Re: [QE-users] Error during calculation U parameter with hp.x
Dear Maria, I think that the error should disappear if you use the corresponding `ibrav` (plus the other associated parameters, like a,b,c,…), and NOT the more general ibrav=0. There are several programs allowing you to get these parameters. For instance, `cell2ibrav.x` which comes with QE; but there are also other tools, e.g. in Python (I believe `qe_tools` or similar). Hope this helps. Best, Lorenzo Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 3 giu 2024, alle ore 10:18, Maria Francisca Coelho Queirós > ha scritto: > > Dear QE Members, > > I have been trying to calculate the U parameter for my system (Pr2O2SO4) > using hp.x. However the calculation suddenly crashes and the error reported > below comes up: >== > SOLVE THE LINEAR SYSTEM > == > > atom # 1 q point # 1 iter # 1 > > Pert. # 1: Fermi energy shift (Ry) = 9.3031E-01-2.3327E-11 > Message from routine scale_sym_ops: > found rotation not compatible with FFT grid > Message from routine scale_sym_ops: > found rotation not compatible with FFT grid > Message from routine scale_sym_ops: > found rotation not compatible with FFT grid > Message from routine scale_sym_ops: > found rotation not compatible with FFT grid > Message from routine scale_sym_ops: > found rotation not compatible with FFT grid > Message from routine scale_sym_ops: > found rotation not compatible with FFT grid > Message from routine scale_sym_ops: > found rotation not compatible with FFT grid > Message from routine scale_sym_ops: > found rotation not compatible with FFT grid > > > %% > Error in routine scale_sym_ops (8): > incompatible FFT grid > > %% > > stopping ... > > > > I have already checked with an example of the LiCoO2 system and it worked, > without any error problem. > I attach the input files, and appreciate any help regarding this issue: > > calculation = 'scf' > etot_conv_thr = 9.00d-05 > forc_conv_thr = 1.00d-04 > outdir = './tmp' > prefix = 'I_42m' > pseudo_dir = '/mnt/beegfs/users/mqueiros/workdir/praseodymium' > tprnfor = .true. > tstress = .true. > !verbosity = 'high' > / > > degauss = 0.001 > ecutwfc = 130 > ecutrho = 780 > ibrav = 0 > nat = 9 > nosym = .false. > nspin = 2 > ntyp = 3 > occupations = 'smearing' > smearing = 'mp' > !starting_magnetization(2) = 1.00d-01 > starting_magnetization(1) = 5.3846153846d-01 > !starting_magnetization(3) = 1.00d-01 > / > > conv_thr = 1.80d-12 > electron_maxstep = 500 > mixing_beta = 0.4 > / > >ion_dynamics = 'bfgs' > !!ion_dynamics='damp' > / > >!cell_dynamics='damp-w' >cell_dynamics='bfgs' > / > ATOMIC_SPECIES > Pr 140.90765 Pr.pbe-spdfn-kjpaw_psl.1.0.0.UPF > O 15.9994 O.pbe-n-kjpaw_psl.1.0.0.UPF > S 32.065 S.pbe-n-kjpaw_psl.1.0.0.UPF > ATOMIC_POSITIONS crystal > Pr0.32636312140.32636312140.6527262428 > Pr0.67363687860.67363687860.3472737572 > O 0.14448753850.73612298550.8806105240 > O 0.73612298550.14448753850.8806105240 > O 0.85551246150.85551246150.1193894760 > O 0.26387701450.26387701450.1193894760 > O 0.750.250.50 > O 0.250.750.50 > S 0.000.000.00 > K_POINTS automatic > 8 8 6 0 0 0 > > CELL_PARAMETERS (angstrom) >4.167827013 -0.0 0.0 >-0.0 4.167827013 0.0 >-2.083913507 -2.083913507 7.361906677 > HUBBARD (ortho-atomic) > U Pr-4f 1.d-8 > > >prefix = 'I_42m' >outdir = './tmp/' >iverbosity = 1 >nq1 = 2, nq2 = 2, nq3 = 1 >conv_thr_chi = 1.0d-3 >dist_thr = 5.D-3 >alpha_mix(1)=0.1 > / > > I have already tried to define the values
Re: [QE-users] Error while phonon calculation (ph.x)
Dear Hasan,As correctly suggested, do use fixed occupations for insulators, as you don’t need to use smearing (and you should not for phonons on insulators).Best,LorenzoInviato da iPhoneIl giorno 26 mag 2024, alle ore 06:16, Md. Jahid Hasan Sagor ha scritto:Dear Lorenzo,I have changed the pbesol potential too in my next run. Usually, I use (https://www.materialscloud.org/work/tools/qeinputgenerator) the website to generate input files and download pseudopotential from there. Would you let me know how I can know which type of potential is best for semiconductors/Insulators?Thank youBestHasan On Fri, May 24, 2024 at 5:13 PM Lorenzo Bastonero <lbast...@uni-bremen.de> wrote:If you’re using PBEsol there is a chance that it (wrongly) describes it as a (band) metal.Lorenzo Lorenzo BastoneroPhD StudentU Bremen Excellence Chair,Bremen Center for Computational Materials Science,and MAPEX Center for Materials and ProcessesUniversity of BremenFaculty of Production EngineeringTAB-Building, Room 3.32Am Fallturm 128359 Bremen, DEhttp://www.hmi.uni-bremen.de/ Il giorno 24 mag 2024, alle ore 22:53, Md. Jahid Hasan Sagor <md.sa...@maine.edu> ha scritto:Hi Lorenzo,Thank you for your response. But Here I am using GaAs, it's a semiconductor, but still I am getting the message.ThanksBestHasanOn Fri, May 24, 2024 at 4:51 PM Lorenzo Bastonero <lbast...@uni-bremen.de> wrote:Hello,Dielectric constant in metals is “infinite”. There’s no static response to a static electric field (you would generate an electric current).As such, any long-range correction to the force-constants is not needed in metals, if that is the reason why you were trying to compute these quantities.Best,Lorenzo Lorenzo BastoneroPhD StudentU Bremen Excellence Chair,Bremen Center for Computational Materials Science,and MAPEX Center for Materials and ProcessesUniversity of BremenFaculty of Production EngineeringTAB-Building, Room 3.32Am Fallturm 128359 Bremen, DEhttp://www.hmi.uni-bremen.de/ Il giorno 24 mag 2024, alle ore 22:20, Md. Jahid Hasan Sagor <md.sa...@maine.edu> ha scritto:Hi Dear QE Experts,I am trying to calculate second order force constants for GaAs. I successfully ran pw.x, but after that when I ran ph.x, I got the following error. ask # 0 from phq_readin : error # 1 no electric. field with metalsInput files are attached here. I have used epsil = .true. for phonon calculation to calculate born charge and dielectric constant. Would you please give some solutions for this?BestMd Jahid HasanMechanical EnggPhD StudentUniversity of Maine <ph.in><scf.in>___The Quantum ESPRESSO community stands by the Ukrainianpeople and expresses its concerns about the devastatingeffects that the Russian military offensive has on theircountry and on the free and peaceful scientific, cultural,and economic cooperation amongst peoples___Quantum ESPRESSO is supported by MaX (www.max-centre.eu)users mailing list users@lists.quantum-espresso.orghttps://lists.quantum-espresso.org/mailman/listinfo/users___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Error while phonon calculation (ph.x)
