Content uploaded by Halldor Gudfinnur Svavarsson
Author content
All content in this area was uploaded by Halldor Gudfinnur Svavarsson on Oct 11, 2017
Content may be subject to copyright.
1
Adsorption applications of unmodified geothermal silica
Halldor G. Svavarssona,b,*, Sigurbjorn Einarssona, Asa Brynjolfsdottira
a Blue Lagoon Ltd. Iceland
b School of Science and Engineering, Reykjavík University, Iceland
* Corresponding author at: School of Science and Engineering, Reykjavík University, Iceland.
Tel.: +354 5996200; fax: +354 5996201. E-mail address: halldorsv@ru.is (H.G. Svavarsson).
ABSTRACT
Silica, precipitated out of geothermal fluid discharged from a geothermal powerplant in
Svartsengi on the Reykjanes peninsula in Iceland, was used as a chromatographic adsorbent to
extract blue colored protein, C-Phycocyanin, from coccoid blue-green algae. The only
supplement used was salt obtained by evaporating the geothermal fluid. Analysis of the silica,
using scanning electron microscopy, X-ray diffractometry and Brunauer-Emmett-Teller
(BET) adsorption confirmed it has a high specific surface area and is amorphous. Upon
adsorption and subsequent elution the purity of the extracted protein, measured as the ratio of
the light absorbance of 620 and 280 nm, increased from 0.5 to above 2.0. Our results could
facilitate utilization of a mostly unused byproduct of geothermal powerplants as
chromatographic material.
Keywords:
Geothermal silica
Chromatographic
Adsorption
2
Protein separation
Phycocyanin
1. Introduction
1.1 Geothermal Silica
Geothermal resources are widespread throughout the world although generally associated
with areas of volcanic activity. The HS-Energy geothermal powerplant is located in
Svartsengi on the Reykjanes peninsula, south-west Iceland on a sequence of lava flows, the
youngest being roughly 800 years old (Saemundsson et al., 2010). The geological structure is
further characterised by interlayers of scoria and hyaloclastite reflecting interglacial and
glacial periods. Since the lava flows, scoria and interlayers of hyaloclastite are highly porous
and permeable, they allow seawater to percolate deep into their aquifers where it heats up and
mixes with meteoric water (Arnorsson, 1995). Geothermal wells drilled through the lava
flows to depths of up to 2,000 m discharge a mixture (here referred to as geothermal fluid) of
2/3 seawater and 1/3 meteoric water with a temperature of about 240°C. Due to leaching, the
hot geothermal fluid contains a high concentration of silicon (Si) when it enters the wells.
Originally, the silicon is present in the hot geothermal fluid as silicic acid (SiOx(OH)4-2x)n.
Upon cooling, the silicic acid precipitates as a three-dimensional network of coagulated
primary silica (SiO2) particles. The primary particles grow up to some nanometers in size
before they coagulate to form aggregated clusters. Such a small particle size gives rise to high
specific surface area, which makes the SiO2 a suitable candidate for adsorption and catalytic
applications.
Steam from the flashed geothermal fluid is used to produce electricity (output power of ~75
MWe). The residual liquid is used in a heat exchange process (output power of ~150 MWt) to
heat up freshwater for district heating of local communities of roughly 20,000 habitants. This
heat exchange process limits the minimum temperature for heat extraction of the geothermal
fluid to about 90°C. Most of the spent geothermal fluid is reinjected into the geothermal
reservoir (~6×106 m3 annually) but some of it (~1.2×106 m3 annually) is discharged on the
surface where it forms the Blue Lagoon (Grether-Beck et al., 2008; Petursdottir et al., 2009).
A small fraction of the discharged fluid is bypassed to sedimentation tanks at the Blue Lagoon
where it cools from 90°C to ambient temperature. The cooling causes supersaturation of the
silicic acid which in turn precipitates as amorphous SiO2. The pH of the resulting supernatant
3
is 7.7 ±0.2, sligthly higher than the pH of the Blue Lagoon which is 7.5 ±0.2. At 90°C, the
geothermal fluid contains about 600 ppm SiO2 and thus the 7×106 m3 of liquid being
discharged and reinjected annually carries about 4,000 tonnes. At 10-15°C, a realistic ambient
temperature, the SiO2 concentration has dropped by roughly an order of magnitude (Fleming
and Crerar, 1982) and thus a precipitation of over 3,000 tonnes could potentially be harnessed
annually from fluid discharged from the HS-Energy facility alone.
