Page 123 - Geothermal Energy Systems Exploration, Development, and Utilization

2.5 Geochemistry 99
have little effect on the solubility of quartz and amorphous silica (Fournier and
Rowe, 1977). The effects of added salts are significant only for concentrations
greater than 2–3 wt% approximately (Marshall, 1980; Fleming and Crerar, 1979;
◦
Fournier, 1985). However, above 300 C small changes in pressure and salinity
become important. The solubility of silica is also affected by pH, for pH values
above 7.8–9.3, depending on temperature. However, since pH of geothermal
reservoir liquids is generally constrained at values of 5–7 by water–rock reactions,
corrections for pH effects are rarely needed in geothermometric calculations. For
these reasons, dissolved silica in solutions of near neutral pH from geothermal
wells is a reliable geothermometer. The interpretation of dissolved silica from
hot springs is somewhat ambiguous because of uncertainties about the mineral
controlling dissolved silica and the amount of steam possibly separated (Fournier,
1991).
Relating the solubility of quartz to enthalpy instead of temperature has several
advantages. At a given enthalpy, there is only one value of dissolved silica, while at
a given temperature there are two values of dissolved silica, one for the liquid and
the other for the steam. The enthalpy can be treated similar to mass, which makes
the silica versus enthalpy plot a good tool to investigate isoenthalpic mixing and/or
boiling.

2.5.7.3 Ionic Solutes Geothermometers
A general decrease in Na–K ratios of thermal waters with increasing temperatures
was observed long ago (White, 1957; Ellis and Mahon, 1964). The initial attempts
to derive, from these observations, an empirical Na–K geothermometer led to
equations with relatively small temperature dependences due to the inclusion in
the data sets of poorly equilibrated spring waters. It was recognized long ago
that Mg contents of thermal waters are strongly dependent on temperature, and
this relationship was early attributed to equilibration of geothermal liquids with
chlorites (Ellis, 1970) or other Mg-bearing minerals, for example, montmorillonites
and saponites.
The Na–K–Ca geothermometer (Fournier and Truesdell, 1973) is probably the
most popular and used ionic solute geothermometer. The Na–K–Ca function is
entirely empirical and assumes two different exchange reactions. The Na–K–Ca
◦
function gives erratic results below 200 C due to high partial pressures of carbon
dioxide (Paces, 1975) and due to the occurrence of exchange reactions involving
also Mg. Fournier and Potter (1979) proposed a quite complex Mg correction to the
Na–K–Ca function. In addition, precipitation of calcite causes an overestimation
of the equilibrium temperature obtained by means of the Na–K–Ca function.
However, Reed and Spycher (1984) have suggested that the best estimate of
reservoir temperature can be attained by considering simultaneously the state
of equilibrium between specific components in water and many hydrothermal
minerals as a function of temperature. Therefore, if a group of minerals converges
to equilibrium at a particular temperature, this temperature corresponds to the
most likely reservoir temperature.