Isotopic and Chemical Evidence for Primitive Aqueous Alteration in the Tagish Lake Meteorite

Finding high Cs/Ba phases in a sample is required to establish ¹³⁵Cs–¹³⁵Ba chronometry. In this study, a sequential acid-leaching experiment was performed as described previously (Hidaka et al. 2001). About 400 mg of powdered sample was leached using 5 mL of 0.1 M CH₃COOH-CH₃COONH₄, 0.1 M HCl, 2 M HCl, and aqua regia, successively. In each leaching step, the sample was ultrasonicated for 1 hr and kept at room temperature for 23 hr after adding each reagent. The acid residue was finally decomposed by HF–HClO₄ for 5 days at 130–150 °C. The complete dissolution was confirmed visually. Separately from the above leaching treatment, about 40 mg of the powdered sample was decomposed by HF–HClO₄, and treated as a whole rock for analysis. Each fraction was evaporated to dryness, and redissolved in 2 M HCl of 1 mL. These samples of five leaching fractions and whole rock were designated as L1, L2, L3, L4, L5, and WR, respectively. Each fraction recovered from individual leaching steps was divided into two portions for Sr and Ba isotopic analyses and for the determination of Rb, Sr, Cs, and Ba elemental abundance.

Minor portions (∼10% of the total amount) of L1–5 and WR fractions were evaporated to dryness once and redissolved into 5 ml of 2% HNO₃. An Agilent 7700× inductively coupled plasma mass spectrometry was used for the determination of elemental abundances of Rb, Sr, Cs, and Ba.

Major portions (∼90%) of L1–5 and WR fractions were used for the isotopic studies after chemical separation with conventional resin chemistry (Hidaka & Yoneda 2014). The method for Sr and Ba chemical separation was carried out as follows: Each sample solution was loaded onto cation-exchange resin packed column (AG50WX8, 200–400 mesh, H⁺ form, 50 mm length ×4.0 mm diameter). The column was washed with 3.5 mL of 2 M HCl for the elution of major elements, and then it was washed with 3.5 mL of 2 M HCl for the elution of the Sr fraction. Finally, the column was washed with 2 mL of 3 M HNO₃ for the elution of the Ba fraction. The recover yields of Sr and Ba were more than 90%. The Sr and Ba fractions were evaporated. Sr fractions recovered by the cation-exchange resin method often include Rb, which produces isobaric interference in the Sr isotopic measurement. For further purification, the Sr fraction was loaded onto a Sr-resin packed column (Eichrom, Sr resin, particle size of 100–150 μm, 100 mm length × 2.5 mm diameter). The column was washed with 2.5 mL of 3 M HNO₃ for the elution of Rb, and was washed with 3 mL of ultrapure water for the elution of Sr.

A TRITON-Plus thermal ionization mass spectrometer equipped with nine Faraday cups linked to 10¹¹ Ω resistors was used for Sr and Ba isotopic analyses. All analyses were performed in static mode with the amplifier rotation system in use (Hidaka & Yoneda 2013).

The Sr fraction was loaded on a single Re filament with a Ta₂O₅ activator. Strontium has four isotopes: ⁸⁴Sr is the p-process isotope, ⁸⁶Sr is the s-process isotope, and ⁸⁷Sr and ⁸⁸Sr are the s-process and r-process isotopes, respectively. In addition, ⁸⁷Sr includes radiogenic ⁸⁷Sr produced from the decay of ⁸⁷Rb (t_1/2 = 4.88 Ga). Strontium isotopic analyses were performed to monitor the mass not only at ⁸⁴Sr, ⁸⁶Sr, ⁸⁷Sr, and ⁸⁸Sr, but also at ⁸⁵Rb to evaluate the isobaric interference of ⁸⁷Rb for ⁸⁷Sr. All Sr isotopic data were referenced to ⁸⁶Sr. For the correction of instrumental mass fractionation, Sr isotopic data were normalized by ⁸⁸Sr/⁸⁶Sr = 8.375209 using the exponential law (Nier 1938). An ⁸⁸Sr⁺ ion beam of (0.16–16) × 10⁻¹¹ A from 30–2000 ng of Sr loaded on a filament was obtained for more than 5 hr from individual fractions.

