Formation and decay of the E1′ center and its precursor in natural quartz: basics and applications

Formation and decay of the E1′ center and its precursor in natural quartz: basics and applications

ARTICLE IN PRESS Applied Radiation and Isotopes 62 (2005) 325–330 www.elsevier.com/locate/apradiso Formation and decay of the E01 center and its pre...

238KB Sizes 0 Downloads 31 Views

ARTICLE IN PRESS

Applied Radiation and Isotopes 62 (2005) 325–330 www.elsevier.com/locate/apradiso

Formation and decay of the E01 center and its precursor in natural quartz: basics and applications Shin Toyoda Department of Applied Physics, Okayama University of Science, 1-1 Ridai, Okayama, 700-0005, Japan

Abstract The E01 center is a paramagnetic defect in quartz, where an unpaired electron is at an oxygen vacancy. The intensity of the E01 center after heat treatment at 300 1C for 15 min (the heat treated E01 center) indicates the amount of oxygen vacancies, which are precursors of the E01 center. Using this technique, oxygen vacancies were found to be much more stable than other impurity centers, and to be useful for new applications, such as dating of granites, dating of uranium ores, estimating the temperature of heat treatment for stone implements, and provenance of eolian dust. The formation process of the oxygen vacancies in natural quartz has been in debate. Our recent pulsed ESR study indicates that they are probably created by natural beta and gamma rays. r 2004 Elsevier Ltd. All rights reserved.

1. Introduction Quartz is one of the most abundant minerals on the Earth’s surface. Since quartz was first used for dating of fault movements with ESR (Ikeya et al., 1982), the method has been applied to tephra, fault gouge, and sediments. Dating of heated flint should be discussed in the same framework because it has the same ESR signals. The E01 center is one of the best characterized paramagnetic defects in quartz where an unpaired electron is at an oxygen vacancy (Silsbee, 1961). The model and electron energy levels have been studied by Rudra and Fowler (1987). According to Rudra and Fowler (1987), the most stable state of the oxygen vacancy is with two electrons (i.e. Si=Si bond). This neutral oxygen vacancy traps an electronic hole released on heating from hole centers such as Al centers and becomes paramagnetic, i.e. the E01 center (Jani et al., 1983). This process explains why its Tel.:+81-86-256-9608; fax:+81-86-256-9702.

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

signal intensity increases with heating in quartz, which had been reported at the initial stage of its study (Weeks and Nelson, 1960).

2. The E01 center and its precursor Fig. 1 shows the signal intensities in a stepwise heating experiment on granitic quartz (Toyoda and Ikeya, 1991a). The signal intensities of Al and Ti(-Li) centers decrease with increasing temperature while that of the E01 center increases up to 300 1C due to the increasing supply of electronic holes, and then decreases above this temperature. When the amount of holes to be supplied is small, the increase of the E01 center intensity is also small as shown in Fig. 2 (Toyoda and Hattori, 2000). In this experiment, the granitic quartz is once heated at 430 1C to erase both the E01 center and Al center signals. Subsequently, gamma ray doses shown in the horizontal axis are given, the signal intensity of the Al center is measured, and then, that of the E01 center is measured after heating

0969-8043/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2004.08.014

ARTICLE IN PRESS 326

S. Toyoda / Applied Radiation and Isotopes 62 (2005) 325–330

Toyoda and Ikeya (1991a) and Toyoda et al. (1992) proposed that the intensity of the E01 center after gamma ray irradiation to more than 200 Gy and subsequent heating at 300 1C for 15 min (heat-treated E01 center) indicates the amount of oxygen vacancies. As shown in Fig. 5 of Toyoda and Ikeya (1991a), the amount of oxygen vacancies decreases with heating above 450 1C. The thermal stability of the oxygen vacancies was evaluated from these results to be much greater than other paramagnetic ESR signals in quartz with a characteristic decay time of more than 109 y.

3. ESR dating for early earth

Fig. 1. Change of intensity of the ESR signals of the E01 ; Al and Ti-Li centers in quartz (Toyoda and Ikeya, 1991a). The intensity of the E01 center increases with heating due to transfer of holes to neutral oxygen vacancies.

