Chemical Geology, 55 (1986) 233--254 Elsevier Science Publishers B.V.

, Amsterdam -- Printed in The Netherlands

233

AN ANALOGUE VALIDATION STUDY OF NATURAL RADIONUCLIDE MIGRATION IN CRYSTALLINE ROCKS USING URANIUM-SERIES DISEQUILIBRIUM STUDIES
J.A.T. SMELLIE I, A.B. MACKENZIE: and R.D. SCOTT 2
~Swedish Geological Company, 751 44 Uppsala (Sweden) Scottish Universities Research Reactor Centre, Glasgow G75 OQU, Scotland (Great Britain)
(Accepted for publication January 15, 1986)

Abstract
Smellie, J.A.T., MacKenzie, A.B. and Scott, R.D., 1986. An analogue validation study of natural radionuclide migration in crystalline rocks using uranium-series disequilibrium studies. In: N.A. Chapman and J.A.T. Smellie (Guest-Editors), Natural Analogues to the Conditions around a Final Repository for High-level Radioactive Waste. Chem. Geol., 55: 233--254. Concentrations and isotope ratios of natural decay series radionuclides have been studied in three contrasting crystalline rock drillcore sections intersecting water-conducting fractures deep in the bedrock. Radioactive disequilibria resulting from rock--water interactions were observed in two of the cores. These indicated U migration along distances of 40 cm or more on a time-scale of 1 Myr. in conjunction with Th immobility under the same conditions. Fracture surface minerals showed a high affinity for radionuclide retardation and a limit of - 3 cm is suggested for the migration of radionuclides from fracture fluids into the saturated rock. This limit may correspond to enhanced matrix porosities resulting from earlier hydrothermal activity along the same channels.

1. Introduction Safety assessment evaluations for the disposal o f high-level radioactive waste characteristically assume a "worst case" situation. This usually means that progressive corrosion of the spent fuel metal container will occur, resulting in a slow release of radionuclide material into circulating meteoric waters present within the surrounding bedrock repository. It is highly probable that crystalline rocks at repository depths will he fractured, even prior to excavation, and that fracture permeability will be the dominant m o d e of flow. Thus, based on
0009-2541/86/$03.50

the premise that a cumulative failure of the near-field engineered barriers will occur, far-field retardation processes in the geosphere would appear to be qualitatively the most important factor in controlling releases o f radionuclides to the biosphere. During far-field radionuclide transport along hydraulically connected fracture systems, radionuclides in solution will be subject to retardation mechanisms such as reversible and irreversible sorption onto fracture and fissure mineral coating phases. In addition, radionuclide diffusion into a water-saturated rock matrix could be an important retardation factor. To date, much of the work and

1986 Elsevier Science Publishers B.V.

234 model derivation involving radionuclide migration has been experimentally based. An obvious means of validating the experimental data is to examine naturally occurring radionuclide migration (or lack of migration) within the bedrock environment of the chosen repository. Such an analogue study allows, most importantly, an appraisal of migration on the scale of geological time. Naturally occurring U and Th are of particular interest in radioactive waste disposal since they represent useful analogues of the actinides. Furthermore, the daughter nuclides in the 238U and 232Th decay chains show ranges of chemical properties and half-lives such that the degree of equilibrium between parent--daughter pairs can be used to provide information a b o u t rock-water interactions. Such studies can indicate the degree of radionuclide mobility and can provide information on the rates of processes occurring over time intervals of up to 1 Myr. As part of a jointly-funded programme of research between the Swedish Nuclear Fuel and Waste Management Co. (SKB) and the Swiss Nationale Genossenschaft fiir die Lagerung Radioaktiver Abf~lle (NAGRA), drillcore lengths intersecting suitable water-conducting fracture zones were chosen from three different hydrogeological environments; two in Switzerland (BSttstein and Grimsel) and one in Sweden (Krhkem~da), all located in granite bedrock. The samples represent profiles from the intersected fracture face into the bedrock. Sections cut from the profiles have been investigated using the uranium decay series (23sU--234U--:3Th--226Ra), in conjunction with rare~earth element (REE) and Fe2+/ Fe 3 analyses, a-track autoradiography, fission4rack analysis and detailed mineralogy. To the extent that the U-series disequilibrium is a p r o d u c t of rock--water interaction, these measurements may help to evaluate the effective addition or removal of U within the near vicinity of an open waterbearing fracture zone. Evidence of such a migration profile occurring into the bedrock under ambient temperatures, should lend or detract support to models involving farfield radionuclide retardation processes.

2. Uranium-series disequilibrium in rocks Studies of the U decay series have now become a fairly standard investigative method within the field of geochemistry. Pioneered by Cherdynstev (e.g., 1955, 1971) the methods have been subsequently improved and applied b y many workers to a wide range of geological environments and their related problems; see, for example, Rosholt (1959, 1983), Thurber (1962), Rosholt et al. (1963, 1966), Koide and Goldberg (1965), Kighoshi (1971), Kronfeld (1974), Osmond and Cowart {1972, 1976), and Fleischer and Raabe (1978). In closed geological systems the nuclides 238U--234U---:3Th attain radioactive equilibrium after ~ 1.7 Myr., i.e. the respective activity ratios 234U/238U, 23Th/234U and 23Th/238U all equal unity. However, if the systems are exposed to weathering and groundwater circulation, and assuming that 23Th is immobile under normal groundwater conditions (Langmuir and Herman, 1980), the different physico-chemical conditions affecting 238U and 234U will result in their fractionation and thus isotopic disequilibrium. The respective activity ratios will therefore be greater or less than unity, depending on whether there is an excess or deficiency of 238U and/or 234U. Fractionation of :34U and 238U arises from the radiogenically produced 234U daughter which originates through the three-step a- and ~-decay from its 238U parent: 238U ~ 234Th -~ :34pa -~ :3a U 4.51 Gyr. 24.1 days 1.18 rain.

235 In general, the isotopic variation of U has been recognized as being produced by t w o mechanisms: (1) selective leaching of 234U itself, (2) the direct a-recoil transfer of a 234U precursor (i.e. 234Th), and a combination of these processes. For selective leaching to occur, :34U atoms must be more susceptible to leaching and removal than ~3sU atoms. This is facilitated b y lattice damage resulting from the radioactive decay of 23~U. The resultant daughter atoms will be located in weakly bonded or interstitial sites, and oxidation of the daughter (i.e. :34U) to the 6+ valency state as a result of the recoil within the lattice, will render it more vulnerable to removal by water. Direct a-recoil of ~34Th can occur from near the surface of a mineral grain into the intergranular pore water, whereupon it is trapped and further decays to :34U. This results in the enrichment of :34U in water and a :34U depletion in the mineral (i.e. solid-to-solution 234U recoil gain). An additional mechanism recognizes the process of daughter product emplacement from pore waters containing dissolved ~38U and 234U. Subsequent decay of these radionuclides results in the recoiling nuclides of :3Th, 234Th, 234Pa and 234U being absorbed or e m b e d d e d in particulate matter at the solid--liquid interface. After sufficient geological time, this mechanism results in solids which are enriched in 234U and 23Th relative to :38U (i.e. solution to solid 234U + 23Th recoil gain). For crystalline rocks, such as these described in this paper, the U-decay system has been used as a sensitive indicator of recent U mobility (Zielinski et al., 1981; Schwarcz et al., 1982; Rosholt, 1983; Gascoyne, 1984; Smellie and Rosholt, 1984). H o w recently this migration has occurred is determined from the half-life of the nuclide involved. In this present study the nuclides :3sU, 234U, 23Th and 226Ra have been measured. Using the daughter/parent activity ratios ~34U/:3sU, 23Th/234U and ::6Ra/ :3Th, radionuclide migration over timescales of 1 Myr., 300,000 yr. and 8000 yr., respectively, can be evaluated. 3. Sample selection and analysis

Drillcore lengths of 40--55 cm were chosen from the respective areas; these varied in diameter from 56 mm (Swedish sample) to 76 mm (Swiss samples). A thin rock slice was removed along the drillcore length for mineralogical and fission-track studies. A portion of the core was retained for reference and the remainder was sawn (Swiss samples) or cleaved (Swedish sample) into 1-cm thick slices for analysis. The mineral coatings scraped from the inside surface of the fracture zone constituted another sample. For U-series measurements samples of the powdered granite were spiked with 232U in equilibrium with ~28Th and then digested with aqua regia and hydrofluoric acid and the small residue of insoluble material was further digested with nitric and hydrofluoric acids in a Parr high-pressure digestion b o m b . This process was repeated if necessary until the residue totally dissolved in 6 M hydrochloric acid. U and Th were recovered from half of the combined acid leachings b y anion-exchange techniques (Bacon and Rosholt, 1982) and were then electrodeposited onto stainless-steel planchettes for aspectrometry measurements using Si surface barrier detectors. The precision of the results is better than +4% for the 234U/238Uactivity ratios, better than +7% for the 23OTh/234U and 23Th/~32Th activity ratios and +15% for the 226Ra/23Th activity ratio. 2~6Ra analyses were initially attempted on the remaining half of the acid-leaching solution by the standard technique of collection and a-counting of the daughter 222Ra (Scott and MacKenzie, 1984, 1985). Since there is no spike available to act as a yield tracer in this method, the process depends upon 100% transfer of 226Ra to the Rn emanation vessel. Replicate analyses and analyses of samples to which a known quantity of 226Ra was added indicated that the above condition was n o t being satisfied and that loss of 226Ra was occurring before transfer to the emanation vessel. No satisfactory explana-