The Ba fraction was loaded on the outer filament of a double Re filament assembly, without any activators. Barium isotopic analyses were performed to monitor the mass at ¹³⁰Ba, ¹³²Ba, ¹³⁴Ba, ¹³⁵Ba, ¹³⁶Ba, ¹³⁷Ba, and ¹³⁸Ba in addition to monitor ¹³⁹La and ¹⁴⁰Ce for evaluation of isobaric interferences of ¹³⁸La, ¹³⁶Ce, and ¹³⁸Ce in the Ba mass spectra. During any analyses in this study, detectable interferences from La and Ce were not found. All Ba isotopic data were referenced to ¹³⁶Ba. For the correction of instrumental mass fractionation, Ba isotopic data was normalized by ¹³⁴Ba/¹³⁶Ba = 0.307776 (Hidaka et al. 2003) using the exponential law. ¹³⁴Ba and ¹³⁶Ba are s-only process isotopes because r-process isotopes of ¹³⁴Xe and ¹³⁶Xe are shielded (Hidaka et al. 2003). The normalizing factor of ¹³⁴Ba/¹³⁶Ba = 0.307776 is useful to discuss additional nucleosynthetic components (Hidaka et al. 2003; Hidaka & Yoneda 2011, 2013). A ¹³⁸Ba⁺ ion beam of (1.0–14) × 10⁻¹¹ A from 60–3000 ng of Ba loaded on a filament was obtained for more than 5 hr from individual fractions. There are no shifts in isotopic compositions observed when different amounts of Sr and Ba materials were measured.

In this study, chemical reagents (Sr standard solution produced by NIST SRM-987; Ba standard solution produced by SPEX CertiPrep) were used as standard materials for Sr and Ba.

Rubidium, Sr, Cs, and Ba elemental abundances and Rb/Sr and Cs/Ba elemental ratios of L1–5 and WR fractions are listed in Table 1. The correlation diagram between Rb/Sr and Cs/Ba elemental ratios of L1–5 and WR of the TL sample based on the chemical data given in Table 1 is shown in Figure 1. In this figure, the reference data of the whole rocks of CI and CM chondrites and the TL meteorite are also plotted for comparison (CI from Anders & Grevesse 1989; Friedrich et al. 2002 CM from Friedrich et al. 2002; Yokoyama et al. 2015 TL from Brown et al. 2000; Friedrich et al. 2002; Yokoyama et al. 2015). The CI and CM chondrites and the TL meteorite are known to show evidence for aqueous alteration on their parent bodies in the early solar system. The Cs/Ba ratios of whole rock samples for the TL meteorite and the CI and CM chondrites are more variable than their Rb/Sr ratios (Cs/Ba, 0.011 ∼ 0.085; Rb/Sr, 0.14 ∼ 0.35), which suggests that the Cs/Ba ratio is more sensitive to aqueous activity on a meteorite parent body than the Rb/Sr ratio. This is likely due to the difference in mobility between Cs and Rb in solid and liquid phases. The mobility of the element in solid and liquid phases can be experimentally defined as partition coefficients (${K}_{D}$). It is reported that the ${K}_{D}$ value of Cs in solid and liquid phases shows a larger variation than that of Rb, and that it greatly depends on the type of solids (Sanchez et al. 2002; Sheppard et al. 2009). These data are often used to discuss the long-termed geochemical behaviors of radionuclides in geological media from the view point of radioactive waste disposal, and may also provide a hint to consider the redistribution behaviors of alkaline elements under aqueous activity in this study. Considering the large variations of Rb/Sr and Cs/Ba ratios in the L1–5 fractions collected from TL (Rb/Sr, 0.14 ∼ 1.3; Cs/Ba, 0.0011 ∼ 0.060), the sequential acid-leaching technique used in this study is useful for chronological purposes to collect several phases with the variable Cs/Ba elemental ratio from the whole rock. Our data are consistent with previous studies using the same leaching-procedures to collect a wide range of Cs/Ba fractions in samples (Hidaka et al. 2001, 2003; Hidaka & Yoneda 2011, 2013). The sequential acid-leaching procedure used in this study was partly modified from the previous method (Shima & Honda 1967). The specific phases dissolved in each leachate are speculated to be water-soluble minerals (L1 and L2), olivine (L3), phosphides (L4), and acid-resistant materials (L5) (Shima & Honda 1967; Hidaka et al. 2001). In particular, Cs/Ba ratios of L1, 2, 3, and 4 fractions are higher than that of WR. The results from the whole rock analyses show that the Cs/Ba ratio of whole rock of the TL meteorite (0.011–0.049) is the lowest among other CI (0.080–0.085) and CM (0.031–0.055) chondrites (Figure 1). In previous studies, 0.1 M CH₃COOH-CH₃COONH₄ leachate (L1) show high Cs abundance (Hidaka et al. 2001) and 2M HCl leachates (L3) also show high Cs abundance (Hidaka & Yoneda 2013). It would be reasonable to consider that the Cs/Ba ratios in TL, CI, and CM reflect the magnitude of aqueous alteration on the meteorite parent bodies. The Cs/Ba ratios from the TL leachates are further evidence for intensive aqueous alteration on the TL parent body.

