确定流体缺失的固相线温度在活塞气缸实验的一种新方法毕业论文外文翻译.doc

《确定流体缺失的固相线温度在活塞气缸实验的一种新方法毕业论文外文翻译.doc》由会员分享，可在线阅读，更多相关《确定流体缺失的固相线温度在活塞气缸实验的一种新方法毕业论文外文翻译.doc（11页珍藏版）》请在三一办公上搜索。

1、外文译文确定流体缺失的固相线温度在活塞气缸实验的一种新方法摘 要 我们描述了一种新的活塞气缸用于确定流体缺失的地质材料的固相线的方法。样品加热增量不到一小时的期间为320千巴的压力。测量油压融化的影响虽小，但检测和压力 - 时间数据的解释产生一个准确的斜线。该技术成功地再现了氯化锂的固相线。为Ollo de Sapo泥质片麻岩的结果是比以前使用的更传统的技术获得。简 介在地球的地幔和地壳岩浆源识别促进实验岩石学的结果。 PT深部原岩的固相线的位置，以了解岩浆代的过程是至关重要的。潜在的地壳源材料，如角闪岩和泥质岩，固相线已经确定实验（例如，1977年威利辛格和1996年约翰内斯皮托1976;杜

2、丝帕蒂诺和比尔德1995年，1996年;帕蒂诺杜丝和哈里斯1998; Vielzeuf和1992年克莱门斯1994年Vielzeuf及Montel）。这些结果表明，在矿物组合的变化，或涉及在熔融反应（例如，云母和长石）的矿物的固溶体的组合物，可以产生强大的固相线温度的影响。因此，为了正确地鉴定浆生产过程中，在特定环境中，它是必不可少的，以确定特定的地壳原岩的固相线曲线。然而，由于工作繁重，需要许多费时的实验熔体的存在或不存在是由观察的扫描型电子显微镜（SEM）图像的运行产品。下面描述的技术可用于确定通过直接观察的活塞 - 气缸装置中的油压的影响熔融的固相线曲线在PT空间的位置。这些影响是非常小

3、的，但可检测到的增量加热所产生的压力 - 时间曲线（Pt）的检查。基本原理 活塞缸实验，我们已经注意到，在5-10的温度变化有一个可衡量的影响压力。这是一个实验期间观察到的第一分钟时，温度上升，配有温度控制器。测得的压力对时间作图时获得相应的斜坡。因此，在加热过程中，在测得的压力变化影响的压力 - 时间曲线的斜率的压实可以被检测到的胶囊型容器和/或样品日本。在胶囊内的熔化的样品将产生一个可能被检测到的压力，如果非常精确地测量和数据被连续拍摄的压实效果。应用这种方法，我们已进行了多次的熔融实验来确定的固相线温度诱导加热样品在活塞缸中的压力 - 时间曲线变化。实验程序和材料实验进行装载，用12.7

4、毫米的固体介质活塞圆筒装置（0.5英寸）直径的NaCl石墨电池组件。测量和pt100-pt87rh13热电偶连接到欧陆808控制器控制温度。电子德鲁克PTX 1400压力变送器测量机油压力，连接到欧姆龙e5ck控制器。采用天然岩石样品和合成材料。20毫克的细碎粉末（1050毫米）被密封在一个黄金胶囊，随后被嵌入在压力下的NaCl插头在室温下。由于固相线温度的确定主要依靠检测样品压实，该技术只适用于缺乏流体实验。含水样品不合适因为水会在发生熔化填充多孔骨料。体积的影响减少由熔化的样品引起的压实可以放大以多孔骨料作为熔体的陷阱。我们使用的石英和钻石聚集在这两种情况下，取得了良好的效果。与过量的石英

5、岩样品（例如，泥质岩和片麻岩），一个石英陷阱引入的熔融温度没有影响。二氧化硅可怜的组合物（例如，斜长角闪岩），一个钻石陷阱使用更合适。由于样品的压实压力测量的变化通常是非常小的（小于1条油压力），一般只持续很短的时间（小于30秒）。检测由于融化在胶囊内的任何变化，我们进行了多次实验，应用热，观察系统在很短的时间间隔响应小的增量。大会通过增加功率手动预定百分比的总功率每30秒。我们选择这个时间间隔是因为它足以确保热平衡在装配之前，下一个输入能量达到加热。用金刚石或石英陷阱的实验中，压实可以慢因为熔体必须迁移通过多孔骨料。这些实验加热的增量是延长到60秒期间的第一个10秒加热后增，斜坡产生温度和压

