Assessment of the radiological impacts of a zircon sand processing plant

Journal of Environmental Radioactivity 82 (2005) 237e250

www.elsevier.com/locate/jenvrad

Assessment of the radiological impactsof a zircon sand processing plant

Serena Righia,*, Massimo Andrettab, Luigi Bruzzia

aInterdepartmental Centre for Research in Environmental Science, University of Bologna,

via dell’Agricoltura 5, 48100 Ravenna, ItalybMontecatini Environmental Research Centre, Edison Group, viale Ciro Menotti 48,

48023 Marina di Ravenna, Italy

Received 29 June 2004; received in revised form 4 January 2005; accepted 20 January 2005

Abstract

The paper presents the results of a study on radiological impacts resulting from a zirconsand processing plant located in the North-Eastern part of Italy.

Activity concentrations of radionuclides found in materials associated with this industrial

process are presented as well as the results of the assessment of the annual effective doses to theworkers and the members of the public.

g-Spectrometric analyses were performed on ‘‘raw’’ sands, end-products, and soils samplednear the plant. Thermoluminescent dosimeters, electric pumps and electret ion chambers were

used to measure the external irradiation, the indoor dust concentrations and the radonconcentrations, respectively.

The ground-level air concentration of radioactive particulate near the plant and the

deposition of particulate matter were estimated by a Gaussian model (ISCLT3).Finally, the annual effective doses, calculated as provided for by Directive 96/29/Euratom,

were estimated to be 1.7 mSv y�1 for workers and 4.4 mSv y�1 for members of the public.

� 2005 Elsevier Ltd. All rights reserved.

Keywords: Zircon sand plant; Occupational and public exposure; Naturally occurring radioactive

materials; Radiation dose; Gaussian model

* Corresponding author. Tel.: C39 0544 937306; fax: C39 0544 937303.

E-mail address: [email protected] (S. Righi).

0265-931X/$ - see front matter � 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jenvrad.2005.01.010

mailto:[email protected]

http://www.elsevier.com/locate/jenvrad

238 S. Righi et al. / J. Environ. Radioactivity 82 (2005) 237e250

1. Introduction

Elemental zirconium is not found in nature. Instead, zirconium bonds withsodium, calcium, iron, silicon, titanium, thorium and oxygen to form a numberof different zirconium-bearing minerals. The most common zirconium ores arebaddeleyite (ZrO2) and zircon (ZrSiO4). Most of the economic useful deposits ofzirconium ore are found as beach sands which represent secondary detrital masses ofheavy minerals (densityO 2.9 g cm�3). Zircon sands contain a significant concen-tration of natural radioactivity because thorium and uranium may substitutezirconium in the zircon crystal lattice. The concentration of radioactivity in zirconlies typically within the ranges 0.5e1 Bq g�1 and 1e5 Bq g�1 for 232Th and 238U,respectively (NRPB, 1993).

Some of the more common uses of zircon sands are in foundry, investmentcasting, ceramic, refractory brick and zirconium metal industries. Zircon in thefoundry and ceramic industry is used in two forms, sand and flour. Zircon flour isa fine powder obtained by zircon sand grinding. In Italy, the ceramic industrygenerally uses zircon flour as an opacifier, thanks to the whitening properties ofzirconium (Bruzzi et al., 2000; Righi et al., 2000a). For effective opacifying, zirconsand must be ground to a particle size close to 5 mm.

The Directive 29/96/Euratom ‘‘European Basic Safety Standards for the pro-tection of health of workers and the general public against the dangers arising fromionising radiation’’ (Council Directive, 1996) has, for the first time, integrated workactivities involving the presence of natural radiation sources into the scope of anEuratom Directive. The whole Title VII of the Directive deals with natural radiationsources. One of the work activities identified in Title VII is the processing ofnaturally occurring radioactive materials (NORM). The Directive draws a distinctionbetween NORM processed for their radioactive, fissile or fertile, properties andNORM processed for other purposes. In the latter case the process is subject toa more flexible control regime. To assist Member States with the implementation ofTitle VII, the European Commission has published a number of guidance documentsdealing with general implementation issues (Radiation Protection 88) (EC, 1997), theestablishment of reference levels for workplaces processing NORM (RadiationProtection 95) (EC, 1999) and the application of the concept of exemption andclearance to natural radiation sources (Radiation Protection 122 part II) (EC, 2000).The Directive requires that arrangements shall be made with regard to all workplaceswhere there is a possibility of exposure to ionising radiation in excess of 1 mSv y�1. Ifdoses exceed 1 mSv but are less than 6 mSv, the guidance document ‘‘RadiationProtection 88’’ suggests to consider whether normal commonsense precautionsshould be taken to avoid all unnecessary exposures to radiation and doses couldeffectively be reduced.