Zoom In Zoom Out Reset image size

All of the Sr and Ba isotopic data for the leaching fraction (L1–5) and whole rock (WR) are listed in Tables 2 and 3, respectively. The results of repeated analyses of standard materials for Sr and Ba are also listed in the same tables. The data are expressed as relative deviations of the isotopic ratios between samples and individual standard materials in units defined as follows:

$$\begin{eqnarray*}\varepsilon {}^{{\rm{i}}}\mathrm{Sr} & = & \left\{\tfrac{{(}^{{\rm{i}}}\mathrm{Sr}{/}^{86}\mathrm{Sr})\mathrm{sample}}{{(}^{{\rm{i}}}\mathrm{Sr}{/}^{86}\mathrm{Sr})\mathrm{standard}}-1\right\}\times {10}^{4}\\ \varepsilon {}^{{\rm{i}}}\mathrm{Ba} & = & \left\{\tfrac{{(}^{{\rm{i}}}\mathrm{Ba}{/}^{136}\mathrm{Ba})\mathrm{sample}}{{(}^{{\rm{i}}}\mathrm{Ba}{/}^{136}\mathrm{Ba})\mathrm{standard}}-1\right\}\times {10}^{4}.\end{eqnarray*}$$

The Ba isotopic deviation patterns of L1–5 and WR fractions are shown in Figure 2. The WR pattern shows isotopic excess of ¹³⁵Ba and ¹³⁷Ba (ε¹³⁵Ba = 0.45 ± 0.06 and ε¹³⁷Ba = 0.19 ± 0.09). The Ba isotopic data for the CI and CM chondrites also showed isotopic excess of coupled ¹³⁵Ba and ¹³⁷Ba (0 < ε^135,137Ba < 1), caused by additional nucleosynthetic components of the r-process isotopes (Andreasen & Sharma 2007; Carlson et al. 2007) or deficit of nucleosynthetic components of the s-process isotope (Bermingham et al. 2016). The Ba isotopic pattern for the whole rock of the TL meteorite was similar to those of the CI and CM chondrites (Andreasen & Sharma 2007; Carlson et al. 2007; Bermingham et al. 2016), which were interpreted to reflect nucleosynthetic components of the s- or r-process.

Zoom In Zoom Out Reset image size

On the other hand, the Ba isotopic deviation patterns of L1–5 show variable and large deviations compared with that of WR. The patterns of L1–4 show a variable isotopic excesses of ¹³⁰Ba, ¹³²Ba, ¹³⁵Ba, ¹³⁷Ba, and ¹³⁸Ba, while that of L5 shows isotopic deficits of ¹³⁰Ba, ¹³²Ba, ¹³⁵Ba, ¹³⁷Ba, and ¹³⁸Ba.

Isotopic data for several phases collected from leaching experiments in primitive meteorites provide information about isotopic heterogeneity in the early solar systems that cannot be deduced from isotopic data of WR. In previous Ba isotopic studies using leaching experiments in CI and CM chondrites (Hidaka et al. 2001, 2003; Hidaka & Yoneda 2011, 2013), the Ba isotopic data of the acid residues showed large isotopic anomalies because of the enrichment of presolar SiC grains, which is a representative carrier of s-process isotopes (Ott & Begemann 1990; Prombo et al. 1993).