Few authors have reported on practical applications of modified geothermal SiO2 and still
fewer on applications of unmodified SiO2. A possible use of geothermal SiO2 as a filler in
paper (Johnston et al., 2004) and as a precursor for silicates (Gallup et al., 2003) has been
described. In both cases the precipitation conditions had to be controlled. The use of natural
SiO2, modified with organosilicate materials, as a chromatographic material has also been
discussed (Tarasevich et al., 1990).
1.2 Chromatographic silica
Silica gel (SiO2∙xH2O) is widely used as an adsorbent in chromatographic columns for
isolation and purification of compounds from a mixture. One of the most common methods
for the analysis of basic pharmaceuticals is liquid chromatography (McKeown et al., 2001),
which conventionally relies on synthetic silica and silica derivatives as the stationary phase.
The production of synthetic chromatographic silica typically involves several chemical
reaction steps followed by a series of after-treatment processes (Hoffmann et al., 2006; Zhang
et al., 2009). In this paper we discuss the chromatographic application of unmodified
geothermal silica. A comparison to sintered geothermal silica is also made.
1.3 Phycocyanin
The Blue Lagoon is a specific geothermal biotope known for its unique microbial ecosystem
(Petursdottir and Kristjansson, 1997; Petursdottir et al., 2009). It contains about 6,000 m3 of
geothermal fluid that is replenished every 40 h. The lagoon′s temperature remain constant at
38 ±1°C. The coccoid blue-green algae, Cyanobacterium aponinum, one of the dominating
species in the microbial ecosystem, is used in this research. A water-soluble photosynthetic
pigment in blue-green algae, Phycocyanin (C-PC), is a protein that belongs to a family of
phycobiliproteins. It is an accessory pigment to the green-colored pigment chlorophyll, also
found in blue-green algae. Together they are an essential component of the algae′s light
harvesting system (McCall, 1998). Phycobiliproteins have in common a similar three-
dimensional structure in addition to their hydrophilic nature. Among the many interesting
4
applications of C-PC is its use as a fluorescent marker of cells and macromolecules (Ramos et
al., 2010) and as a natural colorant in food and cosmetic products, replacing synthetic dyes,
which are often unsafe or even toxic. C-PC has also been shown to exhibit bio-activity
(Eriksen, 2008) which makes it an excellent choice as an additive in food and pharmaceutical
products. However, the use of C-PC in these products is dependent on obtaining the
appropriate grade of purity. The purity of C-PC can be evaluated as the ratio between the light
absorbance at
λ
= 620 nm and 280 nm (A620/A280), where A620nm is the maximum absorbance
of C-PC and A280nm is the total absorbance of proteins. A purity of 0.7 is considered food
grade, 3.9 reactive grade and greater than 4.0 analytical grade (Rito-Palomares et al., 2001).
Despite the many possible applications of phycobiliproteins, their use is limited by the high
cost of extraction and purification. Most of the methods used for purification of C-PC involve
a sequence of operations that include precipitation, centrifugation, dialysis, ion-exchange and
gel filtration chromatography and chromatography on hydroxyapatite (Rito-Palomares et al.,
2001). The purification cost has been estimated at 50-90% of the total production cost (Patil et
al., 2006). Thus, improvements in the purification procedure can lead to a significant
reduction in the production cost. C-PC is unstable to heat and light in an aqueous solution and
denatures at temperatures above 45°C (Jespersen et al., 2005). This instability puts constraints
on the possible processing methods that can be used. In the experiments described in this
paper, unmodified geothermal SiO2 precipitated from Blue Lagoon geothermal fluid was used
as an alternative to other chromatographic materials to extract C-PC from the coccoid blue-
green alga, cyanobacterium aponinum. When SiO2 powder was soaked in a saline solution of
a ruptured cell mass, the C-PC was selectively adsorbed in contrast to other hydrophobic
constituents (such as chlorophyll). The attached C-PC was released from the SiO2 adsorbent
by washing with deionized water.