The Commission guidance documents list the zircon sand processing plants amongindustries where processing of NORM can cause a significant increase in exposure(EC, 1997, 1999). These documents also mention, among the most important routesof radiation exposure from zircon sand processing for workers, external gammasand inhalation of dust. Exposures of the public may arise from atmospheric or

239S. Righi et al. / J. Environ. Radioactivity 82 (2005) 237e250

liquid discharges, from re-use of by-product material or from disposal of solidwaste.

In Italy, Directive 96/29/Euratom has been assimilated by Legislative Decree 241/2000 (Italian Parliament, 2000).

A preliminary analysis of the radiological effects of a zircon sand processing plantwas carried out in 2001 as MSc. thesis (Simonetto, 2001). Then, the survey has beenenhanced by studying the proper parameters for applying in the study areaa Gaussian plume model of atmospheric transport and diffusion (Andretta et al.,2002) and the correct methodologies for determining the occupational dose(Righi et al., 2002). The earlier results have been used as baseline for the presentwork.

In this paper, after the presentation of the results of the g-spectrometric measurescarried out on the raw materials and end-products involved into the industrialprocess, the authors report the estimation of the increase in the exposure of workersand members of the public to natural ionising radiation due to this industrialactivity.

2. Materials and methods

2.1. Zircon sand processing plant

The plant is located in an industrial area in the North-Eastern part of Italy. The‘‘raw’’ zircon sands are imported by ship from Australia (5000 tons per year), SouthAfrica (3000 tons per year) and Ukraine (9000 tons per year), transported to theplant by truck and finally stored in warehouses and silos, waiting to be processed.The industrial process consists of grinding ‘‘raw’’ sands, with a particle size of 100e200 mm diameter, into finer particle sizes (!100 mm). The plant produces sands andflours. Milling is performed by a dry grinding system equipped with suction filters.Particles caught by filters pass through a dust removal system and are re-conveyedinto the milling system. Finished products are stored and transported in bulk orbagged.

2.2. Raw material and end-products: sampling and radioactivity measurements

Radioactivity contents were determined in samples of raw materials and end-products. According to the company, the main used and produced materials weresampled. Two samples of each raw material and end-product were collected; therespective granulometry are given in Table 1.

All samples were dried at a temperature of 105 �C for 24 h and analytical resultswere calculated on the basis of oven-dry (105 �C) weight. They were stored in450 mL Marinelli beakers and sealed for at least three weeks to establish secularequilibrium between 226Ra and its daughter nuclides 214Pb and 214Bi.

Radionuclide activities were determined by g-spectrometry using a high-puritygermanium detector with standard electronics and commercial software (Silena


International GAMMA2000). The main characteristics of the detector are reportedin previous works (Bruzzi et al., 2000; Righi et al., 2000b). Measurements wereperformed in a low background shield, consisting of 10 cm thick lead, with a 3 mmthick cadmium and 0.5 mm thick copper inner lining.

Efficiency calibration was carried out by a standard solution distributed andcertified by Laboratoire de Mesure des Rayonnements Ionisants (Gif sur YvetteCedex, France) containing 57Co, 60Co, 85Sr, 88Y, 109Cd, 113Sn, 139Ce, 137Cs, 241Amwith activity concentrations ranging from 0.5 to 10 Bq g�1. Samples and standardswere measured using the same geometry. Reference material IAEA-375 was used foranalytical quality control.

Uranium-238 and 232Th activity concentrations were obtained by measuring theactivities of their daughter products and assuming secular equilibrium conditions,hypothesis supported by previous works (Bergamini et al., 1985; Kerrigan andO’Connor, 1990). The 238U concentration was determined as a mean value ofthe results from 295.21 keV and 351.92 keV gamma lines of 214Pb, the 232Thconcentration as a mean value of the results from 338.40 keV, 911.07 keV and968.90 keV gamma lines of 228Ac, the 235U concentration from its 143.76 keVgamma line, and the 40K concentration from its 1460.83 keV gamma line. Gammaline values and relative yields used for the calculation are those recommended by theCommissariat A l’Energie Atomique (CEA, 1990). Count times were 60 000 s givinga measurement precision of ca. G10% at the 95% level of confidence.

