Quartz deformation fabrics reflect stress and strain conditions in mylonites, and their interpretation has become a mainstay of kinematic and structural analysis. Quantification of grain size and shape and interpretation of textures reflecting deformation mechanisms can provide estimates of flow stress, strain rate, kinematic vorticity, and deformation temperatures. Empirical calibration and determination of quartz flow laws is based on laboratory experiments of pure samples; however, pure quartzite mylonites are relatively uncommon. In particular, phyllosilicates may localize and partition strain that can inhibit or enhance different deformation mechanisms. Experimental results demonstrate that even minor phyllosilicate content (<15 vol%) can dramatically alter the strain behavior of quartz; however, few field studies have demonstrated these effects in a natural setting.To investigate the role of phyllosilicates on quartz strain fabrics, we quantify phyllosilicate content and distribution in quartzite mylonites from the Miocene Raft River detachment shear zone (NW Utah, USA). We use microstructural analysis and electron backscatter diffraction to quantify quartz deformation fabrics and muscovite spatial distribution, and X-ray computed tomography to quantify muscovite content in samples with varying amounts of muscovite collected across the detachment shear zone. Phyllosilicate content has a direct control on quartz deformation mechanisms, and application of piezometers and flow laws based on quartz deformation fabrics yield strain rates and flow stresses that vary by up to two orders of magnitude across our samples. These findings have important implications for the application of flow laws in quartzite mylonites and strain localization mechanisms in mid-crustal shear zones.Quartz mylonites commonly contain secondary phases, such as phyllosilicates (e.g., Taylor and McLennan, 1985), which may play an important role in the dynamic recrystallization of quartz and other minerals (e.g., Song and Ree, 2007; Herwegh et al., 2011; Hunter et al., 2016; Wehrens et al., 2017). Depending on the quantity and arrangement of phyllosilicates, quartz may change deformation processes (dislocation to diffusion creep), thereby affecting the rheology of the crust (e.g., Tullis and Yund, 1991). During the development of S-C fabrics associated with mylonitization, phyllosilicates form interconnected networks, causing a change in rheological properties in which the weak phase—phyllosilicates—accommodates strain, resulting in a weaker aggregate (e.g., Holyoke and Tullis, 2006). In addition, phyllosilicates inhibit dynamic recrystallization by grain boundary migration, preventing grain growth and encouraging other mechanisms such as dissolution creep, dislocation glide in phyllosilicates, and grain boundary sliding in quartz (e.g., Olgaard, 1990; Song and Ree, 2007; Hunter et al., 2016; Wehrens et al., 2017). Inhibition of grain boundary migration leads to reduced dynamic grain growth, resulting in diminished grain size, which may cause the dominant deformation mechanism of the main phase to switch from grain-size-insensitive dislocation creep to grain-size-sensitive diffusion creep (e.g., Etheridge