Reduced Viscosity of Mg2GeO4 with Minor MgGeO3 between 1000 and 1150 °C Suggests Solid-State Lubrication at the Lithosphere–Asthenosphere Boundary
Round 1
Reviewer 1 Report
The manuscript ”Unexpected softening of Mg2GeO4 with minor MgGeO3 between 1000 and 1150 C: a transient metastable phase at the lithosphere-asthenosphere boundary?” introduces a new concept for mechanical softening taken from material science - Transformation induced plasticity (TRIP) - which is described as a possible mechanism to produce decoupling between the Lithosphere and the asthenosphere below it (the LAB). Synthesized samples of germanite based grains of Mg2GeO4 and minor component of Mg2GeO3 were used as analogue for mantle rock. Samples were deformed at atmospheric pressure and high temperatures (950 – 1250 deg C) using spark plasma sintering (SPS) at Nagaoka (Japan) following the method used and described by Ghosh et al. (2021). Five additional germanite samples from different labs and sources were used as references to the used synthesized samples. Although no microstructure is found to suggest phase transformation the authors rely on the rheological-mechanical data to suggest that phase transformation did occur and was responsible for softening.
The experimental methodology is at the state of the art and the use of the analogue material may allow for achieving conditions otherwise difficult to achieve in laboratory settings. Since I am not an expert on the details of synthesizing such samples or the use of SPS I assume that the first author T. Ferrand experimental experience (e.g., Ferrand et al., 2017) and the operation of the SPS in Japan as described recently in Ghosh et al., (2021) testify for the experimental method and synthesizing samples that were used in this study (due to sample widening).
The manuscript describes a number of interesting ideas and ‘out-of-the-box’ kind of thinking together with concepts from material science that are not often used in geophysics. Although unique voices should be encouraged, the current manuscript and the interpretation of the results lead me to suggest to reject the manuscript as currently constructed.
Main comments:
(1) TRIP is actually associated with hardening not softening. Following Poirier (Creep of crystals, 1985, p. 213-214) - Transformational plasticity is a mechanism for softening while TRIP is associated with hardening and necking (increased strain-rate). In addition, transformation plasticity is characterized on the stress-strain graph (Poirier, fig. 8.9 p. 224) as an abrupt decrease of stress in a stress-strain curve.
(2) The main point of the paper as described numerous times throughout the manuscript is the ‘extreme softening’ described as the TRIP. However, the softening (described as TRIP although comment #1 above) in Fig. 5 shows rather mild gradual weakening with strain (Fig. 5a) without an abrupt drop. In addition, the compilation of results in Fig. 7 shows that only 2 out of ~15-20 of the experiments have a negative stress-strain ratio (‘apparent Young modulus’). Most of the experiments actually show mild strengthening and some show purely plastic behavior where stress is independent of strain.
It is not clear why the traditional way to interpret the stress-strain curve - where the elastic stage followed by a stage of plastic yield and plastic deformation (and later hardening). (again, for reference, Poirier’s book, the sigmoidal creep in figure 1.9(d)). Since there is no evidence for phase transformation and the rheological data does not show that transformation-related softening, the manuscript becomes hypothetical and speculative.
(3) Olivine in the LAB is not expected to be metastable or go through phase transformation (as discussed in the discussion). Therefore, the parallelism between the transition zone and the LAB is hypothetical (lines 624-626).
(4) Fig. 5b – This graph shows only the “creep test” part, where the load is constant (and stress slightly decreases because of the sample widening). Therefore, this graph looks at the changes in strain-rate (strain vs. time). There is a single line with no apparent changes in strain-rate but different parts are labeled as “transient creep” and “cavitation creep”. They seem to be randomly selected, or, if not, it is not clear from the figure how these sections were interpreted as different stages of deformation.
(5) Fig. 12. Besides the rupture of samples in fig. 12b I do not see any difference in the stress-strain curve between the two sets of experiments described as with TRIP (Fig. 12a) and without TRIP (Fig. 12b).
In conclusion, I think there isn’t enough consistency throughout the manuscript and, more importantly, the main argument is not well supported by the results relying on speculations and overinterpretations.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper “Unexpected softening of Mg2GeO4 with minor MgGeO3 between 1000 and 1150°C: a transient metastable phase at the lithosphere-asthenosphere boundary?” is interesting, and provides an important contribution for understanding mechanisms governing the Earth, plate tectonics
I suggest minor changes as follows:
Fig. 1 and related text illustrate discussion about subduction at island arc boundary. May be another fig and description of ocean-continent boundary would be beneficial for readers. Also discussion about differences fast and slow plate motion would add value to the paper.