6、力类似的斜坡。温度和压力是那么的平衡在接下来的功率增加。如果体积减少发生通过压力斜坡斜坡或转折变化表示的时间间隔内。加热的增量增加功率手动从温度控制器的感应。这个程序我们确保相同的能量增量用于每个时间间隔。能量的增加必须尽可能准确地确定熔化发生时的温度。能量的增加也必须足够大，以产生大于测量误差的温度升高。我们估计，在0.4%的电力供应增量足以产生约1020C.温度的升高。测量设备和数据记录检测P和T的小的变化，在这些实验中产生的，它是需要使用高精度的测量设备和记录的压力和温度不断。将所得的数据被用来构建P-T和T-T曲线。在我们的实验中，温度控制和连续使用一个欧陆808控制器监控，有一个内部

7、的冰点补偿器，连接到计算机的数据采集和存储。该控制器具有手动模式，我们使用所需的时间间隔内产生的温度升高。通过电子德鲁克PTX 1400压力变送器测量活塞的RAM大门油压力，并通过欧姆龙继电器e5ck控制器监视。该变送器的精度允许我们检测0.1条油压力的变化（约5巴在胶囊）。在欧姆龙继电器控制器还连接到计算机，并测量压力的连续记录和实时显示。实验所用的材料 在泥质片麻岩不同压力下测定了固相线温度，和氯化锂。采用经典方法先前被确定为片麻岩固相线；即，检测存在或不存在的熔体在运行几天后的反应。这块岩石的矿物成分是约42%的石英，20%毫秒，10%转，18% PL，和8%个。对其组成和融化的关系在卡

8、斯特罗等人给出的细节。1999。其固相线的石英，KFS MS的存在故障的控制，和PL氯化锂在温度低于氯化钠介质压力融化，和氯化锂的固相线温度的立场是众所周知的（克拉克1959；循环1984）。由于锂是在室温下压实，钻石陷阱（例如，广濑和钏路1993）在这些实验中使用。结 论图2显示了从熔融实验与泥质片麻岩进行曲线。在P-T曲线趋势表现出不同的风格。压力下降发生在6.5巴，770C。请注意，P-T曲线恢复初始斜率几秒钟后，一旦熔体填充毛孔。第二期的结果是一个高原压力几乎是恒定的。这是在6.1巴的片麻岩solidii确定的情况下，752C，和11巴，800C。高原发生在约150秒的时间间隔，来填补

9、的熔体陷阱在这两个实验的孔隙的时间（D B和石英颗粒钻石）。三分之一结果的P-T曲线斜率的变化。一个例子是在7.8巴决定片麻岩固相线，770C。得到了用钻石陷阱氯化锂粉实验类似的结果。钻石陷阱填充慢慢产生的P-T曲线的轻微变化。效果的在图3d的P-T曲线是斜率S形的拐点在9.6巴，805C.综 述压力时间曲线的解释三个实验，P10巴的泥质片麻岩产生相似的P-T曲线进行。压力增加的速度比较慢，在实验和熔点附近的速度开始。压力高原后（或压力下降），率降低到原来的值接近。这些波动是系统性的，并反映在装配在加热过程中的力学效应。例如，在6.1巴的实验，压力升高率 5103bar/s在最初的500秒，和

10、 3102bar/s从那时直到达到高原。在高原，在，在500秒的第一类似的结果，观察在6.5巴进行实验得到的速度接近压力增加。的温度时间曲线几乎恒定的斜率表明，热平衡是在每个功率增量达到。然而，这是没有压力的情况下。在P-T曲线斜率的变化表明力学效应（扩张和变形）不是由组件完全吸收在每个时间间隔的功率增量。显然，一些影响传播到随后的功率增加，造成的P-T曲线的斜率增加。这些累积的效应被完全吸收由装配在高原，和压力的增加恢复的初始速率。我们的观点是，这些系统的边坡的变化不是我们的实验文物；相反，它们是由于该组件的力学响应。显然，通过时间间隔长，机械效应可以通过该组件变形完全消退。然而，结果并不完

