From: Joel Polowin [joelp@agiss.com] Sent: Monday, February 02, 1998 11:02 AM To: 'turek@farmacja.amg.gda.pl' Cc: 'hyperchem@hyper.com' Subject: Re: hyper 5 bugs Hi, Maciej. > "There were several trivial mistakes that were introduced to > the scripts, probably by the text editor that was used, such as broken > comment lines and uncorrected paths in the hin-files". If so, why it > does work in Hyper 4.5 without problems? These problems were almost certainly caused by your E-mail system breaking long lines into parts when you sent the script file(s) by E-mail. (Some E-mail systems do this, others don't.) That is why you do not see that problem with your copy of the file. Broken comment lines and bad paths could have a number of interesting effects, but not the ones you describe-they would prevent the script from running, instead of giving bizarre results during calculations. Your E-mail system appears to do a few other strange things, at least by the time your messages get here via the Hypercube server. It allows you to use 8-bit characters such as the curled quotation marks, instead of the 7-bit "'" and "'" characters; what I see at this end is that the characters are translated into things like "shouldn´t" instead of "shouldn't". What this means is that you need = to be careful when you send or receive computer-readable text (such as a program or script or structure file) by E-mail, to make sure that these kinds of changes haven't been made. I seem to recall that one of the bug fixes in 5.02 was supposed to correct an effect similar to what you describe. It might be worth checking that the program files you are using were actually changed by the update-click on "Help / About HyperChem" while the program is running, and "About HyperNDO" in the taskbar icon for that module while a calculation is in progress. Joel Polowin joelp@agiss.com From: Hypercube Evaluation Software [evaluation@hyper.com] Sent: Tuesday, February 10, 1998 1:31 PM To: 'hyperchem@hyper.com' Subject: HyperChem 5.02 evaluation Dear HyperChem users, A free evaluation copy of HyperChem 5.02 is now downloadable from http://www.hyper.com/products/evaluation/hyper5/eval5download.htm It can be used for 14 days and will expire on March 31, 1998. From: Brian Williams [williams@bucknell.edu] Sent: Wednesday, February 11, 1998 9:04 AM To: hyperchem@hyper.com Subject: Eigenvectors in Log Files In looking at a a more detailed log file printout on some calculations, I became curious about the coefficients associated with different atomic orbitals for each eigenvector. My assumption was that if I took these coefficients, squared them and added them up over all the eigenvectors/MOs for a particular atomic orbital, I would get a value of one. In some cases, however, I noticed that coefficients greater than one are reported. Is my assumption that these coefficents squared should add up to one incorrect? What exactly is the best way to interpret these coefficients? Thanks. Brian Williams Chemistry, Bucknell University From: Stavrev, Krassimir [stavrev@hyper.com] Sent: Thursday, February 12, 1998 9:25 AM To: 'williams@bucknell.edu' Cc: 'hyperchem@hyper.com' Subject: RE: Eigenvectors in Log Files Brian: You are probably referring to ab-initio results where sums not equal to 1 are commonly due to basis set non-orthogonality. When a non-orthogonal basis set is used, overlap terms appear in the Roothaan's equations leading to coefficients that can be greater than one in absolute value. In general, such coefficients squared do not sum to one. There are techniques that can be used to impose orthogonality conditions on the wavefunction, look in the literature for symmetric or canonical orthogonalization, for example. HyperChem does not use orthogonalization techniques, the eigenvectors that are reported are those calculated over unorthogonalized basis set. Regards, Krassimir --- Krassimir Stavrev, PhD, Director of Scientific Support Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 --- --Original Message----- From: Brian Williams [SMTP:williams@bucknell.edu] Sent: Wednesday, February 11, 1998 9:04 AM To: hyperchem@hyper.com Subject: Eigenvectors in Log Files In looking at a a more detailed log file printout on some calculations, I became curious about the coefficients associated with different atomic orbitals for each eigenvector. My assumption was that if I took these coefficients, squared them and added them up over all the eigenvectors/MOs for a particular atomic orbital, I would get a value of one. In some cases, however, I noticed that coefficients greater than one are reported. Is my assumption that these coefficents squared should add up to one incorrect? What exactly is the best way to interpret these coefficients? Thanks. Brian Williams Chemistry, Bucknell University From: Brian Williams [williams@bucknell.edu] Sent: Thursday, February 12, 1998 11:10 AM To: Stavrev, Krassimir Subject: RE: Eigenvectors in Log Files Thanks- I suspected something like this might be the case, but it helps to hear it from someone more knowledgable. The main reason I am looking at these log files and coefficients is to try to convince my students there is some connection between the atomic orbitals associated with atoms and the molecular orbitals, and the log files certain do that, especially with their symmetry labels. At 09:24 AM 2/12/98 -0500, you wrote: >Brian: You are probably referring to ab-initio results where sums not >equal to 1 are commonly due to basis set non-orthogonality. When a >non-orthogonal basis set is used, overlap terms appear in the Roothaan's >equations leading to coefficients that can be greater than one in >absolute value. In general, such coefficients squared do not sum to >one. There are techniques that can be used to impose orthogonality >conditions on the wavefunction, look in the literature for symmetric or >canonical orthogonalization, for example. HyperChem does not use >orthogonalization techniques, the eigenvectors that are reported are >those calculated over unorthogonalized basis set. > >Regards, >Krassimir >--- >Krassimir Stavrev, PhD, Director of Scientific Support >Hypercube, Inc. Florida Science and Technology Park >1115 N.W. 4th Street Gainesville, Florida 32601 >Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 > >-----Original Message----- >From: Brian Williams [SMTP:williams@bucknell.edu] >Sent: Wednesday, February 11, 1998 9:04 AM >To: hyperchem@hyper.com >Subject: Eigenvectors in Log Files > >In looking at a a more detailed log file printout on some calculations, >I became curious about the coefficients associated with different atomic >orbitals for each eigenvector. My assumption was that if I took these >coefficients, squared them and added them up over all the >eigenvectors/MOs for a particular atomic orbital, I would get a value of >one. In some cases, however, I noticed that coefficients greater than >one are reported. Is my assumption that these coefficents squared >should add up to one incorrect? What exactly is the best way to >interpret these coefficients? Thanks. > >Brian Williams >Chemistry, Bucknell University > > > From: Stavrev, Krassimir Sent: Thursday, February 12, 1998 11:17 AM To: 'Brian Williams' Subject: RE: Eigenvectors in Log Files Yes, I understand - you can still use the semiempirical methods (except Extended Huckel) for illustrating the AO contributions to the MOs. These eigenvectors squared should sum to one. Regards, Krassimir --- Krassimir Stavrev, PhD, Director of Scientific Support Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 -----Original Message----- From: Brian Williams [SMTP:williams@bucknell.edu] Sent: Thursday, February 12, 1998 11:10 AM To: Stavrev, Krassimir Subject: RE: Eigenvectors in Log Files Thanks- I suspected something like this might be the case, but it helps to hear it from someone more knowledgable. The main reason I am looking at these log files and coefficients is to try to convince my students there is some connection between the atomic orbitals associated with atoms and the molecular orbitals, and the log files certain do that, especially with their symmetry labels. At 09:24 AM 2/12/98 -0500, you wrote: >Brian: You are probably referring to ab-initio results where sums not >equal to 1 are commonly due to basis set non-orthogonality. When a >non-orthogonal basis set is used, overlap terms appear in the Roothaan's >equations leading to coefficients that can be greater than one in >absolute value. In general, such coefficients squared do not sum to >one. There are techniques that can be used to impose orthogonality >conditions on the wavefunction, look in the literature for symmetric or >canonical orthogonalization, for example. HyperChem does not use >orthogonalization techniques, the eigenvectors that are reported are >those calculated over unorthogonalized basis set. > >Regards, >Krassimir >--- >Krassimir Stavrev, PhD, Director of Scientific Support >Hypercube, Inc. Florida Science and Technology Park >1115 N.W. 4th Street Gainesville, Florida 32601 >Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 > >-----Original Message----- >From: Brian Williams [SMTP:williams@bucknell.edu] >Sent: Wednesday, February 11, 1998 9:04 AM >To: hyperchem@hyper.com >Subject: Eigenvectors in Log Files > >In looking at a a more detailed log file printout on some calculations, >I became curious about the coefficients associated with different atomic >orbitals for each eigenvector. My assumption was that if I took these >coefficients, squared them and added them up over all the >eigenvectors/MOs for a particular atomic orbital, I would get a value of >one. In some cases, however, I noticed that coefficients greater than >one are reported. Is my assumption that these coefficents squared >should