To evaluate the additional s- and r-process nucleosynthetic components from the isotopic anomalies of ¹³⁵Ba, ¹³⁷Ba, and ¹³⁸Ba, the Ba isotopic data in this study were the two-component mixing model between average of solar component and the s-process component given by Hidaka et al. (2003). Figures 3(a) and (b) show the correlation diagrams of ε¹³⁵Ba versus ε¹³⁷Ba, and ε¹³⁸Ba versus ε¹³⁷Ba for measured Ba isotopic data from the chemical separates of TL in this study and theoretical data calculated from the Ba isotopic data of s-process components. The Ba isotopic data of s-process components are used in the isotopic data of presolar SiC grains (Ott & Begemann 1990; Zinner et al. 1991; Prombo et al. 1993). The shaded zones in the figures indicate the theoretical Ba isotopic data using s-process components (Ott & Begemann 1990; Zinner et al. 1991; Prombo et al. 1993). For comparison, the line in Figure 3 indicates mixing the Ba isotopic data between the solar value and additional nucleosynthetic components based on the stellar model (¹³⁵Ba:¹³⁷Ba:¹³⁸Ba = 34.8:31.9:33.4; Bisterzo et al. 2015).

Zoom In Zoom Out Reset image size

The Ba isotopic deviation pattern of acid residue (L5) in the TL meteorite can be explained by a two-component mixing model between the average of the solar and additional nucleosynthetic s-process components derived from presolar SiC grains (Ott & Begemann 1990; Zinner et al. 1991; Prombo et al. 1993; Savina et al. 2003; A'vila et al. 2013). It is reasonable that the L5 data point is plotted on the shaded zones in Figures 3(a) and (b), considering that acid-resistant presolar SiC components have survived through the sequential leaching processes, and are enriched in the acid-residual fraction L5.

The Ba isotopic patterns of the L1–4 fractions show variable isotopic excesses (Figure 2). In our previous studies, Ba isotopic data except for ¹³⁰Ba and ¹³²Ba of leachates in carbonaceous chondrites could not be explained only from the addition of s- and r-process nucleosynthetic components (Hidaka et al. 2003; Hidaka & Yoneda 2011, 2013). All six data points in the ε¹³⁵Ba versus ε¹³⁷Ba diagram (Figure 3(a)) fall in the mixing zone, while three of the six data points from the L1-3 in ε¹³⁷Ba versus ε¹³⁸Ba diagram (Figure 3(b)) are plotted off the zone. Considering that the ¹³⁸Ba isotope has the least r-process component comprising it, it is unlikely to interpret that the addition of an r-process component caused the offset shown in Figure 3(b). As one of the possibilities, there is an addition of radiogenic ¹³⁸Ba decayed from ¹³⁸La (t_1/2 = 105 Ga) (Brennecka et al. 2013). Our estimates from the chemical data of La/Ba elemental ratios in the leachates provide the (¹³⁸Ba* /¹³⁸Ba) = 0.4 to 4.5 ppm, which concludes that the production of ¹³⁸Ba* is not resolvable by the Ba isotopic measurements in the current TIMS techniques. The Ba isotopic data of leachates reveal the existence of several additional nucleosynthetic components that originated not only from acid-resistant SiC but also from acid-soluble materials in the TL meteorite.

The ⁸⁷Rb–⁸⁷Sr evolution diagram for the TL meteorite is shown in Figure 4(a). The line in Figure 4(a) is a 4.568 Ga-old reference line corresponding to the formation age of CAIs from the CV3 meteorite (Bouvier & Wadhwa 2010). In the figure, the L2 data point is consistent with the reference line, while the other four data points of L1, L3, L4, L5, and WR fall off the line. The ⁸⁷Rb–⁸⁷Sr chronometer of TL might have been disturbed by late activities on the parent body, such as igneous, aqueous, metamorphic, or impact events. It is plausible that the ⁸⁷Rb–⁸⁷Sr chronometer had been disturbed by aqueous activity on the TL meteorite parent body, considering the Rb/Sr ratios of the TL L1–5 fractions shown in Figure 1. Less correlation between Rb/Sr and ⁸⁷Sr/⁸⁶Sr from the sequential leachates as shown in Figure 4(a) is also reported in other primitive chondrites (Qin et al. 2011; Yokoyama et al. 2015). It is reasonable to interpret that Rb and Sr are contained in different phases as the result of chemical redistribution by primitive aqueous alteration, and that they are differentially leached from these phases by the different acids.