If you’re using PBEsol there is a chance that it (wrongly) describes it as a (band) metal. Lorenzo Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 24 mag 2024, alle ore 22:53, Md. Jahid Hasan Sagor > ha scritto: > > Hi Lorenzo, > > Thank you for your response. But Here I am using GaAs, it's a semiconductor, > but still I am getting the message. > > Thanks > > Best > Hasan > > On Fri, May 24, 2024 at 4:51 PM Lorenzo Bastonero <mailto:lbast...@uni-bremen.de>> wrote: >> Hello, >> >> Dielectric constant in metals is “infinite”. There’s no static response to a >> static electric field (you would generate an electric current). >> As such, any long-range correction to the force-constants is not needed in >> metals, if that is the reason why you were trying to compute these >> quantities. >> >> Best, >> Lorenzo >> >> >> Lorenzo Bastonero >> >> PhD Student >> U Bremen Excellence Chair, >> Bremen Center for Computational Materials Science, >> and MAPEX Center for Materials and Processes >> >> University of Bremen >> Faculty of Production Engineering >> TAB-Building, Room 3.32 >> Am Fallturm 1 >> 28359 Bremen, DE >> >> http://www.hmi.uni-bremen.de/ >> >>> Il giorno 24 mag 2024, alle ore 22:20, Md. Jahid Hasan Sagor >>> mailto:md.sa...@maine.edu>> ha scritto: >>> >>> Hi Dear QE Experts, >>> >>> I am trying to calculate second order force constants for GaAs. I >>> successfully ran pw.x, but after that when I ran ph.x, I got the following >>> error. >>> >>> ask # 0 from phq_readin : error # 1 no electric. field with metals >>> >>> Input files are attached here. I have used epsil = .true. for phonon >>> calculation to calculate born charge and dielectric constant. >>> Would you please give some solutions for this? >>> >>> Best >>> Md Jahid Hasan >>> Mechanical Engg >>> PhD Student >>> University of Maine >>> >>> http://ph.in/>>>> <http://scf.in/>>___ >>> The Quantum ESPRESSO community stands by the Ukrainian >>> people and expresses its concerns about the devastating >>> effects that the Russian military offensive has on their >>> country and on the free and peaceful scientific, cultural, >>> and economic cooperation amongst peoples >>> ___ >>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu >>> <http://www.max-centre.eu/>) >>> users mailing list users@lists.quantum-espresso.org >>> <mailto:users@lists.quantum-espresso.org> >>> https://lists.quantum-espresso.org/mailman/listinfo/users >> >> ___ >> The Quantum ESPRESSO community stands by the Ukrainian >> people and expresses its concerns about the devastating >> effects that the Russian military offensive has on their >> country and on the free and peaceful scientific, cultural, >> and economic cooperation amongst peoples >> ___ >> Quantum ESPRESSO is supported by MaX (www.max-centre.eu >> <http://www.max-centre.eu/>) >> users mailing list users@lists.quantum-espresso.org >> <mailto:users@lists.quantum-espresso.org> >> https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Error while phonon calculation (ph.x)
Hello, Dielectric constant in metals is “infinite”. There’s no static response to a static electric field (you would generate an electric current). As such, any long-range correction to the force-constants is not needed in metals, if that is the reason why you were trying to compute these quantities. Best, Lorenzo Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 24 mag 2024, alle ore 22:20, Md. Jahid Hasan Sagor > ha scritto: > > Hi Dear QE Experts, > > I am trying to calculate second order force constants for GaAs. I > successfully ran pw.x, but after that when I ran ph.x, I got the following > error. > > ask # 0 from phq_readin : error # 1 no electric. field with metals > > Input files are attached here. I have used epsil = .true. for phonon > calculation to calculate born charge and dielectric constant. > Would you please give some solutions for this? > > Best > Md Jahid Hasan > Mechanical Engg > PhD Student > University of Maine > > ___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Naive question of choosing the number of unit cell to calculate vibrational frequencies
Dear Hy Lu, Gamma phonons are obtained within the unit cell only, so no need of supercells (for the neighborhood you add the non-analytical correction, so to account for long range interactions; see eg Gonze 1997 PRB or Baroni 2001 Rev Mod Phys). As it is a perovskite structure though, I guess you might have some tilting of octahedra, which is something not accounted for in the unit cell, thus I assume the need of supercell (although I admit I didn’t read the references you sent). If the latter is the case, then it really depends a lot on the material. Moreover, if you are trying to match some experiments you would need to carefully understand the experimental conditions and model your structure accordingly. I think you can start with the unit cell and work your understanding of the problem up. The use of DFPT (ph.x code) will allow you to compute the frequencies in a cheap and user friendly manner. Note that no temperature effects are included in the harmonic approximation (“bare” phonons). HTH, Lorenzo Bastonero Inviato da iPhone > Il giorno 24 ott 2023, alle ore 17:31, ludwigboltzmann.s...