Transcript
Page 1: Slab melting versus slab dehydration in subduction-zone ... · Slab melting versus slab dehydration in subduction-zone magmatism Kenji Mibea,b,c,1, Tatsuhiko Kawamotod, Kyoko N. Matsukagee,2,

Slab melting versus slab dehydrationin subduction-zone magmatismKenji Mibea,b,c,1, Tatsuhiko Kawamotod, Kyoko N. Matsukagee,2, Yingwei Feib, and Shigeaki Onoa,f

aEarthquake Research Institute, University of Tokyo, Yayoi, Tokyo 113-0032, Japan; bGeophysical Laboratory, Carnegie Institution of Washington,Washington, DC 20015-1305; cDepartment of Earth and Atmospheric Sciences, Cornell University, Ithaca, NY 14850-1504; dInstitute for GeothermalSciences, Graduate School of Science, Kyoto University, Beppu 874-0903, Japan; eDepartment of Environmental Science, Ibaraki University, Mito,Ibaraki 310-0056, Japan; and fInstitute for Research on Earth Evolution, Japan Agency for Marine-Earth Science and Technology, Yokosuka,Kanagawa 237-0061, Japan

Edited by Bruce Watson, Room 1C25, Science Center, Troy, NY, and approved April 7, 2011 (received for review July 28, 2010)

The second critical endpoint in the basalt-H2O system was directlydetermined by a high-pressure and high-temperature X-ray radio-graphy technique. We found that the second critical endpointoccurs at around 3.4 GPa and 770 °C (corresponding to a depthof approximately 100 km in a subducting slab), which is muchshallower than the previously estimated conditions. Our resultsindicate that the melting temperature of the subducting oceaniccrust can no longer be defined beyond this critical condition andthat the fluid released from subducting oceanic crust at depthsgreater than 100 km under volcanic arcs is supercritical fluid ratherthan aqueous fluid and/or hydrous melts. The position of thesecond critical endpoint explains why there is a limitation to theslab depth at which adakitic magmas are produced, as well as theorigin of across-arc geochemical variations of trace elements involcanic rocks in subduction zones.

water ∣ island arc ∣ silicate melt ∣ synchrotron X-ray ∣high-pressure research

Water plays an important role in subduction-zone magmatismbecause it can reduce the melting temperature of rocks in

subduction zones and hence can generate magmas (1–7). There isa long-standing debate about whether the fluids released from asubducting slab are aqueous fluid, hydrous silicate melt, supercri-tical fluid (SCF), or a combination of these (2, 4, 8–12). Whetherthe subducting slab melts or dehydrates depends on the thermalstructure and the phase relation of slab materials under hydrousconditions. Therefore, this long-standing question cannot beanswered until the detailed stability fields of these fluids areclarified.

Under high-pressure (P) and high-temperature (T) conditions,it has been shown that the solubility of both water in silicate melt(13–16) and silicate in aqueous fluid (13, 17–25) increases withincreasing P. As a result, silicate melt and aqueous fluid in theinterior of the Earth are expected to become SCF, and thehydrous solidus of the system can no longer be defined beyond acertain critical condition (26–36). This condition is called thesecond (or upper) critical endpoint (26) and is the point of inter-section between the critical curve and hydrous solidus. Basalt isone of the dominant constituents of a subducting slab and is con-sidered the main carrier of water that triggers the melting ofrocks in subduction zones. Therefore, the second critical end-point in the basalt-H2O system has to be determined in order tounderstand fully the subduction-zone magmatism.

In some silicic silicate-H2O systems, the location of thesecond critical endpoint has been reported [e.g., 1.0 GPa,1,080 °C in the SiO2-H2O system (13); 1.5 GPa, 670 °C in thesystem NaAlSi3O8-H2O (21); 1.5 GPa, 800 °C in the systemKAlSi3O8-H2O (36)]. In mafic systems, however, the determina-tion of the second critical endpoints is not easy because ofthe difficulty in identifying phases (aqueous fluid versus hydroussilicate melt) in the recovered samples quenched from high-Pand high-T conditions (25). The stability fields of fluid phasesin mafic systems have to be determined by observing phases

directly under high-P and high-Tconditions. In the present study,therefore, we determined the second critical endpoint in a basalt-H2O system using a high-P and high-T X-ray radiography techni-que (37) that enabled us to observe aqueous fluid and hydroussilicate melt directly. In addition to the experiments with X-rayradiography, we also conducted long-duration constant-T quenchexperiments in order to confirm that the equilibrium was attainedin our radiographic observations.

ResultsExperimental conditions and results are listed in Table 1. Thestability fields of observed minerals (especially amphibole in therun S1176) in the recovered samples are consistent with previousstudies (38, 39) within the experimental uncertainties, indicatingthat our P-T calibrations (30) are reliable. When experimental Pis below the second critical endpoint, and both aqueous fluid andhydrous silicate melt are present, round shapes can be observedin the radiographic images because of the differences in theirinterfacial tension. In this case, the two-phase boundary betweenthese two fluids is enhanced in the radiographic images becauseof the X-ray refraction contrast. During heating, this two-phaseboundary migrates without exception, because the fluid/meltratio changes with changing T (Movie S1). Therefore, two fluidsare easily identified if they coexist under the experimental P-Tconditions. Above the second critical endpoint, on the otherhand, the two-phase boundary cannot be observed, because onlythe SCF is a stable fluid phase (Movie S2). Instead, crystals(mostly garnet) are sometimes observed falling uniformlythrough the entire sample capsule (Movie S3), which is furtherevidence that only the SCF is a stable fluid phase under theexperimental P-T conditions. This is because not all the crystalsfall to the bottom of the sample capsule when two fluids coexist;some are sustained in melt. It was difficult to determine thestability fields of minerals from the radiographic observations.