2.3. Calculation of effective doses

The effective doses (E ) for workers and members of the public have beencalculated as provided for by Directive 96/29/Euratom (Council Directive, 1996).The effective dose turns out to be the summation of all the contributions dueto different exposure sources (external exposure in a solar year (Eext), committed

Table 1

Granulometry of raw material and end-product samples

Sample description Granulometry (mm)

‘‘Raw’’ sands

Australian zircon sand 100e200South African zircon sand 100e200

Ukrainian zircon sand 100e200

Zircon end-products

Zircon sand %100

Zircon flour ‘‘type A’’ %44

Zircon flour ‘‘type B’’ %9

Zircon flour ‘‘type C’’ %6

Zircon flour ‘‘type D’’ %5


effective dose for inhalation (P

j h( g)j,inhJj,inh) and for ingestion (P

j h( g)j,ing Jj,ing)from intakes in the same period):

EZEextCX

j

hðgÞj;inh Jj;inhCX

j

hðgÞj;ing Jj;ing ð1Þ

where h( g)j,inh and h( g)j,ing are the committed effective dose per unit-intake forinhaled or ingested radionuclide j (SvBq�1) by an individual in the group of age g;Jj,inh and Jj,ing are, respectively, the relevant intake via inhalation or ingestion of theradionuclide j in a year (Bq).

Except for radon progeny and thoron progeny, values of the committed effectivedose per unit-intake for inhalation and ingestion are given for members of the publicand for workers, respectively, in table A, B and C1 in Annex III of the Directive.

2.4. Occupational exposure

According to Eq. (1), occupational exposure has been calculated by adding to themeasurement of external doses to the workers, the effective dose for inhalation (

Pj

h( g)j,inh Jj,inh). The last term has been obtained by summing the dose deriving fromdust inhalation to the dose due to inhalation of radon gas. The effective dose foringestion (

Pj h( g)j,ing Jj,ing) has been omitted because, due to the type of activity

carried out and work methods adopted, it may be considered negligible.

2.4.1. Determination of external doseThermoluminescent dosimeters (TLDs) were used to detect and measure external

radiation exposures to occupationally exposed workers using a LiF:Mg, Ti detector.Among the different advantages of this type of dosimeters, the wide linearity rangebetween the signal and the absorbed dose, and an extremely limited fading.

All personnel who could be occupationally exposed to ionising radiation wasequipped with a TLD dosimeter. The period of monitoring of TLDs was about 45days. The TLDs were replaced after 45 days with a fresh set so that each worker wasmonitored for a whole year.

2.4.2. Determination of inhaled dust particlesTo estimate the airborne dust mass concentrations in the plant environment,

electric pumps having an adjustable flow rate (operated at 10 Lmin�1), were used.Dust samples were collected on the 45-mm diameter cellulose nitrate filters witha pore size of 0.8 mm. The sampling was carried out in the plant environment withthe highest dust rate (milling department).

The determination of the total airborne dust mass inhaled by the workers hasbeen done using the following formulas from the EPA Risk Based Corrective Action(RBCA) methodology (US EPA, 1997):

ADDZPa! IRa! ET! EF!Rf ð2Þ

where ADD is the annual adsorbed airborne dust from inhalation (mg y�1); Pa is theconcentration of particulate in air (mg m�3); IRa is the inhalation rate (1.25 m3 h�1,


corresponding to light exercise); ET is the daily exposure time (8 h day�1); EF is theexposure frequency (250 days y�1). In order not to underestimate the inhalationdose, the inhalable fraction (Rf) of the dust was set to 1 in the calculations.

The activity concentrations of airborne dust in the plant environmentwere assumedto be the same of zirconmaterials involved in the industrial process. This assumption isbased on two hypotheses: (a) the productive process e a mechanical grinding e doesnotmodify the radioactivity concentration in the rawmineralmaterials; (b) the state ofsecular equilibrium among all daughters in the natural decay chains is maintained.

2.4.3. Determination of radon concentrationFor the derivation of the occupational dose due to inhalation of radon gas,

integrated radon measurements have been carried out in indoor workingenvironment (offices, grinding department, packing and storing department) forperiods of about 6 months. The survey has been conducted for one year.

Integrated radon measurements were taken using electret ion chambers (EIC).This technology is based on the electrical potential reduction of an electrostaticallycharged Teflon disk held within a chamber by the ions generated by radon decay. Inusing the EIC method, a measurement is required to correct the electrical potentialreduction caused by ionisation in air by g radiation.

Chambers of nominal 50 mL volume (designated ‘‘L’’) and electrets of lowsensitivity (designated ‘‘LT’’ for long-term deployment) were used in this work.

The following formula was used to calculate the radon concentration, CRn, inBqm�3:

CRnZ

�Vi �Vf

�

Fc!T�BG

where Vi and Vf are the measured initial and final electret voltages, respectively; Fc isthe calibration factor in units of V per Bqm3 d; T is the exposure period in days; andBG is the radon concentration equivalent of natural g radiation background.

