lable at ScienceDirect

Journal of Environmental Radioactivity 129 (2014) 133e139

Contents lists avai

Journal of Environmental Radioactivity

journal homepage: www.elsevier .com/locate/ jenvrad

The enrichment of natural radionuclides in oil shale-fired powerplants in Estonia e The impact of new circulating fluidized bedtechnology

Taavi Vaasma a,*, Madis Kiisk a, Tõnis Meriste b, Alan Henry Tkaczyk a

a Institute of Physics, University of Tartu, Riia 142, 51014 Tartu, Estoniab Tallinn University of Technology, Ehitajate tee 5, 19086 Tallinn, Estonia

a r t i c l e i n f o

Article history:Received 20 May 2013Received in revised form28 December 2013Accepted 2 January 2014Available online 21 January 2014

Keywords:Oil shaleNatural radionuclideCirculating fluidized bed boilerRadionuclide enrichmentActivity balanceAtmospheric emission

* Corresponding author. Tel.: þ372 737 4622.E-mail address: taavi.vaasma@ut.ee (T. Vaasma).

0265-931X/$ e see front matter � 2014 Elsevier Ltd.http://dx.doi.org/10.1016/j.jenvrad.2014.01.002

a b s t r a c t

Burning oil shale to produce electricity has a dominant position in Estonia’s energy sector. Around 90% ofthe overall electric energy production originates from the Narva Power Plants. The technology in use hasbeen significantly renovated e two older types of pulverized fuel burning (PF) energy production unitswere replaced with new circulating fluidized bed (CFB) technology. Additional filter systems have beenadded to PF boilers to reduce emissions.

Oil shale contains various amounts of natural radionuclides. These radionuclides concentrate andbecome enriched in different boiler ash fractions. More volatile isotopes will be partially emitted to theatmosphere via flue gases and fly ash. To our knowledge, there has been no previous study for CFB boilersystems on natural radionuclide enrichment and their atmospheric emissions.

Ash samples were collected from Eesti Power Plant’s CFB boiler. These samples were processed andanalyzed with gamma spectrometry. Activity concentrations (Bq/kg) and enrichment factors werecalculated for the 238U (238U, 226Ra, 210Pb) and 232Th (232Th, 228Ra) family radionuclides and for 40K indifferent CFB boiler ash fractions. Results from the CFB boiler ash sample analysis showed an increase inthe activity concentrations and enrichment factors (up to 4.5) from the furnace toward the electrostaticprecipitator block. The volatile radionuclide (210Pb and 40K) activity concentrations in CFB boilers wereevenly distributed in finer ash fractions.

Activity balance calculations showed discrepancies between input (via oil shale) and output (via ashfractions) activities for some radionuclides (238U, 226Ra, 210Pb). This refers to a situation where themissing part of the activity (around 20% for these radionuclides) is emitted to the atmosphere. Alsodifferent behavior patterns were detected for the two Ra isotopes, 226Ra and 228Ra. A part of 226Ra inputactivity, unlike 228Ra, was undetectable in the solid ash fractions of the boiler. Most probably it is releasedto the surrounding environment.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Burning oil shale in Narva Power Plants provides about 90% ofthe overall electric energy produced in Estonia. In a similar way tocoal burning (Gür and Yaprak, 2010; Papastefanou, 2010), oil shalepower plants have a radiological impact on the environment. This isdue to natural radionuclides, found in oil shale, which concentrateand enrich in the boiler ash fractions during the burning process.Radionuclide activity concentrations and migration to the sur-rounding environment from coal-fired power plants has been

All rights reserved.

studied extensively (Papastefanou, 2010; Pandit et al., 2011), butthe data concerning oil shale power plants is inadequate. Renova-tion of older technological units at Narva Power Plants and con-struction of a new power plant clearly indicates that energyproduction from oil shale will remain an integral part of Estonia’senergy sector in the coming decades, so it is important to under-stand the potential radiological impact of such power plants.