and Wilkie, 1979; Olgaard, 1990; De Bresser et al., 1998, 2001; Fukuda et al., 2018; Richter et al., 2018). Because of this critical dependency of rheology on grain size, it is important to understand how weaker secondary phases such as phyllosilicates can affect grain size and thus, in turn, rheology.The role of phyllosilicate in strain localization has been observed in experimental work (Tullis and Wenk, 1994) and proposed theoretically (Johnson et al., 2004; Gerbi et al. 2010; Montési, 2013; Rast and Ruh, 2021), but only a few studies have explored this in naturally deformed rocks (e.g., Kronenberg, 1981; Song and Ree, 2007; Herwegh et al., 2011; Wehrens et al., 2017). Results from lab experiments conducted on quartz aggregates containing various amounts of muscovite indicate that when quartz deforms by dislocation creep, addition of as little as 10% muscovite causes a mechanical transition from a strong phase that forms a load-bearing framework to an aggregate with an interconnected weak phase, allowing strain partitioning into the weak muscovite network, thereby reducing the composite strength of the aggregate (e.g., Tokle et al., 2023).This paper investigates the role of muscovite in quartz deformation. We focus on the Miocene Raft River detachment shear zone (RRDSZ) associated with the Raft River metamorphic core complex, NW Utah, USA (e.g., Compton, 1980). Quartz microstructural and electron backscatter diffraction (EBSD; e.g., Prior et al., 1999) analyses are combined with X-ray computed tomography (XRCT; e.g., Mees et al., 2003) to quantify muscovite content and its impact on quartz recrystallization mechanisms. Our results suggest that quartz recrystallized grain size (RxGS) varies significantly not only between different samples but also within each sample depending on the presence or absence of muscovite. These findings have important implications for the application of RxGS piezometers and dislocation creep flow law used to constrain the mechanical evolution of detachment shear zones.The Raft River Mountains form the eastern limb of the larger Albion–Raft River–Grouse Creek metamorphic core complex in NW Utah (Fig. S1 in the Supplemental Material1; e.g., Compton, 1980). Exhumation of the RRDSZ initiated ca. 25–20 Ma and shows muscovite 40Ar/39Ar ages as young as 15 Ma (Wells et al., 2000). The RRDSZ is best exposed along Clear Creek Canyon, incised along the transport direction, providing vertical exposures of continuous sections of mylonitic rocks (Fig. S1). The RRDSZ is localized in the ~100-m-thick Proterozoic Elba Quartzite (Compton, 1980). The mylonitic fabric is constant throughout the RRDSZ, characterized by sub-horizontal foliation, defined by flattened and elongated muscovite grains, and eastward stretching lineation (Compton, 1980). Quartz microstructures and stable isotope geothermometry suggest that Miocene mylonitization occurred under greenschist-facies conditions (345–485 °C; Gottardi et al., 2011).Samples were collected along a vertical transect across the RRDSZ at Clear Creek Canyon, which contains a muscovite-rich horizon located ~15–22 m above the basement (Fig. 1A). The base of this horizon is characterized by a transition from