Preprint

UCRL-JC-153721

Silica Scale Management:

Lowering Operating Costs
through Improved Scale
Control, and Adding Value
by Extracting Marketable
By-Products

E. A. Button, W. L. Bourcier, A. Wallace, C. J. Bruton, R.

Leif

U.S. Department of Energy This article was submitted to

Geothermal Resources Council 2003 Annual Meeting, Morelia,
Michoacan, Menico, October 12, 15, 2003

Laboratory

June 18,2003

Approved for public release; further dissemination unlimited

 DISCLAIMER

This document was prepared as an account of work sponsored by an agency of the United States
Government. Neither the United States Government nor the University of California nor any of their
employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for
the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or
represents that its use would not infringe privately owned rights. Reference herein to any specific
commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not
necessarily constitute or imply its endorsement, recommendation, or favoring by the United States
Government or the University of California. The views and opinions of authors expressed herein do not
necessarily state or reflect those of the United States Government or the University of California, and
shall not be used for advertising or product endorsement purposes.

This is a preprint of a paper intended for publication in a journal or proceedings. Since changes may be
made before publication, this preprint is made available with the understanding that it will not be cited
or reproduced without the permission of the author.

This work was performed under the auspices of the United States Department of Energy by the
University of California, Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48.

This report has been reproduced directly from the best available copy.

Available electronically at htb: / /www.doc.gov/bridge

Available for a processing fee to U.S. Department of Energy

And its contractors in paper from
U.S. Department of Energy
Office of Scientific and Technical Information
P.O. Box 62
Oak Ridge, TN 37831-0062
Telephone: (865) 576-8401
Facsimile: (865) 576-5728
E-mail: reDorts@adonis.osti.eov

Available for the sale to the public from

U.S. Department of Commerce
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: (800) 553-6847
Facsimile: (703) 605-6900
E-mail: orders@ntis.fedworld.gov
Online ordering: htb: / /www.ntis.eov/orderine.htm

OR

Lawrence Livermore National Laboratory

Technical Information Department’s Digital Library
http:/ /www.llnl.gov/tid/Library.html
 Burton, Bourcier, Wallace, Bruton, and Leif, LLNL

Silica Scale Management: Lowering Operating Costs through Improved Scale Control, and
Adding Value by Extracting Marketable By-products

Elizabeth A. Burton, William L. Bourcier, Adam Wallace, Carol J. Bruton and Roald Leif
Lawrence Livermore National Laboratory

Key Words: silica, scale, silica extraction, pH-modification, Mammoth

Abstract
We are using laboratory and field experiments to design modeling tools and technology
that will improve silica scale management practices in geothermal plants. Our work will help to
lower operating costs through improved scale prediction and add new revenue streams from sale
of mineral byproducts extracted from geothermal fluids.

Improving the economics and effectiveness of scale control programs and/or extraction
systems in geothermal operations requires a coupled kinetic-thermodynamic model of silica
behavior. Silica scale precipitation is a multi-step process, involving a nucleation-related
induction period, aqueous polymerization, condensation of polymers to form colloids, and
deposition onto a solid surface. Many chemical and physical variables influence the rates of
these steps and their impacts must be quantified and predictable in order to optimally control
silica behavior. To date, in laboratory studies, we have quantified the effects on silica
polymerization of the following set of chemical variables: Na at 500 and 2000 ppm, pH values
from 5 to 9, temperatures of 25 and 5OoC, and silica saturation values from 1.2 to 6 at initial
dissolved silica concentrations of 600 ppm. Lowering pH both increases the induction time prior
to polymerization and decreases the polymerization rate. We have successfully used a multiple
regression model to predict polymerization rates from these variables.

Geothermal fluids contain significant dissolved concentrations of potentially valuable

mineral resources such as zinc, lithium, cesium and rubidium, but silica fouling interferes with
traditional extraction methods. We are developing customized and new technologies to extract
the silica as a commercial-grade commodity as well as the valuable metals. We are conducting
field testing of some of these techniques at a Mammoth, CA geothermal plant using a reverse
osmosis unit to concentrate the fluid, adding a commercial agglomerating agent to promote silica
precipitation, and then removing the silica using a tangential flow ultrafilter. The particle size,
surface area and trace impurities of the silica are characterized for comparison with commercial-
grade silica products. We are also testing ion exchange resins and other hctionalized materials
to extract potentially economic concentrations of lithium, cesium, and rubidium that are enriched
in the reverse osmosis concentrate.