According to the 2009/2010 yearly economic report (EestiEnergia AS, 2007), the installed power capacity of Narva PowerPlants was 2380 MWe, of which Eesti Power Plant (EPP) constituted1615 MWe and Balti Power Plant (BPP) 765 MWe. Different types ofoil shale burning technologies are implemented in Eesti and BaltiPower Plants e PF (pulverized fuel) and CFB (circulating fluidizedbed) boilers. During the past 15 years, Narva Power Plants have

Fig. 1. CFB boiler in Eesti and Balti Power Plant. Numbers indicate various boiler elements. 1 e furnace chamber; 2 e separating chamber; 3 e convective superheater and reheater;4 e economizer; 5 e air preheater; 6 e heat exchanger (INTREX); 7 e electrostatic precipitator (ESP) block (B e ESP 1; C e ESP 2; D e ESP 3 and E e ESP 4). Letters indicate ashsample collection points.

T. Vaasma et al. / Journal of Environmental Radioactivity 129 (2014) 133e139134

been renovated significantly e new boilers, filter systems andtechnological innovations have been put in place, including shiftingproduction from PF to CFB units. CFB boilers have a higher fuelburning efficiency and the amount of gaseous compounds releasedto the atmosphere is lower compared to PF boilers. These techno-logical changes combined with the differences in burned oil shalevolumes have lessened the overall atmospheric emission rates.

Approximately 11e12 million tonnes of oil shale are burnedannually in Narva Power Plants to produce electricity. This createsapproximately 6 million tonnes of oil shale ash which are trans-ported to nearby ash fields. During the burning process, oil shalenatural radionuclides are redistributed and enriched in different ashfractions. The majority of these radionuclides are bound to ashparticles, but a detectable amount of the volatile radionuclides arereleased to the atmosphere. Currently there is inadequate amount ofdata available on how radionuclide dynamics and activity concen-tration fluctuations depend on fuel and ash chemical compositionand boiler parameters. From the perspective of environmental andradiation protection, it is important to gather relevant and up-to-date radiological data about the new CFB technology.

Ash samples were collected from different parts of Eesti PowerPlant CFB boiler’s burning tract. Gamma spectrometry was used todetermine activity concentrations and enrichment factors of thefollowing radionuclides: 238U, 226Ra, 210Pb, 232Th, 228Ra and 40K. Theenrichment factor is defined as a ratio of a radionuclide activityconcentration in a given ash fraction to that measured in the dry oilshale. The main purpose was to determine how industrial usage ofoil shale affects natural radionuclide migration and enrichmentprocesses in different parts of the boiler burning tract. Activitybalance calculations were done to determine whether the radio-nuclides are bound to ash fractions or released to the environment.The activity concentrations and enrichment factors of 238U and232Th family radionuclides and 40K determined in this study canprovide an important platform for future studies to assess the oilshale industry’s radiological impact on the population and envi-ronment. Radiological impact could be a new parameter to takeinto account in future oil shale power plant environmental impactevaluations.

2. Description of the object

2.1. Burning oil shale in CFB and PF boilers

Twenty PF and four CFB (Fig. 1) boilers operating in NarvaPower Plants produce most of the electricity used in Estonia.Although the number of CFB boilers is noticeably smaller than thePF ones, the future development of oil shale burning technologyis based on them. Enormous ash quantities produced by the oilshale industry are due to the high mineral composition andrelatively modest organic fraction (10e65%) of the oil shale (EestiEnergia AS, 2010). Oil shale has a lower heating value between 7.8and 8.9 MJ kg�1, with an average of 8.4 MJ kg�1 (Plamus et al.,2011a). For comparison, the heating value of coal that is mostcommonly used in power plants usually remains between 9 and27 MJ kg�1.

The furnace temperature range is an important difference be-tween the two technologies. In PF boilers, the furnace tempera-ture range is between 1400 and 1500 �C, but in CFB boilers, it staysbetween 750 and 950 �C. Due to lower temperatures, the oil shalecarbonaceous mineral part, mostly limestone and dolomite, willbe subject to less intense decomposition (Ots, 2006; Pihu et al.,2006).