cliff-forming Elba Quartzite to a zone where muscovite makes up 80%–100% of the rock. Microstructural analysis was conducted on thin sections oriented perpendicular to foliation and parallel to lineation. XRCT and EBSD analyses were conducted on ~5 × 5 × 5 mm cubes cut from the thin section billets with the same orientation. Muscovite content was measured by XRCT at the University of Texas at Austin (USA) High Resolution XRCT Facility (see File S1 in the Supplemental Material). Quartz crystallographic preferred orientation was investigated by EBSD at Appalachian State University (Boone, North Carolina, USA) on the cubes imaged by XRCT (see File S1).For each sample, the average RxGS (root mean square of the mean diameter) of the samples was estimated for (1) the entire sample surface, (2) a smaller muscovite-free area, and (3) a muscovite-rich area on each sample surface (Fig. 1; Table S1 [see footnote 1]). Differential stress was estimated based on the quartz RxGS piezometer of Cross et al. (2017). Strain rate was calculated using the quartzite dislocation creep flow law of Hirth et al. (2001), with a stress exponent n of 4, an activation energy Q of 135 kJ/mol, and a temperature of 400 °C (Gottardi et al., 2011).The quartzite is characterized by two quartz grain populations: coarse elongate (>500 µm long) “relict” grains and finer recrystallized grains (20–100 µm) (Fig. 2). The relict grains define the macroscopic fabric and exhibit undulose extinction and deformation lamellae (Fig. 2). Dynamic recrystallization textures reflect dominantly sub-grain rotation as described by Hirth and Tullis (1992) and consistent with previous observations of the RRDSZ (e.g., Gottardi and Teyssier, 2013). Petrographic observations and XRCT results demonstrate that the amount of muscovite varies among the samples, ranging from 4–8 vol% to 14–18 vol% (Fig. 3).Quartz c-axes show Type I cross girdles indicative of basal, prism, and rhomb slip (Fig. 1). The structurally lowest three samples display a narrow girdle with a strong rhomb maxima and a slight dextral asymmetry indicative of top-to-the-east sense of shear.The average RxGSs across all samples range from 20.6 to 60.5 µm. We observe a negative correlation between grain size and muscovite content where the smallest RxGS is associated with the samples containing the most muscovite and vice versa (Fig. 3). Similarly, grain size analysis of muscovite-free and muscovite-rich areas reveals that the RxGS is systematically smaller in the muscovite-rich area than in the muscovite-free area or than the average grain size of all samples except for RR09-10. The grain size spread between muscovite-free and muscovite-rich areas ranges from 0.9 to 22.4 µm across our sample suite (2%–62% variation with respect to bulk grain size) (Fig. 3; Table S1). Based on our RxGS, we use the Cross et al. (2017) quartz RxGS piezometer to estimate differential stress, which ranges from 32.3 to 69.4 MPa, and estimate the strain rate to be between 1.1 × 10−14 to 2.4 × 10−13 s−1 using the Hirth et al. (2001) quartzite dislocation creep flow law (Fig. 3; Table S1).Our findings are consistent with lab results that suggest even minor phyllosilicate content (<10%) exerts