TECHNICAL COMMENT

HIGH-PRESSURE PHYSICS

Comment on ldquoObservation of theWigner-Huntington transitionto metallic hydrogenrdquoXiao-Di Liu1 Philip Dalladay-Simpson2 Ross T Howie2 Bing Li23 Eugene Gregoryanz124

Dias and Silvera (Research Article 17 February 2017 p 715) claim the observation of theWigner-Huntington transition to metallic hydrogen at 495 gigapascals We show thatneither the claims of the record pressure nor the phase transition to a metallic state aresupported by data and that the data contradict the authorsrsquo own unconfirmed previous results

Dias and Silvera (1) present only one ex-perimental run claiming a record pres-sure of 495 GPa The paper presents twofigures of the phase diagram iPhone photosof the sample and the deduced reflectiv-

ity of four wavelengths only at the highest pres-sure point The supplementary materials provideprocessed and fitted infrared (IR) absorptionspectra from 135 to 335 GPa the pressure versusforce curve assuming a linear dependence andthe Raman spectra of the stressed diamond from1200 to 2200 cmminus1 The absence of data com-bined with the uncritical claims have led to theunprecedented three comments (2ndash4) writtenwithin a month of publication of (1)In the past 5 years we have conducted ~120 ex-

periments on hydrogen reaching above 200 GPa(5ndash10) In ~30 runs out of 120 the pressure ex-ceeded 300 GPa and in only 5 out of 120 thepressure exceeded 350 GPa The extensive statis-tics show that the diamond culet sizes of 30 mmdiameter [used in (1)] could be used to reachmaximum pressures of ~315 plusmn 10 GPa with theprobability of 20 To reach pressures close to400 GPa with lower probability of 10 theculet sizes of 15 mm must be used with the sam-ple contracting to 2 to 3 mm at the highest pres-sure Our statistics are in excellent agreementwith other groups working on hydrogen at highpressure (11ndash13) The authors of (1) had two ex-perimental runs in the past year using culets of30 mm diameter (1 14) claiming the unsubstan-tiated pressures of 420 and 495 GPaDias and Silvera (1) claim that a combina-

tion of annealing fine polishing and coatingthe culet with Al2O3 has led to an increase inthe maximum pressure with larger culets (and

samples) compared with previous studies (5ndash13)Those techniques would likely decrease the prob-ability of the premature failure due to hydrogendiffusion but there is no supportive evidencethat these techniques would improve the me-chanical stability of diamond The record pres-sure of 600 GPa was recently achieved on metalsusing a novel double-stage approach with the

pressure-generating area having a diameter of3 mm (15) [smaller by a factor of 100 than thearea in (1)]Figure 1 shows the pressure-versus-load curve

from (1) (4) and our own data Dias and Silvera(1) use three different nonoverlapping calibra-tion methods at ~100 and ~300 GPa Measuringpressure by estimating the load is not a directmethod as it does not probe the sample andorcalibrant in situ The loading curve is unique foreach experimental run and depends on the sizeof the culets angles of the bevels and compress-ibility of the gasket and sample The dependencycannot be linear as shown in (1) but alwaysconsists of three distinct regimes each of whichis sublinear with differing gradients (i) plasticdeformation of the gasket (ii) sharp rise of pres-sure beyond the plastic deformation followedby (iii) the much slower pressure increase dueto the bending of the diamondsTo make the dependence linear Dias and

Silvera (1) take a point from a different exper-iment (14) rescaling it from 420 to 400 GPa andplot four additional meaningless points of ldquovisualobservationrdquo Pressures of 495 and 420 GPa werededuced from a single Raman spectrum whichdoes not cover a wide energy range showing thesignal from hydrogen andor pressure-inducedfluorescence This rules out any critical assessment