2.2. Ash flow rate in PF and CFB boilers

The boiler ash flow rate values are important parameters inactivity balance calculations. The average size of the fuel particlestransported to the PF boiler furnace is between 35 and 60 mm. Theparticle size in CFB boilers stays between 1 and 10 mm. Frictionbetween fuel particles in the circulating layer of the CFB boilerfurnace significantly decreases the particle size (Kuusik et al., 2005;Ots, 2006).

The ash flow proportions (Fig. 2) between PF and CFB boilersalso vary.

CFB boilers do not have a cyclone block so the majority of fly ashthat exits the furnace chamber is precipitated in the first field of theelectrostatic precipitator (ESP).

Fig. 2. Distribution of ash flow rates in PF and CFB boilers. Data from (Ots, 2006; Plamus et al., 2011b).

T. Vaasma et al. / Journal of Environmental Radioactivity 129 (2014) 133e139 135

2.3. Fly ash

Ash particles are finer in CFB boiler ash fractions (Fig. 3)compared to the ones in PF boilers. This is due to the intense me-chanical friction of the oil shale mineral part (Loosaar et al., 2005;Ots, 2006). The boiler’s operating regime and fuel burning pro-cess can be considered constant enough to presume that the ashparticle granulometry does not vary extensively with time. There-fore, some of the data gathered in the past can be used in currentanalysis.

Although the chemical composition of CFB and PF boiler asheshas no remarkable dissimilarities (Ots, 2006; Plamus et al., 2011a),the shape and structure of the ash particles have distinctive dif-ferences (Bityukova et al., 2010). Ash particles generated in the PFboiler are spherical, in contrast to the CFB boiler ashes which havean irregular and porous structure (Bityukova et al., 2010).

High temperatures in the furnace cause some of the trace ele-ments to vaporize. As the temperature declines toward the end ofthe burning tract, volatilized particles combine with one another,condense onto ash and aerosol particles, or stay in the gaseousphase to be emitted to the atmosphere via flue gases or fly ash(depending on the efficiency of the ESP fields) (Ots, 2006). Similarbehavior can be expected from radionuclides. The trace elements(Ots, 2006) as well as radionuclides are also partly precipitated onthe heating surfaces.

Fly ash concentrations emitted to the atmosphere by CFB boilersare between 25 and 30 mg m�3 and around 200 mg m�3 from PF

Fig. 3. Size distribution (mm) of ash particles in different parts of the CFB bo

boilers (Pihu et al., 2006). The fly ash carried through the ESP blockin Narva Power Plant CFB boilers are fine in size: 55e65% (masspercentage) particles are with size up to 2.5 mm, 33e43% remainbetween 2.5 and 10 mm and 2e4% of the particles are larger than10 mm (mass concentrations were measured with Johnas II cascadeimpactor) (Parve et al., 2011). The efficiency of the ESP block issignificantly higher catching solid particles with size greater than10 mm than under 2.5 mm (Parve et al., 2011). The main path for thetrace elements to be emitted to the environment is most likely inthe complex of small particles and/or via flue gases.

The size of the ash particles decreases noticeably from thefurnace toward the ESP block. Ash particles in the ESP block havethe smallest particle diameter (Fig. 3), around 90% of the size re-mains under 0.045 mm.

2.4. Natural radionuclides from fossil fuel-fired power plants

For the PF boiler, the natural radionuclide activity concentra-tions and enrichment factors in ash fractions as well as radionuclideatmospheric emissions have been previously evaluated to someextent (Realo et al., 1996; Realo and Realo, 1997; 1999). Radionu-clide activity concentrations in ashes and radionuclide emissions tothe surrounding environment have been previously evaluated forcoal-fired power plants (Gür and Yaprak, 2010; Karangelos et al.,2001; Pandit et al., 2011). These studies show that burning fossilfuels is associated with radionuclide enrichment in the boiler sys-tem and migration processes to the environment. The radionuclide

iler burning tract (%), Eesti Power Plant. Data from (Kuusik et al., 2005).

Table 1Radionuclide activity concentrations in CFB boiler ash fractions (Bq kg�1 �1s) (EestiPower Plant e energy production unit 8, boiler 1).