significant control on quartz deformation (e.g., Tokle et al., 2023). Samples sharing similar microstructures and dislocation creep deformation processes collected within a 50-m vertical transect show an inverse relationship between RxGS and muscovite content (Fig. 3), ranging from 53.3 µm to 23.1 µm correlating to ~6 vol% to ~16 vol% muscovite, respectively. Additionally, RxGS varies within each sample, to as large as 22.4 µm (Table S1; Fig. 3). The observed correlation between muscovite content and quartz deformation textures has substantial implications for the application of piezometry and strain rate estimates, which are ultimately based on quartz RxGS. The observed RxGS variability translates to a ~32–70 MPa range in flow stress and a range in strain rates of more than one order magnitude (~1.1 × 10−14 to 2.4 × 10−13 s−1) across all samples (Table S1; Fig. 3). Intra-sample grain size variation correlated to muscovite content results in a difference of flow stress of ~10 MPa and more than an order of magnitude in strain rate.Microscale bands of contrasting grain sizes have been interpreted elsewhere to record progressive superposition of deformation fabrics and used to track the evolution of local deformation or metamorphic conditions (e.g., Behr and Platt, 2011). Textural variation within individual samples correlates to a strain rate difference of more than one order of magnitude (Fig. 3), which could similarly be interpreted as being due to superposed fabrics recording an evolving detachment shear zone. However, this interpretation requires extreme deformation temperature gradients and strain rate and/or flow stress localization, or reactivation of the detachment shear zone.If we assume that differences in RxGS are due to temperature, then our observed RxGS range would translate to a difference in temperature of 75–100 °C across our samples. Detachment shear zones can show apparent large temperature ranges because high-strain zones can telescope rocks that have experienced broadly different thermal conditions (e.g., Law et al., 2011). However, there is no evidence of such structures that would juxtapose rocks with different thermal histories in the RRDSZ. Quartz microstructures and dominant deformation mechanism are consistent across all our samples and the broader RRDSZ (Fig. 1; Gottardi and Teyssier, 2013), suggesting that there is no microstructural evidence for such temperature variations. Additionally, geochronological studies demonstrate that deformation and exhumation of the RRDSZ occurred contemporaneously and rapidly and was not temporally partitioned (~5 m.y.; Wells et al., 2000; Gottardi et al., 2015). Finally, there is no field evidence for structural telescoping; the RRDSZ does not exhibit any subsidiary detachment surfaces (e.g., Compton, 1980; Wells, 1997; Wells et al., 2000). Therefore, we assume that all studied samples experienced the same temperature conditions at the same time during exhumation, and that changes in RxGS are not related to temperature variations.The observed RxGS differences could also be explained by differences in strain rate. The strain rate of the RRDSZ has previously been estimated to range between 10−14 to 10−13 s−1 based on