CNL Report:  Factor for Na in reconstructed mass 

                             March 8, 2001

                             Author:  Robert Eldred

The CNL codes use a multiplicative factor of 2.5 for Na, which assumes that Na is present as NaCl. 

To determine if this is reasonable, I compared the measured and reconstructed fine masses for those samples in which 2.5*Na contributes at least 50% of the reconstructed mass.  The results for samples collected between 3/97 and 11/99 are shown in Figure 1.  The reconstructed mass accounts for 94% of the measured mass.  This is closer to unity than for the entire data set, which has a slope of 87%.  To give a slope of unity, the factor would have to be 2.75.  The conclusion is that 2.5 is a reasonable value. 

Note 1:  NaCl is very hygroscopic, so we would expect the measured mass to be greater than for dry NaCl.  

Note 2:  The accuracy for Na is considerably worse than for most elements, since it is at the low energy extreme of the x-ray spectrum, and has the largest x-ray absorption matrix factor of any element.  (This matrix factor compensates for Na x-rays absorbed in the particles.)  The fact that we get good agreement with measured mass indicates that the accuracy error cannot be major. 

Note 3:  All of the samples were from marine sites.  The factor might be different at non-marine sites.  However, Na is a major contributor to reconstructed mass only at marine sites.

Figure 1.  Comparison of measured and reconstructed mass for samples in which 2.5*Na accounted for over 50% of the reconstructed mass.  Samples collected between 3/97 and 11/99.  All of the samples were from marine sites.