Odom and Rink (1989) reported that the signal intensities of the untreated (natural) E01 center in granites are correlated with the ages of the host rocks. This observation implies that ESR dating can date quartz in an age range much older than previously dated (Gru¨n, 1989). However, from our point of view, it is not clear whether the correlation is due to increase in the amount of precursors (oxygen vacancies) in quartz with age or to increase in the portion of paramagnetic states (the E01 center) of oxygen vacancies where the amount of precursors stays at the same amount. Our subsequent papers (Toyoda, 1992; Toyoda et al., 1992; Toyoda and Hattori, 2000) showed that the amount of oxygen vacancies increase with age of host rocks as in Fig. 3, indicating that it is the former case.

Fig. 2. Dose responses of the Al center and the heat-treated E01 center (Toyoda and Hattori, 2000). The intensity of the Al center increases while the E01 center saturates above 200 Gy. It indicates that the saturation is due to a limited number of E01 centers.

at 300 1C for 15 min. The regenerated E01 center intensity increases with doses up to 200 Gy and then saturates while that of the Al center intensity continues to increase with doses. It indicates that the E01 center intensity saturates due to the limited number of oxygen vacancies.

Fig. 3. Intensity of the heat-treated E01 center in granitic/ volcanic quartz as a function of the age of the host rock. There is a good correlation, indicating that the number of oxygen vacancies increases with age (Toyoda, 1992; Toyoda and Hattori, 2000).

ARTICLE IN PRESS S. Toyoda / Applied Radiation and Isotopes 62 (2005) 325–330

4. Mechanisms in nature that create oxygen vacancies in quartz It is essential to identify the formation mechanism of oxygen vacancies in order to establish it as a dating method. Rink and Odom (1991) proposed that internal alpha recoil nuclei emitted by uranium and thorium and by their daughter nuclei contained in quartz matrix in ppb amounts would produce oxygen vacancies and showed the results of numerical simulation based on the uranium and thorium concentrations they measured. On the other hand, Wieser et al. (1989) showed that high-dose gamma rays create the E01 center in quartz. Toyoda et al. (1996) repeated a similar irradiation experiment quantitatively to show, as in Fig. 4, that the amounts of the heat-treated E01 center created by gamma ray irradiation are consistent with those in natural granitic quartz when plotted against calculated accumulated natural beta and gamma ray doses based on the uranium, thorium, and potassium concentrations and ages of the host rocks obtained by other radiometric dating methods. Toyoda et al. (2001) examined whether internal particle irradiation could actually create the amount of oxygen vacancies calculated by numerical simulation. They irradiated boron-doped quartz with thermal neutrons. Through an n-alpha reaction, internal alpha and Li particles were irradiated in quartz to create oxygen vacancies. The amount of oxygen vacancies

327

measured as heat-treated E01 centers was consistent with numerical simulation within a factor of 2.

5. Characteristics of the E01 centers in quartz created by He ion implantation and by electrons We recently performed experiments to examine the properties of the E01 center in quartz created by He ion implantation, in comparison with that by electrons. Hydrothermal quartz from Takatori mine, Ibaraki, Japan was sliced to a thickness of 1 mm for electron irradiation and to 0.2–0.4 mm for He ion implantation, both performed at Takasaki research laboratory of Japan Atomic Energy Research Institute. 5.1. Signal intensity change on heating (Asai et al., 2000) As results of stepwise heating experiments, the intensity of the E01 center formed by electron irradiation increased on heating up to 300 1C and then decreased as previously observed in granitic quartz, while that by He ion implantation did not show any increase but only monotonic decrease starting around 250 1C. 5.2. Pulsed ESR measurements Spin–spin relaxation times were measured by pulsed ESR measurements for the above irradiated samples

Fig. 4. Intensity of the heat-treated E01 center as a function of gamma ray dose given after heating the sample at 600 1C for 30 min to erase oxygen vacancies (Toyoda et al., 1996). The amount is consistent with the natural amount of heat-treated E01 center where the doses correspond to natural accumulated beta and gamma rays.

