substances (e.g. PAN) are collected with the same efficiencies in D1 and D2. Correspondingly:
NO3-N(particles) =
(F1+F2+2.5*D2)*10/V,
NH4-N(particles) =
(F1+F3),
SO4-S(particles) = (F1+2.5*D2).
Note that if the absorption capacity of denuder D3 for ammonia is exceeded, F3 will retain gaseous ammonia, which has passed through the denuder system. Therefore, if this amount of ammonium-N collected on F1 and F3 is significantly larger than the equivalent amount needed to balance the nitrate and sulphate on F1 and F2, the calculated excess should be added to the NH3-N concentration determined from D3.
Sampling
Written step-by-step instructions for the handling of denuders and the operation of the sampling equipment should be available at the site, together with appropriate sampling forms for the registering of sampling flow rates, change of samples, and air sample volumes. The gas meter and flow control device must be properly calibrated in the laboratory. Field calibration of the flow rate and sample volume recording apparatus should be carried out at least once every year.
Chemical analysis
Field blanks should be analysed regularly in order to check on the possible contamination. It is recommended to prepare, and analyse, one complete sampling train every week. Field blanks are to be prepared in the same way as the other samples, sent to the site, and returned unexposed to the laboratory.
Control samples should be included in lack batch of chemical analyses. The control samples should contain the same reagents as the leachates, and known realistic concentrations of the analytes. 5% control samples will generally be sufficient.
Data consistency
Sulphur dioxide and sulphate aerosol will generally be determined by a separate filter-pack sampler, as described in Section 3.2.1. Sulphur dioxide is also quantitatively retained in D1 and D2, and the sum of nitric acid and nitrate in airborne particles may be determined from the filter pack data. Comparison of the results will give a good indication of the performance of the complete sampling and analysis procedure. The results should generally be within ± 10-15%.
This sampling technique is technically demanding and requires good control of chemical analyses and particularly of blank values. The sampling set-up is relatively simple, but needs to be defined in relation to the sampling site and practical arrangements in connection with transport of unexposed denuder sampling trains and filter packs. The recommended denuder tubes are both expensive and brittle. Even more expensive denuder tubes are available, which are unbreakable. More information with respect to the theoretical and practical aspects of denuder sampling for the phase-separated determination of nitrogen species in air can be found in the literature described in the general introduction, reference is also made to a general review article by Ali et al. (1989).
The denuder sampling technique is applicable also to the determination of atmospheric HCl, HF and HNO2, but these components are not included in the EMEP measurement programme.
The simple tubular denuder systems described by Ferm (1979, 1982, 1986) require much less expensive equipment. If blank values (including field blanks) and chemical analyses can be controlled, this system is an alternative to the system described above.
More advanced systems are also available, e.g. for continuous monitoring of gaseous NH3 (Keuken et al., 1988)
Ali, Z., Thomas, C.L.P. and Alder, J.F. (1989) Denuder tubes for sampling og gaseous species. Analyst, 114, 759-769.
Allegrini, I., de Santis, F., di Paolo, V., Febo, A., Perrino, C. and Pozzanzini, M. (1987) Annular denuder method for sampling reactive gases and aerosols in the atmosphere. Sci. Tot. Environ., 67, 1-16.
Allegrini, I., Febo, A., Perrino, C., eds. (1989) Field intercomparison exercise on nitric acid and nitrate measurements. Rome, September 18-24, 1988. Brussels, CEC (Air Pollution Research Report, 22).
Allegrini, I., Febo, A., Perrino, C., eds. (1992) Field intercomparison exercise on ammonium measurement. Rome, April 29-May 4, 1990. Brussels, CEC (Air Pollution Research Report, 37).
Ferm, M. (1979) Method for determination of atmospheric ammonia. Atmos. Environ., 13, 1385-1393.
Ferm, M. (1982) Method for determination of gaseous nitric acid and particulate nitrate in the atmosphere. EMEP Expert meeting on chemical matters, Geneva 10-12 March.
Ferm, M. (1986) A Na2CO3-coated denuder and filter for determination of gaseous HNO3 and particulate NO3- in the atmosphere. Atmos. Environ., 20, 1193-1201.
Hering, S.V. et al. (1988) The nitric acid shootout: field comparison of measurement methods. Atmos. Environ. 17, 2605-2610.
Keuken, M.P., Schoonebeek, C.A.M., van Wensveen-Louter, A. and Slanina, J. (1988) Simulteneous sampling of NH3, HNO3, HCl, SO2and H2O2 in ambient air by a wet annular denuder system. Atmos. Environ., 22, 2541-2548.
Perrino, C. and Gherardi, M. (1999) Optimization of the coating layer for the measurement of ammonia by diffusion denuders. Atmos. Environ., 33, 4579-4587.
Stelson, A.W.
and Seinfeld, J.H. (1982) Relative humidity and temperature dependence of the
ammonium nitrate dissociation constant. Atmos.
Environ., 16, 993-1000.