Bottom ash ESP 1 ESP 2 ESP 3 ESP 4

Radionuclide238U 22.4 � 3.2 60.4 � 5.9 64.1 � 6.2 74.8 � 9.1 75.8 � 7.8226Ra 18.7 � 1.0 58.8 � 1.6 61.7 � 1.5 75.0 � 1.7 90.7 � 2.2210Pb 17.9 � 4.2 68.7 � 6.6 68.4 � 7.8 79.7 � 11.8 75.4 � 11.5228Ra 13.4 � 1.3 36.0 � 2.0 37.9 � 1.8 45.1 � 2.1 52.4 � 2.6232Th 11.9 � 1.1 35.0 � 1.7 34.9 � 1.6 36.9 � 1.7 40.4 � 2.140K 245.0 � 7.4 1251.6 � 16.5 1223.4 � 15.5 1303.8 � 16.8 1201.0 � 18.6

T. Vaasma et al. / Journal of Environmental Radioactivity 129 (2014) 133e139136

propagation area depends on the fuel type, technology in use andatmospheric conditions. Waste volumes and other environmentalimpacts that oil shale power plants bring about have been exten-sively researched, but yet to be studied is the radiological impact ofsuch power plants.

3. Materials and methods

3.1. Sample collection

Oil shale and ash samples were collected from the CFB boiler(Fig. 1) of Eesti Power Plant. Samples were taken from the main ashprecipitation points. This enabled us to determine the activityconcentrations in various ash fractions and to evaluate theirenrichment factors.

All samples were taken during the same day, a few hours apart.Samples were collected in a way which ensured that the sampledoil shale would be represented in collected ash fractions. Due tovariations in oil shale composition and the amounts of oil shaleburned annually, the analysis results of ash samples collected atdifferent times will probably not coincide precisely.

It was not possible to collect samples from the heat exchanger(INTREX), superheater, economizer and air preheater. The absenceof these samples is corrected with calculations. Radionuclide ac-tivity concentrations of the missing ash fractions are assumed to beequal to the corresponding activities in ESP 1 field (as the collectionpoint with most similar properties). All together, the missingfractions constitute around 9% of total ash mass flow of the boiler.

3.2. Sample preparation

Samples were preprocessed by drying and compacting. Dryingtook place at 105 �C for 24 h. Afterward, the samples were placed incylindrically shaped metal containers with a diameter of 62 mmand a height of 20 mm for gamma spectrometric analysis.

Oil shale and bottom ash samples needed to be ground to ach-ieve the necessary size for analysis. The grinding was done using ahand mill to a particle size of 0.5 mm or less. The ESP samples werehomogenous.

To enlarge the mass, all analyzed samples were compressedwith a hydraulic press and placed into containers. The sample massdepended on the porosity and compactability of the ash andremained between 50 and 95 g.

To avoid radon (222Rn) emanation, all samples were sealedhermetically. This was done using plasticine and insulating tape tohermetically seal the container and lid together. Our laboratorytesting has confirmed the functionality of this method. To achievesecular equilibrium between radium and its progenies (222Rn andits short lived daughters), the sealed samples were stored for atleast 23e27 days (6e7 half lives of 222Rn) before gamma spectro-metric measurements.

3.3. Sample analysis

High-purity Germanium detectors, planar BSI GPD-50400 (BSI,Latvia) and coaxial RG Ortec GEM-35200 (EG&G ORTEC, USA), wereused to measure the activities of selected natural radionuclides.GammaVision-32 (version 6.07, EG&G ORTEC) software was used toanalyze the gamma ray spectra.

Activity concentrations and enrichment factors were calculatedfor 238U (234Th e 63.3 keV), 226Ra (214Bi e 609.3 keV and 214Pb e

295.2 keV), 210Pb (46.5 keV), 232Th (208TI e 583.2 keV), 228Ra (228Ace 911.2 keV), 40K (1460 keV).

The measurement energy calibration was done using RGUsource. Radionuclide activity concentration from the spectrumwas

determined separately for each nuclide (peak to peak method).RGU, RGTh and RGK sources were used in efficiency calculations.RGU, RGTh and RGK are IAEA certified reference materials(International Atomic Energy Agency, 1987). According to the cer-tification, the reference material consists of diluted reference ore(uranium e RGU and thorium e RGTh) within a silica matrix. TheRGK source contains potassium sulphate (99.8%).