11、善，因为主要目的是检测在高原的温度，或压力降低，在P-T曲线观察。因此，测得的压力在高原区间样品中真正的压力，因此，在该压力的样品融化。对于泥质片麻岩和氯化锂固相线为片麻岩和氯化锂的固相线的测定可以用来定位的固相线的曲线在PT空间。为Ollo de Sapo片麻岩曲线与以前的实验结果相比，其固相线温度是通过经典的技术与扫描电镜观察抛光实验确定。在这个经典的技术的固相线温度测定的精度取决于必要的实验次数，支架的熔点在一个狭窄的区间。以文中描述的测定是准确的 10C.在本研究中所用的实际固相线接近无熔点在10巴和接近熔点在6巴的泥质片麻岩增量加热技术。这里描述的技术更精确，因为它依赖于由熔化的胶囊

12、内开始引起的力学效应。基于对实验运行在近固相线的条件下观察，熔体仅约1体积%的需要产生所观察到的力学效应。该技术还密切再现由克拉克氯化锂的固相线（1959）（参见波伦1984）。这种材料的行为差别很大，从硅酸盐。压实作用是在泥质片麻岩更容易检测。然而，我们注意到，通过应用技术的氯化锂熔炼得到的固相线点非常接近先前确定的固相线的曲线。参考文献1波伦，S.R.（1984）精确压力校准平衡活塞缸装置和摩擦小炉装置。9,404-412.2Castro,A.,Patio Douce,A.E., Corretg, L.G., de la Rosa, J.D., El-Biad, M., and El-Hm

13、idi, H.（1999）花岗岩和花岗闪长岩，成因伊比利亚地块，西班牙。花岗岩成因的实验测试。矿物学、岩石学贡献255,135-276。3克拉克（1959），公司对八碱金属卤化物熔点压力的影响。化学物理学报，31，1526-1531。4Hirose, K. and Kushiro, I.(1993）在高压下干periodotites部分熔融的组合物的测定：从使用聚集体的钻石periodotites隔离。地球和行星科学通讯，114，477-489。5Patio Douce, A.E. and Beard, J.S.（1995）黑云母片麻岩和石英岩从3到15巴的脱水熔融。岩石学杂志36，70773

14、8。6Patio Douce, A.E. and Beard, J.S.（1996）P的影响，f（O2）和Mg / Fe比在合金脱水熔融模型。岩石学杂志，37，999-1024。7Patio Douce, A.E. and Harris, N.(1998）对喜马拉雅深熔作用的实验约束。岩石学杂志，39，689-710。8Singh, J. and Johannes, W.（1996）熔融英云脱水。第一部分：开始融化。矿物学、岩石学贡献16，25125。9Vielzeuf, D. and Clemens, J.D.（1992）云母+石英缺乏流体熔融：实验与模型。美国矿物，77，12061222。

15、10Vielzeuf, D. and Montel, J.M.（1994）部分熔融的铝合金。第一部分：缺乏流体实验和相位关系。矿物学、岩石学贡献375，393117。11威利，P.J.（1977）地壳深熔作用：实验观察。构造物理，43，4171。外文原文A new method for determining the fluid-absent solidus temperature inpiston-cylinder experimentsABSTRACTWe describe a new piston-cylinder method for determining the fluid-abs

16、ent solidus curves of geological materials. Samples are heated incrementally at pressures from 3 to 20 kbar for periods less than one hour. The effects of melting on the measured oil pressure are small but detectable and yield an accurate solidus by interpretation of pressure-time data. The techniqu

17、e successfully reproduces the solidus curve for LiCl. Results for the Ollo de Sapo pelitic gneiss are superior to those obtained previously using a more conventional technique.INTRODUCTIONThe identification of magma sources in the Earths mantle and crust is facilitated by the results of experimental