The total error (TE) was calculated by adding quadratically all the random errors:error (E1) associated with the chamber parameters, error (E2) associated withreading of electrets and error (E3) associated with uncertainty of the natural gradiation background (Kotrappa et al., 1988).

The lower limit of detection (LLD) of the EIC method depends upon the period ofexposure and the ambient g radiation at the deployment location. Typically, theLLD for a period of about 180 days is approximately 10 Bqm�3.

2.5. Exposure of the members of public

The milling of zircon sands is performed through a dry grinding system; particlescaught by filters are re-conveyed into the grinding system so there are no waste streamsother than atmospheric emissions. Consequently, in order to evaluate the radiologicalimpact of the zircon sand processing plant on members of the public, only theparticulate matter release into the atmosphere and its dispersion and deposition onneighbouring soil have been considered possible significant exposure routes.


2.5.1. Study of the dispersion into the atmosphere and deposition on soilof the particulate matter

A Gaussian plume model (ISCLT3 e Industrial Source Complex Long Term,Version 3) developed by US EPA (1978, 1986) was used to estimate the con-centration of suspended particulate matter emitted by the zircon milling plant intothe atmosphere and its deposition on neighbouring soil. The ISCLT3 can performforecasts of the diffusion and deposition of pollutants emitted from many differentkinds of sources such as point, area and volumetric type. The simulations of ISCLT3model can range from 1 h or less (episodic or short-term simulations) up to somemonths or years (long-term average simulations). The possibility to perform long-time simulations make the Gaussian model, like ISCLT3, particularly suited forusing its results in relation with many air quality standards. Long-term simulationscan be performed using either hourly specific meteorological data, or site specificjoint frequency functions (JFF). JFF are tables that collect the monthly, seasonal orannual frequency distributions of the meteorological data useful for Gaussianmodels application (e.g. wind speed and direction, air temperature, atmosphericstability category, height of the mixing layer). In this work, the authors have used theJFF, elaborated using 15 years data, of a meteorological station that is located somekilometres only far from the investigated industrial plant, and, according to US EPA(1986) indications, they have used the Briggs rural type sigma coefficient. For thesimulations, the authors have assumed a typical average temperature of 20 �C. As tothe temperature and exit velocity of stack emissions, the project data recorded for theemission authorisations have been used. As to dust concentration in the emissions,the authors have used the authorisation values, too, in order to make the mostconservative estimation as possible. Fig. 1 shows the average wind rose of the area ofinterest, elaborated on the basis of the 15 years collected data.

N

NE

E

SE

S

SW

W

NW

0 - 1 (m/s)

1 - 3 (m/s)

3 - 4 (m/s)

4 - 6 (m/s)

6 - 12 (m/s)

Wind speed

Fig. 1. Average wind rose of the investigated area. Note that the ‘‘dead calms’’ are 36.7%.


The annual airborne dust mass inhaled by the people living near the plant hasbeen obtained from the dust air concentrations predicted by the model through theapplication of Eq. (2) (Subsection 2.4.2). In this equation, the inhalation rate (IRa)was assumed equal to 0.89 m3 h�1, the exposure frequency (EF) equal to 365 days y�1

and the daily exposure time (ET) equal to 8 h day�1. Also, in order not tounderestimate the inhalation dose, the inhalable fraction (Rf) of the dust was set to 1in the calculations.

The ISCLT3 model can also estimate the deposition of the particulate emittedfrom the sources, on the basis of the meteorological data of the area of concern andof the physical characteristics of the particulate matter, like its density, roughnessand granular size. These capabilities make the model useful for the estimation of thedeposition on soils of radioactive particulate emitted by the zircon sand processingplant. Depositions have been calculated using the deposition velocities, as reportedby Hanna et al. (1982). The deposition velocities have been evaluated in function ofthe density and diameter of the dust particles emitted by the plant, as reported inTable 2. A soil roughness height of 0.1 cm has been assumed, in accordance with thecharacteristics of the area where the plant is located.

The ISCLT3 model can also take into account the building downwash that is theeffect of the neighbour building on the dynamics of the stack emissions. Buildingscan produce turbulent downstream flows that perturb the motion of the plumes andinduce high ground-level pollutant concentrations. For these reasons, buildingdownwash represents a very important phenomenon for the correct evaluation of theexposition of receptors near the plant. In this work, the building downwash effect hasbeen considered by using a washout coefficient of 6! 105 (Hanna et al., 1982) andan average precipitation of 900 mm y�1, obtained from the local meteorologicaldata.