In experiments up to 3.3 GPa (Fig. 1 A and B and Movie S1),both aqueous fluid and hydrous silicate melt were observed inradiographic images, if the water content of the starting materialswas within the solvus between aqueous fluid and silicate melt atthe experimental P. On the other hand, these two fluids could notbe distinguished in radiographic images (Fig. 1C and Movies S2and S3) in any of the runs with various water contents above3.6 GPa. When the samples were quenched under conditions

Author contributions: K.M. and S.O. designed research; K.M., T.K., K.N.M., and S.O.performed research; K.M., Y.F., and S.O. contributed new reagents/analytic tools; K.M.analyzed data; and K.M. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Freely available online through the PNAS open access option.1To whom correspondence should be addressed. E-mail: [email protected] address: Geodynamics Research Center, Ehime University, Matsuyama, Ehime790-8577, Japan.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1010968108/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1010968108 PNAS ∣ May 17, 2011 ∣ vol. 108 ∣ no. 20 ∣ 8177–8182

GEO

LOGY

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 8,

202

0

Page 2: Slab melting versus slab dehydration in subduction-zone ... · Slab melting versus slab dehydration in subduction-zone magmatism Kenji Mibea,b,c,1, Tatsuhiko Kawamotod, Kyoko N. Matsukagee,2,

where the two fluids could be observed, void spaces and tiny glassspheres (approximately 10-μm diameter) were observed in therecovered samples (Fig. 1 D–K). These void spaces and glassspheres are considered to be quenched aqueous fluid. Hydroussilicate melts were quenched into glasses at lower P (Fig. 1 E,G, and H), whereas they were quenched into a mixture of quenchcrystals and glasses at higher P (Fig. 1J). No voids were found in

the recovered samples quenched from >3.6 GPa (Fig. 1 L–P).Instead, quench crystals were distributed throughout the entiresample capsules (Fig. 1 L–P), which is consistent with the factthat the two fluids cannot be distinguished in radiographic imagesabove 3.6 GPa.

Quench products from long-duration constant-T quenchexperiments also revealed that aqueous fluid and silicate melt

Table 1. Experimental conditions and results

Run no.* P, GPa TM∕TQ, °C H2O, wt %† Observed fluid phases in radiography Quench products

S968 0.5 1;350∕1;350 59 F + M (790–1,350)‡ QM + QF + VdS969 1.8 1;320∕1;320 60 F + M (950–1,320) NDS1176 2.0 1;240∕920 60 F + M (820–1,240) Am + Cpx + QM + QF + VdQB-1 2.0 1;200∕1;200 63 — QM + QF + VdS997 2.8 1;350∕1;350 55 F + M (1,010–1,350) QM + QF + VdS998 2.8 1;350∕1;240 47 F + M (790–1,240) QM + QF + VdS1195 2.8 1;240∕1;240 62 F + M (850–910) NDS1194 3.0 1;240∕1;010 60 F + M (1,010–1,125) QM + QF + VdS965 3.0 1;300∕1;300 62 F + M (900–1,240) NDS963 3.0 1;350∕1;350 63 F + M (840–1,040) NDS996 3.0 1;300∕1;300 63 F + M (880–960) NDS1083 3.0 1;350∕1;130 65 SCF QSS1902 3.3 1;200∕1;150 62 F + M (1,150) QM + QF + VdS1193 3.3 1;180∕1;180 68 SCF QSQB-2 3.4 1;000∕1;000 57 — Gt + QSQB-3 3.4 1;000∕1;000 79 — Gt + QSQB-5 3.4 1;150∕1;150 85 — QSS1076 3.6 1;350∕1;130 58 SCF QSS1077 3.6 1;350∕1;130 52 SCF QSS1078 3.6 1;350∕1;130 45 SCF QSS1081 3.6 1;320∕1;130 21 SCF Gt + QSQB-4 3.6 1;150∕1;150 84 — QSS966 4.0 1;190∕1;190 60 SCF QS

The absolute error in the temperature determination is about �50 °C and that of pressure is about �0.2 GPa.TM, themaximum temperature during the run; TQ, quenched temperature; ND, not determined; QF, fine quenchmaterials from aqueous fluid; QM, glass

and/or quenchmaterials frommelt; QS, quenchmaterials from supercritical fluid; Am, amphibole; Cpx, clinopyroxene; Gt, garnet; Vd, void; F, aqueous fluid;M, silicate melt; SCF, supercritical fluid.*S series: radiography, QB series: long-duration constant-T quench experiments.†Amount of H2O in starting materials. Error: �10%.‡Two fluids were observed during temperature range shown in parentheses.

Fig. 1. X-ray radiographic images and photomicrographs ofrecovered samples. Abbreviations are the same as in Table 1.(A) X-ray radiographic image taken at 0.5 GPa and 1,120 °C(run S968). Both aqueous fluid and silicate melt can be seen.(B) The interface between aqueous fluid and silicate melt inA is indicated by the dotted curve. (C) X-ray radiographic imagetaken at 3.6 GPa and 1,125 °C (S1077). Only the supercriticalfluid phase is observed. The blocky and granular textures inradiographic images (A–C) are instrumental artifacts. (D) Recov-ered sample quenched from 0.5 GPa and 1,350 °C (S968). A largevoid space, which was filled with aqueous fluid under the ex-perimental conditions, can be seen. The surface of the QM glassis covered with tiny glass spheres, which are considered to bequenched products from aqueous fluid. (E) Polished surfaceof D, which is parallel to the X-ray path. Many tiny glass spherescan be seen in the epoxy (dark matrix). (F) Recovered samplequenched from 2.0 GPa and 920 °C (S1176). (G) Polished surfaceof F. (H) Enlargement of the rectangular area in G. (I) Recoveredsample quenched from 2.8 GPa and 1,350 °C (S997). (J) Polishedsurface of I. (K) Enlargement of QF in the recovered sampleS997. (L) Recovered sample quenched from 3.6 GPa and1,130 °C (S1077). (M) Polished surface of L. (N) Enlargementof QS in the recovered sample S1077. (O) Recovered samplequenched from 3.6 GPa and 1,130 °C (S1081). (P) Enlargementof the rectangular area in O. Note that there is no void spacein L,M, N, O, or P, and dense quenched materials are distributedin the entire region in these recovered capsules. The classifica-tion of quenched materials into two types (i.e., QF vs. QM) is nolonger possible in L–P, indicating that the conditions of theseruns are above the second critical endpoint. (E) Secondary elec-tron image. (G, H, J, K, N, and P) Backscattered electron images.