18、 petrology.The P-T positions of solidus curves for deep protoliths are critical to understanding the processes of magma generation. Solidus curves for potential crustal source materials, such as amphibolites and pelites, have been determined experimentally (e.g., Wyllie 1977; Singh and Johannes 1996

19、; Pet 1976; Patio Douce and Beard 1995, 1996; Patio Douce and Harris 1998;Vielzeuf and Clemens 1992; Vielzeuf and Montel 1994). These results have shown that variations in the mineral assemblages,or in the compositions of the mineral solid solutions involved in the melting reactions (e.g., mica and

20、plagioclase), can exert a strong influence on solidus temperatures. Consequently, to properly characterize magma production processes in a particular environment it is essential to determine the solidus curves for specific crustal protoliths. However, the work is arduous, requiring many time-consumi

21、ng experiments in which the presence or absence of melt is determined by observation of scanning electron microscope (SEM) images of the run products. The technique described below can be used to determine the position of a solidus curve in P-T space by direct observation of the effects of melting o

22、n the oil pressure in piston-cylinder apparatus. These effects are very small but can be detected by examination of a pressure-time (P-t) curve produced by incremental heating.RATIONALEWe have noticed in piston-cylinder experiments that 510 C variations in temperature have a measurable effect on pre

23、ssure. This is observed during the first minutes of an experiment when temperature is ramped up with a temperature controller. A corresponding ramp is obtained when measured pressure is plotted against time. Consequently, compaction of the capsule container and/or the sample capsule can be detected

24、by variations in the measured pressure during heating, which affect the slope of the pressure-time curve. Melting of the sample within the capsule will produce a compaction effect that may be detected if pressure is measured very accurately and data are taken continuously. Applying this method, we h

25、ave conducted several melting experiments to identify the solidus temperature by changes in the pressure-time curve induced by heating of the sample in the piston cylinder.EXPERIMENTAL PROCEDURES AND STARTING MATERIALSExperiments were performed in end-loaded, solid-medium piston-cylinder apparatus w

26、ith 12.7 mm (0.5 inch) diameter NaCl-graphite cell assemblies. Temperatures were measured and controlled with Pt100-Pt87Rh13 thermocouples wired to Eurotherm 808 controllers. Oil pressures were measured with electronic DRUCK PTX 1400 pressure transmitters, connected to OMRON E5CK controllers. We use

27、d both natural rock samples and synthetic materials. About 20 mg of finely crushed powder (1050 mm) were sealed within a gold capsule, which was subsequently embedded under pressure in an NaCl plug at room temperature. Because determination of the solidus temperature relies on detecting sample compa

28、ction, the technique only works for fluid-absent experiments. Water-bearing samples are not appropriate because water will fill the porous aggregate before melting occurs. The effects of volume reduction by melting-induced compaction of the sample can be magnified by using a porous aggregate that ac

29、ts as a melt trap. We used quartz and diamond aggregates and obtained good results in both cases. For rock samples with excess quartz (e.g., pelites and gneisses), the introduction of a quartz trap had no influence on the melting temperature. For silica-poor compositions (e.g., amphibolites), the us

30、e of a diamond trap is more appropriate. Measured changes in pressure due to compaction of the sample are normally very small (1 bar oil pressure), and generally last only for a short time (30 s). To detect any change due to melting within the capsule, we performed several experiments applying small

31、 increments of heat and observing the response of the system over short time intervals. The assembly was heated by increasing the power manually a predetermined percentage of the total power every 30 s. We chose this time interval because it is sufficient to ensure that thermal equilibrium within th

32、e assembly was reached before the next energy input (Fig. 1a). In experiments using a diamond or quartz trap, compaction may be slower because melt must migrate through the porous aggregate. For those experiments heating increments were lengthened to 60 s. During the first 10 s after a heating incre

33、ment, a ramp is produced for temperature (Fig. 1b) and a similar ramp for pressure (Fig. 1c). Temperature and pressure are then restabilized before the next power increase. If volume reduction occurs within a time interval it is indicated by a change in the slope or an inflection in the pressure ram

34、ps. The heating increment is induced by increasing the power manually from the temperature controller. With this procedure we ensure that the same energy increment is applied for each time interval (Fig. 1). The energy increment must be as small as possible to determine accurately the temperature at