@nycu.edu.tw ha > scritto: > > > "We have 144 normal modes, corresponding to thenumber of degrees of freedom > of 4 CH3NH3PbI3units in ourunit cell" > > > Ref:1. https://pubs.acs.org/doi/epdf/10.1021/acs.jpcc.5b07432 > > 2.https://scholar.google.com/citations?view_op=view_citation=en=4Ncn-GUJ=pubdate_for_view=4Ncn-GUJ:9yKSN-GCB0IC > > > 在 2023年10月24日 星期二 下午11:25:48 [GMT+8]， > ludwigboltzmann.s...@nycu.edu.tw 寫道： > > > Dear QE users: > I have a naive question of how to determine the number of unit cell to I need > to calculate vibrational frequencies.(at Gamma point) > In the below 2 paper, I saw authors choosing 4 unit cells of perovskite > (MAPbI3) for phonon calculation. > > However, I do not know what number of unit cell should I use? Is their any > way to determine it ? > > Sincerely, > > National Yang Ming Chiao Tung University > HY Lu > > > > <1698160383227blob.jpg> > > Ref:1. https://pubs.acs.org/doi/epdf/10.1021/acs.jpcc.5b07432 > 2.https://scholar.google.com/citations?view_op=view_citation=en=4Ncn-GUJ=pubdate_for_view=4Ncn-GUJ:9yKSN-GCB0IC > <1698160383227blob.jpg> > ___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Naive questions about input files 2D perovskite phonon calculation
Dear Hy Lu, It seems to me that your procedure and inputs (at least for pw.x) are correct. If I am not mistaken, the correction will work along the z direction, so make sure that you have your vacuum direction correctly aligned. Then, just note that the (non-resonant) Raman intensities can be obtained only using LDA and norm conserving pseudo in ph.x. An alternative way in this case could be finite differences. A possible package you can use for this case is the one I developed (search for aiida-vibroscopy). There are also others, but since you cell seems rather big, it seems that could be convenient to use finite field (as I use in the code), instead of finite dielectric tensors. If you are interested in the resonant case, there is a new package for that, called QERaman. I believe it works also for PBE, but only norm-conserving pseudo (not super sure, but it won’t be a problem for you anyways). HTH, Lorenzo Inviato da iPhone > Il giorno 16 ott 2023, alle ore 10:25, ludwigboltzmann.s...@nycu.edu.tw ha > scritto: > > > Dear users: > > I am going to calculate do phonon calculation and Raman spectra calculation > of 2D perovskite (only at Gamma point), > > My procedure is like this: doint: vc-relax & scf→phonon calculation→dynmat > > 1.Should I put assume_isolated='2D' in my vc-relax & scf input? > Ref: q-e/PHonon/examples/example17/reference > > Cause I'm not quite sure what is it mean here. > The extension of DFPT to two dimensional crystals, in "particular gated > two-dimensional heterostructure", > > Ref: > https://www.quantum-espresso.org/wp-content/uploads/2022/03/ph_user_guide.pdf > > > Below is my vc-relax input example. > > > calculation = 'vc-relax' > restart_mode='from_scratch' > prefix ='2D_perovskite' !test100tri > pseudo_dir = " pseudo_dir" > outdir=" outdir" > / > > ibrav=14, !triclinic > > celldm(1)=16.1377035000,celldm(2)=0.99156550541343,celldm(3)=1.87963300392941,celldm(4)=-1.94983007225937e-03,celldm(5)=0.13312393987385,celldm(6)=5.68819706737211e-04 > nat=94, ntyp=5, > ecutwfc=150 > ecutrho=600 > vdw_corr=grimme-d2 > / > > conv_thr=1e-14 > / > ATOMIC_SPECIES > Sn118.71000 Sn_ONCV_PBE-1.2.upf > H 1.00784 H_ONCV_PBE-1.2.upf > C 12.011C_ONCV_PBE-1.2.upf > I 126.90447 I_ONCV_PBE-1.2.upf > N 14.0067 N_ONCV_PBE-1.2.upf > ATOMIC_POSITIONS (crystal) > I 0.81037313470.68964899730.4992596099 > I 0.18962686530.31035100280.5007403901 > ... > > K_POINTS (automatic) > 5 5 1 1 1 1 > > > > Sincerely, > > National Yang Ming Chiao Tung University > HY Lu > > > ___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] asr in dynmat.x
Dear Juliana, Indeed the ASR in the old QE version did not include the correct rotational invariants for 2D systems, i.e. you should have 4 acoustic modes at Gamma that are zero. Although, having a closer look to your inputs, I would suggest you to pay attention to: 1. Why have you commented out ‘assume_isolated’? That is usually important to get rid of the long-range planar interaction between images. 2. The threshold for phonons seem a bit too low. Did you do some tests? In my experience that threshold should be much lower, e.g. 1e-16/18. HTH, Lorenzo Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 8 giu 2023, alle ore 11:08, Paolo Giannozzi > ha scritto: > > I have heard that in 2D systems some acoustic sum rules may yield funny > results for acoustic modes close to q=0. The latest 7.2 version of the code > should contain an improved ASR. See > https://gitlab.com/QEF/q-e/-/merge_requests/1978 > <https://gitlab.com/QEF/q-e/-/merge_requests/1978> > > Paolo > > On 6/8/23 00:12, Juliana Maria Abreu Da Silva Morbec wrote: >> >> You don't often get email from j.mor...@keele.ac.uk >> <mailto:j.mor...@keele.ac.uk>. Learn why this is important >> <https://aka.ms/LearnAboutSenderIdentification >> <https://aka.ms/LearnAboutSenderIdentification>> >> >> Dear Lorenzo. >> Thanks for your response. >> I am attaching here the input files for scf and phonon calculations of bulk >> MoS2, as well as output files of dynmat.x using asr='simple' and >> asr='crystal'. I also noticed that asr='no' give similar results to >> asr='crystal'. >> Regards, >> Juliana >> - >> *Dr Juliana Morbec* >> Lecturer in Physics >> Keele University, UK >> >> *From:* users on behalf of >> Lorenzo Bastonero >> *Sent:* 06 June 2023 14:28 >> *To:* Quantum ESPRESSO users Forum >> *Subject:* Re: [QE-users] asr in dynmat.x >> Dear Juliana, >> In principle at Gamma the acoustic sum rules should only affect the 3 >> acoustic modes, setting them to 0, and leave the other modes close to their >> values ( ~< 1/0.1 cm-1). >> If you are experiencing huge change in the wavenumber, it may probably be a >> sign of under converged calculations. Hard to say though w/o inputs/outputs. >> HTH, >> Lorenzo >> >> Lorenzo Bastonero >> PhD Student >> U Bremen Excellence Chair, >> Bremen Center for Computational Materials