Evaluation of the Memory Effect on Gold-Coated Silica Adsorption Tubes Used for the Analysis of Gaseous Mercury by Cold Vapor Atomic Absorption Spectrometry

Table 3

Comparison of the detailed experimental design for short- and long-term memory effects of Hg sampler and differences between the two studies considered.

Order	Exp. ID	Exp. name	Major Exp. parameters	This study	Brown et al. [14]	Relative comparison
			(a) Amount of STD Hg (ng) injected	5, 10, 30, and 50	5, 15, 20, 35, and 50	In both studies, findings were similar, maximum mercury was liberated in the first heating cycle, and tubes were almost blank by the 6th heating cycle. However, liberation of excess mass from 2nd BHCs was about 0.39 (±0.20) ng higher in the investigation of Brown et al.
			(b) Blank heating cycles (BHC)	5^a (day 0)	6
			(c) Daily interval between STD and blank analysis	None (1-day study)	None (1-day study)
1	Exp. 1	Short-term memory effect	(d) Total tube used	4 mass × 3 tube (triplicate analysis per mass) = 12	5 mass × 5 tube (five times analysis per mass) = 25
			(e) Number of blank run^b	5^a (consecutive run (3 tube/mass)) × 4 (mass) = 20	5 (consecutive run (5 tube/mass)) × 5 (mass) = 25
			(f) Findings	Majority of Hg was liberated in the first three heating cycles and was almost blank by the 6th heating cycle. For higher masses, more heating cycles were needed.	Tubes were almost blank by the 5th heating cycle, irrespective of initial mass loading.
			(a) Amount of STD Hg (ng) injected	5, 10, 30, and 50	5	Four different masses of mercury were injected in our study, while Brown et al.'s study used only one mass and found liberation of excess mass at day 8, which was absent in our study.
			(b) BHC	6 (days 1, 8, and 15)	4 (days 1, 8, and 15)
			(c) Total duration (no. of days)	15	15
2	Exp. 2	Intermediate-term A	(d) Total tube used	4 mass × 3 tube (triplicate analysis per mass) = 12	1 mass × 5 tube (five mass analysis per mass) = 12
			(e) Number of blank run^b	6 (consecutive blanks (3 tube/mass)) × 4 (mass) × 3 (interval days) = 72	5 (consecutive blanks (3 tube/mass)) × 1 (mass) × 3 (interval days) = 15
			(f) Findings	Day 1 analysis was similar to Exp. 1, while insignificant blank level at day 8 and 15.	There were some memory effects at day 8, especially in the first three heating cycle. However, day-15 tubes were blank.
			(a) Amount of STD Hg (ng) injected	5, 10, 30, and 50	5, 12, 28, 40, and 45	In our study, after 20 ng initial injection, gradual increment was seen. In contrast, Brown et al. found linear increment from the beginning.
			(b) BHC	1 (days 8 and 15)	1 (days 8 and 15)
			(c) Total duration (no. of days)	15	15
			(d) Total tube used	4 mass × 3 tube (triplicate analysis per mass) = 12	1 mass × 5 tube (five times analysis per mass) = 5
3	Exp. 3	Intermediate-term B	(e) Number of blank run data^b	1 (blank (3 tube/mass)) × 4 (mass) × 2 (interval days) = 8	1 (blank (5 tube/mass)) × 5 (mass) × 2 (interval days) = 10
			(f) Findings	Up until 10 ng of initial injection, average of 2nd analysis was around blank level. Then the liberation increased gradually until 30 ng. After that, sharp increment was seen. Consistently 3rd heating cycle was around blank level throughout the study.	Linear increment was seen with increasing initial mass of mercury, irrespective of initial STD injection.
			(a) Amount of STD Hg (ng) injected	5, 10, and 30	5	Similarly to Exp. 2, we used more initial standard injections. Brown et al. showed linear and steady state condition; however, we found different pattern for different initial injections. In both studies, similar trends were seen from third heating cycle.
			(b) BHC	2 (days 1, 8, 15, and 45)	2 (days 1, 2, 4, 7, 14, 28, and 35)
			(c) Total duration (no. of days)	45	35
			(d) Total tube used	3 mass × 3 tube (triplicate analysis per mass) = 9	1 mass × 5 tube (five times analysis per mass) = 5
4	Exp. 4	Long term	(e) Number of blank run data^b	2 (blank (3 tube/mass)) × 3 (mass) × 4 (interval days) = 24	2 (blank (3 tube/mass)) × 1 (mass) × 7 (interval days) = 14
			(f) Findings	Initial 5 ng injection showed high memory effect in 2nd heating cycle which was increased with reanalysis time, while higher injection mass (30 and 50 ng) showed lower liberation amount with time. During 3rd analysis tubes were almost blank over the study period.	In 2nd heating cycle, up to 20 days interval increment was linear with time. After that increment was steady. Third heating cycle was near to the blank level.

^aOut of 6 consecutive runs, only 5 blank runs (except the 1st standard analysis) are counted in Exp 1.
^bTo count the number of blank run data only, all the standards measured in the 1st run are not counted in intermediate- and long-term experiments (Exps. 2, 3, and 4).