236 tion is at present available for this loss but attempts at 226Ra analysis by this method were abandoned in favour of direct counting of the 609-keV peak in the T-spectrum of 226Ra decay products in a sub-sample of the untreated powdered granite. This peak, produced by 21+Bi decay, can only be used if 222Rn does not escape to any significant extent from the sample and initial results appear to justify this assumption. REE and total Fe analyses were carried out on 0.14 samples using instrumental neutron activation analysis (INAA). This involved a 6-hr. irradiation in a thermal neutron flux of 3.1012 n cm -2 s -I. Gamma counting using co-axial and planar Ge(Li) detectors was performed at ~ 3 days and ~ 3 weeks after irradiation (MacKenzie et al., 1983). Fe 2+ was analysed using standard wet chemical techniques and fission-track distributions from polished thin sections of the granite were recorded on polycarbonate by thermal neutron irradiation (MacKenzie et al., 1984). 4. Geological setting and mineralogy of the drillcores is macroscopically homogeneous. The chosen core is traversed by a single open fracture zone at 45 to the vertical drillcore length; from geophysical logging investigations (Magnusson and Duran, 1984) the fracture zone is considered water-bearing. This is partly supported by fracture plane coatings of hematite and FeOOH-oxides. In addition, the feldspathic content marginal to the fracture zone shows evidence of alteration extending to 2--3 cm from the fracture plane. The drillcore is representative of the GStemar coarse-grained granite and is macroscopically homogeneous. At greater resolution the rock is mineralogically variable consisting primarily of quartz {15--45 vol.%), plagioclase (30--42 vol.%) and K-feldspar (30--50 vol.%) with subordinate biotite, muscovite and chlorite. The plagioclase (An < 10%) occurs mostly as euhedral to subhedral laths, as a matrix constituent, and included in K-feldspar. Accessory minerals include interstitial amounts of magnetite, apatite, zircon, fluorite, sphene and monazite; isolated grains of galena, pyrite, and more rarely molybdenite, also occur. The nonsulphide accessory phases usually occur as intergranular aggregates sporadically dispersed throughout the rock, resulting in considerable inhomogeneity on the scale of a thin section. Throughout the length of the drillcore sample the granite is weakly but uniformly altered; this is generally typical to the depths of maximum drilling in the massif (Smellie and Stuckless, 1985). In most sections the partial breakdown of the feldspars to muscovite/sericite with minor quartz occurs. Magnetite sporadically shows limited alteration to hematite, or partial to complete pseudomorphing by chlorite. Biotite is sometimes partly altered to chlorite, and muscovite, normally clear and unaltered, has occasionally altered to chlorite and less frequently to epidote and opaque phases (mostly Fe--Ti-oxides). The fractured area occupies a width of 1 cm and is composed of a series of paral-

4.1. Kr~kem~la drillcore (KI: 317.85-318.40 m)

The Kr~kemiila specimen was selected at a depth of ~ 318 m from one of three test holes drilled in the GStemar granite, located along the southeastern coast of Sweden. The Kriikem~ila area within the GStemar granite has been the focus of geological investigations as part of an earlier programme concerning feasibility studies for the disposal of radioactive waste material in crystalline rocks (Scherman et al., 1977). The geology, geochemistry and geochronology of the GStemar granite has been previously described by Kresten and Chyssler (1976), Aberg (1978) and Smellie and Stuckless (1985). The sample location level is characterized by the coarse-grained granite variety which

237 lel to sub-parallel microfissures easily distinguished b y FeOOH-oxide coatings. The central and largest fracture (max. 1.5 m m wide) is also partly infilled with hematite and subordinate chlorite. Macroscopically, the influence of the fracture zone extends up to 2 cm into the host granite. This is apparent as FeOOH-oxide dustings present b o t h interstitially within the rock matrix and included within altered to partly altered K:feldspar phases; biotite and magnetite are also more markedly altered to chlorite and hematite, respectively. matite (up to 40 cm wide) and aplite (up to 1.5 m wide); the fracture occurs in the pegmatite only a few centimetres from the granite contact. Intersection of the fracture with the core axis is 37 and its water-conducting properties were indicated by its open character and idiomorphic quartz fillings. Fluid logging identified a major water inflow between 618 and 621 m, a zone which comprises a potentially waterconducting kakirite horizon. Intersection between these two horizons at a short distance from the borehole is suspected as both exhibit similar dips combined with strikes perpendicular to each other. The selected drillcore length investigated consists of ~ 7 cm of pegmatite in contact with 13 cm of biotite-rich granite which is representative of the BSttstein porphyritic granite variety; the granite at the pegmatite contact has been hydrothermally altered to the extent of ~ 8 cm. In terms of average modal content (vol.%), the granite comprises 27% quartz, 26% plagioclase, 38% K-feldspar and 8% biotite. The K-feldspar, which is seen as large phenocrysts in hand specimen, is perthitic in thin section and the plagioclase is normally zoned with a core of oligoclase (An12-:0) and a rim of albite. Some late magmatic muscovite has formed from the breakdown of K-feldspar and biotite, and pseudomorphs of pinite after cordierite are common. Apatite and zircon are the most frequent accessories present. The pegmatite containing the fracture zone is fresh, coarse-grained, and consists of quartz, K-feldspar and albite, together with muscovite and tourmaline. Regarding the fracture faces, the dominant mineral phases are calcite, fine-grained quartz and clays (mostly interlayerings of Hlite and smectite) and sporadic apatite. Hydrothermal alteration of the granite extends for ~ 8 cm from the pegmatite contact. This is characterized b y strong argiUisation with illite/ smectite as the dominant alteration phases; some calcite is contained in K-feldspar and

4.2. B6ttstein drillcore (BOE: 618.34-618.70 m)

The BSttstein area is located in centralnorthern Switzerland ~ 30 km NW of Ziirich and 5 km from the German border. Investigations carried o u t by N A G R A began in October 1982 and are still in progress. A major part of this work has involved the drilling of a deep borehole to ~ 1500 m: the fracture sample selected for this study has been taken from ~ 618 m depth. The geology, geochemistry and geochronology of the intercepted granite, which forms part of a much wider complex, are summarised b y N A G R A (Technical Reports: NTB 84-25 and NTB 85-01). Essentially t w o hydrothermal events affected the granite: an early-stage coeval with granite emplacement which occurred ~ 300 Myr. ago, and a later (Permian?) event which resulted in alteration along fracture and fault zones. Drilling in the area encountered the crystalline basement at a depth of ~ 300 m. Down to the final depth of 1501 m the core consisted of the BSttstein granite. This granite is typically coarse-grained, rich in biotite, and porphyritic in texture accompanied b y large phenocrysts of K-feldspar. The chosen drillcore section was selected from a fracture intersection which was suspected as being water-conducting. The fracture was located near the contact between the granite and thin intruded horizons of peg-

238 finely dispersed FeOOH-oxides give the rock a reddish tinge. A sealed fissure (0.5-2.0 mm wide) containing some calcite and clay minerals occurs at 8 cm and thereafter the granite is less altered (although the feldspars are still pigmented with Fe-oxide) and the biotite content is higher. The granite contains two zones ~ 4 cm wide characterized by an increased mafic content. One of these zones constitutes the contact area with the pegmatite, and the other is located ~ 8 cm from the contact. Aar granite which is thought to be late Hercynian. Soon after emplacement these granitoids were intruded first by aplitic dykes, and then b y a variety of mafic lamprophyres. The interfaces between these latter dykes and the Aar granite mainly control the groundwater conductive system in the bedrock (Ohse, 1983). The granite is typically grey-coloured, fresh, mineralogically homogeneous in hand specimen, and medium- to coarse-grained in texture; the granite is also characteristically gneissose. The chosen fracture is fresh and intersects the core axis at 40; no alteration on a macro-scale was observed and there was no evidence of any fracture filling minerals. Its water-conducting properties are suspected on the basis of a large release of groundwater encountered while penetrating the fracture zone during drilling. This initially resulted in a flow of ~ 24 1 min. -1, which subsequently decreased to a steady rate around 2 1 min. -1. However, the selected fracture was only one of several present over a short distance (frequency of fracturing in this part of the core is 10 fractures per metre). The age of these fractures is uncertain, at least late Alpine to Recent; the area is still considered tectonically unstable and the continuous formation of new fractures/joints cannot be excluded. Petrographical and mineralogical descriptions of the Grimsel area are given b y Stalder (1964). The average modal content of the granite core (expressed in vol.%) is 33% quartz, 34% K-feldspar, 21% plagioclase, 7% biotite/chlorite, 1--2% muscovite, 1% hornblende and 2% epidote. The associated accessory phases are sphene, orthite, apatite, zircon, calcite and opaques. The feldspars have undergone some alteration to sericite, and the biotite to chlorite; magnetite is partly oxidised. These alterations are products of the complex metamorphic history of the area and should not be considered as due to low-temperature rock--water reactions during recent geological times.