Introduction
Silica scaling is a common problem in geothermal facilities. Geothermal operators need
improved tools to evaluate and mitigate the impact of silica scaling. It is also proving feasible to
develop technologies to economically convert silica and other dissolved elements in geothermal
fluids from a problem to a resource. The benefits to a geothermal operation of controlling silica
in geothermal fluids include:

1
 Burton, Bourcier, Wallace, Bruton, and Leif, LLNL

a. Lower operating and maintenance costs, including mitigation costs associated with
silica scaling;
b. Improved heat transfer performance;
c. Decreased costs and minimized environmental impact associated with chemical
additives;
d. The addition of new revenue streams from mineral extraction, when appropriate.

To achieve these benefits, we are developing a predictive model of silica deposition and the
technologies to economically remove silica with appropriate properties for sale in industrial
markets. The removal of silica also is a necessary precursor to extracting other valuable
dissolved elements from geothermal fluids owing to its fouling of extraction equipment. Our
studies include both laboratory testing and field trials. When deployed, the results will allow
plant operators and technical staff to:
a. Predict silica scaling rates and magnitudes for conditions covering the range of water
chemistries found in geothermal plants and potential resources;
b. Determine treatment type and minimum dosage of chemical agents (PH modification
or various types of mhibitors) to control scaling;
c. Evaluate the potential for creating new revenue from silica and other elements and to
identify the appropriate technologies for their extraction.

Background
Predictive modeling of scale deposition: Effective prediction of silica scaling and
design of scale control programs and/or extraction systems in geothermal operations requires a
coupled kinetic-thermodynamic model of silica behavior. Silica scale precipitation is a time-
dependent, multi-step process, involving aqueous polymerization, condensation of polymers to
form colloids, and deposition onto a solid surface (e.g., Iler, 1979; Weres et al., 1980; Weres et
al., 1981). While thermodynamic models exist that predict the potential magnitude of scaling and
qualitatively capture the overall dependence of reaction rates on the degree of disequilibrium,
they do not capture the overall and step-specific rate effects of various other solution parameters,
including sodium and fluoride concentrations, and pH.

Scale control programs involve the use of chemical inhibitors that retard or disrupt
specific steps and/or change fluid chemistry to retard overall rates such that the fluid moves
beyond scale-sensitive plant facilities before scale precipitates. Optimizing silica extraction
techniques and economics depends critically on understanding how to change conditions to
control rates of reaction and the physical and chemical properties of precipitates. Consequently,
understanding and incorporating the kinetics of precipitation is critical to efficient and cost-
effective management of silica scaling.

Extraction of marketable silica and minerals: Geothermal fluids contain significant

dissolved concentrations of potentially valuable mineral resources such as silica, zinc, lithium
and cesium (Gallup, 1998). This mineral content has often been considered more of a nuisance
than an asset, but interest is increasing in co-producing and marketing some of these
constituents. However, the geothermal industry lacks simple, cost-effective methods to extract
mineral by-products from geothermal fluids. We are examining the potential for adaptation of

2
 Burton, Bourcier, Wallace, Bruton, and Leif, LLNL

existing methods from the hydrometallurgical industry, as well as the use of new technologies,
such as metal-specific membranes.

To allow high-value fluid components to be removed, the ubiquitous dissolved silica in

geothermal fluids first must be removed or reduced in concentration. Most geothermal fluids
contain dissolved silica concentrations in equilibrium with quartz at reservoir temperature. In
general, the hotter the reservoir is, the higher its dissolved silica concentration. By purposefully
precipitating silica as a high surface area, porous material with properties similar to those of
commercially precipitated silica, the silica-scaling problem is solved, and at the same time a
marketable silica by-product is produced. Other benefits may also be realized, such as additional
energy extraction that would not otherwise be economic due to scaling. At the Mammoth, C A
binary power plant, silica scaling becomes a problem when the spent geothermal fluid is used in
an evaporator to cool the working fluid. Recent field tests (discussed below) show that extracting
silica from the Mammoth fluids should minimize scale formation on the evaporative cooling
panels (Fig. la). The use of processed geothermal fluid would eliminate the need to purchase
tertiary water for cooling and thus lower operating costs.

Results and Interpretation

Predictive modeling of scale deposition:Using laboratory tests, we have quantified the
effects on silica polymerization of the following set of chemical variables: Na at 500 and 2000
ppm, pH values from 5 to 9, temperatures of 25 and 50°C, and silica saturation values fiom 1.2 to
6. Total initial dissolved silica concentrations were kept constant at 600 ppm. Silica was added
as sodium metasilicate, and pH was adjusted by addition of hydrochloric acid or sodium
hydroxide. The final total dissolved solids (TDS) contents of the solutions were 1300 and 5200
ppm. Experimental solutions were held at run temperature -cO.5"C by a circulating water bath.
The pH was monitored throughout each run. Silica polymerization was followed over time by
withdrawing aliquots of the solution for spectrophotometric measurement using the
silicomolybdate method as modified by Iler (1979).