ARTICLE IN PRESS 328

S. Toyoda / Applied Radiation and Isotopes 62 (2005) 325–330

rather than by external alpha or by internal alpha and/ or alpha recoil nuclei. The results of stepwise heating experiments also supported this idea. Granitic quartz and quartz from uranium ore showed an intensity change on heating similar to the sample irradiated by electrons having a peak at 300 1C. The signal intensity of the E01 center in amorphous quartz glass shows monotonic decrease on heating. The crystalline lattice of the area in quartz damaged by He ions might become amorphous due to its high LET.

6. Applications Based on the above studies, new types of applications using the characteristics of the E01 center/oxygen vacancies have been proposed. 6.1. Dating of uranium ores Fig. 5. Spin–spin relaxation times measured by pulsed ESR measurements (Toyoda et al., unpublished). The E01 center signal created by high LET radiation shows shorter relaxation time, indicating locally concentrated distribution. The relaxation times for granitic quartz and quartz from uranium ore are rather longer, implying that those are created by external beta and gamma rays.

together with granitic quartz and quartz from a uranium ore. The obtained relaxation times were plotted as a function of averaged spin concentrations as shown in Fig. 5. The relaxation times of hydrothermal He ion-implanted quartz were 1.0–2.0 ms, decreasing with the increasing dose, while it was much longer (7.1–15.7 ms) for electron irradiated samples, also decreasing with increasing dose. The relaxation times of quartz from the Kanyemba uranium deposit were 3.5–7.1 ms and that of granitic quartz was 6.2 ms, both closer to electronirradiated samples.

5.3. Formation process of the oxygen vacancies in quartz in nature The shorter relaxation time corresponds to larger magnetic interaction between spins due to higher local spin concentrations and to shorter distance between spins. Therefore, the above observations are consistent with the theoretical consideration that high LET radiation creates locally concentrated oxygen vacancies while low LET creates those with lower concentration. The experimental results on the relaxation times for granitic quartz and quartz from a uranium ore indicate that the oxygen vacancies in these samples have been most probably created by external beta and gamma rays

Since the sensitivity of the oxygen vacancies in quartz is very low, they have linear response to high gamma ray doses, at least to 50 MGy (Toyoda et al., 1996). Toyoda et al. (1998) tried ESR dating of the Kanyemba uranium deposit in Zimbabwe. They observed two isochrones in the plot between the dose rates and the amount of oxygen vacancies in quartz. Although absolute age determination was not possible because the formation process of the oxygen vacancies was still in debate, the ratio of the ages between two formation processes of uranium was estimated to be 50 using the ratio of the slopes of the isochrones. 6.2. Estimation of the temperatures of heat treatment for stone implements Toyoda and Ikeya (1993) and Dunnell et al. (1994) noted that the ratio between natural and heat-treated E01 center intensity is a function of heat treatment, when ‘‘natural’’ intensity has not been affected by radiation after the heat treatment. The method was successfully applied to stone implements made by American Indians in the 7th century, to discuss the recycling of the implements. 6.3. Variations in paleo-monsoon: origins of eolian dust It was found that the quartz in eolian dust deposited in marine isotope stage 2 (the last glacial period) in the northern part of the Japanese Islands has a heat-treated E01 center intensity higher than that in the southern part, while the values are rather uniform in Holocene, and consistent with the lower values in the stage 2 sediments. The higher value is consistent with pre-Cambrian granite (Naruse et al., 1997; Ono et al., 1998; Toyoda

ARTICLE IN PRESS S. Toyoda / Applied Radiation and Isotopes 62 (2005) 325–330

and Naruse, 2002). They suggest the quartz in eolian dust having a lower value came from the Chinese loess plateau while the one having a higher value originated from the pre-Cambrian basement in Siberia during a colder period.

Fig. 6. Change in signal shape of the E01 center on gamma ray irradiation to granitic quartz (Toyoda and Schwarcz, 1997a). (a) No irradiation, (b) 3 kGy, (c) 6 kGy, (d) 15 kGy. The signal turns one peak with increasing dose due to the contribution of the counterfeit E01 signal.