4.3. Grimsel drillcore (FLG: 94.12--94.52 m)

The Grimsel area is located near the Grimsel Pass situated in central-south Switzerland ~ 15--20 km from the Italian border. The site, which n o w constitutes the N A G R A underground test laboratory, was initially developed as an underground hydroelectric power scheme which had tunneled through the Juchlistock mountain chain. In 1979 an exploratory drilling programme was started b y N A G R A which was completed in 1984. The drillcore sample investigated was selected from a near-horizontal exploration borehole (SB 80.001:94.12--94.52 m) drilled into the bedrock from the access tunnel to the p o w e r station: the altitude of the borehole level is 1730 m. The geology of the region is summarised in recent N A G R A publications (NTB 81-07 and NTB 85-34). The Aar massif of Hercynian age dominates the area; it is aligned in a SW--NE direction with a m a x i m u m extension of 115 km and a width of 23 km. This massif, which is the largest in the Swiss Alps, is surrounded b y metasediments (mostly schists) and is itself mainly c o m p o s e d of plutonic rocks of granitic/granodioritic composition. These granitoid bodies were intruded during the Hercynian orogeny (~ 280 Myr.). The Gastern granite was probably early Hercynian (located to the NW) and this was later followed b y the Grimsel granite/granodiorite. The test site is located within the Central

239 5. Chemistry of the drillcores

5.1. Krakemala drillcore

The drillcore contents and distributions of U and its daughter decay products, Fe and t w o REE's, La and Ce, are listed in Table I and illustrated in Fig. 1. Other elements analysed b u t n o t illustrated include Na, K, Sc, Rb, Cs, Ta and the remaining REE's. The U contents range from 12.51 to 23.26 p p m (mean 15.67 ppm) and the Th from 46.37 to 89.42 p p m (mean 67.04 ppm); the mean Th/U ratio is 4.38. These U and Th values, which are roughly in agreem e n t with the average values from the granite as a whole (Smellie and Stuckless, 1985), are still anomalously high when compared to either the average contents for worldwide granites {Rogers and Adams, 1969a, b) or a compilation of 2500 granites mainly from the U.S.A. (Stuckless and Van Trump, 1982). Furthermore, in crustal rocks the average Th/U ratio is 3.5 (Rogers and Adams, 1969a, b), i.e. Th concentrations usually correlate with U b y a factor of 3. Thus the high Th/U ratio recorded from the GStemar granite would suggest a general removal of U, which has also been demonstrated isotopically b y Smellie and Stuckless {1985). From fission-track studies of the G5temar granite (Kresten and Chyssler, 1976; Smellie and Stuckless, 1985) most of the U is associated with zircon and monazite with subordinate amounts contained in sphene, magnetite, hematite, and ilmenorutile-type phases. Small amounts are also associated with sericite, epidote and FeOOH-oxide dispersions. Although no data are available for minor Th, the distribution of monazite, which can contain up to 10 wt.% Th and perhaps as high as 26 wt.% Th (Frondel et al., 1967), almost certainly controls the Th distributions in the GStemar granite. The distribution of U and Th along the drillcore from the fracture zone is shown in Fig. 1. Apart from the fracture face which

indicates an increase, the U content is fairly uniform for a distance of ~ 8 cm into the granite until a broad peak occurs over a length of 4 cm. In comparison, Th indicates an antipathetic decrease at the fracture face, and then shows an extremely irregular fluctuation until it t o o forms a broad-peak sympathetic to that observed for U. On the basis of mineralogy, the known geochemical behaviour of U and Th, and the similarity of the Th profile with those of the REE's {discussed on p. 243), these profiles suggest that Th has remained relatively immobile throughout the history of the granite, and its irregular distribution thus reflects the irregular distribution of the dominant Th-bearing mineral phase, monazite, which is of primary origin. As the first sample represents fracture-surface scrapings of mostly hematite, chlorite, clay mineral and FeOOH-oxides, the lower Th value recorded here is perhaps not so surprising. With regards to U, which in comparison is easily mobilised and leached from granite provided the system has been open to fluid movement, Fig. 1 shows some interesting features. If the granite had remained a closed chemical system since crystallisation, then most of the U would be associated with primary resistant radioactive minerals (often together with Th and REE) with more minor amounts along intergranular boundaries (Tieh et al., 1980). For the case in question therefore, the distribution of major U would be expected to be sympathetic to Th, which in turn would be a reflection of the monazite content in the rock, as suggested above. In fact there is a sympathetic increase in U and Th from ~ 8--13 cm along the drillcore which shows that this part of the rock has probably retained its original U and Th signature, although the intergranular labile U has almost certainly been removed. In contrast, the absence of any sympathetic U--Th relationship within the initial 8 cm o f the drillcore indicates that U has at some stage been leached from even the more resistant radioactive mineral phases pos-

bO O

TABLE I I s o t o p i c and chemical analyses of u r a n i u m and t h o r i u m f r o m drillcore K1, Kr$kem$la, S w e d e n Specmen Distance f r o m fract u r e face U(*) (ppm) Th(*) (ppm) Th/U 23~U ~4U ~4U/2~U :~Th 2~OTh/~34U 2~6Ra :~6Ra/230Th ~32Th

(cm)
KIA-1 KIA-2 KIA-3 KIA-4 KIA-5 KIA-6 KIA-7 KIA-8 KIA-9 KIA-10 KIA-11 KIA-12 KIA-13 KIA-14 KIA-15 surface scrapings 1 2 3 4 5 6 7 8 9 10 11 12 13 43 20.13 12.51 13.19 12.78 12.92 13.46 13.46 12.65 14.55 14.42 18.63 19.31 23.26 19.31 14.14 18.90 0.54 0.41 0.27 +_ 0 . 4 1 0.41 0.41 0.27 0.27 0.27 + 0.27 0.54 0.41 0.68 0.54 0.41 0.27 46,37 57.96 74.52 65.83 84.46 61.69 64.17 69.13 48.02 69.97 + + -- 2.07 1.24 2.07 0.83 2.07 1.24 1.24 1.24 1.24 1.24 2.3 4.6 5.6 5.2 6.5 4.6 4.8 5.5 3.3 4.9 --3.8 4.3 2.7 3.8 14.8 9.2 9.7 9.4 9.5 9.9 9.9 9.3 10.7 10.6 13.7 14.2 17.1 14.2 10.4 13.9 0.4 0.3 0.2 0.3 0.3 0.3 0.2 + 0.2 0.2 0.2 0.4 0.3 0.5 +- 0.4 0.3 0.2 33.9 16.7 10.9 9.0 8.5 8.7 8.6 7.9 9.4 8.9 11.4 11.4 14.7 12.5 8.5 12.1 + 0.8 0.3 0.2 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.3 0.2 0.5 0.4 0.3 0.2 2.29 1.81 1.12 0.96 0.89 0.88 0.87 0.85 0.88 0.84 0.83 0.80 0.86 0.88 0.82 0.87 0.51 0.05 0.02 0.03 0.04 0.03 0.03 0.03 0.02 0.02 0.03 0.02 0.03 0.03 0.03 0.02 31.9 14.8 9.3 9.1 8.5 8.6 8.5 7.8 9.8 9.3 -- 1.0 0.3 0.3 0.2 0.3 0.2 0.2 0.3 0.3 0.2 0.94 0.89 0.85 1.01 1.00 0.99 0.99 0.99 1.04 1.04 -- 0.06 0.05 0.05 0.05 0.06 0.06 0.05 0.07 0.05 0.05 32.7 16.6 9.7 7.8+ 10.4 + 8.6 6.8 7.5 7.6 8.6 9.6 10.8+ 9.2 8.7 9.2 4.9 2.5 1.5 1.2 1.6 1.3 1.0 1.1 1.1 1.3 1.4 1.6 1.4 1.3 1.4 1.03 1.12 1.04+ 0.86 1.22 + 1.00 0.80 0.96 0.78 0.92 0.15 0.17 0.16 0.13 0.18 0.15 0.12 0.14 0.12 0.14 11.2 14.0 18.0 15.9 20.4 14.9 15.5 16.7 + 11.6+ 16.9 0.5 0.3 0.5 0.2 0.5 0.3 0.3 0.3 0.3 0.3

89.42 83.21 51.34 72.45

2.48 2.07 0.83 2.48

16.1 11.4 8.8 11.4

0.6 0.3 0.2 0.4

1.10 0.91 1.04 0.94

0.07 0.06 0.06 0.06

0.67 0.81 0.99 0.81

0.10 0.12 0.15 0.12

21.6+ 20.1 12.4+17.5 +

0.6 0.5 0.2 0.6

*Determined by radioisotope dilution and a-spectrometry.