8178 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1010968108 Mibe et al.

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 8,

202

0

Page 3: Slab melting versus slab dehydration in subduction-zone ... · Slab melting versus slab dehydration in subduction-zone magmatism Kenji Mibea,b,c,1, Tatsuhiko Kawamotod, Kyoko N. Matsukagee,2,

can be distinguished at P less than 3.3 GPa (Table 1). In thesamples quenched at 2.0 GPa, void spaces and tiny glass spheres(approximately 10-μm diameter) were observed around spheri-cally shaped globules of a mixture composed of quench crystalsand glass. We consider the former void and tiny glass spheres tobe quenched from the aqueous fluids, and the mixture globulesfrom melts. In the recovered samples quenched from P valuesabove 3.4 GPa, the two-type discrete quenched materials wereno longer recognized; quench crystals were distributed through-out the entire sample capsule. These characteristics are consis-tent with the textures of the recovered samples quenched afterthe X-ray radiography. It can be concluded, therefore, that ourradiographic studies were done under equilibrium conditions.

Representative chemical compositions of the phases in therecovered products are listed in Table 2. The composition ofthe supercritical fluid of the run S1077 is close to that of the start-ing material, which confirms that our experiments were doneunder closed-system conditions. It was impossible to determinethe composition of aqueous fluid in the experimental runs atlower P, because the size and amount of quenched materials werenot sufficient to permit analysis. With increasing P, however, theamount of quenched material from the aqueous fluid increased.Compositions of aqueous fluid could be determined for the runsS997 and S1194, and they are roughly in the range of adakiticdacite on a dry basis (Table 2). Compositions of silicate meltsin these runs S997 and S1194 are basaltic on a dry basis, indicatingthat hydrous basaltic melts can coexist with adakitic dacite underthe experimental conditions. Because quenched materials fromboth the aqueous fluid and SCF are heterogeneously distributedin the epoxy matrix, the difference from 100 wt % does notnecessarily represent the water content in the measured phase.

DiscussionSecond Critical Endpoint in Basalt-H2O System. Our direct observa-tions by X-ray radiography and careful examination of thequenched samples indicate that the second critical endpoint

in the basalt-H2O system should occur at 3.4� 0.2 GPa (Fig. 2).Because the second critical endpoint occurs on the hydroussolidus, the T of the second critical endpoint is estimated to beat around 770� 50 °C, based on previous studies (38, 39). Fromthe runs S996 and S1083 at 3.0 GPa and also runs S1902 andS1193 at 3.3 GPa (Table 1), the water content of the aqueous fluidon the solvus between the aqueous fluid and the silicate meltat P between 3.0 and 3.3 GPa must be between 62 and 68 wt %H2O. Therefore, the critical water content at the second criticalendpoint is considered to be less than and probably close to65� 3 wt % H2O. Our results are consistent with most of theprevious studies: None of the previous experimental studies haveshown any direct evidence that the aqueous fluid and hydroussilicate melt could coexist in the basalt-H2O system above 3 GPa.

Kessel et al. (40) reported that the second critical endpoint in aK-free basalt-H2O system occurs at around 5.5 GPa, which is2.1 GPa higher than the result determined in the present study.Kessel et al. (40) determined the solubility of silicate componentsin H2Owith changing Tunder constant P. The solubility of silicatecomponents changes continuously with Twhen the experimentalP is above the second critical endpoint, whereas the solubilitychanges discontinuously with Twhen the experimental P is belowthe second critical endpoint. However, as can be seen from theflat of the solubility curve in Fig. 2C (XH2O ∼ 30 to 80), thechange in solubility is drastic in the case when the experimentalP is above the second critical endpoint. It is, therefore, difficultto determine which type of phase diagram (i.e., Fig. 2 B versus C)is valid under the experimental P by determining the solubilityonly. Using their indirect method (40), which involved someuncertainty in determining the solubility of silicates, it is consid-ered that Kessel et al. (40) were unable to determine preciselythe P of the second critical endpoint.

Depths of the Second Critical Endpoints and Their Bearing on theSubduction-Zone Magmatism. In most of the Earth’s present “nor-mal-type” subduction zones, basaltic magmas are considered to

Table 2. Composition of phases (wt %)

Run no. S968 S1176 S997P, T* 0.5, 1350 2.0, 920 2.8, 1350

Phase† M M100‡ Am Cpx M M100 M M100 F F100

n§ 10 — 7 5 11 — 35 — 34 —SiO2 45.40 (0.47) 50.81 44.42 (0.48) 47.68 (0.97) 48.50 (0.89) 57.81 36.33 (1.94) 49.98 18.21 (5.45) 63.49Al2O3 16.19 (0.13) 18.12 15.89 (0.49) 12.39 (1.63) 19.10 (0.38) 22.77 12.77 (0.48) 17.57 4.53 (1.40) 15.80FeO 5.03 (0.03) 5.62 9.22 (0.63) 8.54 (0.66) 4.42 (0.21) 5.27 7.40 (0.36) 10.18 1.82 (0.55) 6.35MgO 8.29 (0.14) 9.28 14.77 (0.94) 11.12 (1.64) 2.90 (0.16) 3.45 6.66 (0.37) 9.16 1.80 (0.56) 6.27CaO 12.41 (0.18) 13.89 10.88 (0.37) 19.68 (1.50) 8.10 (0.35) 9.65 8.32 (0.78) 11.44 1.78 (1.39) 6.21Na2O 1.96 (0.08) 2.19 2.00 (0.11) 0.44 (0.25) 0.80 (0.14) 0.95 1.16 (0.49) 1.59 0.48 (0.21) 1.67K2O 0.08 (0.01) 0.09 0.07 (0.01) 0.01 (0.01) 0.08 (0.02) 0.10 0.06 (0.02) 0.08 0.06 (0.02) 0.21Total 89.36 (0.56) 100.00 97.25 (0.40) 99.86 (1.42) 83.90 (1.29) 100.00 72.70 (2.32) 100.00 28.68 (8.19) 100.00