Science, >> and MAPEX Center for Materials and Processes >> University of Bremen >> Faculty of Production Engineering >> TAB-Building, Room 3.32 >> Am Fallturm 1 >> 28359 Bremen, DE >> http://www.hmi.uni-bremen.de/ <http://www.hmi.uni-bremen.de/> >> <https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.hmi.uni-bremen.de%2F=05%7C01%7Cpaolo.giannozzi%40uniud.it%7C5e34b0384c774338a46708db67a47dc1%7C6e6ade15296c4224ac581c8ec2fd53a8%7C0%7C0%7C638217728437942031%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=MZqrRwY0Bzamr%2Fsf%2BViS9AHXAao%2FIhu%2BlACgy46NgV8%3D=0 >> >> <https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.hmi.uni-bremen.de%2F=05%7C01%7Cpaolo.giannozzi%40uniud.it%7C5e34b0384c774338a46708db67a47dc1%7C6e6ade15296c4224ac581c8ec2fd53a8%7C0%7C0%7C638217728437942031%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C=MZqrRwY0Bzamr%2Fsf%2BViS9AHXAao%2FIhu%2BlACgy46NgV8%3D=0>> >>> Il giorno 6 giu 2023, alle ore 15:14, Juliana Maria Abreu Da Silva Morbec >>> mailto:j.mor...@keele.ac.uk> >>> <mailto:j.mor...@keele.ac.uk <mailto:j.mor...@keele.ac.uk>>> ha scritto: >>> >>> Hello. >>> >>> I have a question regarding the 'asr' feature in dynmat.x. I am computing >>> Raman spectra of MoS2 bulk and I noticed that asr='simple' and >>> asr='crystal' give different frequencies but similar amplitude for Raman. >>> And asr='crystal' seems to be in better agreement with experiments. How to >>> decide which one to use? >>> >>> Thank you, >>> >>> Juliana Morbec >>> Lecturer in Physics >>> Keele University, UK >>> -- >>> P
Re: [QE-users] asr in dynmat.x
Dear Juliana, In principle at Gamma the acoustic sum rules should only affect the 3 acoustic modes, setting them to 0, and leave the other modes close to their values ( ~< 1/0.1 cm-1). If you are experiencing huge change in the wavenumber, it may probably be a sign of under converged calculations. Hard to say though w/o inputs/outputs. HTH, Lorenzo **** Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 6 giu 2023, alle ore 15:14, Juliana Maria Abreu Da Silva Morbec > ha scritto: > > Hello. > > I have a question regarding the 'asr' feature in dynmat.x. I am computing > Raman spectra of MoS2 bulk and I noticed that asr='simple' and asr='crystal' > give different frequencies but similar amplitude for Raman. And asr='crystal' > seems to be in better agreement with experiments. How to decide which one to > use? > > Thank you, > > Juliana Morbec > Lecturer in Physics > Keele University, UK > -- > Please consider the environment before printing this email.This email and its > contents may be legally privileged and /or confidential.If it has come to you > in error you should not copy or show it to anyone; nor should you take any > action based on it, other than to reply to the sender to notify them of the > error immediately and to then delete the email from your inbox and deleted > items. Keele University staff and students are required to abide by the > University's IT Policies when sending and receiving email. Keele University > email is hosted by a cloud provider and may be stored outside of the UK. > ___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu > <http://www.max-centre.eu/>) > users mailing list users@lists.quantum-espresso.org > <mailto:users@lists.quantum-espresso.org> > https://lists.quantum-espresso.org/mailman/listinfo/users > <https://lists.quantum-espresso.org/mailman/listinfo/users> ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] [EXTERNAL] The phonon code with Raman and Hubbard U is not implemented
Dear Rijan, To warn you: these steps are involved, and you might need to code quite a bit and have to know the theory behind. Refer to the before mentioned paper. If you are willing to do so, you can find the first steps in example 10 of PW (https://gitlab.com/QEF/q-e/-/tree/develop/PW/examples/example10 <https://gitlab.com/QEF/q-e/-/tree/develop/PW/examples/example10>). Then you might try to have a look in the PHonon/FD module (https://gitlab.com/QEF/q-e/-/tree/develop/PHonon/FD <https://gitlab.com/QEF/q-e/-/tree/develop/PHonon/FD>). All the best, Lorenzo ******** Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 11 mag 2023, alle ore 16:13, Karkee, Rijan ha > scritto: > > Hi Lorenzo, > > Thanks for the information. May I know, for now, what steps are involved so > that I can compute those steps by hand? > I want to know which module to run first and what would be the input file > should include and likewise for the next steps. > > Thank you so much. > > Best > Rijan Karkee > > > From: Lorenzo Bastonero > Sent: Thursday, May 11, 2023 2:11:28 AM > To: Karkee, Rijan; Quantum ESPRESSO users Forum > Subject: [EXTERNAL] Re: [QE-users] The phonon code with Raman and Hubbard U > is not implemented > > Dear Rijan, > > Indeed the computation of Raman intensities is limited to LDA and > Norm-conserving pseudopotentials. > > There is an alternative way to compute both phonons (i.e. peak positions) and > intensities (of each mode). > You have to use finite differences for the phonons and finite electric fields > for the intensities (if you want to know the theory, look at Umari and > Pasquarello, Diamonds and Related Materials, 2005). > > These routines are in QE, nevertheless you will need to do the steps by hand. > > Currently, we have an automated workflow for doing such calculations in > AiiDA, which is a workflow manager that automates complex workflows, for QE > and other quantum codes. > The code will be published very soon, and we plan to send an email here once > it is published. > > So, at the moment, if you are interested, you can have a look at the > demonstration of AiiDA + QE (https://aiida-qe-demo.readthedocs.io/en/latest/ > <https://urldefense.com/v3/__https://aiida-qe-demo.readthedocs.io/en/latest/__;!!Bt8fGhp8LhKGRg!DbwBeLMmYIrnyC7VLd4-orYpVCYS7spvVhPfEFwHggBVJv_I2o5dKspdMRW5DKNYoiAGAqSmJ1WuS7chcG1yAe8$>), > and then get your local implementation and install aiida (https://aiida.net > <https://urldefense.com/v3/__https://aiida.net__;!!Bt8fGhp8LhKGRg!DbwBeLMmYIrnyC7VLd4-orYpVCYS7spvVhPfEFwHggBVJv_I2o5dKspdMRW5DKNYoiAGAqSmJ1WuS7ch19tOXhM$>) > aiida-quantum espresso, which is the main block for the future package we > will release. > > HTH, > Lorenzo > > Inviato da iPhone > >> Il giorno 11 mag 2023, alle ore 07:55, Karkee, Rijan via users >> ha scritto: >> >> >> Hello Developers, >> >> I saw some plots of Raman spectra in the tutorial on DFT+U. I was wondering >> if Raman spectra with DFT+U and/or DFT+U+V will be implemented anytime soon >> or if is there any simpler way to add in the code in order for it to have >> Raman with DFT+U? >> As of now, I found that Raman is not implemented with DFT+U. >> >> Thank you, >> Best >> Rijan Karkee >> ___ >> The Quantum ESPRESSO community stands by the Ukrainian >> people and expresses its concerns about the devastating >> effects that the Russian military offensive has on their >> country and on the free and peaceful scientific, cultural, >> and economic cooperation amongst peoples >> ___ >> Quantum ESPRESSO is supported by MaX (www.max-centre.eu >> <https://urldefense.com/v3/__http://www.max-centre.eu__;!!Bt8fGhp8LhKGRg!DbwBeLMmYIrnyC7VLd4-orYpVCYS7spvVhPfEFwHggBVJv_I2o5dKspdMRW5DKNYoiAGAqSmJ1WuS7chO_kspeg$>) >> users mailing list users@lists.quantum-espresso.org >> https://lists.quantum-espresso.org/mailman/listinfo/users >> <https://urldefense.com/v3/__https://lists.quantum-espresso.org/mailman/listinfo/users__;!!Bt8fGhp8LhKGRg!DbwBeLMmYIrnyC7VLd4-orYpVCYS7spvVhPfEFwHggBVJv_I2o5dKspdMRW5DKNYoiAGAqSmJ1WuS7chrxJ0fqQ$> ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effec
Re: [QE-users] The phonon code with Raman and Hubbard U is not implemented
Dear Rijan, Indeed the computation of Raman intensities is limited to LDA and Norm-conserving pseudopotentials. There is an alternative way to compute both phonons (i.e. peak positions) and intensities (of each mode). You have to use finite differences for the phonons and finite electric fields for the intensities (if you want to know the theory, look at Umari and Pasquarello, Diamonds and Related Materials, 2005). These routines are in QE, nevertheless you will need to do the steps by hand. Currently, we have an automated workflow for doing such calculations in AiiDA, which is a workflow manager that automates complex workflows, for QE and other quantum codes. The code will be published very soon, and we plan to send an email here once it is published. So, at the moment, if you are interested, you can have a look at the demonstration of AiiDA + QE (https://aiida-qe-demo.readthedocs.io/en/latest/), and then get your local implementation and install aiida (https://aiida.net) aiida-quantum espresso, which is the main block for the future package we will release. HTH, Lorenzo Inviato da iPhone > Il giorno 11 mag 2023, alle ore 07:55, Karkee, Rijan via users > ha scritto: > > > Hello Developers, > > > > I saw some plots of Raman spectra in the tutorial on DFT+U. I was wondering > if Raman spectra with DFT+U and/or DFT+U+V will be implemented anytime soon > or if is there any simpler way to add in the code in order for it to have > Raman with DFT+U? > > As of now, I found that Raman is not implemented with DFT+U. > > > > Thank you, > > Best > > Rijan Karkee > > ___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] The phonon code for this Hubbard projectors type is not implemented
Dear Rijan,
As stated by the error message, it is not implemented. This means, it will not
work/run.
You may try with “atomic” projectors, which are instead implemented (only for
U, not U+V). Nevertheless, this will most probably give you different results.
The other option is to use finite differences, e.g. using Phonopy.
Best,
Lorenzo Bastonero
Inviato da iPhone
> Il giorno 10 mag 2023, alle ore 20:29, Karkee, Rijan via users
> ha scritto:
>
>
> Dear Quantum ESPRESSO users and developers,
>
> I am using Hubbard U+V to calculate Raman spectra in version 7.1. I am
> getting the following error in ph.x calculation:
> %%
> Error in routine phq_readin (1):
> The phonon code for this Hubbard projectors type is not implemented
>
>
> My input file is following:
> At first I tried using both U and V (please see input file below) and got the
> error. Then I looked into example 18 of PHonon folder for running ph.x
> calculation using Hubbard parameters, I found only U was used so instead of
> V, I changed to U of Hf-5d but still got the same error.
>
>
> calculation = 'scf'
> pseudo_dir = '/lustre/scratch4/turquoise/rkarkee/pseudo'
> outdir = './'
> prefix = 'scf'
> tprnfor = .true.
> tstress=.true.
> wf_collect=.true.
> nstep=200
> verbosity='high'
> /
>
>
> ibrav=0,
> ecutwfc = 70,
> nat = 12,
> ntyp= 2,
> /
>
>
> conv_thr = 1.0e-10
> electron_maxstep = 200
> mixing_beta = 0.2
> diagonalization = "david"
>
>
> /
>
> K_POINTS {automatic}
> 4 4 2 1 1 1
>
>
> ATOMIC_SPECIES
> Hf 178.49 Hf.upf
> Te 127.60 Te.upf
>
> CELL_PARAMETERS (angstrom)
>1.981384438 7.143912880 0.0
> -1.981384438 7.143912880 0.0
>0.0 0.0 13.392710047
>
> ATOMIC_POSITIONS (crystal)
> Hf0.81684878350.81684878350.75
> Hf0.18315121650.18315121650.25
> Te0.70795294700.70795294700.5662574855
> Te0.29204705300.29204705300.0662574855
> Te0.70795294700.70795294700.9337425145
> Te0.29204705300.29204705300.4337425145
> Te0.42933858200.42933858200.8534412009
> Te0.57066141800.57066141800.3534412009
> Te0.42933858200.42933858200.6465587991
> Te0.57066141800.57066141800.1465587991
> Te0.16348148180.16348148180.75
> Te0.83651851820.83651851820.25
>
>
> HUBBARD {ortho-atomic}
> V Hf-5d Hf-5d 1 1 2.0477
> V Hf-5d Te-5p 1 201 0.1286
>
>
> And ph.x input is:
> HfTe5 bulk Raman
>
> tr2_ph=1.0d-18
> prefix='scf'
> trans=.true.
> lraman=.true.
> epsil=.true.
> outdir='./'
> fildyn='HfTe.dynG'
> alpha_mix=0.3
> max_seconds=269200
> /
>
> 0.0 0.0 0.0
>
>
> I also tried with the following but got the same error.