Kr~kem~to DriLLcore(K1:317.85-318.40m)
U

241
Th
100

U (ppm)

90
80

Th I ppm)

70 60 50 40

2.2 234 u / 2 3 8 U Activity Ratio 2.0

1.8

1.6 1.t, 1.2 1.0 0.8

i

Ir

][

230Th/234 U. Activity Ratio

1.1 1.0 0.9

0.8

}t

ttttt

t t
t

1.6 1./~ 1.2 Activity Ratio 1~0

0.8

tt t ttt t

0.6

1.2
Fe

1.0 0.8 0.6 0.~,

(%)

O.S
0

Fe 2. (Fe 2*+ Fe 3*)

0.4

0 0

03

0.2

220 200 180 1 60 (ppm) 140 120100 80 60

40200
1 2 3 4 5 6 7 8 9 10 11 12 13 40 41 42 43

FRACTURE FACE

DriLIcoro Length [ cm )

Fig. 1. Distribution of uranium and its daughter decay products, thorium, REE (La and Ce), total iron and iron oxide ratios characteristic of the K1 drillcore from Kr~kem~da, Sweden.

242 sibly during some hydrothermal event, the suggestion of such a process being supported mineralogically by the partial alteration of monazite along this part of the core. However, U removal is n o t indicated from the fracture surface sample which instead shows a marked increase. This suggests that in the groundwater--fracture system U accumulation rather than leaching is taking place. Similar marginal increases in U have been reported in altered and fractured samples of granite from the Eye--Dashwa Lakes pluton, Ontario, Canada (Kamineni, 1986, in this special issue), which has been interpreted as preferential sorption or coprecipitation of U with hematite. to prevent re-equilibration of the 23Th daughter with :34U; this would account for the overall isotopic disequilibrium observed between :34U and 23Th in the three samples comprising the marginal zone to the fracture. The question n o w arises as to whether these observed isotopic values at the fract u r e - g r o u n d w a t e r interface zone are due to continuous isotopic movement from the granite to the interface zone, followed b y subsequent sorption and/or isotopic fractionation onto the fracture minerals or, if these isotopic products have been introduced from the groundwaters to the fracture face b y direct sorption or through daughter emplacement processes (i.e. solution-to-solid :34U + 23Th recoil gain) as suggested by Rosholt (1983). Unfortunately, it is n o t possible to say if one or a combination of these two processes is responsible. In any case it underlines the importance of the physico-chemical properties of the fracture minerals in radionuclide retardation processes. The work of Airey (1986, in this special issue) clearly illustrates that the subsequent immobility of the radionuclides after up-take from solution will largely depend on the isotopic fractionation between the fracture zone minerals. It was shown that U and Th tend to concentrate principally on the Fe phases, whilst Ra seems to be more often in association with the clay--quartz fraction where its activity exceeds that of its parent Th. Furthermore, radionuclides taken up by crystalline Fe phases such as goethite are believed to be inaccessible to removal by solution. Work b y Kamineni (1986, in this special issue) also records the importance of goethite to explain enrichments of Th along fractured zones located in the Eye--Dasha pluton, Ontario. Unfortunately, from Krhkemlila there is no isotopic analysis of the groundwater in contact with the fracture zone that might indicate which U daughters are being removed from or added to the solid fracture surfaces. Groundwaters at the depth of the drillcore

5.1.1. Uranium-series disequilibrium. Plots of the activity ratios 234U/:38U, :3Th/234U and :26Ra/23Th are presented in Fig. 1. The 234U/238U ratios indicate general disequilibriu m within a 2a analytical error, with values ranging from a m a x i m u m of 2.29 at the fracture surface to a mean value of 0.86 for the major length of the drillcore. In contrast, most of the 23Th/234U and 22~Ra/ 23Th ratios lie within 2a of secular equilibrium. Considering the major length of the drillcore (i.e. excluding the fracture edge) the isotope data show depletions of both :34U and 23Th relative to 23sU. This indicates a significant :34U loss (most likely by recoil processes) with little :38U loss, followed b y a re-equilibration of the :3Th daughter with :34U. In other words, throughout the granite represented b y these samples, the loss of 234U (probably continuous) has been occurring over at least 1 Myr. b y solid-tosolution a-recoil processes. In strong contrast, the granite marginal to the fracture zone is characterized b y enrichments of all four isotopes (Table I); these fall off rapidly over a distance of 1--3 cm into the granite. The 234U/238U ratios indicate an accumulation of 234U which must be occurring sufficiently rapidly

243 sample (~ 318 m) should be reducing in character. According to secondary U accumulation processes as discussed by Ogmond et al. (1983), the isotopic data from the fracture--groundwater interface sample indicates a position furthest down,flow whereu p o n the groundwater is strongly reducing and 234U exceeds 238U, as does 23Th, its in-grown daughter. It would appear reasonable to assume, therefore, that ~the groundwaters at a depth of ~ 350--400 m a t Krhkem~la are characterized by an excess of 234U and are strongly reducing. Along the fracture surfaces sorption and/or coprecipitation of 234U with Fe-oxides is taking place; solutionto-solid 234U + 23Th recoil gain processes may also be contributing. The continuous loss of 234U suggested from the overall length of the drillcore indicates that there is also some radionuclide m o v e m e n t from the fracture edge to the groundwater. The degree of radionuclide retardation and retention from both processes will be dependent on the physico-chemical properties of the fracture zone minerals, which consist mostly of hematite, chlorite, FeOOH-oxides and clay minerals.
5.1.2. Iron oxidation states. Fig. 1 illustrates the distribution of Fe and the variation of the Fe2/(Fe 2 + Fe 3) oxidation ratio along the drillcore. Fe shows a slightly irregular b u t distinct increase towards the fracture face; this is apparent mineralogically by an increase o f h y d r o x y Fe-oxide dispersions, throughout the rock, together with the presence of chlorite and hematite at and near the fracture face. The similarity of the Fe distribution pattern, excluding the fracture face, with that of Th, is due to the presence of magnetite grains which often a c c o m p a n y the monazite aggregates in the granite. An a t t e m p t to assess the extent of any oxidation effects in the rock was crudely made using whole-rock Fe 2 and Fe 3* analysis; the degree of oxidation is expressed as the ratio Fe2+/(Fe : + Fe3). Towards the

fracture face there is some suggestion of increased oxidation.

5.1.3. Rare-earth (REE) distributions. Plots of La and Ce (which illustrate those trends c o m m o n to all the REE's and Ta) are presented in Fig. 1. From the general similarity of the element distribution patterns within the drillcore, and in comparison with Th (Fig. 1), it can be concluded that no obvious mobilisation of these elements has taken place, and that their distribution is dominantly a reflection of the occurrence of monazite in the granite. 5.1.4. Minor uranium distributions. The distribution of minor U {< 1 ppm) has been investigated by fission-track analysis. The results show that even though there is a background dispersion of matrix U (i.e. U present within and interstitial to the main rock-forming minerals) throughout the total length of the core examined, there is a significant increase in density and penetration within the - 3-cm wide margin adjacent to the fracture face; this corresponds to the observed uranium/Fe-oxide increase described above. The impregnation pattern corresponds to the dispersion of FeOOH-oxides and mineral breakdown products such as sericite, which in turn corresponds to the degree of alteration. Alteration is pervasive b u t irregular in extent due to microstructural controls on penetration, which progresses along structural weaknesses such as fissures, intergranular boundaries, and crystallographic planes (mostly affecting the feldspars). 5.2. B6ttstein drillcore

The distributions of U and its daughter decay products, Fe and the two REE's La and Ce, are listed in Table II and illustrated in Fig. 2. Although n o t presented, higher amounts of Na and lower a m o u n t s of K and Rb support the mineralogical observations and help t o distinguish the pegmatite from the

t~

TABLE

II and thorium Th (.1) (ppm) from drillcore BOE (618.34--618.70 Th/U 33SU 2S4U m), Bottstein, 234U/23SU Switzerland 23Th/234U 22SRa 226Ra/Z3Th 232Th

Isotopic and chemical analyses of uranium Specimen Distance from fracture face
(cm)

U (*l) (ppm)
"

23Th

BOE-3396 BOE-1 BOE-2 BOE-3 BOE-4 BOE-5 BOE-6 BOE-7 BOE-8 BOE-9 BOE-10 BOE-11 BOE-12 ,1 ,2 ,3 ,4

surface s c r a p i n g s *~ fracture .3 slice 1 . 5 *3 3 . 0 *3 5 . 0 *3 6 . 5 *4 8 . 0 *4 1 0 . 0 *4 1 2 . 1 *4 13.9 15.6 17.6 20.0 *4 *4 *4 *4

2.42 4.20 3.98 3.41 2.68 2.71 5.73 3.59 3.60 3.60 3.40 3.55 3.79 3.37 4.09

+ +

0.11 0.11 0.08 0.05 0.04 0.08 0.07 0.10 0.10 0.14 0.07 0.05 0.07 0.08

0.99 0.95 0.74 3.27 7.32 12.58 34.69 23.27 24.63 21.44 23.52 25.50 22.27 25.30 24.18