2.5.2. Soils: sampling and radioactivity measurementsThe soil sampling sites were selected in the area of maximum particulate matter

deposition determined by the dispersion model described in the previous subsec-tion. Four samples were collected. Each soil sample was taken from an area of20! 20 cm, 10e15 cm deep, dried at a temperature of 105 �C for 24 h, mechanicallycrushed, sieved (1 mm) and homogenised. Analytical results were calculated on thebasis of oven-dry (105 �C) weight. The soil samples were stored in 450 mL Marinelli

Table 2

Particle characteristics and related deposition velocities

Particle diameter (mm) Mass fraction Particle density (g cm�3) Deposition velocity (cm s�1)

0.5 0.20 4.7 1.5! 10�2

1 0.30 4.7 3! 10�2

2 0.20 4.7 1! 10�1

3 0.20 4.7 2! 10�1

4 0.08 4.7 3! 10�1

5 0.02 4.7 4! 10�1


beakers and sealed for at least three weeks to establish secular equilibrium between226Ra and its daughter nuclides 214Pb and 214Bi and radionuclides activities weredetermined by g-spectrometry using the equipment and the methodology describedin the Section 2.1.

3. Results and discussion

3.1. Raw materials and end-products: natural radioactivity concentrations

The results of g-spectrometric analyses are presented in Table 3. Zircon sands andflours show activity concentrations of 238U of about 2e3 kBq kg�1 and activityconcentrations of 232Th of about 0.4e0.5 kBq kg�1. The values obtained for 238Uand 232Th series in zircon materials are in good agreement with those reported byother authors (UNSCEAR, 1988, 1993; NRPB, 1993). It is clear from these data thaturanium and thorium concentrations in these materials are much higher than

Table 3

Activity concentration and total error of 238U series, 232Th series, 235U series and 40K in kBq kg�1 dry

weight in raw material, end-product and soil samples

Sample description 238U 232Th 235U 40K

‘‘Raw’’ sands

Australian zircon sand 2.4G 0.2 0.52G 0.04 0.107G 0.009 0.034G 0.005

Australian zircon sand 2.2G 0.2 0.48G 0.04 0.102G 0.009 0.033G 0.004

South African zircon sand 3.2G 0.3 0.52G 0.04 0.140G 0.012 0.032G 0.006

South African zircon sand 2.9G 0.2 0.45G 0.04 0.128G 0.011 0.032G 0.005

Ukrainian zircon sand 1.83G 0.15 0.37G 0.03 0.084G 0.007 0.026G 0.004

Ukrainian zircon sand 1.86G 0.16 0.38G 0.03 0.087G 0.008 0.032G 0.004

Zircon end-products

Zircon sand 2.3G 0.2 0.45G 0.04 0.103G 0.009 0.032G 0.005

Zircon sand 2.3G 0.2 0.47G 0.04 0.099G 0.009 0.032G 0.006

Zircon flour ‘‘type A’’ 2.5G 0.2 0.47G 0.04 0.125G 0.011 0.029G 0.004

Zircon flour ‘‘type A’’ 2.6G 0.2 0.49G 0.04 0.139G 0.012 0.029G 0.004

Zircon flour ‘‘type B’’ 2.7G 0.2 0.54G 0.05 0.153G 0.014 0.033G 0.006

Zircon flour ‘‘type B’’ 2.8G 0.3 0.54G 0.05 0.148G 0.014 0.032G 0.004

Zircon flour ‘‘type C’’ 2.8G 0.2 0.51G 0.04 0.155G 0.014 0.027G 0.004

Zircon flour ‘‘type C’’ 2.7G 0.2 0.52G 0.04 0.147G 0.014 0.026G 0.004

Zircon flour ‘‘type D’’ 2.6G 0.2 0.48G 0.04 0.154G 0.014 0.030G 0.004

Zircon flour ‘‘type D’’ 2.8G 0.3 0.50G 0.04 0.156G 0.014 0.031G 0.004

Soils

Sample no 1 0.0153G 0.0014 0.0162G 0.0013 !dl 0.45G 0.04

Sample no 2 0.024G 0.002 0.023G 0.002 !dl 0.46G 0.04

Sample no 3 0.023G 0.002 0.022G 0.002 !dl 0.47G 0.04

Sample no 4 0.0159G 0.0013 0.0154G 0.0014 !dl 0.45G 0.04

Worldwide averagea 0.035 0.030 0.400

a UNSCEAR, 2000.


worldwide average concentrations in soil. Note that 238U series activity concen-trations are about 1000 times higher than normal soil confirming the necessity tocarry out a radiological impact assessment of the zircon sand processing plant.