> HUBBARD {ortho-atomic}
> U Hf-5d 2.0477
>
>
>
> Thank you,
> Rijan Karkee
>
> ___
> The Quantum ESPRESSO community stands by the Ukrainian
> people and expresses its concerns about the devastating
> effects that the Russian military offensive has on their
> country and on the free and peaceful scientific, cultural,
> and economic cooperation amongst peoples
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] QE-GPU and libxc compile error
Dear Michele, If I am not mistaken, as from QE v.7.0 you would need to have git v.2.13 or later (see here <https://gitlab.com/QEF/q-e/-/wikis/Support/Prerequisites>). Could you check if this applies to you? I had a similar error when compiling with an older version of git. Plus, also make sure you have the m4 package installed. Let me know if that helps, Lorenzo **** Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 21 apr 2023, alle ore 12:28, Michele Re Fiorentin > ha scritto: > > Dear QE developers and users, > > I’m trying to install QE on a GPU-accelerated system (M100) while linking it > to libxc. I’ve compiled both libxc (version 7.0.0) and QE (version 7.2) with > the same compilers and libraries: > > LIBXC: ./configure CC=pgcc F77=pgf90 FC=pgf90 F90=pgf90 MPIF90=mpipgifort > --prefix=/m100/home/userexternal/mrefiore/libxc > > QE: ./configure CC=pgcc F77=pgf90 FC=pgf90 F90=pgf90 MPIF90=mpipgifort > --enable-openmp --with-cuda=/cineca/prod/opt/compilers/cuda/11.0/none > --with-cuda-runtime=11.0 --with-cuda-cc=70 --with-scalapack=no > --enable-cuda-env-check=no --with-libxc=yes > --with-libxc-prefix=/m100/home/userexternal/mrefiore/libxc > --with-libxc-include=/m100/home/userexternal/mrefiore/libxc/include > > Libxc compiles very well without problems. > QE configure runs smoothly, the library is found and the linker does not > complain. > However, when compiling pw I get the following error > > make[1]: Entering directory > '/m100/home/userexternal/mrefiore/qe-7.1_libxc/XClib' > mpipgifort -fast -Mcache_align -Mpreprocess -Mlarge_arrays -mp -D__PGI > -D__CUDA -D__FFTW3 -D__LIBXC -D__MPI -Mcuda=cc70,cuda11.0 > -I/m100/home/userexternal/mrefiore/qe-7.1_libxc//external/devxlib/src > -I/m100/home/userexternal/mrefiore/qe-7.1_libxc//external/devxlib/include > -acc -I/m100/home/userexternal/mrefiore/qe-7.1_libxc//external/devxlib/src > -I. -I/m100/home/userexternal/mrefiore/qe-7.1_libxc//include > -I/cineca/prod/opt/libraries/fftw/3.3.8/gnu--8.4.0/include > -I/m100/home/userexternal/mrefiore/libxc/include > -I/m100/home/userexternal/mrefiore/qe-7.1_libxc//upflib -c > dft_setting_routines.f90 > NVFORTRAN-S-0038-Symbol, xc_family_hyb_gga, has not been explicitly declared > (dft_setting_routines.f90) > NVFORTRAN-S-0038-Symbol, xc_family_hyb_mgga, has not been explicitly declared > (dft_setting_routines.f90) > 0 inform, 0 warnings, 2 severes, 0 fatal for matching_shortids > NVFORTRAN-S-0038-Symbol, xc_family_hyb_lda, has not been explicitly declared > (dft_setting_routines.f90) > NVFORTRAN-S-0038-Symbol, xc_family_hyb_gga, has not been explicitly declared > (dft_setting_routines.f90) > NVFORTRAN-S-0038-Symbol, xc_family_hyb_mgga, has not been explicitly declared > (dft_setting_routines.f90) > 0 inform, 0 warnings, 3 severes, 0 fatal for xclib_set_auxiliary_flags > make[1]: *** [../make.inc:16: dft_setting_routines.o] Error 2 > make[1]: Leaving directory > '/m100/home/userexternal/mrefiore/qe-7.1_libxc/XClib' > make: *** [Makefile:206: librxc] Error 1 > > > Clearly the compilation fails. I can’t figure out how to solve the issues > with those symbols “not declared”. Is it a compiler incompatibility? > Any help would be much appreciated! > > > Michele > > > > > -- > Michele Re Fiorentin, PhD > > Department of Applied Science and Technology (DISAT) > Politecnico di Torino > corso Duca degli Abruzzi 24, 10129 Torino (Italy) > tel: +39 0110904333 > > ___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Query regarding ecutfock
Dear Arini, As a final remark, this is what experience tells us, but remember that proper testing should be directly performed on the desired property with a desired accuracy. This means that a better practice could be e.g. to test ``ecutfock`` on phonon frequencies at Gamma (i.e. only displacements in UNITCELL). This already gives a better idea of convergence, while still not going for expensive supercells. If you are interested in more details, I’d suggest you the following paper, which I found particularly instructive: https://www.nature.com/articles/s41524-018-0127-2 <https://www.nature.com/articles/s41524-018-0127-2> Bests, Lorenzo **** Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 12 apr 2023, alle ore 14:08, Arini Kar <16i170...@iitb.ac.in> ha > scritto: > > Dear Lorenzo, > > Your suggestion was helpful. The forces on the pristine unit cell did not > change after reducing ecutfock to 2*ecutwfc. Thanks a lot. > > > > Regards > Arini > > > > On 2023-04-12 13:26, Lorenzo Bastonero wrote: > >> Dear Arini, >> >> You should do some test first in the pristine unit cell structure before >> going forward to the supercell calculations. >> >> In principle, you could lower down even more that cutoff (and the qpoint for >> the fock operator as well). Usually a good threshold is ~10^-5 Ry/bohr on >> every atomic force component. >> >> Note: this threshold is different from the one of the relax. So you should >> check that when changing your variables (ecutfock in this case) all your >> forces don't change within this threshold. >> >> HTH, >> Lorenzo >> >> Inviato da iPhone >> >>> Il giorno 11 apr 2023, alle ore 20:45, Arini Kar via users >>> ha scritto: >>> >>> Dear users, >>> >>> I am trying to run phonon calculations using phonopy. I am using norm >>> conserving pseudopotential. Can I use ecutfock = 2*ecutwfc instead of the >>> default value in the scf calculations? Would it affect the accuracy of >>> force calculations? >>> >>> Thanking you. >>> >>> Regards >>> >>> Arini Kar >>> >>> PhD student >>> >>> Indian Institute of Technology Bombay, India >>> >>> >>> >>> ___ >>> The Quantum ESPRESSO community stands by the Ukrainian >>> people and expresses its concerns about the devastating >>> effects that the Russian military offensive has on their >>> country and on the free and peaceful scientific, cultural, >>> and economic cooperation amongst peoples >>> ___ >>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu) >>> users mailing list users@lists.quantum-espresso.org >>> https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Query regarding ecutfock
