太古宙的軟糊地函
實驗數據顯示地球的地函比之前認為的更容易熔化，因此直到20億至30億年以前或許都一直處在軟糊糊的狀態。
地球在45.6億年前形成時是處於完全熔化的狀態，自此之後便持續冷卻下來。時至今日，大部分的地球已經完全固化。金屬鐵地核的最外圍仍然為液體――其為最初熔化的地球殘存下來的最後一部份；但包覆在地核外面，由岩石組成的矽質地函卻幾乎全為固體。地函在地球歷史的大半時期可能都是處於固態；或者，這不過是我們的想法。Andrault和其同事發表在《自然―地質科學》的論文中，呈現的高壓熔化試驗結果指出地函熔化的溫度可能比我們以為的還要低200到250℃。因此在20億至30億年前地球比現在還要高溫的時候，地函可能大部分都處於部分熔融的狀態。地函從早期的軟弱狀態轉變成固體時或許產生了某些重大變化，包括板塊構造運動的改變在內。
地球的溫度從地表至核心一路從近乎零度升高至攝氏數千度。但此地溫梯度並不是一條單純而平滑的曲線。化學成分的不均、相變以及熱傳輸方式的差異(傳導或對流)結合起來把地溫梯度彎曲成一條斜率不一且有多次躍變的曲線。隨著深度增加，固態地函開始熔化時的溫度――即固相線――也會跟著上升。固相線的斜率通常比地溫梯度還要和緩，在低壓時固態線會跟地溫梯度交叉而讓熔融發生。
確實，地球表面的火山爆發即為現今地函在某些地點還是會熔化的證據。今日的中洋脊標示出地函固相線跟地溫梯度有所交會的主要地點。在這些環繞全球的中洋脊所發生的熔融現象和岩漿活動產生了地球三分之二的地殼。不過，相較於整個地函來說發生熔化的只占一小部分，因為地函大部分區域的溫度還是低於固相線。
測量地函固相線的實驗是把代表地球內部成分的岩石樣品在不同壓力範圍下加熱。但這不是一件簡單的任務，數十年來實驗人員的研究結果有很大的出入而帶給他們相當艱鉅的挑戰。在多數實驗中，實驗人員是在實驗結束，樣品冷卻到室溫時立刻仔細檢查岩石是否有發生熔融。然而，要用這種方式找出極為少量的熔融物質不啻是在大海撈針。通常熔融物質在樣品冷卻時就已經結晶成固體，要把它們辨識出來幾乎是不可能的任務。
Andrault和其同事運用一種不同方式來測量地函的固相線：他們在極端高溫高壓的實驗環境下就地分析樣品是否有熔融產生。雖然就技術層面來說相當困難，但研究人員利用X光繞射和電導率成功測量出他們的樣品中不到整體體積1%的熔融物質，使得他們呈現的地函熔點或許是目前為止最貼近真實情況的。在較低的壓力下，他們的實驗結果跟過往的研究相當符合。然而，在壓力到達7 GPa以上，相當於地下大約200公里以下的時候，他們發現地函的固相線比之前的估計還要低200到250°C。
今日大部分地函所處的溫度較此新測定的固相線還低。但是早期地球的溫度比現在要高。雖然科學家還尚未確切界定出地球的冷卻速率，不過可以合理認為在20億至30億年前，地函大部分的溫度都在這條固相線之上，因此同時具有固態晶體和熔融成岩漿的液體而處在軟糊糊的狀態。地函大部分區域都相當軟弱會使得地函整體的黏滯性降低，造成地球火山活動的數量和類型有所改變。
許多地質現象指出地球在20億至30億年前的這段時期歷經了重大變化：地殼年齡分布在這段時期有個高峰，代表地殼的產量增加或者是有更多地殼被保存下來；有些種類的岩漿停止生成；大氣中的氧氣含量首度增加；同時板塊構造運動的訊號首度在地球各處出現。雖然驅動這些變化的因素仍尚未定論，但是地函的狀態從軟弱變成堅硬提供了一個簡潔且巧妙的合理解釋。精確來說，Andrault和其同事提出地函中的軟弱層位會讓地函跟上覆的岩石圈各自為政。隨著地球逐漸冷卻，軟弱層位凝固下來，地函和岩石圈板塊會形成連結，因而啟動隱沒作用、板塊構造運動，以及其他種種變化。
雖然Andrault和其同事運用的就地測量方法對於含量極少的熔融物質相當敏銳，但卻無法迅速定出究竟有多少熔融物質。未來研究還需要測出當溫度超過固相線時熔融物質的增加量有多快(即熔融速率)。過往研究提出熔融速率可能是高度非線性的，當溫度在固相線附近時熔融產生的速率相當慢，但溫度到達某一個閾值時產量就會急遽上升。在過往其他敏銳度較低的實驗方式中得到的固相線溫度較高，事實上可能就是閾值，此時熔融的體積百分比提升到1%以上。若是如此，則地函現今與過往的差異可能就只是因為少量的熔融物質所造成，這是否足以解釋20億至30億年前發生的巨大變化仍然不清楚。
Andrault和其同事利用高壓實驗顯示地函的固相線比之前預估的還要低，也暗示了地函在太古宙的多數時間都是處於部分熔融的狀態。地球在此時發生的重大變化，包括板塊構造運動的啟動在內，可能都是因為地函較晚固化所導致。
An Archaean mushy mantle
Stephen Parman
Experimental data
reveal that Earth’s mantle melts more readily than previously thought, and may
have remained mushy until two to three billion years ago.