329

7. Precautions concerning the counterfeit E01 signal Conventionally, the signal of the E01 center has been used for dating of quartz in fault gouge (Ikeya et al., 1982; Fukuchi, 1986; Lee and Schwarcz, 1994) and tephra (Toyoda and Ikeya, 1991b), and for dating of flint (Porat et al., 1994) where the equivalent doses have been obtained by extrapolating the increase of the E01 center to the zero ordinate. However, the increase may have been due to the counterfeit ESR signal (Toyoda and Schwarcz, 1997a). This signal was found to be created by gamma ray irradiation and to overlap the ‘‘real’’ E01 center signal. It is recognized by the shape of the signal as shown in Fig. 6 where the two-peak shape of the E01 center signal turns one-peak with increasing gamma ray dose given. The signal is unstable so that the signal disappears on heating at 170 1C for 15 min. Fig. 7 shows the results of stepwise heating experiments where the intensity decreases up to 170 1C in irradiated sample’s while no such intensity drop was observed in samples without irradiation, and then, increases above this temperature up to 300 1C as usual in both samples. The signal shape observed in the irradiated sample returns the original two-peak shape above 170 1C. It is possible, when the E01 center is used for dating like other impurity signals, that the dating procedure looks successful because the intensity increases with gamma ray dose, but actually it is due to production of the counterfeit E01 signal. This signal is actually confirmed in irradiated natural fault gouge (Toyoda and Schwarcz,

Fig. 7. Nominal intensity change of the E01 signal in irradiated/ unirradiated quartz on stepwise heating (extracted from Toyoda and Schwarcz, 1997a). The drop of intensity for the irradiated quartz at 170 1C is due to decay of the counterfeit E01 signal.

ARTICLE IN PRESS 330

S. Toyoda / Applied Radiation and Isotopes 62 (2005) 325–330

1997b). It is not recommended to use the E01 center signal for dating like other impurity center signals. In the case of tephra (Toyoda and Ikeya, 1991b), it is most probable that that tephra in marine deposit was contaminated by old sedimentary quartz which showed the E01 center signal.

8. Conclusions The precursors must be considered when looking at the E01 center signal in quartz. The precursors, oxygen vacancies have been created by external beta and gamma rays and accumulated in natural quartz for more than 10 Ma. New applications are possible using the specific characteristics between the precursors and the E01 center, while using this signal should be avoided for conventional dating of Quaternary samples.

References Asai, T., Toyoda, S., Komuro, K., Horikawa, Y., 2000. The process creating lattice defects in quartz of uranium ores: Investigation by ion implantation and electron beam irradiation. Adv. ESR Appl. 16, 15–23 (in Japanese with English abstract). Dunnell, R.D., Ikeya, M., McCutcheon, P.T., Toyoda, S., 1994. Heat treatment of Mill Creek and Dover cherts on the Malden Plain. southwest Missouri. J. Archaeol. Sci. 21, 79–89. Fukuchi, T., 1986. ESR dating of fault movement using various defect centers in quartz; the case in the western south Fossa Magna, Japan, Earth Planet. Sci. Lett. 78, 121–128. Grun, R., 1989. ESR dating for the early Earth. Nature 338, 543–544. Ikeya, M., Miki, T., Tanaka, K., 1982. Dating of Fault by electron spin resonance on intrafault materials. Science 215, 1392–1393. Jani, M.G., Bossoli, R.B., Halliburton, L.E., 1983. Further characterization of the E0 1 center in crystalline SiO2. Phys. Rev. B 27, 2285–2293. Lee, H.-K., Schwarcz, H.P., 1994. Criteria for complete zeroing of ESR signals during faulting of the San Gabriel fault zone, southern California. Tectonophys. 35, 317–337. Naruse, T., Ono, Y., Hirakawa, K., Okashita, M., Ikeya, M., 1997. Source areas of eolian dust quartz in East Asia tentative reconstruction of prevailing winds in isotope stage 2 using electron spin resonance. Geogr. Rev. Japan 70A, 15–27. Odom, A.L., Rink, W.J., 1989. Natural accumulation of Shottky-Frenkel defects: implications for a quartz geochronometer. Geology 17, 55–58. Ono, Y., Naruse, T., Ikeya, M., Kohno, H., Toyoda, S., 1998. Origin and derived courses of eolian dust quartz deposited during marine isotope stage 2 in East Asia, suggested by