0.12 0.25 0.12 0.29 0.29 0.87 0.58 0.75 0.62 0.70 0.75 0.41 0.50 0.58

0.4 0.2 0.2 1.0 2.7 4.6 6.1 6.5 6.8 6.0 6.9 7.2 5.9 7.5 5.9

1.78 3.09 2.93 2.51 1.97 1.99 4.21 2.64 2.65 2.65 2.50 2.61 2.79 2.48 3.01

0.08 0.08 0.05 0.04 + 0.03 0.06 -+ 0 . 0 5 0.07 0.07 0.10 0.05 0.04 0.05 0.06

1.84 3.17 2.65 2.60 2.01 2.04 4.15 2.65 2.67 2.64 2.47 2.63 2.82 2.46 2.94

0.08 0.08 0.05 -+ 0 . 0 4 0.03 0.06 0.05 0.07 0.07 0.10 0.05 0.04 0.05 0.06

1.04 1.03 0.90 1.04 1.02 1.02 0.99 1.00 1.01 1.00 0.99 1.01 1.01 0.99 0.98

0.04 0.04 0.05 0.03 0.03 0.04 -+ 0 . 0 2 0.03 0.04 0.04 0.05 0.03 + 0.03 0.03 0.03

1.78 3.00 2.82 2.65 2.23 2.23 4.63 2.89 3.01 2.52 2.78 2.99 3.13 2.70 3.29

0.12 0.11 0.05 0.08 0.06 0.14 0.08 0.12 0.10 0.08 0.12 0.10 0.07 0.10

0.97 0.95 1.06 1.02 1.11 1.09 1.12 1.09 1.13 0.95 1.13 1.14 1.11 1.10 1.12

0.04 _+ 0 . 0 7 0.08 + 0.04 -+ 0 . 0 6 0.05 0.05 0.05 0.07 0.07 0.07 0.06 0.05 +- 0 . 0 5 0.05

-2.2 0.3 -2.2 0.3 1.9 0.3 2.1 0.3 4 . 7 -+ 0 . 7 3 . 0 -+ 0 . 4 2.8 0.4 2.7 0.4 -2 . 7 _+ 0 . 4 2.6 0.4 2.6 0.4 2.9 0.4

-0.73 0.11 -0 . 8 3 -+ 0 . 1 2 0.85 0.13 0.94 0.14 1 . 0 2 +- 0 . 1 5 1.04 0.16 0 . 9 3 -+ 0 . 1 4 1.07 0.16 -0.90 0.14 0.83 0.12 0.96 0.15 0.88 0.13

0.24 0.24 0.18 0.79 1.77 3.04 8.38 5.62 5.95 5.18 5.68 6.16 5.38 6.11 5.84

0.03 0.06 0.03 0.07 -+ 0 . 0 7 0.21 -+ 0 . 1 4 0.18 0.15 -+ 0 . 1 7 0.18 0.10 0.12 0.14

Determined by radioisotope dilution and ~-spectxometry. Ivanovich and Wilkins (1984b). Pegmatite. Granite.

B~ffsi'ein Dri[lcore (BOE:6183/,-618.70 m)

Pegmafite 6

245
"-----. u.
E~ Th

I ~'~

Gmnife

J L ~ \

(p~n)

5 4 3 2 1 0

34 32 30
28

;~ 22 20 12 10
8 6

26

[ ppm)

Th

/. 2 0 23tU,?~~
Activity

Rotio

1,0 O.g
,,t

230Th734 u Ratio
Activity
1.0

09

1.1

10 226Ro?30Th Activity
Rotio

09
0.8

07

tt
t
I

Tot~ Fe
(.,~)

2.0 1.0 (10

Fe2+
(F. 2~ * Fe 3* )

15 tO
~S

100 9o
80

(ppcrO

60 50 40 3o 20 10
1 2 3 4 5 6 7 e 9 10 11 1z 13 l& 15 16 17 18 19

Lo

20

FRACTUREFACE

Drlitcore Length

{ cm )

Fig. 2. Distribution of uranium and its daughter decay products, thorium, REE (La and Ce), total iron and iron oxide ratios characteristic of the BOE drillcore from BSttstein, Switzerland.

246 granite. From Table II the pegmatite U contents range from 2.42 to 4.43 p p m and the Th from 0.74 to 12.58 p p m (mean Th/U ratio of 0.9). In the adjacent granite the corresponding ranges are 3.37--5.73 ppm U and 21.44--34.69 p p m Th, respectively (mean Th/U ratio of 6.5). Compared to fresh, unfractured granite parts of the same drillcore (NAGRA 85-01), which records U values ranging from 8.6 to 10.4 p p m and Th values from 19.8 to 29.3 ppm {mean Th/U ratio of 2.65), the sampled core section indicates a significant depletion of U; Th appears to have been immobile. The fresh granite compares favourably with world-wide averages for normal granites. Although no further data are available for the pegmatite horizons, aplitic zones througho u t the drillcore are characterized b y exhibiting low Th (7.61--16.62 ppm) and high U (10.88--29.88 ppm) with a mean Th/U ratio of 0.95. Fission-track analyses are n o t available, however most of the U and Th is believed to be associated with the accessory mineral contents which include zircon and apatite. Minor amounts of U and Th may be a c c o m m o d a t e d b y the mafic constituents such as biotite/chlorite, or associated with finely dispersed Fe-oxides. The distribution of U and Th along the drillcore is illustrated in Fig. 2. Within the granite portion of the drillcore the U content is uniformly distributed until the pegmatite is approached whereupon a small increase is observed. In the pegmatite the U decreases only to increase again ~ 3 cm from the fracture surface; a small decrease occurs at the fracture surface. The Th is essentially sympathetic to that of U except that in the pegmatite a continuous decrease is observed. As inferred from the mineralogy, the granite represented by this part of the drillcore has been subjected to hydrothermal temperatures in association with the emplacem e n t of the adjacent pegmatite and aplite horizons. This appears to have remobilised and removed part of the available U in the rock, resulting in low b u t uniform levels within both the granite and pegmatite. The observed increase of U and Th at the granite-pegmatite c o n t a c t may be due to limited remobilisation and concentration of both elements marginal to the pegmatite; this contact zone just within the granite is also characterized b y an increase in mafic mineral content. 5.2.1. Uranium-series disequilibrium. Plots of the activity ratios 234U/238U, 23Th/234U and 22~Ra/23Th are presented in Fig. 2. The 234U/238U ratio is unity for most of the core length although a slight increase to a maximum of 1.04 is observed close to the fracture face. The above discussion of the Kr~kem~la 234U/238U ratio distribution can again be applied in this case but the degree of U mobilisation involved is obviously less. The 23Th/234U results for BSttstein exhibit a value slightly greater than unity except close to the fracture whereu p o n a significant decrease occurs. This presents an obvious contrast to the Kr~kem~la core and suggests solution to rock transfer of 23OTh produced by the decay of 234U in solution. The difference between the Kr~kem~la and BSttstein 23Th/234U results could be partly explained by a much more rapid fluid transport through the Kr~kem~la core than in the BSttstein core. In c o m m o n with Kr~kem~la, the 226Ra/ 23Th ratio is considered to represent equilibrium. 5.2.2. Iron oxidation states. Fig. 2 illustrates the distribution of Fe and the variation of the Fe2+/(Fe 2 + Fe 3) oxidation states. The total Fe contents are significantly higher in the granite than the pegmatite; their respective distributions are relatively homogeneous. These uniform distributions are supported mineralogically in that there is a fairly regular dispersion of FeOOH-oxides throughout the rock matrix. The oxide ratios are almost all less than unity, which indicates a fairly uniform degree

247 o f oxidation t h r o u g h o u t b o t h rock types. This is in accord with this part of the BSttstein drillcore as a whole. It has been found (NTB 85-02) that Fe 2 minerals are c o m m o n in fractures above 950-m depth, b u t are entirely absent below 1050 m. 5.2.3. Rare-earth (REE) distributions. Fig. 2 presents data of La and Ce which illustrate the main REE and Ta trends encountered. Both La and Ce show fairly uniform distributions within the granite until ~ 7 cm from the pegmatite contact. At this point a systematic decrease occurs and then a very marked sharp increase just at the contact. This is followed b y a gradual decrease through the pegmatite to the fracture edge. In addition to being typical of all the measured lanthanides (in contrast Sc, Co, Rb, Ba, Hf and Ta showed small increases), the La and Ce profiles are similar to that of Th. The data strongly suggest that these increased concentrations are the result of limited remobilisation and concentration during hydrothermal activity coeval with the introduction of the pegmatites. collected ~ 1 km from the test-site (Bajo, 1980). These have shown that the main U- and Th-bearing mineral phases are allanite, thorite, zircon and apatite; polycrase is sporadically present sometimes included in feldspar. Although limited fission-track studies have indicated an absence of interstitial U, the slightly lower Th/U ratios for the granite, in comparison with worldwide averages, suggests that small additions of U may have taken place. From Fig. 3 the U distribution varies irregularly between ~ 5 and ~ 9 ppm and this is even more accentuated by the Th which ranges from 16 to 28 ppm. This variation is n o t affected by the fracture and can be explained by mineralogical inhomogeneities. There is no obvious relationship between the U and Th, which suggests that the U and Th contents are controlled by radioactive minerals which are distinctly U- or Th-bearing. Alternatively, it m a y reflect a redistribution of U within the rock. In contrast Th has probably been immobile and has retained its original signature. 5.3.1. Uranium-series disequilibrium. Plots of the 234U/238U, ZSTh/234U and ~26Ra/23Th activity ratios are presented in Fig. 3 and show secular equilibrium within the limits of analytical precision (20). This means an absence of rock--water interaction within the interval of at least the last 1 Myr. 5.3.2. Iron oxidation states. Fig. 3 presents the distribution of total Fe and the Fe oxidation ratios along the sampled drillcore. Approaching the fracture from both ends of the drillcore, t h e total Fe content increases and then decreases marginal to the fracture. The variation in the total Fe content is considered to reflect the genetic inhomogeneities due to the primary mineralogy rather than any redistribution during rock--water interaction. The Fe2+/(Fe 2 + Fe 3) oxidation ratios are generally uniform although there may be a suggestion of slightly enhanced oxidation at one side of the fracture zone.