Dear Arini, You should do some test first in the pristine unit cell structure before going forward to the supercell calculations. In principle, you could lower down even more that cutoff (and the qpoint for the fock operator as well). Usually a good threshold is ~10^-5 Ry/bohr on every atomic force component. Note: this threshold is different from the one of the relax. So you should check that when changing your variables (ecutfock in this case) all your forces don’t change within this threshold. HTH, Lorenzo Inviato da iPhone > Il giorno 11 apr 2023, alle ore 20:45, Arini Kar via users > ha scritto: > > > Dear users, > > I am trying to run phonon calculations using phonopy. I am using norm > conserving pseudopotential. Can I use ecutfock = 2*ecutwfc instead of the > default value in the scf calculations? Would it affect the accuracy of force > calculations? > > Thanking you. > > Regards > > Arini Kar > > PhD student > > Indian Institute of Technology Bombay, India > > > > ___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] About the calculation of dielectric function by using QE package
Dear Purinut, > but I cannot find the NCPP files. I would like to ask 1) Where can I get the > NCPP files? 2) Is there a way to calculate the dielectric function of the > material that applied PAW pseudopotential? 1.) Did you try looking here http://www.pseudo-dojo.org/ <http://www.pseudo-dojo.org/> ? You should be able to find what you need. 2.) If it can be of any help, you can try looking if PAW is implemented in the TDDFT package and compute it from there. HTH, Lorenzo ******** Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] E-field units inconsistent
Dear Joseph, In Ry a.u. the electric charge is sqrt(2), thus the units are correct. Best, Lorenzo Lorenzo Bastonero PhD Student U Bremen Excellence Chair, Bremen Center for Computational Materials Science, and MAPEX Center for Materials and Processes University of Bremen Faculty of Production Engineering TAB-Building, Room 3.32 Am Fallturm 1 28359 Bremen, DE http://www.hmi.uni-bremen.de/ > Il giorno 14 feb 2023, alle ore 23:43, Joseph Frimpong ha > scritto: > > Dear Users and Developers, > > I need clarifications on the finite e-field conversion factor for 1 a.u found > on the input file description at > https://www.quantum-espresso.org/Doc/INPUT_PW.html > <https://www.quantum-espresso.org/Doc/INPUT_PW.html#idm840> and in the source > code at $QEDIR/PW/src/extfield.f90. > > Reading the input file description online: under “efield”, says in Ry a.u.; 1 > a.u. = 36.3609*10^10 V/m > > We know E-field in a.u. is 5.14220674763(78)×10^11 V⋅m−1 from > https://en.wikipedia.org/wiki/Hartree_atomic_units > <https://en.wikipedia.org/wiki/Hartree_atomic_units>; which means in Ry a.u., > it should be 51.422/2 = 25.711, but the 36.3609 is actually 51.422/sqrt(2). > > In the source code: $QEDIR/PW/src/extfield.f90, line #38, in the variable > “eamp”, the comment says “(1 a.u. = 51.44 10^11 V/m)” which matches the > Wikipedia result but not the INPUT_PW webpage. > > Which one is correct? Can someone please resolve this confusion? > > Thanks for your time. > > - > Joseph Frimpong > Graduate Research Assistant > Department of Chemistry > Wayne State University > > > - > Joseph Frimpong > Graduate Research Assistant > Liu's Group > Department of Chemistry > Wayne State University > 5101 Cass Ave, Detroit, MI 48202 > Phone: (313) 577-0552 > https://liugroupwsu.weebly.com/ > <https://liugroupwsu.weebly.com/>___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu > <http://www.max-centre.eu/>) > users mailing list users@lists.quantum-espresso.org > <mailto:users@lists.quantum-espresso.org> > https://lists.quantum-espresso.org/mailman/listinfo/users > <https://lists.quantum-espresso.org/mailman/listinfo/users> ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] The pencil decomposition option
Dear Elie, You were almost there! Try adding literally what is suggested, i.e. “-pd .true.” Best, Lorenzo Bastonero Inviato da iPhone > Il giorno 9 nov 2022, alle ore 15:08, Elio Physics ha > scritto: > > > Dear all, > > I have been running some calculations on a supercomputer with some > constraints on the number of processors I can run QE. I am faced with the > error: > > %% > Error in routine fft_type_set (6): > there are processes with no planes. Use pencil decomposition (-pd .true.) > > %% > > How do I tell the code to use the pencil decomposition? I have tried adding > -pd to the 'mpirun' an adding use_pd=.true. in the PW inout but in vain. > > Thanks in advance > > Elie Moujaes > Adjunct professor > Federal University of Rondonia > Brazil. > > > > ___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] oscillations in G(V) diagrams
Dear Jakob,
Do you mean the Helmoltz free energy (F)?
As by definition, the Gibbs free energy (G) does not depend upon volume, but on
pressure and temperature.
At p=0, then they are the same, but notice always that:
G(p,T) = min_{V}[F(V, T)+pV]
Thus, at p=0, you still need to take the minimum of F at the fixed temperature.
Moreover, I guess you are using the quasi-harmonic approximation. There exist
several flavours. Unless your system is isotropic, scaling the volume
isotropically can be a source of error (due to a physical assumption, not due
to the numerics).
Best,
Lorenzo Bastonero
Lorenzo Bastonero
PhD Student
U Bremen Excellence Chair,
Bremen Center for Computational Materials Science,
and MAPEX Center for Materials and Processes
University of Bremen
Faculty of Production Engineering
TAB-Building, Room 3.32
Am Fallturm 1
28359 Bremen, DE
http://www.hmi.uni-bremen.de/
> Il giorno 4 nov 2022, alle ore 09:20, Jakob Kraus
> ha scritto:
>
> Dear QE users,
>
>
> I am trying to calculate heat capacities, and to this end, I have calculated
> the Gibbs energy as a function of volume
>
> at several temperatures in a range of 0-700 K.
>
> However, the functions that result from this do not show a single minimum,
> but rather a whole lot of oscillations, which
>
> seem to originate from the phonons, since the E(V) diagram at 0 K shows a
> single minimum as expected.
>
> I am thankful for any help you can offer with getting rid of these
> oscillations.
>
> I have attached input files and some plots to showcase the problem.