Polymerization rates increase with increasing Na concentration, pH and temperature (Fig.

2), and decrease with decreasing silica concentration. These variables also affect the degree to
which the polymerization rate varies with the silica concentration variable (slopes in Fig. 2c and
d). The silica concentration variable reflects the control of saturation state on rate. The term for
the rate dependence on saturation state has been expressed in a variety of forms (e.g., Goto,
1956, for silica polymerization; Lasaga, 1981, for mineral precipitatioddissolution). We plotted
silica concentration as log Q (where Q is saturation state expressed as the ratio of dissolved silica
concentration, C, to the equilibrium concentration with respect to amorphous silica at mn
temperature, S); as log (C-S); and as log (1-Q). We achieved the best fit with the log (C-S) term
(Fig. 2c and d).

Lowering pH both increases the induction time prior to polymerization and decreases the
polymerization rate (e.g. Fig. 2a). These results explain field experience where silica scaling has
been effectively prevented by pH-modification (e.g., Rothbaum et al., 1979; Gill, 1993).
However, our results show that the magnitude of the pH effect depends on the other chemical
variables as well. Future experiments will expand the matrix of tests to include the range of

3
 Burton, Bourcier, Wallace, Bruton, and Leif, LLNL

chemistries typical of geothermal plant waters. Longer-term experiments are planned to provide
baseline data for modeling the subsequent condensation and deposition steps.

We used multiple linear regression to model rates of polymerization as a function of the

above chemical variables and elapsed time. Each chemical variable was input as a predictor
variable in log concentration units. This simple model appears to be a good predictor of
polymerization rates for the intermediate range of rate measurements (Fig. 3), but is less reliable
for rates measured early or near the end of each run.The poorer fit is partly an effect of the
induction period and of the detection limits of the molybdate method. Future modeling will
quantify the impact of fluid chemical variables on the length of the induction period before
polymerization begins, and will also incorporate the kinetics of the condensation and deposition
steps.

Extraction of marketable silica and minerals at Mammoth: The relatively low

dissolved silica content of 250 ppm in Mammoth fluids presents a challenge to silica extraction.
Most geothermal fluids from flash plants contain 2500 ppm. Conventional methods for
extracting silica are not effective for the Mammoth fluids because of slow polymerization rates at
low silica concentrations. For this reason, we processed the Mammoth geothermal fluid through
a reverse osmosis (RO) membrane (Fig. lb). The RO unit provides a silica-enriched concentrate
for silica and other metals removal, and a low TDS permeate for use in the evaporative cooler.
The reverse osmosis unit can concentrate the fluid to any desired silica concentration. The
optimum silica concentration allows rapid extraction, but does not foul the reverse osmosis
membranes with precipitated silica. Silica concentrations between 600 and 700 ppm appeared to
satisfy both constraints in our Mammoth field trials.

We used a commercial agglomerating agent to promote silica precipitation and removed

the silica using a tangential flow filter. The silica was characterized using a particle size analyzer
and a gas adsorption (BET) surface area analyzer, and then digested for chemical analysis. The
most promising samples also were sent to commercial laboratories for product testing. We are
comparing these test results with the properties of commercial-grade silica products to guide
further work in optimizing extracted silica properties for specific markets.

The gas adsorption (BET) surface area of silica precipitates obtained to date are in the
range of 40 to 130 m2/g for samples that were air dried at 50°C. These values are near the low
end of the range of surface areas of commercial silica, but should increase as our processing and
post-processing conditions are optimized. The chemical composition of a representative silica
precipitate is given in Table 1. Our Mammoth geothermal silicas are of very high purity, greater
than 98 wt % for untreated silica, and about 99.6 wt % for acid rinsed. These compositions
compare favorably with commercial precipitated silica that often have 1-5 wt% impurities. The
precipitate does contain measurable arsenic. However, the arsenic is less enriched in the silica
precipitate than in bulk fluid. The Si/As ratio of the fluid is about 800 vs. 70 in the silica
precipitate. A simple acid rinse reduced the arsenic content by about 2/3. Better arsenic removal
methods are available and will be investigated.

Of additional interest at Mammoth are the potentially economic concentrations of

lithium, cesium, and rubidium that are enriched in the reverse osmosis concentrate. Work is

4
 Burton, Bourcier, Wallace, Bruton, and Leif, LLNL

underway to extract these elements using commercial ion exchange resins, and functionalized
carbodsilica aerogels.

