PhreeqcUsers Discussion Forum

Registrations currently disabled due to excessive spam. Please email phreeqcusers at gmail.com to request an account.
Welcome Guest
 

  • Forum Home
  • Login
  • Register

  • PhreeqcUsers Discussion Forum »
  • Processes »
  • Dissolution and precipitation »
  • Mass balance issues
« previous next »
  • Print
Pages: [1]   Go Down

Author Topic: Mass balance issues  (Read 2262 times)

Flopi

  • Top Contributor
  • Posts: 26
Mass balance issues
« on: 04/07/17 23:49 »
Dear All,
We are trying to do some calculations with precipitation and dissolution and we face a weird mass balance issue. We are using the basic command PRINT TOT("Ca") to get the total calcium quantity in solution after calcite reacts.

However, if we do TOT("Ca") + EQUI_DELTA("Calcite") we don't obtain the quantity of calcium initially present (1e-10 instead of 0 in this particular case). After spending an afternoon trying to figure it out, we were wondering if the difference wouldn't come from some rounding loosing some digits here and there ?

Could you confirm that it is the case ?

Below is a minimal working example of what I mean

SOLUTION 1
   temp       10
   pH       8
   pe       4
   redox       pe
   units       mol/kgw
   density 1
   Ca      1e-3
   C(4)      1e-3
   -water   1 #kg

EQUILIBRIUM_PHASES
   Calcite 0 10


USER_PRINT
   -start
   10   PRINT TOT("Ca") + EQUI_DELTA("Calcite") -1e-3
   -end
Logged

dlparkhurst

  • Global Moderator
  • *****
  • Posts: 4062
Re: Mass balance issues
« Reply #1 on: 05/07/17 01:10 »
In your example, I get a value of 1e-13, instead of zero. That value is within the value for convergence tolerance, which is 1e-8 by default. That means that the calcium mole balance for the initial solution is solved to be 1e-3 * (1 +/- 1e-8), or plus or minus 1e-11 mol/kg. The reaction calculation will also have a convergence within a convergence tolerance of 1e-8.

You can set

KNOBS
-conv 1e-12

and get a value closer to zero than 1e-13. The convergence tolerance cannot be lowered indefinitely. Eventually machine precision plays a role.
Logged

  • Print
Pages: [1]   Go Up
« previous next »
  • PhreeqcUsers Discussion Forum »
  • Processes »
  • Dissolution and precipitation »
  • Mass balance issues
 

  • SMF 2.0.19 | SMF © 2021, Simple Machines | Terms and Policies
  • XHTML
  • RSS
  • WAP2