HJ 917-2017Stationary source emission--Determination of total gaseous mercury --Carbon sorbent traps / thermal cracking atomic absorption spectrophotometric method (English PDF)

6.1 Sampling System

6.1.1 Sampling system components

Typical gaseous mercury sampling systems typically include activated carbon sorbent tubes, sampling probes, temperature sensors, dehumidification units, Sampling controller, sampling pump, see Figure 1.

Figure 1 Typical gaseous mercury sampling system

6.1.2 Activated carbon adsorption tube

6.1.2.1 Two-stage

The adsorption tube contains two sections of specially prepared activated carbon treated with iodine or other halogen and its compounds and each section can be divided Analysis, see Figure 2.

Figure 2 two-stage activated carbon adsorption tube

6.1.2.2 Three-stage type

Adsorption tube equipped with three sections of special activated carbon, which plus the standard section of activated carbon to add a known quality of mercury, can also be used on the market Activated carbon tubes have been spiked, see Figure 3.

Figure 3 Three-stage activated carbon adsorption tube

6.1.3 Sampling probe

Probe with heating function (heating temperature not less than 120 °C), to prevent flue gas condensation.

6.1.4 Temperature sensor

Measurement accuracy within ± 1.5 °C.

6.1.5 Dehumidification device

Before the airflow enters the sampling controller, the dehumidifier should be used to remove the water vapor from the airflow.

6.1.6 Sampling controller

Used to determine the total volume of dry flue gas, you can control the sampling flow to maintain within ± 10%.

6.1.7 Sampling pump

Power should be able to meet the system flow range requirements.

6.2 Mercury analyzer

Has a thermal cracking unit, the sample of mercury atomized to form gaseous mercury, atomic absorption spectrophotometric quantitative analysis Mercury, the detection limit should be less than 2.0 µg/kg.

6.3 Other

Common laboratory equipment and equipment.

7.1 Sample Collection

Fixed pollution source sampling and laying of sampling points should be consistent with GB/T 16157 in the relevant provisions. Before and after sampling should be Airtight testing. For each sampling, two activated carbon adsorption tubes (6.1.2.1) must be used for parallel and dual sampling, with a flow rate of 0.2 L/min ~ 0.6 L/min, the sampling time should not be less than 30 min. After sampling, record the sampling time, the atmosphere Pressure Remove the sampled activated carbon tube and seal both ends to wipe off the deposits on the outer wall of the tube.

At least one blank activated carbon adsorption tube (6.1.2.1) shall be taken for each sampling, taken to the sampling site as a complete program empty White samples are stored with the collected samples and brought back to the laboratory for analysis.

7.2 Sample Storage

All samples should be stored in sealed sample storage container at room temperature, the sample can be stored for 14 d.

7.3 Dynamic spiked samples

Take a three-stage adsorption tube (6.1.2.2) and perform dynamic spiking of the sample according to the same procedure (7.1) as for sample collection of the collection, the standard addition should be 50% of the actual amount of mercury samples to 150%, each batch should be collected a dynamic spiked samples.

7.4 Laboratory blank samples

Blank activated carbon adsorption tubes (6.1.2.1) were used as laboratory blank samples.

8.1 Instrument analysis reference conditions

Optimize assay conditions or parameters based on instrument user's manual. Thermal cracking temperature should be greater than 650 °C.

8.2 Standard curve established

Pipette Mercury standard solution (5.5) 0 µl, 10.0 µl, 20.0 µl, 40.0 µl, 60.0 µl, 80.0 µl and 100 µl to the sample boat, the mercury contents were 0 ng, 100 ng,.200 ng, 400 ng, 600 ng, 800 ng and 1000 ng Standard series. According to mercury analyzer operating procedures were measured, the instrument response to mercury content to establish a standard curve.

Note. The standard curve can also be weighed with different quality soil or sediment containing mercury standard preparation.

8.3 Sample Determination

The activated carbon adsorption tube at both ends of the seal cap is opened, the interval of quartz wool and each section of activated carbon were transferred to the sample boat , The sample was measured under the same conditions as in the standard curve (8.2).

8.4 Determination of blank samples

8.4.1 Determination of the entire program blank samples

According to the same conditions with the sample (8.3) were measured to determine the entire program blank sample mercury content.

8.4.2 Laboratory blank sample determination

The mercury content of the laboratory blank samples (7.4) was determined according to the same conditions as the samples (8.3).

9.1 Calculation Results

The concentration of gaseous mercury in the exhaust gas of fixed sources is calculated according to formula (1) nd V mmm 1 2 0 -= (1) where. - Gaseous mercury concentration in dry flue gas at standard conditions (273.15 K, 101.325 kPa), µg/m3;

m1 - the mercury content measured in the first paragraph of the adsorption tube, ng;

m2 - measured in the second paragraph of the adsorption tube mercury content, ng;

m0 - laboratory mercury content in blank samples, ng;

Vnd - the standard state of the sample volume of dry gas, L.

9.2 Results indicated

When the test result is less than 100 µg/m3, keep one decimal place; when the test result is greater than or equal to 100 µg/m3, Keep 3 significant digits.

10.1 Precision

In six laboratories, gaseous mercury simulated flue gas was repeatedly collected at concentrations of 5.0 µg/m3, 20.0 µg/m3 and 40.0 µg/m3 Six samples were collected for determination. The relative standard deviations in the laboratory were 4.8% -9.7%, 7.5% -9.9% and 7.1% -9.6% respectively.

The relative standard deviations (RSDs) were 4.5%, 4.1% and 4.7%, respectively. The repeatability limits were 1.1 µg/m3 and 5.4 µg/m3, 8.9 µg/m3; reproducibility limits were 1.2 µg/m3, 5.5 µg/m3 and 9.5 µg/m3, respectively.

10.2 Accuracy

In six laboratories, gaseous mercury simulated flue gas was repeatedly collected at concentrations of 2.0 µg/m3, 20.0 µg/m3 and 50.0 µg/m3 Six samples were collected for determination. the relative errors were -9.3% -7.9%, 0.0% -10.0%, -10.0% -2.3%, respectively;

The final error values were -1.8% ± 6.0%, 6.4% ± 4.3%, - 8.5% ± 3.0%.

11.1 Blank sample

The sampling volume of 10L, the whole program blank sample (8.4.1) should not exceed the detection limit; sample body Product is 10 L, the laboratory blank sample measured value (8.4.2) should not exceed the detection limit.

11.2 Standard curve

The correlation coefficient of the standard curve should be >=0.999.

11.3 Parallel samples

When the mercury concentration > 1 µg/m3, the relative deviation (RD) of the parallel sample determination is less than or equal to 10%; when the mercury concentration is less than or equal to 1 µg/m3, the relative deviation (RD) of parallel samples was <=20%.

11.4 Penetration rate

When the mercury concentration > 1 µg/m3, the absorption tube penetration B <= 10%; when the mercury concentration <= 1 µg/m3, the penetration rate B<=20%. According to the formula (2) Calculate the absorption tube penetration rate.

Where. B-pipe penetration rate,%;

m2 - the second section of the adsorption tube activated carbon measured mercury content, ng;

m1-adsorbed tube in the first paragraph of the activated carbon measured mercury content, ng.

11.5 Dynamic spike recovery

The dynamic spiked sample recovery R is between 85% and 115%.