5.3. Grimsel drillcore The distributions of U and its daughter decay products, Fe and t w o REE's, La and Ce, are listed in Table III and illustrated in Fig. 3. Most of the trace-element data are relatively homogeneous throughout the length of the drillcore specimen; there are no systematic trends that cannot be explained b y mineralogical inconsistencies. A similar pattern is shown by U and Th which vary from 5.26 to 9.25 ppm and 16.06 to 27.53 ppm, respectively (mean Th/U ratio of 3.1). Only a few published analyses are available from the Aar granite as a whole; these record average U and Th contents of 11 and 27 ppm, respectively (Th/U ratio of 2.4). Some autoradiographic studies have been carried o u t on the Grimsel granite from specimens

t~ 0v

TABLE and thorium Th(*) (pPm) T h / U 238U 234U 234U/23SU 23Th 23Th/234U from drillcore FLG (94.12--94.52 m, exploration borehole

III S B 80. 0 0 1 ), G r i m s e l t e s t - s i t e , S w i t z e r l a n d 226Ra 226Ra/23Th 23~Th

Isotopic and chemical analyses of uranium

Specimen

Distance from fracture face (cm) 23.06 19.25 23.14 27.53 17.10 18.96 22.65 21.20 16.06 25.67 25.83 18.92 21.49 23.68 20.12 18.96 20.66 22.65 5.91 5.09 4.36 4.91 5.88 5.42 5.36 5.66 6.46 4.21 4.91 7.28 4.91 4.27 5.69 5.89 4.12 5.04

U (*) (ppm)

FLG-1 FLG-2A FLG-3A FLG-4A FLG-5A FLG-6A FLG-7A FLG-8A FLG-gA FLG-2B FLG-3B FLG-4B

2.0 3.9 5.0 7.0 8.8 10.8 13.0 15.8 18.6 0.4 0.6 2.2

5.2 5.3 5.1 5.6 4.4 5.5 6.2 5.8 5.6 6.5 6.9 6.2 5.7 + 0.03 -+ 0 . 0 3 0.03 0.02 0.03 0.04 0.02 0.02 0.03 0.02 _+ 0 . 0 2 0.02 0.03 0.03 + 0.02 0.03 0.03 0.03 0.03 0.16 0.21 -+ 0 . 1 3 0.31 0.12 0.14 0.30 0.10 0.21 -+ 0 . 1 2 0.31 0.12 0.20 - 0.14 0.17 0.13 0.17 0.14 1.15 1.03 1.10 1.04 1.11 1.05 1.07 1.07 0.89 1.05 1.05 0.96 0.96 1.03 1.07 1.01 0.94 1.07 1.11 0.06 0.06 0.05 -+ 0 . 0 6 0.06 0.06 0.06 0.04 0.07 0.04 0.07 0.05 0.07 0.08 0.05 0.06 0.06 0.06 0.05 5.7 5.7 5.0 5.1 5.9

0.8 0.8 0.8 0.8 0.7 0.8 0.9 0.9 0.8 1.0 1.0 0.9 0.9 0.9 0.9 0.8 0.8 0.9

0.96 0.99 0.90 0.87 1.05 1.12 0.85 1.18 1.31 1.14 1.17 1.50 1.13 0.96 1.12 1.15 1.04 1.00

+ 0.14 0.15 -+ 0 . 1 4 0.13 0.16 0.17 0.13 0.18 0.20 0.17 0.18 0.22 0.17 0.14 0.17 0.17 0.15 0.15

5.57 4.65 5.59 6.65 4.13 4.58 5.47 5.12 3.88 6.20 6.24 4.57 5.19 5.72 4.86 4.58 4.99 5.47

-+ 0 . 1 6 + 0.18 + 0.13 + 0.31 0.12 + 0.13 0.21 0.10 0.20 0.12 0.31 0.13 0.20 0.14 0.16 0.13 0.17 0.13

FLG-SB + 0.12 0.20 0.16 0.15 0.14

3.4

6.62 7.15 6.99 8.38 5.26 6.47 9.25 6.15 6.69 7.36 7.60 5,86 6.80 6.77

0.12 0.15 0.18 0.08 0.14 0.23 + 0.14 0.07 0.15 -+ 0 . 1 5 0.15 0.08 0.18 _+ 0 . 2 7

FLG-6B FLG-7B FLG-8B FLG-9B FLG-10B

4.8 6.4 7.8 9.2 10.8

7.44 6.94 6.31 6.38 7.29

0.66 0.75 0.54 1.28 + 0.50 _+ 0 . 5 4 0.87 0.41 0.83 0.50 1.28 0.54 0.83 - 0.58 0.66 0.54 0.70 0.54

3.5 2.7 3.3 3.3 3.3 2.9 2.4 3.4 2.4 3.5 3.4 3.2 3.2 -3.2 2.9 3.0 3.2 3.1

4.87 5.26 5.14 6.16 3.87 4.76 6.80 4.52 4.92 5.41 5.59 4.31 5.00 4.98 5.47 5.10 4.64 4.65 5.36

0.09 _+ 0 . 1 1 0.13 +- 0 . 0 6 0.10 0.17 0.10 0.05 0.11 0.11 _+ 0 . 1 1 0.06 0.13 0.20 0.09 -+ 0 . 1 5 0.12 0.11 0.10

4.72 5.21 5.15 6.22 3.79 4.66 6.78 4.58 4.82 5.40 5.59 4.29 5.23 4.91 5.53 5.02 4.66 4.60 5.32

+ 0.09 0.11 0.13 0.06 0.10 _+ 0 . 1 7 0.10 0.05 0.11 0.11 0.11 0.06 0.13 -+ 0 . 2 0 0.09 0.15 0.12 0.11 0.10

0.97 0.99 1.00 1.01 0.98 0.98 1.00 1.01 0.98 1.00 1.00 1.00 1.05 0.99 1.01 0.98 1.00 0.98 0.99

*Determined

by radioisotope

dilution and s-spectrometry.

Grimsel Oritic0re(FLG: 94..12-94.52 m}

o ~ tx
'

249
27

10 9
B

I
I

Th

26 .25
-

24 23 21 20 Th ( ppm ]

7 (pp~
R 6

,22
121 Q

19 18

17 1G

1.1

234Lt/238LL
Activity Ratio 1.0 09

I T

} ;

, 'tt~'t,=~-'tztz
f I I f

12 2301~/234 ~ Activity Ratio

1,1

10

}t {tttt i
t{t
I
' "
I% , ,

0.9

1.6 14 226Ra/230. ~ 1.2 Activity Ratio 1.0

O9

Tatar

Fe

(%)

1.0 09

131
!,2
1.1

Fe2 (Fe

2+,Fe 3+)

0.8 0.6

,ot
9080. 70.

I [

I
I
100-

I
Ce

(ppm)

60" 5040302010" 0 2 3 4 5 6 7

#
~10 11 12 13 14 15 16 17 / FRACTURE FACE Drillcore Length (crn)

Fig. 3. Distribution of uranium and its daughter decay products, thorium, R E E (La and Ce), total iron and iron oxide ratios characteristic of the F L G drillcore from Grimsel, Switzerland.

250

5.3.3. Rare-earth (REE) distributions. Fig. 3 presents the distributions of La and Ce which are representative of the REE's as a whole. The patterns are irregular and reflect the distributions of REE-bearing minerals such as allanite and polycrase.