ESR signal intensity. Global and Planetary Change 18, 129–135. Porat, N., Schwarcz, H.P., Valladas, H., Bar-Yosef, O., Vandermeersch, B., 1994. Electron spin resonance dating of burned flint from Kebara cave, Israel. Geoarchaeology 9, 393–407. Rink, J.W., Odom, A.L., 1991. Natural alpha recoil particle radiation and ionizing radiation sensitivities in quartz detected with EPR: implication for geochronometery. Nucl. Tracks. Radiat. Meas. 18, 163–173. Rudra, J.K., Fowler, W.B., 1987. Oxygen vacancy and the E01 center in crystalline SiO2. Phys. Rev. B. 35, 8223–8230. Silsbee, R.H., 1961. Electron spin resonance in neutronirradiated quartz. J. Appl. Phys. 32, 1459–1462. Toyoda, S., Ikeya, M., 1991a. Thermal stabilities of paramagnetic defect and impurity centers in quartz: basis for ESR dating of thermal history. Geochem. J. 25, 437–445. Toyoda, S., Ikeya, M., 1991b. ESR dating of quartz and plagioclase from volcanic ashes using E01 ; Al and Ti centers. Nucl. Tracks. Radiat. Meas. 18, 179–184. Toyoda, S., 1992. Production and decay characteristics of Paramagnetic defects in quartz: application to ESR dating: Ph. D. Thesis, Osaka University, Japan, 106p. Toyoda, S., Ikeya, M., Morikawa, J., Nagatomo, T., 1992. Enhancement of oxygen vacancies in quartz by natural external beta and gamma ray dose application to ESR geochronometer of Ma–Ga range. Geochem. J. 26, 111–115. Toyoda, S., Ikeya, M., 1993. The use of electron spin resonance (ESR) for the determination of prehistoric lithic heat treatment. Appl. Radiat. Isot. 44, 227–231. Toyoda, S., Rink, J.W., Schwarcz, H.P., Ikeya, M., 1996. Formation of E0 1 precursors in quartz: applications to dosimetry and dating. Appl. Radiat. Isot. 47, 1393–1398. Toyoda, S., Schwarcz, H.P., 1997a. Couterfeit E01 signal in quartz. Radiat. Meas. 27, 59–66. Toyoda, S., Schwarcz, H.P., 1997b. The hazard of the Counterfeit E01 signal in quartz to ESR dating of fault movements. Quatern. Geochronol. 16, 483–486. Toyoda, S., Komuro, K., Sato, K., Ikeya, M., Yoshida, H., 1998. ESR and CL observed in quartz grains from uranium deposits: implications for uranium migration in natural hydrogeological environment. Radiochemica Acta. 82, 331–334. Toyoda, S., Hattori, M., 2000. Formation and decay of the E01 center and of its precursor. Appl. Radiat. Isot. 52, 1351–1356. Toyoda, S., Rink, W.J., Yonezawa, C., Kagami, T., 2001. Insitu production of alpha particles and alpha recoil particles in quartz applied to ESR studies of oxygen vacancies. Quatern. Sci. Rev. 20, 1057–1061. Toyoda, S., Naruse, T., 2002. Eolian dust from the Asian deserts to the Japanese Islands since the Last Glacial Maximum; the basis for the ESR method. Transactions, Japanese Geomorphological Union 23, 811–820. Weeks, R.A., Nelson, C.M., 1960. Trapped electrons in irradiated quartz and silica: II. Electron spin resonance. J. Am. Ceram. Soc. 43, 399–404. Wieser, A., Goksu, Y., Regulla, D.F., 1989. ESR dosimetry in the ‘‘Giga-rad’’ range. Appl. Radiat. Isot. 40, 911–913.