Run no. S1194 S1077 S1081P, T 3.0, 1010 3.6, 1130 3.6, 1130

Phase M M100 F F100 SCF SCF100 Gt SCF SCF100

n 19 — 34 — 11 — 9 15 —SiO2 35.62 (1.91) 48.78 16.04 (3.19) 65.23 33.37 (1.65) 51.31 40.70 (0.46) 40.59 (1.92) 55.90Al2O3 12.49 (0.74) 17.10 3.92 (0.79) 15.94 10.62 (0.54) 16.33 22.78 (0.19) 10.87 (0.49) 14.97FeO 6.10 (0.37) 8.36 1.11 (0.35) 4.51 6.51 (0.56) 10.01 14.93 (0.49) 5.78 (0.27) 7.96MgO 6.64 (0.47) 9.09 1.36 (0.49) 5.53 5.35 (0.64) 8.23 13.05 (0.46) 4.64 (0.31) 6.39CaO 10.02 (0.78) 13.73 1.27 (0.30) 5.17 7.43 (1.43) 11.42 8.97 (0.36) 9.00 (0.89) 12.39Na2O 2.05 (0.46) 2.81 0.82 (0.18) 3.34 1.61 (0.28) 2.47 0.02 (0.01) 1.37 (0.26) 1.89K2O 0.09 (0.02) 0.13 0.07 (0.02) 0.28 0.15 (0.03) 0.23 0.00 (0.00) 0.36 (0.07) 0.50Total 73.01 (3.61) 100.00 24.59 (4.44) 100.00 65.04 (2.24) 100.00 100.45 (0.75) 72.61 (2.64) 100.00

Numbers in parentheses represent one standard deviation.*Pressure in GPa, quenched temperature in °C.†Abbreviations are the same as in Table 1.‡Normalized to 100 wt %.§Number of analyses.

Mibe et al. PNAS ∣ May 17, 2011 ∣ vol. 108 ∣ no. 20 ∣ 8179

GEO

LOGY

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 8,

202

0

Page 4: Slab melting versus slab dehydration in subduction-zone ... · Slab melting versus slab dehydration in subduction-zone magmatism Kenji Mibea,b,c,1, Tatsuhiko Kawamotod, Kyoko N. Matsukagee,2,

be generated by the melting of mantle wedge peridotite by theaddition of aqueous fluid released from the subducting slab(1, 3–7). The mean depth below subduction-zone volcanoes to thetop of the deep seismic zone (Wadati–Benioff zone) has beengiven as 124� 38 km (ref. 41) or 108� 14 km (ref. 5). Recentstudies have shown that this depth ranges from 65–130 km(ref. 42) or 72–173 km, with a global average of 105 km (ref. 43).Based on our experimental results, the position of the secondcritical endpoint in the subducting slab should be located ataround 100-km depth, if the dominant fluid component is H2O(Fig. 3). Therefore, our experimental results strongly indicatethat the slab-derived fluid under most subduction-zone volcanoesis SCF rather than a separate aqueous fluid, hydrous melt, or bothof them. Only in the case when dehydration reactions of hydrousminerals in subducting slabs occur at depths shallower than100 km may aqueous fluid and/or hydrous melt be released fromthe slabs, depending on P-T conditions.

It is often suggested that the fluid released from the subductingslab plays an important role in the transport of trace elementsfrom the slab and in the generation of the geochemical charac-teristics of magmas in subduction zones (5, 44). In the fore-arc region, where the P is less than the second critical endpoint,the fluid mobile elements (FMEs) (i.e., aqueous fluid/mineralpartition coefficient, Dfluid∕mineral > 1, and/or fluid/melt partitioncoefficient, Dfluid∕melt > 1), such as B, Sr, and Pb (ref. 44) are

XH2O (wt%) XH2O (wt%)

CB

S+

M

F

P < 3.4 GPa

MSCF

P > 3.4 GPa

T1

T2

(SCF1)(SCF2)

600

1,000

1,400

20 40 60 800 100 20 40 60 800 100

MMM +++ FFF

S + F

SS +++ MMM +++ FFF SS +++ SSSCCCFFF

Tem

per

atu

re(°

C)

600

1,000

1,400

1,200

800

0 1 2 3 4 5Pressure (GPa)

A

Solidus

Critical curve

Tem

per

atu

re(°

C)

60

Pressure (GPa)XH2O (wt%)

3.63.4

3.240

D

600

700

800

900

20

S + SCF

Solidus

Critical curve

S + M

S + M + F

M + F

S + F

Tem

per

atu

re(°

C)

