Comparison of EA drying agents

Problem

All elemental analyzers used in conjunction with Ionplus equipment should use a water trap to remove water from the carrier gas upon combustion. Until recently, the desiccant SICAPENT® was used for this purpose. Since the Sicapent material is no longer available on the market, the replacement is going to be either “Wetsorber” (calcium chloride) or “Wetsorber S” (calcium chloride, aluminium oxide and magnesium perchlorate). How do the new materials compare when it comes to cross-talk in radiocarbon samples?

Measurement

Three different types of desiccants were used for water removal during the combustion of overmodern samples OX2 (NIST SRM 4990C, oxalic acid) and blanks PHA (In house blank – phthalic anhydride) in Elementar’s elemental analyzer. The combustion sequence containing the samples weights is similar for all three materials, as in the below example (Table 1).

Table 1. The order of 7 samples (Ox2 and blanks) within one graphitization using EA-AGE3 system

Therefore a batch of 7 samples was graphitized with each material: Sicapent, Wetsorber and Wetsorber S. The total of three batches were subsequently measured in the same magazine, using a MICADAS system. The fraction modern carbon or F¹⁴C found in the blanks (Figure 1), as well as the blank ages (Figure2), for each blank sample were plotted as a function of their position with respect to the previous OX2:

Figure 1. F¹⁴C levels in the blanks ordered by position with respect to the Ox2 (see Table 1)

Findings

We found that Wetsorber S has an even lower cross-talk, similar to or even better than Sicapent. Wetsorber however is not recommended in applications for ¹⁴C analysis. An effect on the OX2’s was not observed.