To determine 238U activity concentration, its daughter 234Th(63.3 keV) was used. The following presumptions were made: 1) Allof the 234Th in the samples originates from 238U. 2) The two ra-dionuclides have similar behavior properties in the burning pro-cess, which allows the activity concentrations to be considered thesame. Additional measurements were carried out 8 months after-ward, which confirmed the first presumption.

The sample measurement time varied from 12 to 24 h.

3.4. Self-absorption

A correction was applied to activity calculations to account forself-absorption of gamma photons with energies <100 keV, due topartial absorption of emitted photons in the sample matrix. Auranium dioxide and lead sources with smaller sizes and higheractivities were used in the measurement of the attenuated andunattenuated beam intensities. Measurements were done withoutcollimation and in accordance with Cutshall et al. (1983). Theapplied self-absorption correction factors (Fca) were calculated for210Pb and 234Th in the following way:

Fca ¼ Fa1=Fa0 (1)

Fa0 is the absorption factor of a comparison sample. Fa1 is the ab-sorption factor of the ash sample. The comparison sample was anIAEA certified reference material (RGU) which is used for the effi-ciency calculation.

4. Results and discussion

4.1. Activity concentrations

Results of the analyzed ash samples indicated a generally sys-tematic increase in the radionuclide activity concentrations movingfrom the furnace toward the back end of the burning tract (Table 1,Fig. 4) (see Table 2).

The combined uncertainty comprises of both sample andbackground statistical, efficiency and reference material activity.The efficiency uncertainty for 214Pb (295.22 keV) and 214Bi(609.32 keV) gamma energies was approximately 0.5% (at onesigma). The statistical uncertainty was the dominant factor.

Mass reduction due to organicmaterial burn out from oil shale isthe primary cause for the elevated radionuclide activity concen-trations in the collected ash fractions. Activity concentration dif-ferences in ash fractions depend on the characteristics of the boilersand the ash particles themselves. The activity concentration of 40K

Table 2Radionuclide activity concentrations (Bq kg�1 �1s) in oil shale.

Narva quarry Aidu quarry

Radionuclide238U 24.6 � 3.4 21.6 � 3.6226Ra 23.8 � 0.9 24.8 � 1.0210Pb 26.0 � 4.0 23.4 � 5.4228Ra 12.1 � 1.0 12.4 � 1.1232Th 12.0 � 0.9 11.9 � 0.940K 393.4 � 7.4 360.0 � 7.0

T. Vaasma et al. / Journal of Environmental Radioactivity 129 (2014) 133e139 137

is particularly distinguishable as it already has a high value in oilshale and which increases significantly in fly ash fractions.

Ash particles decrease in size from the furnace toward the ESPashes. ESP ashes have a greater specific surface area (Bityukovaet al., 2010) which allows the trace elements and radionuclides toassociate with them to a greater extent. The temperature decreasein the back end of the burning tract favors additional association ofpreviously vaporized particles through condensation processes.

Radionuclides that originate from the same decay chain (238U or232Th) have similar trends for activity concentration increase(Fig. 4). The more volatile radionuclides (210Pb and 40K) have quitestable activity concentrations in ESP fields (1e4). This situationdiffers from the one described for PF boilers by Realo and Realo(1997), where radionuclide activity concentrations have a distinctincreasing trend. These dissimilarities occur due to differences inthe input oil shale and the combustion technology investigated.

Compared to PF boiler (Realo and Realo, 1997) the 210Pb con-centrations in the CFB boiler remain lower in the last two ESP fields,but 40K activity concentration values are higher in all ESP ashes. TheK-40 activity concentration in the CFB boiler is relatively homoge-neously distributed in the different ESP ash fractions. This couldderive from the radionuclide peculiarities, combined with lowertemperatures at the back end of the burning tract which createsbetter conditions for the radionuclide to condense or combine withfly ash particles.