finite strain and thermochronological data (Wells et al., 2000; Gottardi and Teyssier, 2013). These rates are systematically at least one order of magnitude slower than strain rates determined from RxGS analysis, except those for the purest quartzite samples (Fig. 4). Besides demonstrating changes in muscovite content within the mylonites, the RRDSZ is devoid of any evidence of high-strain zones and rather accommodated strain homogenously (e.g., Sullivan, 2008).Variation in RxGS could be related to variations in stress at constant temperature. It is generally accepted that detachment shear zones evolve under constant stress (Behr and Platt, 2011). Subsidiary shear zones and rheological contrast between units may concentrate stress and deformation on various scales. However, as mentioned above, the RRDSZ lacks any subsidiary structures that may have localized stress locally and is localized entirely in the Elba Quartzite. The only lithological variation results from variation in muscovite content.Because exhumation of the Raft River metamorphic core complex occurred rapidly and the RRDSZ is devoid of macro- and microstructural evidence of polyphase deformation or brittle overprint of the ductile microstructures, we thus (1) interpret the observed range of fabrics to reflect one event rather than the sum of several overprinting fabrics, and (2) attribute textural variation to phyllosilicate control over deformation and strain localization processes. This interpretation is consistent with experimental results (Tullis and Wenk, 1994; Tokle et al., 2023) and previous qualitative observations of texture variation in polyphase mylonites (e.g., Song and Ree, 2007; Herwegh et al., 2011; Hunter et al., 2016; Wehrens et al., 2017).Paleopiezometric and strain rate calculations based on microstructural analysis are commonly applied to tectonic inferences of local and regional structures (e.g., Hirth et al., 2001; Gueydan et al., 2005; Behr and Platt, 2011; Gottardi and Teyssier, 2013; Lusk and Platt, 2020). Our results support the conclusions of laboratory studies (e.g., Tokle et al., 2023) that demonstrate that failure to account for phyllosilicate content may lead to erroneous results.In this project, we have investigated the effect of muscovite on quartz deformation in the RRDSZ. Our results demonstrate a strong inverse relationship between quartz RxGS and muscovite content both between samples and within sub-millimeter-scale subregions within individual samples. The range in grain sizes between our proximal samples translates into variation in calculated strain rate of greater than one order of magnitude (~1.1 × 10−14 to 2.4 × 10−13 s−1) and more than doubling of the calculated flow stress from most- to least-pure quartzite (32–70 MPa). These estimates are incompatible with previous field and thermochronology integrated strain rate estimates of the RRDSZ. Elsewhere, similar textural observations have been interpreted as evidence for polyphase deformation, reactivation, and superposition. We offer a simpler explanation that relatively small amounts of mica exert a disproportionate influence on quartz deformation fabrics, which has important implications for the application of flow