– Major element chemistry data is stored as weight percents of elements, not as oxide percents as reported for XRF analyses. Thus, XRF data need to be converted to weight percents of each element. We decided to do it this way because data are collected both ways, and the straight weight percent is a simpler and less ambiguous way to represent chemistry data. Sorry.

– Loss-on-ignition (LOI) data can be saved in the database, but won't be picked up for exposure age calculations because there is no way to determine whether it represents water, hydrocarbons, carbonate, etc. If you want LOI to be interpreted as a specific compound, you need to convert it to weight percents of the elements in question and add those appropriately. The main situation in which this is an issue is for data from carbonate rocks. The oxide composition of CaCO3 is 56% CaO and 44% CO2, but sometimes this is represented as 56% CaO and 44% LOI. If you took this literally, you would enter wt_pct_Ca = 40, wt_pct_O = 16, and LOI = 44. However, this would result in incorrect exposure age calculations because the calculator would not know that LOI should be interpreted as CO2. Instead, you need to enter wt_pct_Ca = 40, wt_pct_O = 48, wt_pct_C = 12, and LOI = 0 to correctly represent the major element chemistry.

– Aliquot handling. Because there are sometimes situations in Cl-36 chemistry where multiple target fractions can be extracted from the same whole rock sample, it is necessary to make sure that Cl-36 measurements on target fractions are matched up with correct chemistries for the target fractions and for the whole rock. The way we handle this in the database is that each entry in the Cl-36 table must contain something in the 'aliquot' field. Then, there must be entries in the major element and trace element tables that contain the same value in the aliquot field. Sometimes, this requires duplicating data. For example, consider a situation in which you made one major element chemistry measurement and one trace element measurement on a rock sample, and then you extracted two target fractions. Each target fraction then has one Cl-36 measurement and one major element measurement. Representing this will require the following: (i) make two entries in the Cl-36 table for aliquots a and b; (ii) make two entries in the major elements table for the 'target' analyses, also identified as aliquots a and b; (iii) enter the whole rock major element data as 'WR' in the major elements table TWICE, with different aliquot names; (iv) enter the trace element data for the whole rock in the trace elements table TWICE, with different aliquot names. Of course this involves duplicating data, which is undesirable, but it is needed to make it unambiguous which chemistry data should be associated with which Cl-36 concentration. This can probably be explained better with a more detailed example.