Enrichment factors (Fig. 5) follow a similar trend as the activityconcentrations, increasing in value toward the finer particles of theburning tract. Pb-210 and 40K are depleted in CFB boiler bottom ash,indicating the volatile character of these radionuclides. Theenrichment factors in the CFB boiler remain lower than thoseestimated by Realo and Realo (1997). The values for 210Pb and 40K in

Fig. 4. Radionuclide activity concentrations (Bq kg�1 �1s) i

the third field of the ESP block are 2.3 and 2.5 times lower,respectively. Activity concentration and enrichment factor differ-ences in ash fractions are mainly caused by the variations in tech-nological features and burning parameters of the PF and CFBboilers. Differences in construction and operation of the two boilertypes cause variations in the circulation and size distribution of ashparticles in various boiler parts.

Oil shale from Narva’s quarry was used for energy productionduring the sampling period. For this reason, activity concentrationvalues fromNarva oil shale samples were used in enrichment factorcalculations.

Oil shale samples were collected from Narva and Aidu quarries.The radionuclide activity concentration values in Narva and Aidusamples correlated (Table 2). A similar oil shale radionuclidecomposition was observed regardless of the oil shale mininglocation.

4.2. Radionuclide activity balance

Radionuclide activity concentrations for CFB boiler weremeasured to determine the extent to which the natural radionu-clides are combined with solid ash fractions.

To understand the radionuclide activity balance we firstconsider the ash balance in the boiler system. The ash flow ratio(Plamus et al., 2011b) of bottom:ESP 1:ESP 2:ESP 3:ESP 4 is corre-spondingly 0.374:0.467:0.068:0.0016:0.0001. In other words, forevery ton of ash produced, 374 kg is precipitated as bottom ash,467 kg as ESP 1 ash and so forth. The ash fractions that we wereunable to collect (discussed in paragraph 3.1) have an ash flow ratioof 0.09. The activity concentration of these missing fractions wasequaled with the value in ESP 1. Here an assumptionwas made thatprior to the ESP 1 field, the missing ash fractions have similar ac-tivity concentration values. An ash content of 45% (Plamus et al.,2011b) was used in calculations.

Approximately 300 tonnes of fly ash per year (in CFB boilers)migrate through the ESP block and are emitted to the atmosphere.This is negligible compared to the total ash volume (6 milliontonnes per year) and has minor importance in the overall ash flow.

An ash to oil shale activity ratio equal to 1 (Fig. 6) indicates thatthe sum of the radionuclide activity is entirely detectable in the ashfractions. All the radionuclide activity (within the limits of uncer-tainty) can be detected for 228Ra and 232Th. K-40 has a ratio of 0.95

n various ash fractions of Eesti Power Plant CFB boiler.

Fig. 5. Radionuclide enrichment factors (with a �1s uncertainty) in different ash fractions of Eesti Power Plant CFB boiler.

T. Vaasma et al. / Journal of Environmental Radioactivity 129 (2014) 133e139138

which indicates that 5% of the radionuclide activity is not detect-able in boiler system ash fractions. Most likely, this missing fractionis emitted to the atmosphere via fly ash and flue gases. The same isobservable, but at higher values, for 238U (19%), 210Pb (19%) and226Ra (20%). Minor atmospheric emissions of 228Ra and 232Th couldoccur, but the measurement and balance calculation accuracy doesnot enable us to determine it precisely.

Previous research on PF boilers (Realo and Realo, 1997) showsdifferent radionuclide migration behavior (Fig. 6) compared to ourCFB boiler results. In the PF boiler, the undetected fraction is 30% for210Pb, around 12% for 238U and 6% for 232Th. Although, within thelimits of uncertainty, radionuclide equilibrium can be presumed forall radionuclides (Fig. 6) except 210Pb (Realo and Realo, 1997). In theyears since the first study there has been remarkable technologicalprogress in power production, influencing radionuclide associationwith fly ash and flue gases as well as activity calculations.