Fig. 2. Phase diagrams in the vicinity of the second critical endpoint in thebasalt-H2O system. (A) P-T diagram showing the experimental results andphase relations in the basalt-H2O system. Open bars represent the P-T con-ditions at which the two fluid phases (aqueous fluid and silicate melt) wereobserved, whereas filled bars represent the P-T conditions at which the twofluids were not observed in the radiographic images. These results indicatethat the P of the second critical endpoint can be bracketed between 3.3 and3.6 GPa. The amount of H2O in the starting materials of data plotted here is62� 6 wt % (Table 1). Solid curve is the hydrous solidus in the basalt-H2Osystem (38). The transitional T from filled bars to open bars does not neces-sarily coincide with the hydrous solidus. Because the spherical melt appearsupon heating when the experimental condition reaches well above the so-lidus and hence the degree of melting becomes quite high, it follows that allthe open bars in A are located above the hydrous solidus. The dotted line andthe star represent the estimated position of the critical curve and second cri-tical endpoint (3.4� 0.2 GPa) in the basalt-H2O system. (B) Isobaric phase dia-gram below the second critical endpoint. S: silicate solid, M: silicate melt, F:aqueous fluid. Note that the stability field of SþMþ F (red field) exists onthe lower-T side of the stability field of Mþ F (blue field), because the basalt-H2O system is the multicomponent system. This stability field of SþMþ Fextends toward the water-poor side. The P-T path in experiments belowthe second critical endpoint should encounter the stability field ofSþMþ F, if only a small amount of water is present in the system. The mis-cibility gap therefore cannot be overlooked, even when the water content inthe starting material is less than the critical water content at the experimen-tal P. (See ref. 35 for a detailed discussion.) The experimental P of data plottedhere ranges from 2.8 to 3.3 GPa. (C) Isobaric phase diagram above the secondcritical endpoint. The meaning of open and filled bars in B and C is the sameas inA. Filled squares in C represent the P-Tconditions at which only the SCF isstable fluid phase observed in the long-duration constant-T quench experi-ments. The experimental P of data plotted here ranges from 3.4 to 3.6 GPa.(D) P − T − XH2O phase diagram showing the position of the second criticalendpoint (star) in the basalt-H2O system. The second critical endpoint occursat around 3.4� 0.2 GPa and 770� 50 °C, where the hydrous solidus (Solidus)intersects the critical curve (dotted curve). Note that the solidus of the hy-drous basalt is no longer defined above the second critical endpoint, andthe aqueous fluid changes continuously to the silicate melt with T. Detailsof these phase diagrams have yet to be determined in future works.

Fig. 3. Schematic illustration showing the types of fluids and related volca-nic rocks in subduction zones. Not to scale. When dehydration occurs likepath A, both aqueous fluid and silicate melt could exist in both the subduct-ing slab and mantle wedge, because path A is shallower than the second cri-tical endpoint in both basalt-H2O (approximately 100 km) and peridotite-H2O (approximately 120 km) (ref. 35) systems. In this case, adakitic magmascould be produced by the slab melting. The depth of path B is between thesecond critical endpoint in the basalt-H2O system and that in the peridotite-H2O system. Therefore, the slab-derived fluid is SCF, whereas both aqueousfluid and silicate melt could exist in the mantle wedge. In path C, only SCF isstable in both the subducting slab and the mantle wedge, because path C isdeeper than the second critical endpoint in both basalt-H2O and peridotite-H2O systems. In paths B and C, normal types of island-arc volcanic rocks ofbasalt, andesite, dacite, and rhyolite (BADR) could be generated. The primi-tive basalt on the volcanic front in normal-type island-arc magmatism (pathB) could become tholeiitic (Th) or calc-alkaline (CA) as a result of the possibleremoval of alkaline-rich aqueous fluid from magma (35). On the other hand,the primitive magma on the back-arc side (path C) should remain alkaline-rich (BADR, A) because the extraction of the alkaline elements by thefluid-melt immiscibility is impossible beyond the second critical endpoint(35). Abbreviations for F, M, and SCF are the same as in Table 1.

8180 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1010968108 Mibe et al.

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 8,

202

0

Page 5: Slab melting versus slab dehydration in subduction-zone ... · Slab melting versus slab dehydration in subduction-zone magmatism Kenji Mibea,b,c,1, Tatsuhiko Kawamotod, Kyoko N. Matsukagee,2,

preferentially extracted from the slab to the mantle wedge. Withincreasing P, however, not only the FMEs but also the fluidimmobile elements (FIEs), such as Nb, Th, and REE, enter intothe SCF (45).

It has been reported that the ratio of FME to FIE in volcanicrocks decreases continuously from the volcanic front to theback-arc side (46, 47). These across-arc geochemical variationshave often been interpreted as a sign that the amount of fluidreleased from a slab decreases continuously with increasing depthof the Wadati–Benioff zone, assuming that the trace elementcomposition of the fluid remains constant with slab depth (46).However, the composition of slab-derived fluids cannot be con-stant with P (45). Our experimental results suggest that theseacross-arc geochemical variations are direct evidence that thefluids released from the subducting slab change from an aqueousfluid to SCF with increasing depth under volcanic arcs.

Concerning subduction zones where young and hot oceaniccrust is subducted, it has been argued that the melting of subduct-ing oceanic crust itself, rather than mantle wedge peridotite,could happen and generate adakitic magmas (48–50). One ofthe interesting features of adakites is that they are mainly situatedwhere the Wadati–Benioff zone is located at 70–90 km beneaththe volcanic arc (51, 52). It has been suggested (52) that thisdepth limitation of 90 km is related to the stability limit ofamphibole in the subducting oceanic crust. However, serpentinein the oceanic lithosphere could also carry water to the depthsof 150–200 km (ref. 53) and hence could be the source of waterthat might trigger the melting of oceanic crust at depths greaterthan 90 km.

Our results lead to an alternative explanation as to whyadakitic magmatism occurrences are generally situated wherethe depth of the Wadati–Benioff zone is shallower than 90 km.According to our phase diagrams (Fig. 2C), it would be easierto generate hydrous melt (i.e., adakites) below 3.4 GPa becausea relatively high T is required to produce melt (i.e., silicate-richSCF2 in Fig. 2C) above 3.4 GPa. Above 3.4 GPa, water-rich SCF(SCF1 in Fig. 2C) would be released successively from the slab tothe mantle wedge before the slab T becomes high enough. In thiscase, this low-T, water-rich SCF (SCF1) could become the sourceof water that generates “normal” basaltic magmas by the hydrousmelting of mantle wedge peridotite, instead of forming adakites.