The 232Th/238U ratios (0.16e0.22) indicate the better crystallographic affinitybetween zirconium and uranium than between zirconium and thorium. As presentedin Table 3, end-product activities from the different decay chains are consistent withthe ones of ‘‘raw’’ sands. These results confirm the hypothesis that the millingprocess does not disturb the secular equilibrium and does not change natural activityconcentrations.

3.2. Occupational exposure

The average external dose of all the monitored people determined through theTLD dosimeters is 0.33 mSv y�1; the values vary from 0.25 to 0.42 mSv y�1. Most ofthe workers have external doses below 0.30 mSv y�1. The fraction of the effectivedose due to external irradiation has been fixed at a value equal to the average dosereceived by the monitored workers.

To assess the intake via inhalation of airborne particles, it is necessary to firstevaluate how much radioactivity enters and deposits in the respiratory tract. Resultsof the dust sampling (Subsection 2.4.2) show airborne dust concentrations in themilling department ranging from 1 to 2 mgm�3. An average value equal to1.5 mgm�3 was adopted to assess inhalation intake and, according to the describedmethodology (Subsection 2.4.2), the inhalation rate of airborne particles is estimatedto be 3750 mg y�1. On the other hand, the activity concentration in airborne particleshas been assumed to be equivalent to the weighted average of activity concentrationsmeasured in raw materials samples. The weighted average was calculated consideringthat 29% of zircon sands used by the company comes from Australia, 18% comesfrom South Africa and 53% comes from Ukraine. The resulting activityconcentrations in airborne particles are 2.2 kBq kg�1 for the 238U chain, 430 Bq kg�1

for the 232Th chain, 100 Bq kg�1 for the 235U chain and 30 Bq kg�1 for 40K. Thus,the total committed effective dose from inhalation of radionuclides in airborneparticulate matter has resulted to be about 0.76 mSv y�1.

The 222Rn concentrations in the different plant environments, measured byelectret ion chambers, showed average values of 230, 70 and 30 Bqm�3 for grindingarea, packing and storing area and offices, respectively. The latest ICRP doseconversion factor of 3! 10�9 Sv h�1 per Bqm�3 (ICRP, 1994) was used to convertradon concentration into dose following its inhalation. The effective dose due toradon and its decay products has been estimated starting from the weighted averageof radon concentrations by assuming that workers spend 2000 h y�1 in indoorworkplaces (1000 h y�1 in sacking and storage departments and 500 h y�1 both inmilling department and office). The estimated effective dose from radon and its decayproducts is approximately 0.6 mSv y�1.

Finally, the total effective dose has been calculated as the summation of threecomponents: external exposure and internal exposure from radon and its progenyand from inhalation of airborne materials, as it was indicated in Section 2.4.


Therefore, total occupational exposure has been estimated to be approximately1.7 mSv y�1. Hence the internal dose due to the inhalation of radioactive dusts is themain fraction of the total dose (about 45% of the total), followed by the internaldose due to inhalation of radon (about 35%), while the dose from external radiationis the minor fraction (about 20%).

It is evident that this study has some inherent limitations. The inhalable fractionof dust was not measured so it is assumed to be 1.0 and the dust sampling wascarried out only in the plant environment with the highest dust rate. In addition, theindoor stay was fixed at 2000 h y�1, without considering the time spent workingoutdoor. Therefore, the dose limit of 1 mSv y�1 is, probably, respected. However, inorder to get a true representation of the occupational exposure, individualassessment would also need to be conducted. If the actual effective dose exceeds1 mSv y�1, the employer should perform an investigation to consider the engineeringcontrols and work practices to determine whether the exposure has been restricted sofar as reasonably practicable. Steps should be taken to reduce occupational exposureto airborne particulate matter, for example, by increasing ventilation continuouslyor during milling events, by locating the milling machine in a separate well-ventilatedroom, or by using automatic controls. Investigating how increased ventilation ratesand other mitigation strategies will affect radon concentrations may lower occu-pational radon exposure. If engineering controls and techniques are not practicable orcannot guarantee the elimination of radiological hazards then Personal ProtectiveEquipment (PPE) represent the most reasonable option to protect workers.

3.3. Exposure of members of the public

The simulation results performed with the Gaussian model (Subsection 2.5.1)indicate that the maximum concentration (4.1 mg m�3) and maximum deposition(114 mgm�2 y�1) of particulate matter occur near the emission point, at about 200 mfrom the stack. These results can be explained with the relative high frequency ofoccurrence, in the area of concern, of high instable atmospheric classes of dispersion(classes A and B) that give rise to this typical diffusion and deposition patterns fromhigh emissions like the ones taken into account.

