Unit conversion issue - Amm ug/l to mol/kg

[patersonb]

I've got a dataset with measured concentrations in μg/L Ammonia. I'm using the Amm.dat database to understand the distribution of dissolved ammonia species.

When I enter these concentrations in phreeqc as μg/L, the concentrations in the output file in molality (mol/kg) are different from what I would expect if I do these calculations manually outside of phreeqc.

For example, 25 μg/l Amm is equivalent to 1.468E-06 mol/L (25 μg/L divided by Amm mol mass =17.031). It's essentially the same value in mol/kg for the solutions I'm interested in.

The Output file states:

------------------------------------
Reading input data for simulation 1.
------------------------------------

   TITLE
   SOLUTION 1
       units      ug/L
       temp      10
       pH                 7.0      charge
       density      0.99970
       Amm        25
   END

Initial solution 1.   

-----------------------------Solution composition------------------------------

   Elements           Molality       Moles

   Amm               1.386e-06   1.386e-06

----------------------------Description of solution----------------------------


The phreeqc molality concentration is clearly lower than the manual calculation.

The SOLUTION_MASTER_SPECIES in Amm.dat states:
Amm      AmmH+   0   AmmH      17.031 (which is the correct molar mass)

What if anything am I doing wrong? Many thanks.

[dlparkhurst]

Every so often, we make a mistake. The definition in the Amm.dat (or phreeqc.dat from Appelo) has an error in the entry for Amm in SOLUTION_MASTER_SPECIES. It should be

Code: [Select]

Amm         AmmH+   0   Amm            17.031

with Amm as the next to the last entry instead of AmmH. So a gfw of ~18 was used, instead of ~17 in converting your concentration.

We will fix the error. In the meantime use the following in your SOLUTION definition.

Code: [Select]

Amm        25 as Amm


[dlparkhurst]

Or, you can make the change in the database file.

[patersonb]

Many thanks. I kind of thought that might be the issue, but had some doubts after seeing the discussion in the following topic:

https://phreeqcusers.org/index.php/topic,2114.msg7705.html#msg7705

Particularly regarding the redox-uncoupled gases and Sg/H2Sg and whether 32 or 34 was the correct GFW.

Thanks again for your help

[dlparkhurst]

Having discussed with Tony the best choice for the default gram formula weight for Amm in the Amm.dat database, we have decided to stay with AmmH. Labs I am familiar with in the US use "as N", whereas the lab he is familiar with uses "as NH4" in reporting ammonia concentrations. Others probably use "as NH3" (Amm).

So, you just have to be careful in defining the ammonia concentrations and pay attention to the reporting units. In your case you should use "as Amm".  For backward compatibilitiy and lack of a clear best choice, the definitions in Amm.dat (and Tony's phreeqc.dat) will remain

Code: [Select]

SOLUTION_MASTER_SPECIES
Amm         AmmH+   0   AmmH            17.031


[patersonb]

Thanks for the update and clarification.