laws in quartzite mylonites and the interpretation of strain localization mechanisms.We thank the University of Texas High-Resolution X-ray Computed Tomography Facility for data acquisition and Hanna Romy for help with data processing. This research was funded through National Science Foundation grant EAR-Tectonics 1849812 (Gottardi and Casale). We thank editor Rob Strachan and reviewers John Platt, Andreas Kronenberg, Ryan Thigpen, and Rick Law for their reviews and constructive advice.

中文翻译：

---

一点云母就能大有帮助：页硅酸盐对自然变形岩石中石英变形结构的影响


石英变形组构反映了糜棱岩中的应力和应变条件，它们的解释已成为运动学和结构分析的支柱。晶粒尺寸和形状的量化以及反映变形机制的纹理的解释可以提供流动应力、应变率、运动涡度和变形温度的估计。石英流动定律的经验校准和确定是基于纯样品的实验室实验；然而，纯石英岩糜棱岩相对不常见。特别地，页硅酸盐可以局部化和分配应变，从而抑制或增强不同的变形机制。实验结果表明，即使少量的页硅酸盐含量（<15 vol%）也能显着改变石英的应变行为；然而，很少有实地研究在自然环境中证明了这些影响。 为了研究页硅酸盐对石英应变织物的作用，我们量化了来自中新世筏河脱离剪切带（美国犹他州西北部）的石英岩糜棱岩中的页硅酸盐含量和分布。我们使用微观结构分析和电子背散射衍射来量化石英变形结构和白云母空间分布，并使用 X 射线计算机断层扫描来量化在分离剪切带收集的具有不同数量的白云母的样品中的白云母含量。页硅酸盐含量可以直接控制石英变形机制，以及基于石英变形织物的压力计和流动定律的应用，屈服应变率和流动应力在我们的样品中变化高达两个数量级。 这些发现对于石英岩糜棱岩中流动定律的应用和中地壳剪切带的应变定位机制具有重要意义。石英糜棱岩通常含有次生相，例如层状硅酸盐（例如，Taylor 和 McLennan，1985），这可能发挥着重要的作用。在石英和其他矿物的动态重结晶中的作用（例如，Song 和 Ree，2007 年；Herwegh 等人，2011 年；Hunter 等人，2016 年；Wehrens 等人，2017 年）。根据页硅酸盐的数量和排列，石英可能会改变变形过程（位错到扩散蠕变），从而影响地壳的流变性（例如，Tullis 和 Yund，1991）。在与糜棱岩化相关的 S-C 织物的开发过程中，页硅酸盐形成互连网络，导致流变特性发生变化，其中弱相（页硅酸盐）适应应变，导致聚集体较弱（例如，Holyoke 和 Tullis，2006）。此外，页硅酸盐通过晶界迁移抑制动态再结晶，防止晶粒生长并促进其他机制，例如页硅酸盐中的溶解蠕变、位错滑移和石英中的晶界滑动（例如，Olgaard，1990；Song 和 Ree，2007；Hunter 等）等人，2016；Wehrens 等人，2017）。晶界迁移的抑制导致动态晶粒生长减少，导致晶粒尺寸减小，这可能导致主相的主要变形机制从晶粒尺寸不敏感的位错蠕变转变为晶粒尺寸敏感的扩散蠕变（例如， Etheridge 和 Wilkie，1979；Olgaard，1990；De Bresser 等，1998，2001；Fukuda 等，2018； 由于流变学对晶粒尺寸的这种关键依赖性，了解页硅酸盐等较弱的次生相如何影响晶粒尺寸并进而影响流变学非常重要。在实验工作中已经观察到了页硅酸盐在应变定位中的作用（Tullis）和 Wenk，1994）并在理论上提出（Johnson 等，2004；Gerbi 等，2010；Montési，2013；Rast 和 Ruh，2021），但只有少数研究在自然变形岩石中对此进行了探索（例如，Kronenberg， 1981；Song 和 Ree，2007；Wehrens 等，2017）。对含有不同量白云母的石英骨料进行的实验室实验结果表明，当石英因位错蠕变而变形时，添加少至 10% 的白云母会导致从形成承重框架的强相到具有承载能力的骨料的机械转变。互连的弱相，允许应变分配到弱白云母网络中，从而降低骨料的复合强度（例如，Tokle 等人，2023）。本文研究了白云母在石英变形中的作用。我们重点关注与美国犹他州西北部拉夫特河变质核复合体相关的中新世拉夫特河脱离剪切带（RRDSZ）（例如，Compton，1980）。石英微观结构和电子背散射衍射（EBSD；例如，Prior 等人，1999）分析与 X 射线计算机断层扫描（XRCT；例如，Mees 等人，2003）相结合，以量化白云母含量及其对石英再结晶机制的影响。我们的结果表明，石英重结晶晶粒尺寸 (RxGS) 不仅在不同样品之间存在显着差异，而且在每个样品内部也存在显着差异，具体取决于白云母的存在与否。 这些发现对于 RxGS 压力计和位错蠕变流定律的应用具有重要意义，这些定律用于约束滑脱剪切带的力学演化。拉夫特河山脉形成了西北地区较大的阿尔比恩-拉夫特河-松鸡溪变质核复合体的东翼犹他州（补充材料1中的图S1；例如，Compton，1980）。 RRDSZ 的挖掘工作大约于 1997 年开始。 25–20 Ma，显示白云母 40Ar/39Ar 年龄早至 15 Ma（Wells 等，2000）。 RRDSZ 最好沿着 Clear Creek Canyon 暴露，沿着运输方向切割，提供糜棱岩连续部分的垂直暴露（图 S1）。 RRDSZ 位于约 100 米厚的元古代厄尔巴石英岩中（Compton，1980）。糜棱岩组构在整个 RRDSZ 中是恒定的，其特征是亚水平叶理，由扁平和拉长的白云母颗粒和向东伸展的线理所定义（Compton，1980）。石英微观结构和稳定同位素地温测量表明，中新世糜棱岩化发生在绿片岩相条件下（345-485 °C；Gottardi 等人，2011）。样品是沿着 Clear Creek 峡谷 RRDSZ 的垂直横断面采集的，其中含有白云母-丰富的地平线位于基底上方~15-22 m（图1A）。该层位底部的特点是从形成悬崖的厄尔巴石英岩过渡到白云母占岩石 80%–100% 的区域。对垂直于叶理和平行于线理的薄片进行微观结构分析。 XRCT 和 EBSD 分析是对从具有相同方向的薄截面坯料切下的约 5 × 5 × 5 mm 立方体进行的。 白云母含量是通过德克萨斯大学奥斯汀分校（美国）高分辨率 XRCT 设施的 XRCT 测量的（参见补充材料中的文件 S1）。阿巴拉契亚州立大学（美国北卡罗来纳州布恩）的 EBSD 在 XRCT 成像的立方体上研究了石英晶体择优取向（参见文件 S1）。对于每个样品，其平均 RxGS（平均直径的均方根）对样品进行了估计：(1) 整个样品表面，(2) 较小的无白云母区域，以及 (3) 每个样品表面上的白云母丰富区域（图 1；表 S1 [参见脚注 1]）。差异应力是根据 Cross 等人的石英 RxGS 压力计估算的。 （2017）。使用 Hirth 等人的石英岩位错蠕变流动定律计算应变率。 (2001)，应力指数 n 为 4，活化能 Q 为 135 kJ/mol，温度为 400 °C (Gottardi 等，2011)。