These results are of about two orders of magnitude lower than the Italianthreshold values for these kinds of pollutants and point out that the effects on theenvironment outside the industrial plant can be considered absolutely negligible. InFig. 2, just for explanation purposes, the estimated iso-concentration levels in thearea near by the plant are presented.

The extremely low impact of the deposited particulate emissions on theneighbouring areas of the processing plant has been confirmed through the resultsobtained in the measurements of the collected soil samples. Soil samples have beencollected at about 200 m north-west from the emission stack, just around the point ofmaximum deposition. Average concentrations in soil are 20 Bq kg�1 for members ofthe 238U series, 19 Bq kg�1 for members of the 232Th series and 460 Bq kg�1 for 40K.The observed concentrations are consistent with the average values generally foundin soils (e.g. 35 Bq kg�1 for 238U series, 30 Bq kg�1 for 232Th series and 400 Bq kg�1


for 40K (UNSCEAR, 2000)) and the 238U and 232Th activity concentrations aresimilar despite different atmospheric fluxes for these radioisotopes. This suggests thatthere were no enhancements of radionuclide concentrations in surface layers of soil,which might indicate recent deposition from the zircon sand milling plant.

Consequently for the population living near the plant, external irradiation andradionuclides ingestion appear negligible pathways of exposure; the only significantpathway could be the inhalation of airborne particles.

In fact, the estimation of the dose received by the population due to releases formthe milling plant has been calculated using formula (1) described above (Section 2.4)considering only the dominating exposure pathways. Thus, formula (1) has beensimplified as shown hereunder:

EZX

j

hðgÞj;inh Jj;inh ð3Þ

The dose due to radionuclide airborne emissions has been calculated from annualaverage dust concentrations predicted by ISCLT3 Gaussian model and from theconsequent inhaled airborne dust (see Subsection 2.5.1). The maximum effective dosefrom airborne emissions is estimated to be 4.4! 10�3 mSv y�1 and occurred atabout 200 m from stacks. Recognizing that the dose estimate is based on theconservative assumption that a person remains outdoors at the plant location forabout 3000 h y�1 (8 h d�1! 365 d y�1), the actual dose received by the public wouldbe substantially lower than 4.4! 10�3 mSv y�1. However, the maximum dose of4.4! 10�3 mSv y�1 is well below the annual limit of 1 mSv.

Fig. 2. Annual particulate iso-concentration levels, in mg m�3.


4. Conclusions

The total occupational exposure has been estimated to be approximately1.7 mSv y�1. The Italian legislation provides for corrective measures to be taken toreduce the annual effective dose below 1 mSv y�1 when the annual occupationaleffective dose exceeds this limit. Because of the numerous conservative assumptionsused in calculating exposure estimate, the actual exposure is likely to be substantiallyless than the estimated exposure and therefore it is likely that the 1 mSv y�1 doselimit is respected. In any case, the authors think it appropriate to carry out individualassessments for workers who, owing to their specific tasks, come mostly into contactwith zircon sands.

Also, for the assessment of exposure of population living in the vicinity of theplant the authors have made very conservative assumptions: a person exposedfor 3000 h y�1 (8 h d�1! 365 d y�1) at the maximum particulate concentration(4.1 mg m�3). Nevertheless, the annual exposure of the general public remains wellbelow the 1 mSv limit (about 4 mSv y�1). This result is consistent with more commonlevels (1e10 mSv y�1) reported by UNSCEAR 2000 Report (UNSCEAR, 2000).

In conclusion, this study indicates that the radiological risks from this zirconmilling plant are very low, but that for small number of workers there might becircumstances in which risks are higher, and it is for this reason that further workshould be undertaken to clarify their extent.

Acknowledgements

The authors express many thanks to Silvano Cazzoli of ANPEQ (NationalProfessional Association of Italian Qualified Experts in Radiological Protection) forthe provision of advice and technical assistance. The authors also thanks GiuseppeGnani and Giorgio Mazzotti of Health Physics Service of Ravenna Hospital andPatrizia Lucialli of ARPA Emilia-Romagna for their kind support.

References

Andretta, M., Righi, S., Simonetto, C., Bruzzi, L., Serra, R., 2002. Environmental impact and risk

assessment of zircon mineral processing plant emission. In: Brebbia, C.A. (Ed.), Third International

Conference on Computer Simulation in Risk Analysis and Hazard Mitigation. WIT Press,

Southampton, pp. 457e466.