2. Experimental Conditions
2.1 Algae cultivation
Blue-green algae, isolated from the Blue Lagoon, were cultivated in a semi-continuous mode
in a 1.2 m3 tubular photobioreactor at 45°C at the Blue Lagoon Research and Development
Center in Iceland. The cultivation media was geothermal fluid with 0.3% mass/vol Cell-hi WP
nutrient. Illumination was provided by a high pressure sodium light (160 µE/m2/s). A fixed
pH of 7.5 was maintained by regulating the CO2 gas feed rate during growth. At harvesting,
5
the algae suspension contained 12.2% wt. dry weight of algae. A 360 ml volume of the
suspension was homogenized for 10 min using a 900 W ultrasonic cell crusher at 20 kHz
(SYJ900-D from Sharpertek) with a duty cycle of 2 s on and 3 s off. Subsequently, the
solution was centrifuged at 3200 × g for 10 min. A 288 ml volume of supernatant (referred to
as crude extract) was obtained and collected.
2.2 Chromatographic silica
Raw materials used for the chromatographic recovery of C-PC consisted of geothermal SiO2
(referred to as BL-silica) and geothermal salt (referred to as BL-salt). The chemical
composition of the BL-silica and the BL-salt was determined by inductively coupled plasma
mass spectrometry (ICP-MS). The BL-silica was removed from the sedimentation tank and
pumped into a filtration press at a pressure of 2 bar. The resulting filter cake was dried at
60°C, crushed manually and sieved. After removal of the BL-silica, the BL-salt was prepared
by drying the supernatant. The characteristics of the BL-silica (ground in a mortar), before
and after sintering at 1000°C for 2 h, was determined using a scanning electron microscope
(SEM) and by measuring the specific surface area (BET), t-plot area and Barrett-Joyner-
Halenda (BJH) average adsorption pore width (Micromeritics TriStar 3000 Surface Area and
Porosity Analyzer). The mineralogy of the BL-silica was determined by X-ray diffraction
analysis (XRD: Bruker AXS).
2.3 Protein extraction
Two different approaches for adsorption and subsequent elution were applied; Method 1 and
Method 2. In the following discussion, we focus primarily on the latter procedure.
2.3.1 Method 1
BL-silica agglomerates, ranging in size from 0.2 to 0.7 mm, were packed in a 30 cm tall
plexiglas column with a diameter of 5 cm. The column was loaded with ~190 ml crude extract
(see 2.1 Algae cultivation above) at a pressure of 0.4 bar. Afterwards, 2.2 l of BL-salt solution
(25% wt. BL-salt in deionized water) was pumped through the column to flush the
chlorophyll. Subsequently the C-PC was eluted using a continuous flow of 2.0 l deionized
water. The pressure was gradually increased from 0.4 to 0.6 bar during pumping of the BL-
salt solution and the eluent.
2.3.2 Method 2
6
BL-silica that passed through a 125 µm mesh screen was mixed with the crude extract and a
BL-salt solution. The C-PC discharge was accomplished batchwise by alternating cycles of
centrifugation and the addition of deionized water. A 100 ml crude extract (see 2.1 Algae
cultivation above) was mixed with 10 ml of saturated BL-salt solution (approximately 0.36 g
salt per ml) and 20 g of BL-silica and then centrifuged. The supernatant was discarded and the
sediment was slurried in 50 ml of saline solution (40 ml of deionized water and 10 ml of
saturated BL-salt solution) and centrifuged again. This step of slurrying the sediment and
centrifuging was repeated with deionized water in discrete volumes of 10 ml. The supernatant,
containing eluted C-PC, was collected after each step. A flow sheet of the extraction and the
purification process is shown in Fig. 1. Absorption values and absorbance spectra of the
supernatant (eluent) and the crude extract were measured using a UV-visible Camspec M350
spectrophotometer. The Purity Ratio (PR) of the C-PC was taken as the A620nm/A280nm ratio.
Methods 1 and 2 were repeated with sintered (1000°C/2h) BL-silica, instead of unsintered
silica, upon which the binding capacity of the silica was completely lost.