I have marked some remarks in the manuscript's PDF file.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper presents experimental data on an extreme softening of a peridotite analogue in the Mg2GeO4−MgGeO3 system at 1 atm between 1000 and 1150°C, followed by a sharp hardening. Within this temperature interval, the apparent viscoelastic Young modulus of the material drops down to zero once a critical stress as low as 30-100 MPa is reached. The hardening is explained by specific diffusion that strengthens thermodynamically stable phases or grain-boundary structures at 1200°C. This experimentally recorded process is similar to commonly observed ones in steels during the shear-induced transformation from austenite to martensite, the final material being significantly harder. This TRansformation-Induced Plasticity, widely observed in TRIP alloys, with subsequent hardening, is evidenced in the paper for the first time in this mantle analogue (line 820). Other hypotheses of the lithosphere-asthenosphere boundary (LAB) origin are also mentioned in introduction. Specifically discussed in the text are 1) a transient metastable phase responsible of the softening, 2) possible plate lubrication without partial melting at the LAB, 3) relationship between the LAB and asthenosphere, and 4) influence of phase transitions in the mantle. As advocated in Section 4.2, the extreme softening observed in the experiments is not related to any melting process. So, the main idea of the authors is to apply the experimental results for explanation of lithospheric plate motions on the asthenosphere through a solid-state mechanism, in contrast to popular understanding of the LAB as a transient zone from a rigid to partly molten medium. Although HP-HT experiments are still required to define exact conditions of such softening in the MgO-FeO-SiO2 system, the submitted paper on phenomenology of this process in a possible peridotite analogue at the LAB should be published.
Comments
- In introduction, some references, for instance, about the origin and evolution of life (Valentine & Moores, 1974; Parnell, 2004; Santosh, 2010) are excessive. More relevant for this study are some modifications of major predictions for present-day understanding of plate tectonics as compared to original ideas proposed in 1960th. These changes would actually highlight the recent LAB concept.
- Convective and low viscosity accents in the “French and Anglo-Saxon definitions” of the asthenosphere, respectively do not seem to be a good idea for a scientific paper. It sounds like “Different rocks for different folks” once suggested by Jerry Wasserburg. Expressions like these are a matter of newspapers and jokes. The term “asthenosphere” was originally introduced to designate the sphere of weakness below the lithosphere by Joseph Barrell in 1914. In addition, convection beneath oceans was proposed by Arthur Holmes in 1929. Of course, the asthenosphere is identified worldwide by different seismic tomography images. Geochemically, volcanic eruptions can be referred to both asthenospheric and lithospheric magma sources. If there is true evidence on convection, asthenospheric properties can be attributed to a convectively unstable unit even beneath continental areas.
- Burnley et al., 1991 (references in lines 350, 367, Table 1); Tasaka & Hiraga, 2013 (line 695); Jung & Green, 2009 (lines 704, 740); Sundberg & Cooper, 2008 (line 709); Maruyama & Hiraga, 2017 (line 710) are not listed.
- In the references, Robbin & Levin, 1959; Tielke et al., 2016 are out of place alphabetically. It should be noted that accepted in Minerals are references in order of appearance in a text.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 4 Report
Dear Authors,
This paper deals with the source of weakening occurring at conditions of the lithosphere-asthenosphere boundary (LAB). For a plate-like behavior to be consistent on Earth, this weakening is required to decouple the lithospheric plates from the underlying asthenospheric mantle, but the mechanism(s) involved remains today very elusive. Based on new experimental data at room pressure, this paper suggests that such a weakening might result from a phase transition within a narrow temperature range, which would fit the conditions of the LAB. However, although the paper is well written and the mechanical data deserve to be published (the observed weakening is very interesting by itself), there are major issues that lead me to do not recommend this paper for publication in minerals, at least in the present form. Actually, the paper looks like a review paper about processes occurring at LAB conditions, but the present version is not presented this way. In contrast, you propose an idea that is not supported by the data, which is too much speculative. Furthermore, the paper needs to be highly revised and several points have to be addressed before reaching a suitable form for publication. You will find my comments/concerns in the appended pdf file.
With my Best Regards,
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The revised manuscript “Unexpected low viscosity of Mg2GeO4 with minor MgGeO3 between 1000 and 1150°C suggests solid-state lubrication at the lithosphere-asthenosphere boundary” includes some improvements to the manuscript in terms of more accurate use of descriptive terms (e.g., ‘extreme softening’ was replaced by ‘enhanced plasticity’) and overall focus. I will restate the significance of the presented experimental study using analogue material to study the rheology and mechanics of the deep mantle, the valuable discussion for the possible deformation mechanism in the mantle, and the importance of having new and different voices heard in the research discourse.
However, the manuscript still needs to be clearer for what it present. Currently, there is a strong emphasis on the experimental results although I do not see clear evidence that suggests the occurrence of TRIP, and the microstructure observations, as discussed, cannot rule out or strongly suggest that a transformation (of any kind) indeed occurs (the authors rely strongly on what they term ‘enhanced plasticity’ stage in the stress-strain curve, but this mechanical behavior does not imply any specific mechanism). It is advised that the authors emphasize their suggested new and alternative way for interpreting the stress-strain.