Zoom In Zoom Out Reset image size

The ¹³⁵Cs–¹³⁵Ba isochron diagram for the TL meteorite is shown in Figure 4(b). Because all of the Ba isotopic compositions from the TL meteorite suffered from the addition or depletion of nucleosynthetic s-process components, it is impossible to detect the presence of radiogenic ¹³⁵Ba directly from the isotopic compositions. Assuming that the ¹³⁵Ba isotopic excesses of TL were derived from two origins, the nucleosynthetic s-process component and the radiogenic component, the amount of radiogenic ¹³⁵Ba, (ε¹³⁵Ba*), can be calculated from the subtraction of the additional nucleosynthetic components using the following equation:

$$\begin{eqnarray*}&&({\varepsilon }^{135}\mathrm{Ba}* )=({\varepsilon }^{135}\mathrm{Ba})-({\varepsilon }^{137}\mathrm{Ba})\times f,\end{eqnarray*}$$

where f is a factor showing the ¹³⁵Ba/¹³⁷Ba isotopic ratio of the additional nucleosynthetic s-process component (Ott & Begemann 1990; Zinner et al. 1991; Prombo et al. 1993; Savina et al. 2003; Bisterzo et al. 2015). As shown in Figure 3(a), the ¹³⁵Ba/¹³⁷Ba isotopic ratio of additional nucleosynthetic components is from 1.517 (${f}_{\min }$) to 1.908 (${f}_{\max }$), as estimated from the isotopic data from presolar SiC grains (${f}_{\min }$: Zinner et al. 1991 ${f}_{\max }$: Prombo et al. 1993). The ¹³⁵Cs–¹³⁵Ba isochron diagram consisting of the Cs/Ba elemental ratios and calculated ε¹³⁵Ba, (ε¹³⁵Ba*), of the TL leachates is shown in Figure 4(b). Since there is no positive correlation in the figure, chronological information cannot be provided from the Cs–Ba data set in this study. Although the data set collected in this study is limited, more data collection in the future may refine this approach.

Although there are several studies on ¹³⁵Cs in the early solar system (McCulloch & Wasserburg 1978a; Hidaka et al. 2001; Hidaka & Yoneda 2011, 2013; Bermingham et al. 2014, 2016; Brennecka & Kleine 2017), the estimated initial ¹³⁵Cs abundance ranges from 2.8 × 10⁻⁶ to 6.8 × 10⁻⁴ but is still disputed. The time interval between the formation of CAIs and the occurrence of primitive aqueous alteration was estimated to be 4.3–5.7 Ma from the Mn–Cr dating of carbonates in chondrites (Fujiya et al. 2012). Assuming the initial ¹³⁵Cs/¹³³Cs ratio ((Brennecka & Kleine 2017) for minimum value = 2.8 × 10⁻⁶ and (Hidaka & Yoneda 2013) for maximum = 6.8 × 10⁻⁴) and the formation interval of the carbonates, the ¹³⁵Ba isotopic excesses produced by radiogenic ¹³⁵Ba of individual TL leaching fractions are estimated as ε¹³⁵Ba = 0.000073–1.5. Considering the analytical precision for Ba isotopic measurements in our current techniques and ¹³⁵Ba isobaric interference from the s-process component, a higher Cs/Ba phase (¹³³Cs/¹³⁶Ba > 10), like an evaporate mineral having high Rb/Sr ratio (Zolensky et al. 1999), must be found in primitive solar materials for the detection of ¹³⁵Ba*.

The ⁸⁷Rb–⁸⁷Sr and ¹³⁵Cs–¹³⁵Ba decay systems on TL in this study do not provide any chronological information. Is there a possibility that any late events, like an impact metamorphism, disturb the TL chronometers? Considering that the shock stage of TL is typical for carbonaceous chondrite (Brown et al. 2000; Zolensky et al. 2002), it is unlikely that a late impact and the related metamorphism disturbed the TL chronometers. As noted at the end of Section 3.1, the data of Cs/Ba ratios help us to exclude the possibility of terrestrial weathering in our TL sample. A reasonable explanation for the absence of a correlation in the isochron diagrams is a reflection of the fact that Cs and Ba are fractionated via the selective dissolution of Cs from mineral structure due to the relatively higher mobility of Cs compared to Ba during fluid mineral interactions, and that the open system behavior of the ¹³⁵Cs–¹³⁵Ba system held in TL during the lifetime of ¹³⁵Cs. This is supported by the larger variability in the Cs/Ba ratios compared to Rb/Sr ratios shown in Figure 1.