Our experimental results lead to the conclusion that the depthlimit of 90 km for the formation of adakitic magmatism iscontrolled by the position of the second critical endpoint in thebasalt-H2O system, rather than the availability of water in theslab. For some subduction zones, it has been reported thatmagmatism changes from an adakite type to a normal subduc-tion-zone type with increasing depth of the Wadati–Benioff zone(51, 54). This change in the major-element chemistry of volcanicrocks is considered further proof that the slab melting is limitedto the fore-arc side, whereas SCF as a result of slab dehydrationcauses normal-type magmatism in most of the subduction zone.

Materials and MethodsExperiments were carried out using an X-ray radiography technique (37)together with a Kawai-type double-stage multianvil high-P apparatus(SPEED-1500) installed at SPring-8 (55). The composition of basalt used asa starting material was close to the average for mid-ocean-ridge basaltglasses (56) (SiO2: 50.4 wt %, Al2O3: 16.8 wt %, FeO�: 10.9 wt %, MgO:7.9 wt %, CaO: 12.0 wt %, Na2O: 1.9 wt %, and K2O: 0.1 wt %). Waterwas added as hydroxides and deionized water. The sample container wasan AuPd tube with a cylindrical single-crystal diamond placed on each end.Because diamonds are transparent to X-rays, the real-time images of thesamples under high-P and high-T conditions could be directly observed usingan X-ray camera.

The Twas monitored with aW5%Re-W26%Re thermocouple. Because thehigh-T junction of the thermocouple was far from the sample chamber for

the radiography experiments, a correction was made to estimate the realsample T from the thermocouple readings. The T values reported here arereal (or estimated) sample T values that can be directly compared to the otherstudies on phase relations. No correction for the effect of P on the thermo-couple electromotive force was applied. Detailed experimental methods andP-T calibrations have been reported elsewhere (30, 35).

The Twas increased or decreased under a constant load. The heating andcooling rates were about 5 °C∕s, which is comparable to those adopted in thestudies on the observation of critical behavior between aqueous fluid andsilicate melt using a hydrothermal diamond-anvil cell (27, 28, 32, 36). TheX-ray camera was on at all times during the experiment, and images wererecorded from the beginning of the heating (i.e., room T) until quenching.In the experiments at P values below the second critical endpoint, the experi-mental P-T path encountered the stability field of aqueous fluidþ silicatemelt with increasing Tunder a constant P. In this case, one fluid phase (eitheraqueous fluid or silicate melt) formed spheres in the other phase because ofthe differences in the interfacial tension between them. A round shape istherefore expected to be observed in the radiographic images. On the otherhand, in experiments beyond the second critical endpoint, a round shapeshould not be observed, because SCF is the only homogeneous fluid phaseexisting at high T. The P of the second critical endpoint can therefore bedetermined by bracketing the upper and the lower P limit where the roundshape is present or absent in the radiographic images. It has been reportedthat no visible change in the size of the silicate melt sphere in an aqueousfluid matrix was observed in the radiographic images while the T was keptat 1,180 °C for 45 min (35). This indicates that the modal abundance ofphases does not change with time during radiographic observation, suggest-ing that chemical equilibrium was attained.

After observation with the X-ray camera, the samples were quenchedunder the desired P-T conditions so that the X-ray radiographic images couldbe compared with the quenched textures. Phases in the recovered sampleswere identified with a JEOL JXA-8900 electron microprobe at the CornellCenter for Materials Research, Cornell University. Chemical composition wasanalyzed by wavelength dispersive spectrometers with a JEOL JXA-8800Relectron microprobe at the Earthquake Research Institute, University ofTokyo. The data were reduced using the correction procedure of Bence andAlbee (57). An acceleration voltage of 15 kV and a beam current of 12 nAwere employed. Counting time was 10 s for all elements. A focused electronbeam was employed for analysis of minerals. A 10- to 30-μm diameter elec-tron beam was used to obtain the composition of quenched materials fromthe silicate melt and aqueous fluid.

Because radiographic studies were done within a relatively short experi-mental duration, such as tens of minutes (30, 35), the equilibrium waschecked by conducting a series of long-duration constant-T quench experi-ments to see if we could reproduce quench textures of the aqueous fluids,hydrous melts, and SCFs, which were observed in the experimental productsquenched after the radiography observations. Experiments were carried outusing a Kawai-type multianvil high-P apparatus installed at the EarthquakeResearch Institute, University of Tokyo (58). In these long-duration constant-Tquench experiments, only one diamond end cap was used with an AuPdsample capsule, and the high-T junction of the thermocouple was placedimmediately adjacent to the AuPd sample capsule. The thermocouple read-ing was therefore exactly the same as the sample T. The duration of the long-duration constant-T quench experiments was 6 h. The recovered sampleswere polished and examined with a JEOL JXA-8800R electron microprobeand a JEOL JSM-5600LV scanning electron microscope at the EarthquakeResearch Institute.

ACKNOWLEDGMENTS. We thank K. Funakoshi, T. Fujii, A. Yasuda, S. Nakada,Y. Orihashi, T. Iizuka, S. Urakawa, Y. Nishihara, E. Takahashi, T. Okada,A. Nozawa, Y. Yoshimura, G. H. Gudfinnsson, M. Frank, H. Watson, S. Keshav,C. Hadidiacos, J. Sinnott, J. Hunt, and W. A. Bassett for helpful discussionsand technical support. This study was partly supported by the ResearchFellowships of the Japan Society for the Promotion of Science (JSPS) forYoung Scientists, the JSPS postdoctoral fellowships for research abroad,and grants from a joint research program at the Institute for Study ofthe Earth’s Interior, Okayama University, the Nissan Science Foundation,and the Japanese Ministry of Education, Science, Sports, and Culture. Thesynchrotron radiation experiments were performed at the BL04B1 in theSPring-8 with the approval of the Japan Synchrotron Radiation ResearchInstitute. Reviews by three anonymous referees significantly improved themanuscript.