石英岩具有两种石英颗粒群的特征：粗细长（>500 µm 长）“残余”晶粒和更细的再结晶晶粒 (20–100 µm)（图 2）。残余颗粒定义了宏观结构并表现出波状消光和变形片层（图2）。动态再结晶织构主要反映了 Hirth 和 Tullis (1992) 所描述的亚晶粒旋转，并且与之前对 RRDSZ 的观察结果一致（例如，Gottardi 和 Teyssier，2013）。岩相观察和 XRCT 结果表明，样品中白云母的含量各不相同，范围从 4-8 vol% 到 14-18 vol%（图 3）。石英 c 轴显示 I 型十字带，指示基底、棱柱、和菱形滑移（图1）。 结构最低的三个样品显示出狭窄的腰带，具有较强的菱形最大值和轻微的右旋不对称性，表明从上到东的剪切感。所有样品的平均 RxGS 范围为 20.6 至 60.5 µm。我们观察到晶粒尺寸和白云母含量之间存在负相关关系，其中最小的 RxGS 与含有最多白云母的样品相关，反之亦然（图 3）。类似地，无白云母和富含白云母区域的晶粒尺寸分析表明，富含白云母区域的 RxGS 系统地小于无白云母区域或除 RR09-10 之外的所有样品的平均晶粒尺寸。在我们的样品组中，无白云母区域和富含白云母区域之间的晶粒尺寸分布范围为 0.9 至 22.4 µm（相对于整体晶粒尺寸的变化为 2%–62%）（图 3；表 S1）。基于我们的 RxGS，我们使用 Cross 等人。 (2017) 石英 RxGS 压力计估计差异应力，范围从 32.3 到 69.4 MPa，并使用 Hirth 等人估计应变率在 1.1 × 10−14 到 2.4 × 10−13 s−1 之间。 (2001) 石英岩位错蠕变流动定律（图 3；表 S1）。我们的研究结果与实验室结果一致，表明即使少量页硅酸盐含量（<10%）也能对石英变形产生显着控制（例如，Tokle 等人，2023） ）。在 50 米垂直横断面内收集的具有相似微观结构和位错蠕变过程的样品显示，RxGS 和白云母含量之间存在反比关系（图 3），范围从 53.3 µm 到 23.1 µm，与 ~6 vol% 到 ~16 vol% 相关分别是白云母。此外，每个样品中的 RxGS 有所不同，最大可达 22.4 µm（表 S1；图 3）。 观察到的白云母含量和石英变形结构之间的相关性对于测压法和应变率估计的应用具有重大意义，这些估计最终基于石英 RxGS。观察到的 RxGS 变异性转化为所有样品中流变应力的 ~32–70 MPa 范围和超过一个数量级的应变率范围（~1.1 × 10−14 至 2.4 × 10−13 s−1）（表S1；图3)。与白云母含量相关的样品内晶粒尺寸变化导致流变应力差异约为 10 MPa，应变率超过一个数量级。对比晶粒尺寸的微尺度带已在其他地方解释，以记录变形织物的渐进叠加和用于跟踪局部变形或变质条件的演变（例如，Behr 和 Platt，2011）。各个样品内的纹理变化与超过一个数量级的应变率差异相关（图3），这可以类似地解释为由于记录了不断变化的分离剪切区的叠加织物所致。然而，这种解释需要极端的变形温度梯度和应变率和/或流动应力局部化，或分离剪切带的重新激活。如果我们假设 RxGS 的差异是由温度引起的，那么我们观察到的 RxGS 范围将转化为我们的样品温度为 75–100 °C。分离剪切带可以显示出明显的大温度范围，因为高应变区域可以伸缩经历了截然不同的热条件的岩石（例如，Law et al., 2011）。然而，没有证据表明此类结构会将具有不同热历史的岩石并置在 RRDSZ 中。 石英微观结构和主要变形机制在我们所有的样品和更广泛的 RRDSZ 中都是一致的（图 1；Gottardi 和 Teyssier，2013），这表明没有微观结构证据表明这种温度变化。此外，地质年代学研究表明，RRDSZ 的变形和折返是同时快速发生的，并且没有时间上的划分（~5 m.y.；Wells 等，2000；Gottardi 等，2015）。最后，没有结构伸缩的现场证据； RRDSZ 不表现出任何辅助脱离面（例如，Compton，1980；Wells，1997；Wells 等人，2000）。因此，我们假设所有研究的样品在折返过程中同时经历相同的温度条件，并且 RxGS 的变化与温度变化无关。观察到的 RxGS 差异也可以通过应变率的差异来解释。之前根据有限应变和热年代学数据估计 RRDSZ 的应变率范围在 10−14 至 10−13 s−1 之间（Wells 等人，2000 年；Gottardi 和 Teyssier，2013 年）。这些速率总体上比 RxGS 分析确定的应变速率慢至少一个数量级，最纯净的石英岩样品除外（图 4）。除了展示糜棱岩内白云母含量的变化外，RRDSZ 没有任何高应变区的证据，而是均匀地调节应变（例如，Sullivan，2008）。RxGS 的变化可能与恒温下应力的变化有关。人们普遍认为，脱离剪切带在恒定应力下演化（Behr 和 Platt，2011）。 单元之间的辅助剪切带和流变对比可能会在不同尺度上集中应力和变形。然而，如上所述，RRDSZ 缺乏任何可能局部产生局部应力的附属结构，并且完全位于厄尔巴岛石英岩中。唯一的岩性变化来自于白云母含量的变化。由于拉夫特河变质核杂岩的折返发生得很快，并且 RRDSZ 缺乏多相变形或韧性微观结构的脆性叠印的宏观和微观结构证据，因此我们 (1) 解释观察到的织物范围反映了一个事件而不是几种套印织物的总和，并且（2）将纹理变化归因于页硅酸盐对变形和应变局部化过程的控制。这种解释与实验结果（Tullis 和 Wenk，1994；Tokle 等，2023）以及之前对多相糜棱岩纹理变化的定性观察（例如，Song 和 Ree，2007；Herwegh 等，2011；Hunter 等）一致。 ., 2016; Wehrens et al., 2017). 基于微观结构分析的古地表测量和应变率计算通常应用于局部和区域结构的构造推断（例如，Hirth et al., 2001; Gueydan et al., 2005; Behr普拉特，2011；戈塔迪和泰西尔，2013；拉斯克和普拉特，2020）。我们的结果支持实验室研究的结论（例如，Tokle 等人，2023），这些研究表明，不考虑页硅酸盐含量可能会导致错误的结果。在这个项目中，我们研究了白云母对 RRDSZ 中石英变形的影响。 我们的结果表明，无论是在样品之间还是在单个样品的亚毫米级子区域内，石英 RxGS 和白云母含量之间都存在很强的反比关系。我们的近端样品之间的晶粒尺寸范围转化为大于一个数量级的计算应变率变化（~1.1 × 10−14 至 2.4 × 10−13 s−1），并且计算的流动应力增加了一倍以上从最高到最低纯度的石英岩 (32–70 MPa)。这些估计与 RRDSZ 先前的现场和热年代学综合应变率估计不相容。在其他地方，类似的结构观察结果被解释为多相变形、重新激活和叠加的证据。我们提供了一个更简单的解释，即相对少量的云母对石英变形织物产生了不成比例的影响，这对于石英岩糜棱岩中流动定律的应用以及应变局部化机制的解释具有重要意义。我们感谢德克萨斯大学高分辨率X 射线计算机断层扫描设备用于数据采集，Hanna Romy 提供数据处理帮助。这项研究由国家科学基金会 EAR-Tectonics 1849812（Gottardi 和 Casale）资助。我们感谢编辑 Rob Strachan 和审稿人 John Platt、Andreas Kronenberg、Ryan Thigpen 和 Rick Law 的审阅和建设性建议。