Variable and detectable isotopic excesses of ⁸⁴Sr, ¹³⁰Ba, and ¹³²Ba are observed in L1–3 (see in Tables 2(b) and 3(b)). On the other hand, the L4 fraction shows significant excesses of ¹³⁰Ba and ¹³²Ba isotopes, but no anomaly of ⁸⁴Sr.

Yokoyama et al. (2015) found a much wider range of ⁸⁴Sr isotopic anomalies (μ⁸⁴Sr = −3427 to 60) in the TL fractions from their original stepwise-leaching method. In particular, very large ⁸⁴Sr isotopic deficits were observed in the last leachate (μ⁸⁴Sr = −3427). Considering the low Sr elemental abundances (<0.5% of the total Sr content) in the last leachate, the acid-resistant presolar SiC grains might have been enriched by the effective exclusion of the solar Sr component during the leaching processes. On the other hand, the last leachate, L5, in this study still has ∼18% of the total Sr content. Comparing the Sr elemental abundance in the last fractions between two different leaching methods, the small variation of ⁸⁴Sr of the L5 fraction in this study seems reasonable.

There are three possibilities for the isotopic excess on lighter isotopes of the elements, such as ⁸⁴Sr, ¹³⁰Ba, and ¹³²Ba. The first possibility is nucleosynthetic origin for p-process isotopes that are produced mainly by type II supernovae (SNe II). Finding anomalies in the pure p-process isotope is very rare, although heterogeneous distribution of p-process isotopic components in the early solar system has been reported (McCulloch & Wasserburg 1978b; Andreasen & Sharma 2006).

Next, the second possibility is the accumulation of spallogenic nuclides produced by cosmic-ray irradiation. Hidaka & Yoneda (2014) reported systematic p-process isotopic excess induced from solar cosmic-ray irradiation, which suggests that this should be discussed further. The Kapoeta meteorite is known as a solar-gas rich meteorite that possibly experienced early irradiation by solar cosmic rays. Systematic isotopic excesses of ⁸⁴Sr, ¹³⁰Ba, ¹³²Ba, ¹³⁶Ce, ¹³⁸Ce, and ¹⁴⁴Sm found in surficial parts of regolith particles of the Kapoeta meteorite suggest the spallogenic nuclides were produced by interaction with a solar cosmic ray from the early Sun. However, considering the cosmic-ray exposure age of TL, the duration of 5.5 Ma (Nakamura et al. 2003) may not be enough to accumulate spallogenic products to be detected by isotopic analysis.

Lastly, the third possibility is apparent anomalies accompanied by enrichment of s-process isotopes. Negative or positive anomalies in pairs of ¹³⁰Ba and ¹³²Ba isotopes are correlated with the excess or deficiency of s-process isotopes, which are also found in Ba isotopes of presolar grains and acid-leachates of carbonaceous chondrites (Ott & Begemann 1990; Qin et al. 2011). Figure 5 shows a correlation diagram between ε¹³⁰Ba and ε¹³²Ba to understand the origin of the ¹³⁰Ba and ¹³²B isotopic excesses found in the TL leachates. The data points of the TL leachates are plotted on the mixing line for solar and presolar components, but off the mixing line for solar and spallogenic components. The results reveal that the isotopic anomalies of ¹³⁰Ba and ¹³²Ba accompany the s-process components. Considering that three of the four Sr isotopes, ⁸⁶Sr, ⁸⁷Sr, and ⁸⁸Sr, are also synthesized by the s-process, the isotopic anomalies of ⁸⁴Sr found in the L1–3 fractions (see Table 2(b)) are also interpreted as the isotopic shift accompanying the depletion of s-process components. It is feasible to consider that the isotopic excesses of ⁸⁴Sr, ¹³⁰Ba, and ¹³²Ba in this study are apparent anomalies accompanied by enrichment of s-process isotopes, because the normalizing isotopic ratios used in this study, ⁸⁸Sr/⁸⁶Sr and ¹³⁴Ba/¹³⁶Ba, to make corrections for the instrumental mass fractionation include s-process isotopes.

Zoom In Zoom Out Reset image size