6. Summary and conclusions

Drillcore sections traversing water-conducting fracture zones have been investigated from the Kr~kem~la area (GStemar granite) in southeast Sweden, the BSttstein area in north-central Switzerland, and the Grimsel area in south-central Switzerland. The cores have been examined using the U decay series together with the analysis of REE's, Fe2/Fe 3 ratios, and other elements such as Na, K, Sc, Co, Rb, Cs, Ba and Ta. Fissiontrack analyses have been carried out on polished thin rock sections from the Krhkem~la granite. Mineralogical studies of the GStemar sample (from ~ 380-m depth) have shown that at some stage the core has been subject to a weak pervasive alteration throughout its length. A more intense alteration is evident within a marginal zone (3--5 cm wide) adjacent to the fracture zone. Here there is widespread alteration of feldspar and magnetite together with impregnations of FeOOH-oxides. The mineral coatings on the fracture face have not been quantitatively determined, but appear to be composed of a mixture of hematite, chlorite, clay minerals and FeOOH-oxides. Of the elements measured, the variation of REE's and Th (together with Ta, Cs and Sc) appear to reflect the primary distribution of those radioactive refractory phases dominated by monazite. U, on the other hand, shows evidence of mobilisation and leaching from monazite up to a distance of 8 cm from the fracture zone; for the remainder of the core length the U distribution is sympathetic with that of Th and the REE's. Work by Smellie and Stuckless (1985) showed a general loss of U in the

GStemar granite as a result of large-scale hydrothermal alteration processes some 420 + 171 Myr. ago. These processes may have con tributed to the breakdown and removal of U from the monazite phases marginal to the fracture zone. Isotopically the 234U/238U ratio along the major length of the drillcore is less than unity, indicating U leaching. In strong contrast, the granite marginal to the fracture zone is characterized by a systematic increase in the 234U/:38U ratio to a value of 2.29 at the fracture face, which indicates an accumulation of U. The 23Th/:34U ratio is unity throughout most of the core with a small decrease being apparent close to the fracture. The 226Ra/23Th ratio values are considered to indicate secular equilibrium. The source of U giving rise to the enhanced concentrations and z34U/:3sU ratio values close to the fracture could be either U leached from the granite (as indicated by the 234U/:3sU ratios for the rest of the core), or from U transported in solution within the fracture migrating into the water-saturated rock. If the former process is responsible, then U migration over at least 40 cm is occurring whereas a constraint of ~ 3 cm can be applied if the latter process is taking place. The two processes are n o t mutually exclusive. The 23Th--234U and 226Ra--23Th equilibrium and 234U--238U disequilibrium suggest that the process involved is either continuous or else indicates a time-scale long enough to allow re-equilibration of 23Th with 234U and 226Ra with 23Th but short enough to maintain 234U disequilibrium with 23sU, i.e. a time-scale of ~ 1 Myr. The mechanism giving rise to U deposition at and near the fracture could be due to any or all of: (a) precipitation at a boundary between differing physico-chemical conditons; (b) sorption on the fracture filling minerals; and (c) solution-to-solid recoil transfer. The degree and retention of radionuclide retardation and retention from

251 b o t h processes will be d e p e n d e n t on the physico-chemical properties o f the fracture zone minerals. The BSttstein sample is inconveniently located near the contact between a pegmatite and aplite horizon. As a result the granite has been subjected to hydrothermal temperatures which are reflected in the mineralogy and the chemistry. U has been mobilised and depleted from the examined core which has resulted in a fairly uniform distribution b o t h within the granite and pegmatite. In contrast, Th and REE mobilisation has been restricted to a 7-cm zone in the granite adjacent to the pegmatite contact, with enrichment of these elements along a zone ~ 1--2 cm wide adjacent to the pegmatite contact. The mineralogy of the fracture zone and the host pegmatite do n o t indicate any low-temperature alteration processes which might have resulted in systematic compositional gradients around the fracture zone. The 234U/238U ratio is unity for most of t h e c o r e length although a slight increase is O b s e ~ e d close to the fracture face; the :26Ra/23Th ratio suggests secular equilibrium. The above discussion of the Kr~ikem~la 234U/:38U ratio distribution can again be applied in this case b u t the degree of U mobilisation involved is obviously less. The 23Th/234U results for B5ttstein exhibit a value slightly greater than unity except close to the fracture. This presents an obvious contrast to the Kr~kem~la core and suggests solution to rock transfer of 23Th produced by the decay of 234U in solution. The difference between the Kr~kem~la and BSttstein :3Th/234U results could be partly explained by a much more rapid fluid transport through the Kr~kem~ila core than in the BSttstein core. Contrastingly to BSttstein the Grimsel drillcore is located within a fairly homogeneous granite. Mineralogically and chemically there is no evidence of any systematic variations around the fracture that might be attributed to hydrothermal or low-temperature processes (in fact the fracture faces are devoid of any mineral coatings). This is supported by the isotopic data which show secular equilibrium for the U daughters measured. Two possibilities therefore exist: either groundwater interactions at this depth are n o t occurring, or, the fracture is not water-conducting. Based on hydrogeological criteria the fracture is considered waterconducting. Thus, to take the first possibility, the groundwater believed to be in contact with this fracture zone is characterized by an excess of 234U [234U/238U activity ratio of 2.12 (Ivanovich and Wilkins, 1984a, b)]. Whether this is the result of chemical or recoil mechanisms, the fracture faces in c o n t a c t with the groundwaters should reflect some isotopic discordance. As this is not the case, the isotopic equilibrium measured would lead one to suspect that the fracture is young, having occurred at any stage from Alpine to Recent times. A fairly recent joint may n o t have had time to react sufficiently with the groundwater which is n o w characterized by a chemical and isotopic signature obtained from other parts of the granite bedrock. In conclusion, the most important points resulting from this study are: (1) Two out of three of the investigated samples indicate that naturally occurring U radionuclides have been remobilised b y rock--water interaction processes within normal c o m p a c t granite marginal to a waterconducting fracture zone at repository depths. Th does n o t exhibit mobilisation under the same conditions. (2) Rock--water interaction has taken place under predominantly reducing conditions within measureable geological time, i.e. within the last 0.5 Myr. as determined by the 238U--234U---:3Th--22~Ra decay series and may still be active at the present time. (3) Isotopic movement has been mostly from the granite to the fracture zone through distances of at least 40 cm. (4) Some radionuclide movement has also taken place from the fracture zone (via

252

the groundwater) into the granite; the extent of penetration has been from 1--3 cm and this measure could therefore act as an upper limit in migration models. However, it may be an important prerequisite for penetration that the fracture system has been previously subject to hydrothermal processes, resulting in a zone of enhanced porosity around the fracture. Furthermore, the mineralogical alteration products would make the rock more conducive to radionuclide sorption/ retardation mechanisms. (5) Sorption and isotopic fractionation has occurred at the fracture--groundwater interface in association with fracture zone minerals such as hematite, chlorite, clay minerals and FeOOH-oxides. The major source of these radionuclides is believed to be the groundwaters which are either known or considered to contain excess 234U. (6) Fission-track analyses of the Kr~kem~la samples indicate that there was a significant increase in the a m o u n t and penetration of U within the ~ 3-cm wide margin adjacent to the fracture face. Penetration has been facilitated b y micro-scale structural weaknesses such as fissures, intergranular boundaries and crystallographic planes, and further aided b y pervasive alteration, especially of the feldspars. Radionuclide movements would appear to be the p r o d u c t of both fluid-flow and matrix diffusion processes. However, m a n y more similar studies need to be carried o u t w i t h i n different geological and groundwater environments before any far-reaching conclusions can be drawn. Still, as demonstrated by the data, such analogue studies can play an important role in validating the importance of far-field radionuclide retardation mechanisms within the context of high-level radioactive waste disposal in crystalline bedrock, whereupon fracture permeability will be the dominant m o d e of groundwater flow. Acknowledgements This project was jointly funded by the

Swedish Nuclear Fuel Supply Company (SKB) and the Swiss Nationale Genossenschaft ftir die Lagerung Radioaktiver Abf~lle (NAGRA), whose support at all stages of the study is greatfully acknowledged. With regards to the Swiss samples, the considerable help of Dr. Urs Frick (NAGRA) is also acknowledged.

References
Aberg, G., 1978. Precambrian geochronology of south-eastern Sweden. Geol. F6ren. Stockholm F6rh., 100: 125--154. Airey, P.L., 1986. Radionuclide migration around uranium ore bodies in the Alligator rivers region of the Northern Territory of Australia -- Analogue of radioactive waste repositories -- A review. In: N.A. Chapman and J.A.T. Smellie (GuestEditors), Natural Analogues to the Conditions around a Final Repository for High-level Radioactive Waste. Chem. Geol., 5 5 : 2 5 5 - - 2 6 8 (this special issue). Bacon, M.P. and Rosholt, J.N., 1982. Accumulation rates of Th-230, Pa-231 and some transition metals in the Bermuda Rise. Geochim. Cosmochim. Acta, 46: 651--666. Bajo, C., 1980. Extraction du Th et de I'U de granites suisses. Ph.D. Thesis, Federal Institute of Technology, Ziirich, No. 6738 (unpublished). Cherdynstev, V.V., 1956. Trans. Proc. 3rd Session Comm. on Absolute Age Determinations of Geological Formations, Moscow, 1954, 175 pp. Cherdynstev, V.V., 1971. Uranium-234. Israel Program for Scientific Translations, Jerusalem, 234 pp. Fleischer, R.L. and Raabe, O.G., 1978. Recoiling alpha-emitting nuclei -- Mechanisms for uraniumseries disequilibrium. Geochim. Cosmochim. Acta, 42: 973--978. Frondel, J.W., Fleischer, M. and Jones, R.S., 1967. Glossary of uranium- and thorium-bearing minerals. U.S. Geol. Surv., Bull. No. 1250, 4th edn. Gascoyne, M., 1984. Uranium-series disequilibrium studies of Shield rocks and groundwaters. 17th Info. Meet., CNFWMP (Can. Nucl. Fuel Waste Prog.), Ottawa, Ont., Feb. 1984. Ivanovich, M. and Wilkins, M.A., 1984a. Harwell report on the analysis of N A G R A groundwater samples from deep boreholes: Uranium disequilibrium measurements. AERE (At. Energy Res. Establish.), Harwell, Nucl. Phys. Div., No. AEREG 3135, 8 pp. Ivanovich, M. and Wilkins, M.A., 1984b. Harwell
o