Acknowledgments
Funding for these projects has been supplied by the Department of Energy, Office of
Geothermal and Wind Technologies, the California Energy Commission, and the Lawrence
Livermore National Laboratory program for Laboratory Directed Research and Development.
Industrial collaborators include ChevronTexaco and Mammoth Pacific LP. We also appreciate
advice and suggestions fiom Paul Hirtz of Thermochem and Jack Harrar.

References
Gallup, D.L., 1998. “Geochemistry of geothermal fluids and well scales, and potential for
mineral recovery.” Ore Geology Reviews, v. 12, p. 225-236.
Gill, J.S., 1993. “Inhibition of silica-silicate deposits in industrial waters.” Colloids and Surfaces
A: Physicochemical and Engineering Aspects, v. 74, p. 101- 106.
Goto, K., 1956. “Effect of pH on Polymerization of Silicic Acid.” Journal of Physical Chemistry,
V. 60, p. 1007-1008.
Iler, R.K., 1979. The Chemistry of Silica. J. Wiley and Sons, New York, 866 p.
Lasaga, A.C., 1981. “Rate Laws of Chemical Reactions.” In Lasaga, A.C. and R.J. Kirkpatrick,
eds., Kinetics of Geochemical Processes. Reviews in Mineralogy, v. 8, p. 1-68.
Rothbaum, H.P., B.H. Anderton, R.F. Harrison, A.G. Rohde, and A. Slatter, 1979, “Effect of
Silica Polymerization and pH on Geothermal Scaling.” Geothermics, v. 8, p. 1-20.
Weres, O., A. Yee, and L. Tsao, 1980. Kinetics of Silica Polymerization. National Technical
Information Service, Dept. of Energy LBL7033,256 p.
Weres, O., A. Yee, and L. Tsao, 1981. “Kinetics of Silica Polymerization.” Journal of Colloid
and Interface Science v. 84, #2, p. 379-402.

5
 ~ ~~ ~~

Burton, Bourcier, Wallace, Bruton, and Leif, LLNL

Table and FiPure Captions

Table 1. Composition of silica precipitate fiom Mammoth geothermal fluid for untreated, water-
rinsed, and dilute HC1-rinsed silica powders.

Figure 1. Schematics of Mammoth site: a) binary power plant in which the geothermal fluid is
used to heat the isobutane “working fluid” that turns the turbine; b) geothermal fluid processing
unit used for silica and metals extraction testing.

Figure 2: Polymerization of silica in test solutions of 600 ppm initial silica at 25OC (a and c) and
5OoC0, and d), at pH values from about 5 to 9, and Na concentrations of 500 or 2000 ppm.
Legend entries show pH and Na concentrations. Upper diagrams show measured values of
molybdate active silica (in ppm) over time; bottom diagrams show the same data expressed as
polymerization rate (change in molybdate active silica concentration per change in time, in
ppm/hr) vs log (C-S) where C is concentration in solution and S is the concentration of silica in
ppm at equilibrium with amorphous silica at run temperature.

Figure 3: Results of multiple linear regression model for silica polymerization predicted fiom Na
concentration, pH, temperature, elapsed time, and degree of disequilibrium for amorphous silica,
expressed as log (C-S).

6
Elizabeth Burton, 6/17/03 9:13 AM -0700, Re: 3

I I 175 I I

Printed for Marty Fleming <flemingll @Ilnl.gov> 3

Elizabeth Burton, 6/17/03 9:13 AM -0700, Re: 4

siIk a
extraction
600-9aQ
ppm SO, Metals
llextraction
-lppmC&Rb
4 5 ppm Li

oisrnosis separation

Printed for Marty Fleming eflemingll @Ilnl.gov> 4

 I

Q
4
400

0
a2 0.4 0.6 0a I
e l a p s d time (hrb

bRC2003 8urto.n -g-Fi _. , .

Printed for Marty Fleming <flemingll @llnl.gov> 5

Elizabeth Burton. 6/17/03 9:13 AM -0700. Re: 6

Printed for Marty Fleming cflemingll @ Ilnl.gov> 6

 i 1
8 c

om
1.5 2.D

(13

Printed for Marty Fleming cflemingll @Ilnl.gov>

 100
I
i m"c

* ;

1.6 2.Q
a
bRC2003;Bwrton Fia 2D.TIF . ,. - . - . . . ~

Printed for Marty Fleming cflemingll @ Ilnl.gov> 8

 _-

Elizabeth Burton, 6/17/03 9:13 AM -0700, Re: 9

AW

Printed for Marty Fleming <flemingll @Ilnl.gov> 9

 hurton.LLNL.do4

Printed for Marty Fleming eflemingll @Ilnl.gov> 10