add up to one incorrect? What exactly is the best way to >interpret these coefficients? Thanks. > >Brian Williams >Chemistry, Bucknell University > > > From: Hypercube Software [evaluation@hyper.com] Sent: Thursday, February 12, 1998 4:35 PM To: 'hyperchem@hyper.com' Subject: HyperChem Lite Electronic Version Dear HyperChem users, For the first time, Hypercube, Inc. is offering a fully electronic version of HyperChem Lite for purchase on the World Wide Web. The program is now downloadable from http://www.hyper.com/sales/electronic/electronic-lite.htm. The HyperChem Lite download is a self-extracting executable that will install a time-limited copy of HyperChem Lite. This copy of HyperChem Lite will install and run under Windows 95, Windows NT, and Windows 3.x. The HyperChem Lite electronic version is a time-limited copy and use is restricted to one (1) day. You must license (register) your product for unlimited use. From: underhill-r@rmc.ca Sent: Wednesday, February 18, 1998 12:11 PM To: hyperchem@hyper.com Subject: selected-atom variable I'm trying to write an excel macro to do a file conversion and I need to beable to obtain the sequence number of an atom in a molecule. I have named all the atoms in the molecule and can select them at will. However, when I issue the DDERequest for "selected-atom 1" which should return the desired info I get the following message back Query-value: Unknown variable 'selected-atom' Any ideas what's wrong? Is there another way to get the desired info? I'm working with release 4 for Windows. Dr. Ross Underhill Royal Military College of Canada Kingston, Ontario K7K 5L0 (613) 541-6000 X6175 From: Stavrev, Krassimir [stavrev@hyper.com] Sent: Wednesday, February 18, 1998 1:29 PM To: 'underhill-r@rmc.ca' Cc: 'hyperchem@hyper.com' Subject: RE: selected-atom variable Check your Control Panel/Regional Settings. Language settings other than English(US) can produce similar error messages. Regards, Krassimir --- Krassimir Stavrev, PhD, Director of Scientific Support Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 -----Original Message----- From: underhill-r@rmc.ca [SMTP:underhill-r@rmc.ca] Sent: Wednesday, February 18, 1998 12:11 PM To: hyperchem@hyper.com Subject: selected-atom variable I'm trying to write an excel macro to do a file conversion and I need to beable to obtain the sequence number of an atom in a molecule. I have named all the atoms in the molecule and can select them at will. However,when I issue the DDERequest for "selected-atom 1" which should return The desired info I get the following message back Query-value: Unknown variable 'selected-atom' Any ideas what's wrong? Is there another way to get the desired info? I'm working with release 4 for Windows. Dr. Ross Underhill Royal Military College of Canada Kingston, Ontario K7K 5L0 (613) 541-6000 X6175 From: Ross Underhill [underhill-r@rmc.ca] Sent: Wednesday, February 18, 1998 3:35 PM To: 'Stavrev, Krassimir' Subject: RE: selected-atom variable That indeed solved the problem ( or at least made it go away). Thanks! I can't say I'm thrilled about having my settings messed up for one program -----Original Message----- From: Stavrev, Krassimir [SMTP:stavrev@hyper.com] Sent: Wednesday, February 18, 1998 1:29 PM To: 'underhill-r@rmc.ca' Cc: 'hyperchem@hyper.com' Subject: RE: selected-atom variable Check your Control Panel/Regional Settings. Language settings other than English(US) can produce similar error messages. Regards, Krassimir --- Krassimir Stavrev, PhD, Director of Scientific Support Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 -----Original Message----- From: underhill-r@rmc.ca [SMTP:underhill-r@rmc.ca] Sent: Wednesday, February 18, 1998 12:11 PM To: hyperchem@hyper.com Subject: selected-atom variable I'm trying to write an excel macro to do a file conversion and I need to beable to obtain the sequence number of an atom in a molecule. I have named all the atoms in the molecule and can select them at will. However, when I issue the DDERequest for "selected-atom 1" which should return the desired info I get the following message back Query-value: Unknown variable 'selected-atom' Any ideas what's wrong? Is there another way to get the desired info? I'm working with release 4 for Windows. Dr. Ross Underhill Royal Military College of Canada Kingston, Ontario K7K 5L0 (613) 541-6000 X6175 From: Mr HG Kruger [kruger@scifs1.und.ac.za] Sent: Friday, February 20, 1998 3:20 AM To: hyperchem@hyper.com Subject: molecular dynamics I am trying to find the chair structure of cyclohexane when starting with the boat structure by using molecular dynamics. Unsuccessful. I believe it should be possible with both molecular mechanics as well as semi empirical methods. Has any one achieved it before? If so, would you be so kind to send me the conditions/settings? Thanks for your time Gert Kruger __________________________________________________________ Dr HG Kruger, Dept Chemistry, University of Natal, PO Box 18091, Dalbridge 4014, Durban, South Afica Tel +27-31-2602181 Fax +27-31-2603091 Email kruger@scifs1.und.ac.za __________________________________________________________ From: Bellnier, David [dbellnie@hs1.buffalo.edu] Sent: Friday, February 20, 1998 2:41 PM To: hyperchem@hyper.com Subject: amphiphilicity I get to ask this same question every year or so, hoping for some insight! Is there any way to computationally determine a "amphiphilic moment", of sorts, for a molecule? My specific interest is in porphyrin-like molecules, where amphiphilic molecules (for example, a porphyrin with a long aliphatic chain on one side and a carboxylic acid function on the opposite side) tend to have more photodynamic activity in vitro and in vivo. Thanks (again). From: Ron Starkey [STARKEYR@GBMS01.UWGB.EDU] Sent: Friday, February 20, 1998 2:16 PM To: Mr HG Kruger Cc: hyperchem@hyper.com Subject: Re: molecular dynamics On Fri, 20 Feb 1998, Mr HG Kruger wrote: > I am trying to find the chair structure of cyclohexane > when starting with the boat structure by using molecular dynamics. > Unsuccessful. I believe it should be possible with both molecular > mechanics as well as semi empirical methods. > > Has any one achieved it before? If so, would you be so kind to send > me the conditions/settings? > > Thanks for your time > Gert Kruger > __________________________________________________________ Gert, It sure can be done. If you do an MD at 300K it may take a while. At 600K it will leave the boat conformation sooner. You, of course, will not always obtain a chair conformation, entropy will favor a twist-boat conformation. With several attempts you will eventualy gat a chair. I have my students repeat a MD at 300K 5 times and record the resulting cyclohexane conformation. Almost everyone will get at least a chair on one of the trials. Ronald Starkey Chemistry Univ Wisconsin - Green Bay From: Pastorek, Christine [pastorekc@quintessence.chem.orst.edu] Sent: Saturday, February 21, 1998 5:38 PM To: 'hyperchem@hyper.com' Subject: question on Model Building substituted cyclohexanes I am using HyperChem v. 5.01 on a PC and would like to know if there is any way to get around the following problem. When I build a substituted cyclohexane molecule and use Model Build to start, I get the axial instead of equatorial position for the substituted group. The structure comes to the correct equatorial position if I go directly from building the structure to geometric optimization without the benefit of Model Build. In general, it is more efficient to use Model Build first, then go on to geometric optimization. I would like to be able to do this with this group of molecules. Is there some way to change the Model Build algorithm? > Dr. Christine Pastorek > Integrated Laboratory Program Director > Gilbert Hall 153 > Department of Chemistry > Oregon State University > Corvallis, OR 97331-4003 > Voice: (541) 737-6732 > FAX: (541) 737-2062 > mailto:pastorekc@chem.orst.edu > From: Stavrev, Krassimir [stavrev@hyper.com] Sent: Monday, February 23, 1998 10:10 AM To: 'Pastorek, Christine' Cc: 'hyperchem@hyper.com' Subject: RE: question on Model Building substituted cyclohexanes Christine: What if you apply Build/Constrain Geometry (tetrahedral) to the substituted carbon prior to the model building? I think this may help. Regards, Krassimir --- Krassimir Stavrev, PhD, Director of Scientific Support Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 -----Original Message----- From: Pastorek, Christine [SMTP:pastorekc@quintessence.chem.orst.edu] Sent: Saturday, February 21, 1998 5:38 PM To: 'hyperchem@hyper.com' Subject: question on Model Building substituted cyclohexanes I am using HyperChem v. 5.01 on a PC and would like to know if there is any way to get around the following problem. When I build a substituted cyclohexane molecule and use Model Build to start, I get the axial instead of equatorial position for the substituted group. The structure comes to the correct equatorial position if I go directly from building the structure to geometric optimization without the benefit of Model Build. In general, it is more efficient to use Model Build first, then go on to geometric optimization. I would like to be able to do this with this group of molecules. Is there some way to change the Model Build algorithm? > Dr. Christine Pastorek > Integrated Laboratory Program Director > Gilbert Hall 153 > Department of Chemistry > Oregon State University > Corvallis, OR 97331-4003 > Voice: (541) 737-6732 > FAX: (541) 737-2062 > mailto:pastorekc@chem.orst.edu > From: Pastorek, Christine [pastorekc@chem.orst.edu] Sent: Monday, February 23, 1998 3:50 PM To: 'Stavrev, Krassimir' Subject: RE: question on Model Building substituted cyclohexanes Krassimir, Yes that does it. Thank you. Why is this