>
>
> Best regards,
>
>
> Jakob Kraus, TU Freiberg
> ___
> The Quantum ESPRESSO community stands by the Ukrainian
> people and expresses its concerns about the devastating
> effects that the Russian military offensive has on their
> country and on the free and peaceful scientific, cultural,
> and economic cooperation amongst peoples
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Convergence problem in vc-relax
Hi, At first look it seems your cutoff is a bit too small, as well as your conv_thr. Pressure usually is rather affected by low cutoff. I’d expect at least 80/90 Ry cutoff for converged results within sensible parameters. Best, Lorenzo Inviato da iPhone > Il giorno 12 lug 2022, alle ore 17:40, Ali Jaberi > ha scritto: > > > Hi, > > I am trying to relax the structure for Li(Ni1/3Mn1/3Co1/3)O2. I could relax > the structure with VASP but not with Quantum ESPRESSO. I have tried changing > so many parameters but could not get any result so end up posting my problem > here. > > Here is my input file: > > - > > calculation='vc-relax' > nstep = 500, > restart_mode = 'from_scratch' , > pseudo_dir = '' , > outdir = './outdir' > prefix = 'NMC', > verbosity ='high' , > tprnfor = .true. > etot_conv_thr= 1.0D-3 > forc_conv_thr = 1.0D-2 > / > > > ibrav = 0, > nat =36, > ntyp =5, > ecutwfc = 34 , > ecutrho = 380, > occupations ='smearing', > smearing='mv', > degauss=0.001, > nspin = 2, > starting_magnetization(2)=0.7, > starting_magnetization(3)=0.7, > starting_magnetization(4)=0.7, > lda_plus_u = .true. > Hubbard_U(2)=4.84, > Hubbard_U(3)=5.0, > Hubbard_U(4)=5.96, > / > > > electron_maxstep = 700 > conv_thr = 1.0d-4, > mixing_beta = 0.3, > diagonalization = 'david', > / > > > ion_dynamics = "bfgs", > / > > >cell_dynamics = 'bfgs' >press_conv_thr = 0.9 > / > > ATOMIC_SPECIES > Li 6.940 Li.pbe-sl-rrkjus_psl.1.0.0.UPF > Mn 54.938 Mn.pbe-spn-rrkjus_psl.0.3.1.UPF > Co 58.933 Co.pbe-spn-rrkjus_psl.0.3.1.UPF > Ni 58.693 Ni.pbe-spn-rrkjus_psl.1.0.0.UPF > O 15.999 O.pbe-n-rrkjus_psl.1.0.0.UPF > > CELL_PARAMETERS angstrom > 4.924047240.0. > -2.462023624.264350000. > 0.0. 14.1456 > > ATOMIC_POSITIONS angstrom > Co -0.819853894.26292856 11.78328480 > Co0.822315911.421449992.34816960 > Co2.462023624.264350007.0728 > Li0.0.0. > Li -0.819853894.262928564.71048480 > Li0.822315911.421449999.42096960 > Li -0.00032.842900010. > Li1.642169782.841478544.71048480 > Li3.284339510.9.42096960 > Li2.462023641.421449990. > Li -0.819853871.420028554.71048480 > Li -1.639707762.842900019.42096960 > Mn -0.00032.842900017.0728 > Mn1.642169782.84147854 11.78328480 > Mn3.284339510.2.34816960 > Ni2.462023641.421449997.0728 > Ni -0.819853871.42002855 11.78328480 > Ni -1.639707762.842900012.34816960 > O 0.0.3.38910187 > O 0.0. 10.75649813 > O-0.819853894.262928568.09958667 > O-0.819853894.262928561.32138293 > O 0.822315911.42144999 12.81007147 > O 0.822315911.421449996.03186773 > O-0.00032.842900013.38910187 > O-0.00032.84290001 10.75649813 > O 1.642169782.841478548.09958667 > O 1.642169782.841478541.32138293 > O 3.284339510. 12.81007147 > O 3.284339510.6.03186773 > O 2.462023641.421449993.38910187 > O 2.462023641.42144999 10.75649813 > O-0.819853871.420028558.09958667 > O-0.819853871.420028551.32138293 > O-1.639707762.84290001 12.81007147 > O-1.639707762.842900016.03186773 > > K_POINTS automatic > 4 4 4 0 0 0 > - > > Ther pressures I got for this relaxation are very high at each step!!! > > total stress (Ry/bohr**3) (kbar) P= > -23972.31 > total stress (Ry/bohr**3) (kbar) P= > -26659.21 > total stress (Ry/bohr**3) (kbar) P= > -24871.74 > total stress (Ry/bohr**3) (kbar) P= > -24165.07 > total stress (Ry/bohr**3) (kbar) P= > -24040.51 > total stress (Ry/bohr**3) (kbar) P= > -23928.12 > total stress (Ry/bohr**3)
Re: [QE-users] restarting the calculation
Dear Alex, What do you mean with higher precision? Decreasing the threshold of the self-consistent cycle (i.e. ‘conv_thr’ in the ‘ELECTRONS’ tag)? In this case you can add in ‘ELECTRONS’ the flags: startingpot=‘file’ startingwfc=‘file’ If you instead mean increasing the cutoff on wfc, I don’t think that you can restart, since the charge density and wfc are defined on a different grid. On the other hand, if you just want to increase the kpoints grid, you can put just the first flag, and leave out the strartingwfc flag. That will work for sure (like doing an nscf). Best wishes, Lorenzo Bastonero > Il giorno 30 apr 2022, alle ore 22:08, Aleksandra Oranskaia > ha scritto: > > > Hello dear users and developers of QE! > > Can please somebody advise on how to restart the calculation when you > successfully finish SCF calculation but want to re-run it with higher > precision starting with the previously generated wavefunctions? > > I tried restarting_mode = 'restart' but as I understand it works either for > structural relaxations or non-complete SCF when max_sexonds was used. What am > I missing? > > Thank you! > -- > Best wishes, > Alex (she/her/hers), > phd candidate in chemical sciences > 'I.. a universe of atoms, an atom in the universe' (Richard P. Feynman) > https://cpms.kaust.edu.sa/ > > This message and its contents, including attachments are intended solely for > the original recipient. If you are not the intended recipient or have > received this message in error, please notify me immediately and delete this > message from your computer system. Any unauthorized use or distribution is > prohibited. Please consider the environment before printing this > email.___ > The Quantum ESPRESSO community stands by the Ukrainian > people and expresses its concerns about the devastating > effects that the Russian military offensive has on their > country and on the free and peaceful scientific, cultural, > and economic cooperation amongst peoples > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] [QE-GPU] nscf calculation problem
Hi, I’m currently running on Marconi100 nscf calculations (qe v 6.6) for the interface of two monolayers (which I already studied separately and got no issues). I need to calculate several bands (i.e. 8000) for GW calculations. Running on 2 nodes with npool 2, the dynamical RAM per process is above 16Gb. I tried out with npool 1, but I encountered the error "S matrix not positive definite”. I decided then to try with 4 nodes and npool 2 with the option diago_david_ndim='2', then the dyn RAM/process is 14.97Gb, so it has to work now. But the calculation stops without the “CRASH” file and in the output I got "2 total processes killed (some possibly by mpirun during cleanup)”. In the slurm-output there are several segmentation fault error messages. I also discussed the problem with my supervisor but we are puzzled. Any suggestions on what to do? You’ll find attached the main outputs. Thanks for your precious time. Kind regards, Lorenzo Bastonero -- Indirizzo istituzionale di posta elettronica degli studenti e dei laureati dell'Università degli Studi di TorinoOfficial University of Turin email address for students and graduates slurm-1280351.out Description: Binary data hetero.PdPt.AA.2L.soc.nscf.out Description: Binary data ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