Mibe et al. PNAS ∣ May 17, 2011 ∣ vol. 108 ∣ no. 20 ∣ 8181

GEO

LOGY

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 8,

202

0

Page 6: Slab melting versus slab dehydration in subduction-zone ... · Slab melting versus slab dehydration in subduction-zone magmatism Kenji Mibea,b,c,1, Tatsuhiko Kawamotod, Kyoko N. Matsukagee,2,

1. McBirney AR (1969) Compositional variations in Cenozoic calc-alkaline suites ofCentral America. Proceedings of the Andesite Conference (Oregon Department ofGeology and Mineral Industries, Prtland, OR), pp 185–189.

2. Wyllie PJ, Sekine T (1982) The formation of mantle phlogopite in subduction zonehybridization. Contrib Mineral Petrol 79:375–380.

3. Kushiro I (1983) On the lateral variations in chemical composition and volume ofQuaternary volcanic rocks across Japanese arcs. J Volcanol Geotherm Res 18:435–447.

4. SakuyamaM, Nesbitt RW (1986) Geochemistry of the Quaternary volcanic rocks of thenortheast Japan arc. J Volcan Geotherm Res 29:413–450.

5. Tatsumi Y, Eggins S (1986) Subduction Zone Magmatism (Blackwell Science, Boston).6. Mibe K, Fujii T, Yasuda A (1999) Control of the location of the volcanic front by

aqueous fluid connectivity in the mantle wedge. Nature 401:259–262.7. Hattori KH, Guillot S (2003) Volcanic fronts form as a consequence of serpentine

dehydration in the forarc mantle wedge. Geology 31:525–528.8. Elliott T, Plank T, Zindler A, White W, Bourdon B (1997) Element transport from slab to

volcanic front at the Mariana arc. J Geophys Res 102:14991–15019.9. Class C, Miller DM, Goldstein SL, Langmuir CH (2000) Distinguishing melt and fluid

subduction components in Umnak Volcanics, Aleutian Arc. Geochem Geophys Geosyst1(6):1004, 10.1029/1999GC000010.

10. Prouteau G, Scaillet B, Pichavant M, Maury R (2001) Evidence for mantle meta-somatism by hydrous silicic melts derived from subducted oceanic crust. Nature410:197–200.

11. Grove TL, Parman SW, Bowring SA, Price RC, Baker MB (2002) The role of an H2O-richfluid component in the generation of primitive basaltic andesites and andesites fromMt. Shasta region, N California. Contrib Mineral Petrol 142:375–396.

12. Wysoczanski RJ, et al. (2006) Volatile contents of Kermadec Arc—Havre Trough pillowglasses: Fingerprinting slab-derived aqueous fluids in the mantle sources of arc andback-arc lavas. J Volcan Geotherm Res 152:51–73.

13. Kennedy GC, Wasserburg GJ, Heard HC, Newton RC (1962) The upper three phaseregion in the system SiO2-H2O. Am J Sci 260:501–521.

14. Burnham CW, Jahns RH (1962) A method for determining the solubility of water insilicate melts. Am J Sci 260:721–745.

15. Hamilton DL, Burnham CW, Osborn EF (1964) The solubility of water and effects ofoxygen fugacity and water content on crystallization in mafic magmas. J Petrol5:21–39.

16. Eggler DH, Burnham CW (1984) Solution of H2O in diopside melts: A thermodynamicmodel. Contrib Mineral Petrol 85:58–66.

17. Anderson GM, Burnham CW (1965) The solubility of quartz in supercritical water. Am JSci 263:494–511.

18. Nakamura Y, Kushiro I (1974) Composition of the gas phase in Mg2SiO4-SiO2-H2O at15 kbar. Year Book Carnegie Inst Washington 73:255–258.

19. Schneider ME, Eggler DH (1986) Fluids in equilibrium with peridotite minerals:Implications for mantle metasomatism. Geochim Cosmochim Acta 50:711–724.

20. Ryabchikov ID (1988) Geochemical Evolution of Earth’s Mantle (Nauka, Moscow), (inRussian), p 37.

21. Paillat O, Elphick SC, Brown WL (1992) The solubility of water in NaAlSi3O8 melts:A re-examination of Ab-H2O phase relationships and critical behaviour at highpressures. Contrib Mineral Petrol 112:490–500.

22. Ryabchikov ID (1993) Fluid transport of ore metals in ultramafic mantle rocks. Proceed-ings of the Eighth Quadrennial International Association on the Genesis of OreDeposits (IAGOD) Symposium (Schweizerbartsche Ferlagbuchhandlung, Stuttgart),pp 425–433.

23. Manning CE (1994) The solubility of quartz in H2O in the lower crust and uppermantle. Geochim Cosmochim Acta 58:4831–4839.

24. Stalder R, Ulmer P, Thompson AB, Gunther D (2001) High pressure fluids in the systemMgO-SiO2-H2O under upper mantle conditions. Contrib Mineral Petrol 140:607–618.

25. Mibe K, Fujii T, Yasuda A (2002) Composition of aqueous fluid coexisting with mantleminerals at high pressure and its bearing on the differentiation of the Earth’s mantle.Geochim Cosmochim Acta 66:2273–2285.