Earth
was completely molten when it formed 4.56 billion years ago, and it has been
cooling ever since. Today, most of the Earth has solidified. The outer part of
the metallic iron core is still liquid — the last survivor of the originally
molten Earth. But the rocky, silicate mantle that surrounds the core is largely
solid. The mantle has probably been solid for most of Earth’s history. Or so we
thought. Writing in Nature Geoscience, Andrault and colleagues present high-pressure
melting experiments that indicate that the mantle can melt at temperatures 200
to 250 °C lower than previously thought. So, 2 to 3 billion years ago, when
Earth was hotter than today, much of the mantle could have been partially
molten. The transition from an early mushy mantle to a solid one may have
generated several significant changes attributed to this time period, including
the shift to plate tectonics.
Earth’s
temperature increases from essentially zero at the surface to thousands of
degrees Celsius at the centre. But this geothermal gradient is not a simple,
smooth curve. Chemical heterogeneities, phase changes and differing modes of
heat transport (conduction versus convection) combine to yield a geothermal
gradient with varying slopes and numerous jumps. The temperature at which the
solid mantle begins to melt — the solidus — also increases with depth. The
slope is generally shallower than the geothermal gradient and, at low
pressures, the solidus can cross the geothermal gradient, causing melting to
occur.
Indeed,
volcanic eruptions at Earth’s surface are evidence that the present-day mantle
does melt in some places. Today, the mid-ocean ridges mark the primary location
where the mantle solidus and geothermal gradient coincide. Melting and
magmatism at these ridges — which encircle the globe — produces two-thirds of
our planet’s crust. Still, the proportion of the total mantle that melts is
small because most of it exists at temperatures below the solidus.
The
mantle solidus is measured experimentally by heating rock samples
representative of Earth’s interior over a range of pressures. But this is no
easy task and has challenged experimentalists for decades, with studies
producing widely disparate results. In most experiments, the presence of melt
is determined by closely examining the sample after the experiment, once it has
cooled to room temperature. However, finding tiny amounts of melt this way can
be like searching for a needle in a haystack. Often the melts crystallize when
the sample is cooled, making their identification nearly impossible.
Andrault
and colleagues1 use a different approach to measure
the mantle solidus. They analyse their samples for the existence of melt in
situ at extreme high pressure and temperature. This is technically quite
challenging, but the researchers use X-ray diffraction and electrical
conductivity to detect less than 1 vol% melt in their samples, thus providing
possibly the best constraints yet on the mantle’s melting point. At relatively
low pressures, their experiments match previous work. However, at pressures
above 7 GPa, which corresponds to a depth of about 200 km, they find that the
mantle solidus is 200 to 250 °C lower than previous estimates.
Today,
most of the mantle exists at temperatures below this newly determined solidus.
But the early Earth was warmer. Although Earth’s exact cooling rate is not well
constrained, it is plausible that 2 to 3 billion years ago, most of the mantle
was above the solidus temperature and could have been mushy — composed of both
magmatic melt and crystals. The existence of extensive mushy mantle would have
lowered the viscosity of the mantle and changed the amount and types of
volcanism on the planet.
Many
geologic observations suggest this period 2 to 3 billion years ago was a time
of great change on Earth: there was a spike in crustal ages, indicative of
either increased crustal production or preservation, some lava types ceased
production, oxygen rose in the atmosphere for the first time, and the first
pervasive signs of plate tectonics appeared. What drove these changes is
debated, but a transition from a mushy to a solid mantle provides a plausible
and elegantly simple mechanism. Specifically, Andrault and colleagues propose
that a mushy layer could have decoupled the mantle from the overlying
lithosphere. As Earth cooled and the mush layer crystallized, the mantle and
lithospheric plates could have become coupled, triggering the onset of
subduction and plate tectonics, among other changes.
Although
the in situ methods used by Andrault and colleagues are sensitive to the
presence of exceedingly small amounts of melt, they do not readily quantify the
amount of melt present. Future work will need to measure how quickly the amount
of melt increases as temperature rises above the solidus (the melting rate).
Previous work suggests that the melting rate can be highly non-linear,
with very low rates of melt production at temperatures near the solidus and a
dramatic increase in melt production only when a threshold temperature is
reached. The higher-temperature solidi found by previous less-sensitive
experimental approaches may actually be this threshold point, where
the melt percent rises above about 1 vol%. If so, the difference between the
modern and ancient mantle may be the presence of only a small amount of melt.
Whether that would be enough to explain the large changes seen 2 to 3 billion
years ago is not yet clear.
Andrault
and colleagues use high-pressure experiments to show that the mantle
solidus is lower than previously estimated, implying that it could have
remained partially molten for much of the Archaean. Significant changes on
Earth in this time period, including the initiation of plate subduction, may
have resulted from this delayed solidification.
原始論文：Denis
Andrault, Giacomo Pesce, Geeth Manthilake, Julien Monteux, Nathalie
Bolfan-Casanova, Julien Chantel, Davide Novella, Nicolas Guignot, Andrew King,
Jean-Paul Itié & Louis Hennet. Deep
and persistent melt layer in the Archaean mantle. Nature Geoscience, 2018; doi:10.1038/s41561-017-0053-9
引用自：Stephen
Parman. An Archaean mushy mantle. Nature Geoscience, 2018; doi:10.1038/s41561-018-0060-5
沒有留言:
張貼留言