The results reveal different behavior and migration mechanismsbetween radium isotopes (226Ra and 228Ra). This may be caused bythe origin of the radionuclides. Ra-226 is part of the 238U decaychain but 228Ra belongs to the 232Th decay chain. Uranium and

Fig. 6. Radionuclide activity ratio, ash/

thorium may not be equally combined with the oil shale mineralfraction. Similarly to coal, uranium may preferably form complexeswith the organic part of the fossil fuel (Papastefanou, 2010). Non-volatile radionuclides that are associated with the mineral frac-tion tend to migrate less in the boiler gas passage. A similarmechanism has been detected in coal fueled power plants(Papastefanou, 2010). Radionuclide behavior mechanisms areextremely complicated at different temperatures and requireadditional chemical analysis for better evaluation.

The atmospheric propagation of volatile 210Pb depends on thearea’s meteorological conditions. Pb-210 has a half-life of 22.3 yearswhich is long enough to cause noticeable accumulation in theenvironment if the emissions form the power plants are contin-uous. Due to the high dose coefficient for 210Pb, inhaling orconsuming the isotope with water or food products could prove tobe a noteworthy source of additional radiation dose.

An additional important dose contributor is 210Po (the daughternuclei of 210Pb). Po-210 has not been measured in the ashes createdby oil shale fired power plants, and is the subject of future work inthis area. This will require selection of suitable methodology,

oil shale (with a �1s uncertainty).

T. Vaasma et al. / Journal of Environmental Radioactivity 129 (2014) 133e139 139

appropriate ash collection methods and extensive laboratory ex-periments contingent on the ash characteristics. The current workconcentrates on the enrichment of specific natural radionuclides inthe CFB boiler, and it will be interesting to consider the impact ofother radionuclides such as 210Po in future research.

5. Conclusions

The oil shale combustion process in Narva Power Plants causesnatural radionuclides (found in the fossil fuel) to become enrichedin boiler ash fractions. The enrichment increases toward smallerparticle sizes and lower temperature values at the back end of theburning tract.

Radionuclide migration processes in the boiler system aredependent on the fuel properties (its origin, composition andorganic/mineral fraction) and boiler operational processes (tem-perature regime, ash circulation mechanisms, ash retention time inboiler). The installed technologies in Narva Power Plants (PF andCFB) have distinctive technological and operational variationswhich affect radionuclide migration processes in the boiler systemdifferently.

Results from the CFB boiler ash sample analysis showed an in-crease in the activity concentrations and enrichment factors fromthe furnace toward the ESP block. Compared to existing data (Realoand Realo, 1997) the volatile radionuclide (210Pb and 40K) activityconcentrations in CFB boilers have noticeably different values andare more evenly distributed in finer ash fractions than in PF boilers.

Radionuclide total association with CFB boiler ash fractions wasin some cases undetectable. Activity balance calculations revealed atendency for radionuclides to be partially emitted to the atmo-sphere. The undetected activity (20% for 226Ra and 19% for 238U and210Pb) is most likely released to the atmosphere via fly ash and fluegases. The radionuclide behavior and migration processes in theCFB boiler differ from those described for PF boilers in previouswork (Realo and Realo, 1997). This study demonstrates that CFBboilers have distinct radiological characteristics compared to PFboilers. It can be concluded that this technology would also beassociatedwith a different influence on the surrounding populationand environment.

The present work gives ground for further studies on radionu-clide enrichment processes in oil shale power plants, their atmo-spheric propagation and precise calculations of additional doses tothe public.

Acknowledgments

The article authors express their gratitude to AS Eesti Energia fortheir co-operation in gathering samples and sharing important andnecessary information. We acknowledge support from Estonian

Research Council award ETF9304 and the CRDF Global and EstonianScience Foundation 2010 Energy Research Competition (CRDF-ETFII), including award ESP1-7030-TR-11 and ETF award 22/2011.

References

Bityukova, L., Mõtlep, R., Kirsimäe, K., 2010. Composition of oil shale ashes frompulverized firing and circulating fluidized-bed boiler in Narva thermal powerplants, Estonia. Oil Shale 27, 339e353.

Cutshall, N.H., Larsen, I.L., Orslen, C.R., 1983. Direct analysis of 210Pb in sedimentsamples: self-absorption corrections. Nucl. Inst. Met. 206, 309e312.