26. Niggli P (1920) Die leichtfluchtigen Bestandteile im Magma (BG Teubner, Leipzig).27. Shen A, Keppler H (1997) Direct observation of complete miscibility in the albite-H2O

system. Nature 385:710–712.28. Bureau H, Keppler H (1999) Complete miscibility between silicate melts and hydrous

fluids in the upper mantle: Experimental evidence and geochemical implications.Earth Planet Sci Lett 165:187–196.

29. Wyllie PJ, Ryabchikov ID (2000) Volatile components, magmas, and critical fluids inupwelling mantle. J Petrol 41:1195–1206.

30. Mibe K, et al. (2004) Determination of the second critical end point in silicate-H2Osystems using high-pressure and high-temperature X-ray radiography. GeochimCosmochim Acta 68:5189–5195.

31. Kawamoto T, et al. (2004) Mg∕Si ratios of aqueous fluids coexisting with forsterite andenstatite based on the phase relations in the Mg2SiO4-SiO2-H2O system. Am Mineral89:1433–1437.

32. Kawamoto T (2006) Hydrous phases and water transport in the subducting slab. RevMineral Geochem 62:273–289.

33. Hermann J, Spandler C, Hack A, Korsatov AV (2006) Aqueous fluids and hydrous meltsin high-pressure and ultra-high pressure rocks: Implications for element transfer insubduction zones. Lithos 92:399–417.

34. Hack AC, Thompson AB, Aerts M (2007) Phase relations involving hydrous silicatemelts, aqueous fluids, and minerals. Rev Mineral Geochem, eds A Liebscher andCA Heinrich 65:129–185 (in Fluid-fluid interactions).

35. Mibe K, et al. (2007) Second critical endpoint in the peridotite-H2O system. J GeophysRes 112:B03201, 10.1029/2005JB004125.

36. Mibe K, Chou I-M, Bassett WA (2008) In situ Raman spectroscopic investigationof the structure of subduction-zone fluids. J Geophys Res 113:B04208, 10.1029/2007JB005179.

37. Kanzaki M, et al. (1987) A new technique to measure the viscosity and density ofsilicate melts at high pressure. High-Pressure Research in Mineral Physics, eds MHManghnani and Y Syono (Am Geophys Union, Washington, DC), pp 195–200.

38. Lambert IB, Wyllie PJ (1972) Melting of gabbro (quartz eclogite) with excess water to35 kilobars, with geological applications. J Geol 80:693–708.

39. Green TH (1982) Anatexis of mafic crust and high pressure crystallization of andesite.Andesites: Orogrenic Andesites and Related Rocks, ed RS Thorpe (Wiley, New York), pp465–487.

40. Kessel R, Ulmer P, Pettke T, Schmidt MW, Thompson AB (2005) The water-basalt systemat 4 to 6 GPa: Phase relations and second critical endpoint in a K-free eclogite at 700 to1400 °C. Earth Planet Sci Lett 237:873–892.

41. Gill JB (1981) Orogenic Andesite and Plate Tectonics (Springer, Berlin).42. England P, Engdahl R, Thatcher W (2004) Systematic variation in the depths of slabs

beneath arc volcanoes. Geophys J Int 156:377–408.43. Syracuse EM, Abers GA (2006) Global compilation of variations in slab depth beneath

arc volcanoes and implications. Geochem Geophys Geosyst 7:Q05017, 10.1029/2005GC001045.

44. Keppler H (1996) Constraints from partitioning experiments on the composition ofsubduction-zone fluids. Nature 380:237–240.

45. Kessel R, SchmidtMW, Ulmer P, Pettke T (2005) Trace element signature of subduction-zone fluids, melts and supercritical liquids at 120–180 km depth. Nature 437:724–727.

46. Ishikawa T, Nakamura E (1994) Origin of the slab component in arc lavas from across-arc variation of B and Pb isotopes. Nature 370:205–208.

47. Ryan JG, Morris J, Tera F, Leeman WP, Tsvetkov A (1995) Cross-arc geochemical varia-tions in the Kurile Arc as a function of slab depth. Science 270:625–627.

48. Kay RW (1978) Aleutian magnesian andesites: Melts from subducted Pacific Oceancrust. J Volcanol Geotherm Res 4:117–132.

49. Defant MJ, Drummond MS (1990) Derivation of some modern arc magmas by meltingof young subducted lithosphere. Nature 347:662–665.

50. Rapp RP, Watson EB (1995) Dehydration melting of metabasalt at 8–32 kbar: Implica-tions for continental growth and crust-mantle recycling. J Petrol 36(4):891–931.

51. Defant MJ, Drummond MS (1993) Mount St. Helens: Potential example of the partialmelting of the subducted lithosphere in a volcanic arc. Geology 21:547–550.

52. Martin H (1999) Adakitic magmas: Modern analogues of Archaean granitoids. Lithos46:411–429.

53. Ulmer P, Trommsdorff V (1995) Serpentine stability to mantle depths and subduction-related magmatism. Science 268:858–861.

54. Samsonov AV, et al. (2005) The relationship between adakitic, calc-alkaline volcanicrocks and TTGs: Implications for the tectonic setting of the Karelian greenstone belts,Baltic Shield. Lithos 79:83–106.

55. Utsumi W, et al. (2002) High-pressure science with a multi-anvil apparatus at SPring-8.J Phys Condens Matter 14:10497, 10.1088/0953-8984/14/44/322.

56. Hofmann AW (1988) Chemical differentiation of the Earth: The relationship betweenmantle. continental crust, and oceanic crust. Earth Planet Sci Lett 90:297–314.

57. Bence AE, Albee AL (1968) Empirical correction factors for the electron microanalysisof silicates and oxides. J Geol 76:382–403.

58. Mibe K, Yoshino T, Ono S, Yasuda A, Fujii T (2003) Connectivity of aqueous fluid ineclogite and its implications for fluid migration in the Earth’s interior. J GeophysRes 108(B6):2295, 10.1029/2002JB001960.

8182 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1010968108 Mibe et al.

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 8,

202

0


Top Related