Bergamini, M., Borio, R., Campos Venuti, G., De Zaiacomo, T., Fabbri, S., Formignani, M., Gazzola, A.,

Giacomelli, R., Lembo, L., Lomberdi, C.C., Melandri, C., Nanni, R., Pallavicini, L., Risica, S.,

Sciocchetti, G., Spezzano, P., Tarroni, G., 1985. Radiation protection aspects of the use of zircon sand.

Science of Total Environment 45, 135e142.Bruzzi, L., Baroni, M., Mazzotti, G., Mele, R., Righi, S., 2000. Radioactivity in raw materials and end-

products in Italian ceramic industry. Journal of Environmental Radioactivity 47, 171e181.

CEA (Commissariat A l’Energie Atomique), 1990. Bibliotheque de donnees nucleaires pour la

spectrometric gamma et alpha, second ed. Gif-Sur-Yvette Cedax, France.


Council Directive, 29/06/1996. Council Directive 96/29/Euratom of 13 May 1996. Laying down basic

standards for the protection of the health of workers and the general public against the dangers arising

from ionising radiation. Official Journal L 15, 0001e0029.EC (European Commission), 1997. Recommendations for the implementation of Title VII of the

European Basic Safety Standards Directive (BSS) concerning significant increase in exposure due to

natural radiation sources. Report Radiation Protection 88. European Commission, Luxembourg.

EC (European Commission), 1999. Reference levels for workplaces processing materials with enhanced

levels of naturally occurring radionuclides. Report Radiation Protection 95. European Commission,

Luxembourg.

EC (European Commission), 2000. Practical Use of the Concepts of Clearance and Exemption. Report

Radiation Protection 122. European Commission, Luxembourg.

Hanna, S., Briggs, G.A., Hosker, R.P., 1982. Handbook on Atmospheric Diffusion, DOE/TIC-11223,

U.S. Department of Energy, pp. 102.

ICRP (International Commission on Radiological Protection), 1994. Protection against radon-222 at

home and at work. ICRP Publication no 65, Annals of ICRP, 23(2). Pergamon Press, Oxford.

Italian Parliament, 2000. Legislative Decree no 241 of 26 May 2000 implementing EU directive 96/29/

Euratom laying down basic safety standards for the radiation protection of workers and the public.

Supplement to Italian Republic’s official journal no 203 of 31 August 2000.

Kerrigan, G.C., O’Connor, B.H., 1990. Evaluation of Th series disequilibrium in western Australian

monazite. Health Physics 58, 157e163.

Kotrappa, P., Dempsey, J.C., Hickey, J.R., Stieff, L.R., 1988. An electret passive environmental 222Rn

monitor based on ionization measurement. Health Physics 54, 27e56.NRPB (National Radiological Protection Board), 1993. Working with Zircon Sands NRPB Broadsheet

Series. United Kingdom, London.

Righi, S., Bruzzi, L., Palmonari, C., 2000a. The ceramics industry and the new European Directive

regulating protection against radiation. Ceramica Acta 12, 74e84.

Righi, S., Betti, M., Bruzzi, L., Mazzotti, G., 2000b. Monitoring of natural radioactivity in working

places. Microchemical Journal 67 (1e3), 119e126.

Righi, S., Simonetto, C., Guerra, R., Bruzzi, L., August 2002. Occupational exposure to natural radiation

at a zircon sand processing plant. In: Strand, P., Jølle, T., Sand, A. (Eds.), International Conference on

Radioactivity in the Environment. Principality of Monaco, ISBN 82-90362-14-5, pp. 224e228.

Simonetto, C., 2001. Radiological impact assessment of zircon sands employed in industrial processes.

MSc thesis, University of Bologna, Italy (in Italian).

UNSCEAR, 1988. Sources, effects and risks of ionizing radiation. United Nations Scientific Committee on

the Effects of Atomic Radiation. Report to the General Assembly with annexes. United Nations

Publication, New York, USA.

UNSCEAR, 1993. Sources and effects of ionizing radiation. United Nations Scientific Committee on the

Effects of Atomic Radiation. Report to the General Assembly with scientific annexes. United Nations


UNSCEAR, 2000. Sources and effects of ionizing radiation. United Nations Scientific Committee on the

Effects of Atomic Radiation. Report to the General Assembly with scientific annexes. United Nations


US EPA (United States Environmental Protection Agency), 1978. Workbook for the comparison of Air

Quality Models, Research Triangle Park, NC, Rep. EPA-450/2-78-028a and b.

US EPA (United States Environmental Protection Agency), 1986. Guidelines on air quality models,

Research Triangle Park, NC, Rep. EPA-450/2-78-027R.

US EPA (United States Environmental Protection Agency), 1997. Exposure Factor, EPA/600/P-95/002F.

Documents

Assessment of the radiological impacts of a zircon sand processing plant