Eesti Energia AS, 2010. Keskkonnaaruanne 2009e2010. Accessible: https://www.energia.ee/c/document_library/get_file?uuid¼bb250184-08c8-43f0-8eca-f429fca63656&groupId¼10187.

Eesti Energia AS, 2007. AS Narva Elektrijaamad Energiakompleksi ArendusprojektiKeskkonnamõju Hindamise Programm. Accessible: http://www.envir.ee/orb.aw/class¼file/action¼preview/id¼651475/EEþKMHþprogramm.pdf.

Gür, F., Yaprak, G., 2010. Natural radionuclide emission from coal-fired power plantsin the southwestern of Turkey and the population exposure to external radia-tion in their vicinity. J. Environ. Sci. Health e Part Toxic/Haz. Subst. Environ. Eng.45, 1900e1908.

International Atomic Energy Agency, 1987. Preparation and Certification of IAEAGamma Ray Spectrometry Reference Materials RGU-1, RGTh-1 and RGK-1. IAEA,Vienna. Report nr IAEA/RL/148.

Karangelos, D.J., Rouni, P.K., Anagnostakis, M.J., Hinis, E.P., Simopoulos, S.E., 2001.Technologically enhanced natural radioactivity of fly-ash produced in lignite-fired power plants and the associated radiological impact. In: ConferencePublication of the Third International Symposium on Naturally OccuringRadioactive Materials; September 17e21, 2001; Brussels, Belgium.

Kuusik, R., Uibu, M., Kirsimae, K., 2005. Characterization of oil shale ashes formed atindustrial-scale CFBC boilers. Oil Shale 22, 407e419.

Loosaar, J., Arro, H., Parve, T., Pihu, T., Prikk, A., Tiikma, T., 2005. New 215 MWel CFBpower units for Estonian oil shale. In: Proceedings of FBC 2005 18th Interna-tional Conference on Fluidized Bed Combustion; May 22e25, 2005; Toronto,Ontario, Canada.

Ots, A., 2006. Oil Shale Fuel Combustion. Technical University of Tallinn, Tallinn.Pandit, G.G., Sahu, S.K., Puranik, V.D., 2011. Natural radionuclides from coal fired

thermal power plants e estimation of atmospheric release and inhalation risk.Radioprotection 46, S173eS179.

Papastefanou, C., 2010. Escaping radioactivity from coal-fired power plants (CPPs)due to coal burning and the associated hazards: a review. J. Environ. Radioact.101, 191e200.

Parve, T., Loosaar, J., Mahhov, M., Konist, A., 2011. Emission of fine particulates fromoil shale fired large boilers. Oil Shale 28, 152e161.

Pihu, T., Arro, H., Prikk, A., Parve, T., Loosaar, J., 2006. Combustion experience ofEstonian oil shale in large power plants. In: International Conference on OilShale: “Recent Trends in Oil Shale”; 7e9 November 2006; Amman, Jordan.

Plamus, K., Soosaar, S., Ots, A., Neshumayev, D., 2011a. Firing estonian oil shale ofhigher quality in CFB boilers e environmental and economic impact. Oil Shale28, 113e126.

Plamus, K., Ots, A., Pihu, T., Neshumayev, D., 2011b. Firing Estonian oil shale in CFBboilers e ash balance and behaviour of carbonate minerals. Oil Shale 28, 58e67.

Realo, K., Realo, E., 1999. 210Pb and 238U in Estonian fuel products and ashes. In:Nordic Society for Radiation Protection/IRPA 12th Ordinary Meeting; 23e27August, 1999. Risoe National Laboratory, Skagen, Denmark, pp. 229e232.

Realo, K., Realo, E., 1997. Radionuclides in Estonian oil-shales and oil-shale ashes. In:Proceedings of the IRPA Regional Symposium on Radiation Protection inNeighbouring Countries of Central Europe; 8e12 September. IRPA, Prague,Czech Republic, pp. 161e164.

Realo, E., Realo, K., Jõgi, J., 1996. Releases of natural radionuclides from oil-shale-fired power plants in Estonia. J. Environ. Radioact. 33, 77e89.