RESEARCH

Liu et al Science 357 eaan2286 (2017) 25 August 2017 1 of 2

1Key Laboratory of Materials Physics Institute of SolidState Physics Chinese Academy of Sciences Hefei China2Center for High Pressure Science and Technology AdvancedResearch Shanghai China 3Center for the Study of Matterat Extreme Conditions Department of Mechanical andMaterials Engineering Florida International University MiamiFL USA 4Centre for Science at Extreme Conditions andSchool of Physics and Astronomy University of EdinburghEdinburgh UKCorresponding author Email egregoryanzedacuk

Inte

nsity

(A

rb U

nits

)

60004000Wavenumbers (cm-1)

272 GPa 83 K

338 GPa 83 K

301 GPa 83 K

314 GPa 83 K

400

300

200

100

0

Pre

ssur

e (G

Pa)

1050Load (kN)

Culet Gasket Sample 20 microm Re None 30 microm W Cr 30 microm Re MgO 30 microm Re O2

25 microm Re H2 (4)

400

300

200

100

Pre

ssur

e (G

Pa)

8006004002000Screw Rotation Angle (deg)

Ruby

IRVibron

VisualObservations

400 GPa Diamond PhononPrevious Experiment (12)

495 GPa Diamond Phonon

Fig 1 Pressure versus load (A) Our data and data from (4) (B) Data from Dias and Silvera (1)(C) Some raw IR absorption spectra provided but not plotted in (1) (in color) The black curvesare normalized spectra taken from figure S1 in (1) We note that the intensity and frequency of theraw spectra at 338 GPa do not match those displayed in figure S1 in (1)

on June 29 2018

httpsciencesciencemagorg

Dow

nloaded from

that the sample did not diffuse out and that thepeak assigned to the stressed diamond is notdue to any other factorsThe IR absorption data are consistent with

the loss of the sample and contradict the authorsrsquoown previous claims (14) In Fig 1 we plot theraw data provided by but not plotted in (1) to-gether with the normalized spectra The samplewas clearly diffusing out between the claimedpressures of 314 and 338 GPa and ultimatelywas completely lost above this The lack of IRtransmission above 338 GPa and the fact thatthe ldquosamplerdquo appears dark at 415 GPa correlatewith the gasket flow scenario we describe belowIn (14) the authors present the same IR data upto 420 GPa but do not provide any explanationfor why their results are so different in two con-secutive experiments Also no explanation isgiven as to why there was no IR transmission

absorption between 335 and 495 GPa the pres-sure range in which hydrogen is semiconductingThe analysis of the photos provided is also

very consistent with the loss of the sample at anearlier stage in the experiment (Fig 2) The sam-ple at 415 GPa is expanded under pressure and isslightly bigger than at 205 GPa whereas it is thesame size at 495 GPa In (8) we demonstratewith photos of the sample up to 280 GPa theflow of the metal gasket in the sample chamberas H2 diffuses out of the chamber As the chambercloses the thin layer of hydrogen between thecollapsed gasket and the diamond gives the darkappearance [compare figure 2B in (1) and figure 3in (8)] With the increasing pressure all hydro-gen diffuses out of the chamber and the darkarea which could partially be rhenium hydridebecomes shiny as it bridges the anvils but wouldhave different reflectivity from the rest of the

gasket as in figure 2C in (1) [see also the photoof ldquometallic hydrogenrdquo in figure 3D in (11)] Be-cause the raw reflectivity spectra or their pres-sure dependence are not given it is not possibleto confirm the self-consistency of the measure-ments as well as the claim of the increased re-flectivity of the ldquosamplerdquo versus the rest of thegasketThe presented phase diagrams [figures 1 and

4 in (1)] show an unconfirmed H2-PRE phasefor which no data are presented in the paperThe existence of this phase was surmised in (14)and even though no supporting data are givenDias and Silvera (1) state ldquoafter the 335-GPa pres-sure point resulted in our sample starting to turnblack (Fig 2B) as it transitioned into the H2-PREphaserdquo which the authors claimed to be trans-parent [see figure 3S in (14)] Furthermore (1)cites one of the authorsrsquo own publications on themelting curve but without any explanation plotsthe smooth melting curve (see Fig 1) contradic-tory to the authorsrsquo earlier claim about the ex-istence of the sharp peak at 60 GPa

