From owner-chemistry@ccl.net Wed Jul 18 01:42:01 2018 From: "Grigoriy Zhurko reg_zhurko./a\.chemcraftprog.com" To: CCL Subject: CCL: PBE for inorganic chemistry Message-Id: <-53398-180717144928-9100-76LhEpaB9XqRzp7YFwl1Gw/a\server.ccl.net> X-Original-From: Grigoriy Zhurko Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Date: Tue, 17 Jul 2018 21:51:26 +0400 MIME-Version: 1.0 Sent to CCL by: Grigoriy Zhurko [reg_zhurko**chemcraftprog.com] I have heard that the PBE functional is often appropriate for inorganic chemistry, while the B3LYP one is usually better for organic chemistry. Can you help me find publications, in which the advantages of PBE for computing inorganic molecules is described (for citing)? Grigoriy Zhurko https://chemcraftprog.com/ From owner-chemistry@ccl.net Wed Jul 18 08:46:01 2018 From: "Yury Minenkov yury.minenkov ~ gmail.com" To: CCL Subject: CCL: PBE for inorganic chemistry Message-Id: <-53399-180718042915-21242-mnuiBbSOqeKA4XxKvHQr7A[A]server.ccl.net> X-Original-From: Yury Minenkov Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="UTF-8" Date: Wed, 18 Jul 2018 11:29:07 +0300 MIME-Version: 1.0 Sent to CCL by: Yury Minenkov [yury.minenkov]~[gmail.com] Dear Grigoriy, First of all, I am not sure whether this statement is entirely correct. Perhaps, in the majority of cases PBE will outperform B3LYP for inorganic species, and B3LYP will provide better results for organic molecules. But, in general, it is hard to say a priori whether B3LYP or PBE is going to perform better for your system. I am sure this topic has been discussed many times. Unfortunately, I remember no references apart from my own/my colleagues works (sorry for self-promotion). Perhaps something can be found in "A Chemist's Guide to Density Functional Theory, " 2nd Edition by Wolfram Koch, Max C. Holthausen. Thus, in this work Dalton Trans., 2012,41, 5526-5541 (DOI: 10.1039/C2DT12232D) we have studied the performance of various DFT methods in reproducing molecular geometries of Organometallic species (which can be considered as both organic/inorganic). Interestingly, PBE functional outperforms B3LYP for all types of bonds (all internuclear distances, all transition metal related distances, etc). The only exception is carbon-carbon/carbon-heteroelements bonded distances for which B3LYP works clearly better than PBE. Please, see Fig. 5. However, in other recent study where we tested the DFT methods in reproducing spatial structures of di- and tri-atomic molecules for which very accurate spectroscopic data were available, B3LYP is general turned out to perform slightly better than PBE. See, ACS Omega, 2017, 2 (11), pp 8373–8387 DOI: 10.1021/acsomega.7b01203 This is about molecular spatial structures ("geometries"). For energetics the situation is similar. In this work (J. Chem. Theory Comput., 2016, 12 (4), pp 1542–1560 DOI: 10.1021/acs.jctc.5b01163) we explored the performance of DFT methods in reproducing reaction enthalpies of inorganic (Transition metals) reactions. Chart 7 in this work indicates that PBE in general outperforms B3LYP. But - the exceptions are very well possible - see the performance for separate transition metals. In another recent study we tested the DFT methods in reproducing heats of formation/formation enthalpies (J. Chem. Theory Comput., 2017, 13 (8), pp 3537–3560 DOI: 10.1021/acs.jctc.7b00335) of ORGANIC compounds. Non-dispersion corrected PBE0 turned out to provide better formation enthalpies than classical B3LYP. See Figure 1. Then, adding the dispersion makes the performance of PBE0 worse, while clear improvement on B3LYP can be noticed. So, dispersion corrected B3LYP-D3 outperforms PBE0-D3. Hence, any generalizations are dangerous. With kind regards, Yury On Tue, Jul 17, 2018 at 8:51 PM, Grigoriy Zhurko reg_zhurko.-*-.chemcraftprog.com wrote: > > Sent to CCL by: Grigoriy Zhurko [reg_zhurko**chemcraftprog.com] > I have heard that the PBE functional is often appropriate for inorganic chemistry, while the B3LYP one is usually better for organic chemistry. Can you help me find publications, in which the advantages of PBE for computing inorganic molecules is described (for citing)? > > Grigoriy Zhurko > https://chemcraftprog.com/> > From owner-chemistry@ccl.net Wed Jul 18 09:21:01 2018 From: "Susi Lehtola susi.lehtola_-_alumni.helsinki.fi" To: CCL Subject: CCL: PBE for inorganic chemistry Message-Id: <-53400-180718054549-4118-PGuDrmIjQmDU7oSNQMmdRQ]_[server.ccl.net> X-Original-From: Susi Lehtola Content-Language: en-US Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=utf-8; format=flowed Date: Wed, 18 Jul 2018 12:45:37 +0300 MIME-Version: 1.0 Sent to CCL by: Susi Lehtola [susi.lehtola]^[alumni.helsinki.fi] On 07/17/2018 08:51 PM, Grigoriy Zhurko reg_zhurko.^^^.chemcraftprog.com wrote: > Sent to CCL by: Grigoriy Zhurko [reg_zhurko**chemcraftprog.com] I > have heard that the PBE functional is often appropriate for inorganic > chemistry, while the B3LYP one is usually better for organic > chemistry. Can you help me find publications, in which the advantages > of PBE for computing inorganic molecules is described (for citing)? Not an expert, but since your functionals are none else than PBE and B3LYP, my guess it's not really a question of "goodness" but rather of popularity. B3LYP/6-31G* became popular within chemistry back in the day (functional dates from 1994) and is still used a lot, even though better functionals and more cost efficient basis sets exist. Part of the phenomenon is experience - people know where it works and where it fails - but the other part is something else. In contrast, in the solid state community, the paradigm was first LDA, which later switched to PBE. The PBE functional is from 1996. Pure functionals are often preferred in the solid state, since the combination of non-local exact exchange and fully local correlation doesn't give good results in extended systems. Also, evaluating exact exchange can be very costly in solid state calculations. Many inorganic compounds form crystals. Presumably, the functional that worked for inorganic crystalline systems (PBE) was also adopted for molecular calculations. -- ------------------------------------------------------------------ Mr. Susi Lehtola, PhD Junior Fellow, Adjunct Professor susi.lehtola^^^alumni.helsinki.fi University of Helsinki http://www.helsinki.fi/~jzlehtol Finland ------------------------------------------------------------------ Susi Lehtola, dosentti, FT tutkijatohtori susi.lehtola^^^alumni.helsinki.fi Helsingin yliopisto http://www.helsinki.fi/~jzlehtol ------------------------------------------------------------------