3. Results and discussion
3.1 Protein extraction - Method 1
The proceeding of the column′s loading and elution is visualized in chronological order in
Fig. 2. As seen, the green colored chlorophyll flows through the column while the C-PC is
retained at the stationary BL-silica. A good separation of C-PC was obtained, with a net 3-
fold increment in the PR. The column, however, degraded due to compression of the
agglomerates during pumping of the eluent. Consequently the flow rate through the column
gradually decreased to ~4 ml/min which is impractically low. The possibility of using the
geothermal SiO2 as a stationary phase in the chromatographic column requires further
evaluation. An attempt to strengthen the SiO2 agglomerates with sintering resulted in a total
loss of performance accompanied by a decrease of its specific surface area from 60 to 1 m2
(see Table 1.). This problem might be overcome by embedding the unsintered SiO2
agglomerates in a rigid matrix, but such speculations are beyond the scope of this paper.
3.2 Protein extraction - Method 2
The concentration of C-PC was determined using Eq. (1) (Mishra et al., 2008), where A620 and
A652 are the absorption values at 620 and 652 nm, respectively.
7
CC-PC = (A620-0.474A652)/(5.34) [mg/ml] (1)
A graphical interpretation of the result is given in Fig. 3. It is apparent from the figure that the
shapes of the concentration and purity curves resemble each other. Both lines rise roughly
linearly with increasing eluent volume up to ~140 ml. At higher volumes the concentration
decreases throughout the elution while the purity line flattens out rather abruptly at volumes
above ~200 ml. The maxima of the PR, ~2.0, and the CC-PC, 6.3 mg/ml, appear at the same
eluent volume (~140 ml). As always, there is a trade-off between the purity of the protein and
its recovery rate. Several different batches were collected. The 10 ml volume with the highest
purity will be referred to as collection #1. The volume fractions with PR ≥ 1.0 were later
combined, making a total volume of 117 ml (collection #2) with 4.9 mg/ml C-PC and a PR of
1.3, which corresponds to 31% recovery of the C-PC. Combining the volume fractions
between 40 and 250 ml resulted in a PR of 1.0 and 50% recovery (collection #3). Absorption
spectra, in the range of 250-700 nm, of collections #1 and #2, are shown in Fig. 4. A spectrum
of the crude extract is shown for comparison.
The PR and CC-PC values for the crude extract are 0.46 and 18.2 mg/ml, respectively. It
therefore contained ~1,800 mg C-PC (100 ml×18.2 mg/ml) whereas the area under the
measured concentration curve (Fig. 3.) corresponds to ~1,100 mg. This equals to a total
recovery of 61% ((1.1/1.8)×100%). A higher BL-silica/crude-extract ratio might increase the
recovery rate of the C-PC but most likely at the expense of its purity.
3.3 Morphology analyses of the chromatographic silica
Values of the specific surface area, micropore surface area, pore volume and average
adsorption pore width of the BL-silica, before and after sintering are given in Table 1. Table
2. shows the chemical composition of the BL-silica and the BL-salt.
The XRD patterns of sintered and unsintered BL-silica are shown in Fig. 5. The amorphous
structure of the latter is clearly demonstrated by the absence of sharp diffraction peaks. The
dry SiO2 powder contains ~7% wt. of salt (mainly NaCl). Sharp peaks in Fig. 5 at 2Ɵ =
31.75° and 45.48° are due to crystalline NaCl present in the BL-silica. After sintering, a
strong and sharp peak has appeared at 2Ɵ = 22°, indicating some of the amorphous SiO2 has
been converted to a crystalline form of SiO2 (cristobalite). The concentration of NaCl,
however, appears to have decreased, as indicated by the lower intensity of the diffraction
peaks.
8
Fig. 6. shows the SEM images at two different magnifications of BL-silica agglomerates
before and after sintering. Prior to sintering we see a porous surface composed of particles
with diameters ranging from 20 nm to 90 nm. According to Icopini et al. (2005), who
investigated oligomerization of SiO2 as a function of its concentration, ionic strength and pH
in natural brine solutions, coagulation occurs when the SiO2 particles reach ~3 nm in size.