necessary and when should I use this option? Chris Pastorek > -----Original Message----- > From: Stavrev, Krassimir [SMTP:stavrev@hyper.com] > Sent: Monday, February 23, 1998 7:10 AM > To: 'Pastorek, Christine' > Cc: 'hyperchem@hyper.com' > Subject: RE: question on Model Building substituted cyclohexanes > > > > Christine: What if you apply Build/Constrain Geometry (tetrahedral) to > the substituted carbon prior to the model building? I think this may > help. > Regards, > Krassimir > --- > Krassimir Stavrev, PhD, Director of Scientific Support > Hypercube, Inc. Florida Science and Technology Park > 1115 N.W. 4th Street Gainesville, Florida 32601 > Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 > > > -----Original Message----- > From: Pastorek, Christine [SMTP:pastorekc@quintessence.chem.orst.edu] > Sent: Saturday, February 21, 1998 5:38 PM > To: 'hyperchem@hyper.com' > Subject: question on Model Building substituted cyclohexanes > > > I am using HyperChem v. 5.01 on a PC and would like to know if there > is > any way to get around the following problem. > When I build a substituted cyclohexane molecule and use Model Build to > start, I get the axial instead of equatorial position for the > substituted group. The structure comes to the correct equatorial > position if I go directly from building the structure to geometric > optimization without the benefit of Model Build. In general, it is > more > efficient to use Model Build first, then go on to geometric > optimization. I would like to be able to do this with this group of > molecules. Is there some way to change the Model Build algorithm? > > Dr. Christine Pastorek > > Integrated Laboratory Program Director > > Gilbert Hall 153 > > Department of Chemistry > > Oregon State University > > Corvallis, OR 97331-4003 > > Voice: (541) 737-6732 > > FAX: (541) 737-2062 > > mailto:pastorekc@chem.orst.edu > > From: Stavrev, Krassimir Sent: Monday, February 23, 1998 4:05 PM To: 'Pastorek, Christine' Subject: RE: question on Model Building substituted cyclohexanes Christine: That's just a way to change the default preferences of the model builder. I discovered it accidentally. Another way of changing the stereochemistry prior to the geometry optimization is to use a Shift-click-Draw technique-> build the substituted cyclohexane first (the group goes in axial position, as you described it). Then click on the drawing tool and activate the drawing cursor, then press the Shift button, HOLD IT and click on the substituted carbon of the cyclohexane ring - the group that is attached to it should switch to equatorial position. Regards, Krassimir --- Krassimir Stavrev, PhD, Director of Scientific Support Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 -----Original Message----- From: Pastorek, Christine [SMTP:pastorekc@chem.orst.edu] Sent: Monday, February 23, 1998 3:50 PM To: 'Stavrev, Krassimir' Subject: RE: question on Model Building substituted cyclohexanes Krassimir, Yes that does it. Thank you. Why is this necessary and when should I use this option? Chris Pastorek > -----Original Message----- > From: Stavrev, Krassimir [SMTP:stavrev@hyper.com] > Sent: Monday, February 23, 1998 7:10 AM > To: 'Pastorek, Christine' > Cc: 'hyperchem@hyper.com' > Subject: RE: question on Model Building substituted cyclohexanes > > > > Christine: What if you apply Build/Constrain Geometry (tetrahedral) to > the substituted carbon prior to the model building? I think this may > help. > Regards, > Krassimir > --- > Krassimir Stavrev, PhD, Director of Scientific Support > Hypercube, Inc. Florida Science and Technology Park > 1115 N.W. 4th Street Gainesville, Florida 32601 > Voice:(352) 371-7744/(800) 960-1871 Fax:(352) 371-3662 > > > -----Original Message----- > From: Pastorek, Christine [SMTP:pastorekc@quintessence.chem.orst.edu] > Sent: Saturday, February 21, 1998 5:38 PM > To: 'hyperchem@hyper.com' > Subject: question on Model Building substituted cyclohexanes > > > I am using HyperChem v. 5.01 on a PC and would like to know if there > is > any way to get around the following problem. > When I build a substituted cyclohexane molecule and use Model Build to > start, I get the axial instead of equatorial position for the > substituted group. The structure comes to the correct equatorial > position if I go directly from building the structure to geometric > optimization without the benefit of Model Build. In general, it is > more > efficient to use Model Build first, then go on to geometric > optimization. I would like to be able to do this with this group of > molecules. Is there some way to change the Model Build algorithm? > > Dr. Christine Pastorek > > Integrated Laboratory Program Director > > Gilbert Hall 153 > > Department of Chemistry > > Oregon State University > > Corvallis, OR 97331-4003 > > Voice: (541) 737-6732 > > FAX: (541) 737-2062 > > mailto:pastorekc@chem.orst.edu > >