35、 which melting occurs. The energy increase must also be large enough to produce temperature increases that are greater than measurement error. We have estimated that an increment in the power supply of 0.4% is sufficient to produce a temperature increase of about 1020 C.Measurement equipment and dat

36、a recordingTo detect the small changes in P and T produced in these experiments, it is necessary to use high-precision measurement equipment and to record pressures and temperatures continuously. The resulting data are used to construct P-t and T-t curves. In our experiments, temperature was control

37、led and monitored continuously using an Eurotherm 808 controller, with an internal ice-point compensator, connected to a computer for data acquisition and storage. The controller has a manual mode that we used to produce temperature increases within the desired time intervals. Oil pressure was measu

38、red at the gate of the piston ram by an electronic DRUCK PTX 1400 pressure transmitter, and monitored by an ONROM E5CK controller. The precision of this transmitter allows us to detect variations of 0.1 bar oil pressure (ca. 5 bar in the capsule). The ONROM controller was also connected to the compu

39、ter, and measured pressures were recorded continuously and displayed in real time.Materials used in the experimentsSolidus temperatures were determined at different pressures for pelitic gneiss, and for LiCl. A solidus curve for the gneiss was determined previously using the classical method; that i

40、s, detecting the presence or absence of melt in runs after reaction for several days. The mineralogical composition of this rock is approximately 42% Qtz, 20% Ms, 10% Bt, 18% Pl, and 8% Kfs. Details on its composition and melting relations are given in Castro et al. (1999). Its solidus curve is cont

41、rolled by the breakdown of Ms in the presence of Qtz, Kfs, and Pl.LiCl melts at temperatures lower than that of the NaCl pressure medium, and the P-T position of the solidus curve for LiCl is well known (Clark 1959; Bohlen 1984). Because LiCl is compacted at room temperature, a diamond trap (e.g., H

42、irose and Kushiro 1993) was used in these experiments.RESULTSFigure 2 shows the curves obtained from the melting experiments performed with the pelitic gneiss. Trends in the P-t curves exhibit distinct styles. A pressure decrease occurred at 6.5 kbar, 770 C (Fig. 2a). Note that the P-t curve resumed

43、 its initial slope after several seconds, once the melt filled the pores. A second P-t result is a plateau during which pressure is nearly constant. This is the case for the gneiss solidii determined at 6.1 kbar, 752 C, and 11 kbar, 800 C (Figs. 2b and 2d). The plateau occurred over a time interval

44、of about 150 s, the time needed to fill the pores of the melt-trap used in these two experiments (diamonds in b and quartz grains in d). A third result is a change in the slope of the P-t curve. An example is the gneiss solidus determined at 7.8 kbar, 770 C (Fig. 2c). Similar results were obtained f

45、or the experiments with LiCl powder using a diamond trap (Fig. 3). The diamond trap filled slowly producing slight changes in the P-t curve. The effect indicated by the P-t curve in Figure 3d is a sigmoidal inflection in the slope at 9.6 kbar, 805 C.DISCUSSIONInterpretation of the pressure-time curv

46、esThe three experiments performed with the pelitic gneiss at P 10 kbar produced similar P-t curves. The rate of pressure increase is slower at the beginning of the experiment and faster near the melting point. After the pressure plateau (or pressure decrease), the rate decreases to a value close to

47、the original one. These fluctuations are systematic, and reflect mechanical effects within the assembly during heating. For example, in the experiment performed at 6.1 kbar (Fig. 2b), the rate of pressure increase was 5 103 bar/s during the first 500 s, and 3 102 bar/s from that point until the plat

48、eau was reached. After the plateau, pressure increased at a rate close to that observed in the first 500 s. Similar results were obtained in the experiment performed at 6.5 kbar. The nearly constant slope of the temperature-time curves suggests that thermal equilibrium was reached within each power

49、increment. However, this is not the case for pressure. The variations in the slopes of the P-t curves suggest that mechanical effects (dilatation and deformation) are not absorbed completely by the assembly within each time interval of power increment. Evidently, some of the effects propagate to the subsequent power increment, causing the slope of the P-t curve to increase. The