Our values of 20-90 nm are much larger than the value they give. However, it must be taken
into account that in our case the silica precipitation was triggered by change in temperature
from whereas in their case, it was triggered by a change in the pH. Upon sintering the
particles have grown and apparently fused together. The porous appearance is no longer
visible; instead a glass-like surface has formed. These structural changes may, in part, be
attributed to the salt present in the BL-silica. As the sintering temperature of 1000°C is well
above the melting point of NaCl (801°C), the SiO2 could have been coated with the molten
salt which solidifies upon cooling. Additional phases, formed by reactions of the SiO2 and the
molten salt, may also be present in a small amount. Crystalline particles, roughly 10 µm on a
side, were observed after sintering (see inset in Fig. 6(b).). A further confirmation of the
morphological changes occurring in the sintered SiO2 is indicated by the decrease in the pore
volume, from 0.24 to 0.001 cm3/g (Table 1.). This is consistent with the observed loss of
performance of the SiO2 after sintering. The average pore width of 23 nm is several times the
size of the C-PC protein molecule (Fisher et al., 1980). These facts could explain the selective
adsorption of C-PC from the crude extract containing chlorophyll and C-PC. It can therefore
be speculated that the chromatographic mechanism is size dependent (Paul-Dauphin et al.,
2007).
4. Conclusions
A possible use of geothermal SiO2 as a chromatographic material has been described. SiO2
deposits have been considered an undesirable byproduct of geothermal power production and
large quantities of it can be extracted from geothermal fluid. Its ability to adsorb protein
molecules was demonstrated by the separation of C-PC from disrupted blue-green algae mass.
A critical factor controlling the adsorption capability is the high specific surface area as
observed by loss of performance upon sintering. The process described here has considerable
potential as a simple, ecological and inexpensive first step in the purification and separation of
natural proteins by utilizing geothermal resources.
9
Acknowledgements
This research was supported in part by the Icelandic Technology Development Fund and the
Energy Fund of the National Energy Authority.
References
Arnorsson S., 1995. Geothermal systems in Iceland: Structure and conceptual models. 1.
High-temperature areas, Geothermics 24, 561-602.
Eriksen N.T., 2008. Production of phycocyanin - a pigment with applications in biology,
Appl. Microbiol. Biot. 80, 1-14.
Fisher R.G., Woods N.E., Fuchs H.E., Sweet R.M., 1980. Three-dimensional structures of C-
Phycocyanin and B-Phycoerythrin at 5-A resolution, J. Biol. Chem. 255, 5082-5089.
Fleming B., Crerar D., 1982. Silicic acid ionization and calculation of silica solubility at
elevated temperature and pH application to geothermal fluid processing and
reinjection, Geothermics 11, 15-29.
Gallup D.L., Sugiaman F., Capuno V., Manceau A., 2003. Laboratory investigation of silica
removal from geothermal brines to control silica scaling and produce usable silicates,
Appl. Geochemi. 18, 1597-1612.
Grether-Beck S., Muhlberg K., Brenden H., Felsner I., Brynjolfsdottir A., Einarssson S.,
Krutmann J., 2008. Bioactive molecules from the Blue Lagoon: in vitro and in vivo
assessment of silica mud and microalgae extracts for their effects on skin barrier
function and prevention of skin ageing, Exp. Dermatol. 17, 771-779.
Hoffmann F., Cornelius M., Morell J., Froeba M., 2006. Silica-based mesoporous organic-
inorganic hybrid materials, Angew. Chem. Int. Edition 45, 3216-3251.
Icopini G.A., Brantley S.L., Heaney P.J., 2005. Kinetics of silica oligomerization and
nanocolloid formation as a function of pH and ionic strength at 25 degrees C,
Geochim. Cosmochim. AC. 69, 293-303.
Jespersen L., Stromdahl L.D., Olsen K., Skibsted L.H., 2005. Heat and light stability of three
natural blue colorants for use in confectionery and beverages, Eur. Food Res. Technol.
220, 261-266.
Johnston J.H., McFarlane A.J., Borrmann T., Moraes J., 2004. Nano-structured silica and
silicates – new materials and their applications in paper, Curr. Appl. Phys. 4, 411-414.
McCall R., 1998. Cyanobacterial phycobilisomes, J. Struct. Biol. 124, 311-334.
McKeown A., Euerby M., Lomax H., Johnson C., Richie H., Woodruff M., 2001. The use of
silica for liquid chromatographic/mass spectrometric, J. Sep. Sci. 24, 835-842.