253 report on the analysis of N A G R A rock and fracture infilling samples from B6ttstein core: uranium series disequilibrium measurements. AERE (At. Energy Res. Establish.), Harwell, Nucl. Phys. Div., No. AERE-G 3371, 10 pp. Kamineni, D.C., 1986. Distribution of uranium, thorium and rare-earth elements in the Eye-Dasha Lakes pluton - - a study of some analogue elements. In: N.A. Chapman and J.A.T. Smellie (Guest-Editors), Natural Analogues to the Conditions around a Final Repository for High-level Radioactive Waste. Chem. Geol., 5 5 : 3 6 1 - - 3 7 3 (this special issue). Kigoshi, K., 1971. Alpha recoil 234Th: dissolution into water and the 234U/23sU disequilibrium in nature. Science, 1973: 47--48. Koide, M. and Goldberg, E.D., 1965. Uranium-234/ uranium-238 ratios in seawater. In: M. Sears (Editors), Progress in Oceanography, Vol. 3. Pergamon, Oxford, pp. 173--177. Kresten, P. and Chyssler, J., 1976. The GStemar massif in south-eastern Sweden: A reconnaissance survey. Geol. F6ren. Stockholm F6rh., 98: 155--161. Kronfeld, J., 1974. Uranium deposition and ~34Th alpha recoil: an explanation for extreme 234U/ 238U fractionation within the Trinity aquifer. Earth Planet. Sci. Lett., 27: 327--330. Langmuir, D. and Herman, J.S., 1980. The mobility of thorium in natural waters at low temperatures. Geochim. Cosmochim. Acta, 44: 1 7 5 3 " 1766. MacKenzie, A.B., Scott, R.D., McKinley, I.G. and West, J.M., 1983. A study of long term ( 1 0 3 104 y) elemental migration in saturated clays and sediments. IGS (Inst. Geol. Sci.), London, Rep. No. FLPU 83-6, 48 pp. MacKenzie, A.B., Bowden, P. and Kinnaird, J.A., 1984. Combined neutron activation and particle track analysis of element distributions in a rock slice of mineralised granite. J. Radioanalyt. Chem., 82: 314--352. Magnusson, K-A., and Duran, O., 1984. Comparison between core log and hydraulic and geophysical measurements in boreholes. Geoexploration, 22: 169--186. NTB 81-07. Sondierbohrungen Juchlistock--Grimsel. N A G R A (Natl. Genoss. Lager. Rad. Abf~lle), Basel, Tech. Bet. NTB 81-07, 79 pp. NTB 84-25. Erl//uterung zur "Geologischen Karte der zentrale Nordschweiz, 1:100 000". N A G R A (Natl. Genoss. Lager. Rad. Abf~/lle), Basel, Tech. Bet. NTB 84-25, 234 pp. NTB 85-01. Nukle~/re Entsorgung Schweiz: Konzept und Ubersicht fiber das Projekt Gew~thr 1985. N A G R A (Natl. Genoss. Lager. Rad. Abf~/lle), Basel, Tech. Ber. NTB 8 5 - 0 1 , 1 9 0 pp. NTB 85-02. Radiometrischer Messungen an Gesteinproben der Tiefbohrung B6ttstein. N A G R A (Natl. Genoss. Lager. Rad. Abf/ille), Basel, Tech. Ber. NTB 85-02, 300 pp. NTB 85-34. Felslabor Grimsel: Rahmenprogram und Statusbericht. N A G R A (Natl. Genoss. Lager. Rad. Abf~]le), Basel, Tech. Bet. NTB 85-34, 148 pp. Ohse, W., 1983. LSsungs- und Fa]lungscheinungen im System oberfl~/chnn//hes unterirdisches Wasser/gesteinsbildende Minerale - - eine Untersuchung auf der Grundlage der chemischen Gleichgewichts-Thermodynamik. Ph.D. Thesis, University of Kiel, Kiel (unpublished). Osmond, J.K. and Cowart, J.B., 1976. The theory and uses of uranium isotopes in hydrology. At. Energy Rev., 14(4): 621---679. Osmond, J.K. and Cowart, J.B., 1982. Ground water. In: M. Ivanovich and R.S. Harmon (Editors), Uranium Series Disequilibrium: Applications to Environmental Problems. Oxford University Press, Oxford, pp. 202--245. Osmond, J.K., Cowart, J.B. and Ivanovich, M., 1983. Uranium isotopic disequilibrium in groundwater as an indicator of anomalies. Int. J. Appl. Radiat. Isot., 1: 283--308. Rogers, J.J.W. and Adams, J.A.S., 1969a. Thorium. In: K.H. Wedepohl (Editor), Handbook of Geochemistry, Vol. 2, No. 4. Springer, Berlin, pp. 90-A-1--90-O-1. Rogers, J.J.W. and Adams, J.A.S., 1969b. Uranium. In: K.H. Wedepohl (Editor), Handbook of Geochemistry, Vol. 2, No. A. Springer, Berlin, pp. 92-A-1--92-O-1. Rosholt, J.N., 1959. Natural radioactive disequilibrium of the uranium series. U.S. Geol. Surv., Bull. No. 1084-A. Rosholt, J.N., 1983. Isotopic composition of uranium and thorium in crystalline rocks. J. Geophys. Res., 88(B-9): 7315--7330. Rosholt, J.N., Shields, W.R. and Garner, E.L., 1963. Isotope fractionation of uranium in sandstone. Science, 139: 224--226. Rosholt, J.N., Doe, B.R. and Tatsumoto, M., 1966. Evolution of the isotopic composition of uranium and thorium in soil profiles. Geol. Soc. Am. Bull., 77: 987--1004. Schwarcz, H.P., Gascoyne, M. and Ford, D.C., 1982. Uranium series disequilibrium studies of granitic rocks. In: G.W. Bird and W.S. Fyfe (Editors), Geochemistry of Radioactive Waste Disposal. Chem. Geol., 3 6 : 8 7 - - 1 0 2 (special issue). Scott, R.D. and MacKenzie, A.B., 1984. Detection of 22:Rn by use of silicon surface barrier detectors. Int. J. Appl. Radiat. Isot., 35: 301--303. Scott, R.D. and MacKenzie, A.B., 1985. Measurement of low 222Rn activities by means of surface barrier detectors. Nucl. Instrum. Methods. A, 238: 160--164. Sherman, S., Olkiewicz, A. and Agrell, I., 1977. Bedrock investigation, Kr~kem~la. In: Handling of Spent Nuclear Fuel and Final Storage of Vitrified High Level Reprocessing Waste, Part II: Geology. SKB (Nucl. Fuel Waste Manage. Co.), Stockholm, Stud. Rep.

254 Smellie, J.A.T. and Rosholt, J.N., 1984. Radioactive disequilibrium in mineralised fracture samples from two uranium occurrences in northern Sweden. Lithos, 17: 215--225. Smellie, J.A.T. and Stuckless, J.S., 1985. Element mobility studies of two drill-cores from the Ghtemar Granite (Kr~tkem~la Test Site), southeast Sweden. Chem. Geol., 51: 55--78. Stalder, H.A., 1964. Petrographische und mineralogische Untersuchungen im Grimselgebiet. Thesis, University of Berne, Berne (published in Schweiz. Mineral. Petrogr. Mitt., 44: 187--398). Stuckless, J.S. and Van Trump, G.J., 1982. A compilation of radioelement concentrations in granite rocks of the contiguous United States. Proc. IAEA/OECD (Int. At. Energy Agency/Org. Educ. Cult. Org.). Symp. on Uranium Exploration Methods. Int. At. Energy Agency, Vienna, pp. 191--208. Thurher, D.L., 1962. Anomalous 234U/23sU in nature. J. Geophys. Res., 67 : 4518--4520. Tieh, T.T., Ledger, E.B. and Rowe, M.W., 1980. Release of uranium from granitic rocks during in situ weathering and initial erosion (central Texas). Chem. Geol., 29: 227--248. Zielinski, R.A., Peterman, Z.E., Stuckless, J.S., Rosholt, J.N. and Nkomo, I.T., 1981. The chemical and isotopic record of rock-water interaction in the Sherman granite, Wyoming and Colorado. Contrib. Mineral. Petrol., 78 : 209--219.