As the authors wrote in the response letter: “The fact that a misunderstood process/phase is not expected cannot be an argument strong enough to discredit a discussion.”. Fair enough, but the speculative nature of the paper should be presented as such while having its worth and value to the community in introducing and discussing a new alternative way of thinking on possible deformation-transformation processes that could potentially affect the mechanical behavior of the mantle and introduce alternative prospective on the LAB and/or the mantle transition zone. This way the reader would not be misled and could judge by himself the proposed concepts as they are at the moment (i.e., lack direct evidence). This can be done with some changes to the organization of the manuscript and by emphasizing the discussion while separate the observations from the interpretations (one example would simply be moving the ‘state of the art’ part to the discussion). Also, to communicate the suggested alternative TRIP, it is essential to relate it to the known and established way of interpreting the stress-strain curve that is used in most cases (an initial linear stress increase with strain, which is determined by the material’s Young modulus, followed by a yield point that transition the deformation to an inelastic, non-reversible, viscous mechanism) whether it fits or does not fit with the new interpretation.
Not overlooking the high scientific merit, and innovation aspects of the manuscript I think that the speculative part of it (that might be larger than most papers) should be presented as such and not masked by descriptions that are mixed with interpretations.
Specific comments:
(1) There is an overall problem in the manuscript with mixing descriptions of results with interpretations, which was not improved significantly (simply relabeling part of the introduction as ‘state of the art’ did not fix the problem). For example, ‘unexpected result’ is used throughout, from the introduction, results, figure caption, and discussion, although it is not clearly stated why is it ‘unexpected’ and what would be otherwise expected for the mechanical behavior.
(2) I do not understand why the authors decide not to make changes to some of the reviewers’ comments. One of the responses says: “we report in this paper… that a “transformation” actually occurs in the material within a narrow temperature window”. The sentence was left as a statement without mentioning what way it shows ‘transformation’ and without giving references directing to a specific figure or text. This kind of statement without relying on something concrete like an argument or specific reference to the text or figure is impossible to relate to.
(3) The term ‘viscoelastic Young’s modulus’ is new to me. That was also part of my comment #2 in the previous review. Is it a new term coined by the authors in this paper? If not can the authors add references where such use of Young’s modulus is used for viscous-plastic deformation? I reiterate this because it seems inaccurate to me. A more descriptive way is to say the derivative or slope of the stress-strain curve (but still, that would ignore that the initial stage is mostly elastic and not part of the viscous-plastic behavior).
(4) Line 16-17: Saying that the Young’ modulus drops down to zero describes what is simply can be regarded as yield point – transition from elastic to anelastic behavior. Therefore, describing the yield point as ‘viscous reduction’ since the stress-strain slope changes abruptly disregard the elastic nature of the initial stage of deformation. This point is fundamental since the authors throughout the manuscript use the term ‘enhanced plasticity’, not pointing out why/how is it different than a simple yield point (where you would also have such a change of stress-strain slope).
(5) Line 386 (and response to comment 1): How is the necking phenomenon that occurs under tension relevant for the compression experiments?? (e.g. the reference of Chow et al., 2005 given by the authors).
Author Response
Please see the attachment.
Best regards,
TPF.
Author Response File: Author Response.pdf
Reviewer 4 Report
Dear authors,
Please, consider my comments below
Line 27: What do you mean by displacive processes. Please clarify.
Line 63: Written this may, it says that the lithosphere is brittle. MOST of the lithosphere is ductile. So please, rephrase!
Line 86 to 89: So, you claim that low-pressure experiments are better than high-pressure ones to explore processes at high pressure. For me, it is non-sense. Furthermore, this sentence suggests that your experiments would be completely different and irrelevant in a Grigg-type apparatus. Of course, you may have some error in stress measurement, but the behavior should remain the same at high pressure (if you’re right). I suggest to delete this sentence. Actually, you don’t really need it.
Line 168: delete the comma between « molten » and « layer »
Line 199 to 200: What is significant for rheology is not necessarily significant for volcanology in terms of H2O content. Please, clarify.
Line 217: Dis-GBS means dislocation-accommodated GBS, not « assisted ».
Lines 239-245: I think this paragraphe is beyond the scope of the paper. Overall, I still think that you can reduce the state-of-the-art section by focusing only on the nature of the LAB.
Line 250 to 252: Please, be aware that, In a solid-medium apparatus, you can quench in a few seconds (and then comes the unloading), preserving all of produced structures.
Figure 11: Electron Backscatter Diffraction, not Backscattered
Author Response
Please see the attachment.
Best regards,
TPF.
Author Response File: Author Response.pdf