10
Mishra S.K., Shrivastav A., Mishra S., 2008. Effect of preservatives for food grade C-PC
from spirulina platensis, Process Biochem. 43, 339-345.
Patil G., Chethana S., Sridevi A., Raghavarao K., 2006. Method to obtain C-Phycocyanin of
high purity, J. Chromatogr. A 1127, 76-81.
Paul-Dauphin S., Karaca F., Morgan T.J., Millan-Agorio M., Herod A.A., Kandiyoti R., 2007.
Probing size exclusion mechanisms of complex hydrocarbon mixtures: The effect of
altering eluent compositions, Energ. Fuel 21, 3484-3489.
Petursdottir S.K., Bjornsdottir S.H., Hreggvidsson G.O., Hjorleifsdottir S., Kristjansson J.K.,
2009. Analysis of the unique geothermal microbial ecosystem of the Blue Lagoon,
FEMS Microbiol. Ecol. 70, 425-432.
Petursdottir S.K., Kristjansson J.K., 1997. Silicibacter lacuscaerulensis gen. nov., sp. nov., a
mesophilic moderately halophilic bacterium characteristic of the Blue Lagoon
geothermal lake in Iceland, Extremophiles 1, 94-99.
Ramos A., Gabriel Acien F., Fernandez-Seville J.M., Gonzalez C.V., Bermejo R., 2010.
Large-scale isolation and purification of C-phycocyanin from the cyanobacteria
Anabaena marina using expanded bed adsorption chromatography, J. Chem. Technol.
Biot. 85, 783-792.
Rito-Palomares M., Nunez L., Amador D., 2001. Practical application of aqueous two-phase
systems for the development of a prototype process for C-Phycocyanin recovery from
spirulina maxima, J. Chem. Technol. Biot. 76, 1273-1280.
Saemundsson K., Johannesson H., Hjartarson A., Kristinsson S.G., Sigurgeirsson M.A., 2010.
Geological map of southwest Iceland, Iceland GeoSurvey, scale 1:100.000.
Tarasevich Y.I., Nazarenko A.V., Bondarenko S.V., 1990. Preparation, properties and
chromatographic applications of natural and synthetic amorphous silica modified by
organosilicate substances, Abstr. Pap. Am. Chem. S. 200, 186.
Zhang A., Xiao C., Xue W., Chai Z., 2009. Chromatographic separation of cesium by a
macroporous silica-based supramolecular recognition agent impregnated material,
Sep. Purif. Technol. 66, 541-548.
Figure captions:
Fig. 1. A flow sheet of the C-PC extraction process.
Fig. 2. Separation of C-PC at different stages in chronological order: (a) at early stage of the
loading, (b) after loading with 190 ml crude extract, (c) after flushing with 2.2 l of brine
solution (25% BL-salt in deionized water), and (d) after elution with 2.0 l deionized water.
11
Fig. 3. Concentration of C-PC (dashed line) and C-PC purity (solid line) as a function of
eluent volume.
Fig. 4. Absorption spectra for algae extracts of different purity: crude extract (dotted line),
collection #1 (dashed line), and collection #2 (solid line).
Fig. 5. X-ray diffraction pattern of BL-silica before (black solid line) and after sintering at
1000°C for 2 hours (red dotted line).
Fig. 6. SEM images at two different magnifications of the BL-silica: a) before sintering; and
b) after sintering. The inset in b) shows crystals formed during the sintering process.
Table 1. Surface characteristics of the BL-silica.
BL-silica
BET spec.
surf. area
[m2/g]
Micropore
surf. area
t-Plot [m2/g]
Vol. of pores
with diameter
1.7 – 300 nm
[cm
3
/g]
BJH adsorption
av. pore width
[nm]
Unsintered 50 6 0.24 23
Sintered
1
0.3
0.001
N/A
Table 2. Chemical composition of BL-silica and BL-salt (wt. %).
Component
SiO2
Cl
Na
Ca
K
Mg
Total
BL-silica
92.9
3.9
2.2
0.49
0.39
< 0.1
99.9
BL-salt
0.02
58.7
34.0
1.83
1.75
< 0.1
96.3