REFERENCES AND NOTES

1 R P Dias I F Silvera Science 355 715ndash718 (2017)2 A Goncharov V Struzhkin Comment on Observation of the

Wigner-Huntington transition to metallic hydrogen arXiv170204246 [cond-matmtrl-sci] (13 February 2017)

3 M Eremets I Frozdov Comments on the claimed observationof the Wigner-Huntington transition to metallic hydrogenarXiv170205125 [cond-matmtrl-sci] (16 February 2017)

4 P Loubeyre F Occelli P Dumas Comment on Observationof the Wigner-Huntington transition to metallic hydrogenarXiv170207192 [cond-matother] (23 February 2017)

5 R T Howie C L Guillaume T Scheler A F GoncharovE Gregoryanz Phys Rev Lett 108 125501 (2012)

6 R Howie T Scheler C L Guillaume E Gregoryanz Phys RevB 86 214104 (2012)

7 R Howie E Gregoryanz A F Goncharov J Appl Phys 114073505 (2013)

8 R T Howie I B Magdău A F Goncharov G J AcklandE Gregoryanz Phys Rev Lett 113 175501 (2014)

9 R T Howie P Dalladay-Simpson E Gregoryanz Nat Mater14 495ndash499 (2015)

10 P Dalladay-Simpson R T Howie E Gregoryanz Nature 52963ndash67 (2016)

11 M I Eremets I A Troyan Nat Mater 10 927ndash931 (2011)12 P Loubeyre F Occelli P Dumas Phys Rev B 87 134101 (2013)13 C S Zha Z Liu M Ahart R Boehler R J Hemley Phys Rev

Lett 110 217402 (2013)14 R Dias O Noked I F Silvera New low temperature phase in

dense hydrogen The phase diagram to 421 GPa arXiv160302162 [cond-matother] (7 March 2016)

15 L Dubrovinsky N Dubrovinskaia V B PrakapenkaA M Abakumov Nat Commun 3 1163 (2012)

ACKNOWLEDGMENTS

The authors are grateful to P Loubeyre for providing his data

101126scienceaan2286

Liu et al Science 357 eaan2286 (2017) 25 August 2017 2 of 2

Fig 2 Photos of the sample from (1) and (14) (A) Photo of the semitransparent sample at420 GPa from (14) (B) Photo of the dark sample at 415 GPa from (1) (C and D) Photos of thesamples at 205 and 495 GPa from (1) We have used the 30 mm stated by (1) as being the sizeof the culet as the scale to estimate the size of the sample In the vertical direction the samplereaches 15 mm which is almost twice as large as the 8 to 10 mm stated in (1) For comparison thesize of the hydrogen sample at 390 GPa was ~2 mm see figure 3 in (10)

RESEARCH | TECHNICAL COMMENTon June 29 2018

httpsciencesciencemagorg

Dow

nloaded from

Comment on Observation of the Wigner-Huntington transition to metallic hydrogenXiao-Di Liu Philip Dalladay-Simpson Ross T Howie Bing Li and Eugene Gregoryanz

DOI 101126scienceaan2286 (6353) eaan2286357Science

ARTICLE TOOLS httpsciencesciencemagorgcontent3576353eaan2286

CONTENTRELATED

httpsciencesciencemagorgcontentsci3576353eaan2671fullhttpsciencesciencemagorgcontentsci3556326715full

REFERENCES

httpsciencesciencemagorgcontent3576353eaan2286BIBLThis article cites 11 articles 1 of which you can access for free

PERMISSIONS httpwwwsciencemagorghelpreprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAASSciencelicensee American Association for the Advancement of Science No claim to original US Government Works The title Science 1200 New York Avenue NW Washington DC 20005 2017 copy The Authors some rights reserved exclusive

(print ISSN 0036-8075 online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

on June 29 2018

httpsciencesciencemagorg

Dow

nloaded from