Search the mailing list, this question has been raised many times before.

2016-01-20 7:10 GMT+01:00 Riya Rogers <[email protected]>:

> Thanks a lot for your valuable comments Dr Nick
>
> But as Rajan pointed out, I am facing the same problem of transiesta
> convergence.
>
> Should I use : TS.ReplicateA2Left/Right   to increase my electrode size
> and may cause in convergence?
>
> Please comment.
>
> With Regards
> Riya
>
> On Wed, Jan 20, 2016 at 10:33 AM, RAJAN SINGH <[email protected]>
> wrote:
>
>> Thank you Dr Nick
>>
>> After doing relaxation from siesta when I am running my transiesta it
>> wont converge. The Qto and Qt values are not same. How to achieve that?
>>
>> Efermi from SIESTA                   :    -3.47054
>> Total charge                  [Qt0]  :   414.00000
>> Charge in update region       [Qc]   :   133.70127
>> Charge outside update region  [Qcn]  :   280.29873
>> Left electrode                [L]    :   133.72936
>> Left electrode/device         [L-C]  :     3.91751
>> Device                        [C]    :   133.70127
>> Device/right electrode        [C-R]  :     4.65178
>> Right electrode               [R]    :   133.34561
>> Other                         [O]    :     4.65448
>>
>>  Reading GF file, with title:
>>    nanoribbon.TSGFL
>>  Title: 'Generated GF file'
>>
>>  Reading GF file, with title:
>>    nanoribbon.TSGFR
>>  Title: 'Generated GF file'
>>
>> ts-charge:        O        L      L-C        C      C-R        R       Qt
>> ts-charge:    4.654  133.729    3.918  329.548    4.652  133.346  609.847
>>
>>
>> With Regards
>> Rajan
>>
>> On Wed, Jan 20, 2016 at 12:06 AM, Nick Papior <[email protected]>
>> wrote:
>>
>>> Please re-read the slides _carefully_. You should probably not relax the
>>> entire device region.
>>>
>>> This is slightly touched in the slides.
>>>
>>> 2016-01-19 19:33 GMT+01:00 RAJAN SINGH <[email protected]>:
>>>
>>>> Thank You Prof. Nick.
>>>>
>>>> I read the slides according to them the steps should be :
>>>>
>>>> 1. Relax the Device atoms using siesta run (CG run)
>>>> 2. After relaxation attach the electrodes.
>>>> 3. Again perform the Siesta relaxation (CG run) but this time put a
>>>> geometry constraints on electrode atoms.
>>>> 4. after step 3 relaxation go for Transiesta Run.
>>>>
>>>> Is that right?
>>>>
>>>> Please help me.
>>>>
>>>> With Regards
>>>> Rajan
>>>>
>>>> On Tue, Jan 19, 2016 at 11:49 PM, Nick Papior <[email protected]>
>>>> wrote:
>>>>
>>>>> Check these slides, and see if they will help you (they are a guiding
>>>>> example!):
>>>>>
>>>>> http://departments.icmab.es/leem/siesta/tlv14/slides12.pdf
>>>>>
>>>>>
>>>>> 2016-01-19 19:04 GMT+01:00 RAJAN SINGH <[email protected]>:
>>>>>
>>>>>> Dear Siesta Users
>>>>>>
>>>>>> Kindly give me advice on the following:
>>>>>>
>>>>>> In order to perform the transiesta calculation on relaxed device
>>>>>> structure the procedure which I am performing is:
>>>>>>
>>>>>> 1. First I write a .fdf file (left electrode + Device + right
>>>>>> electrode).
>>>>>> 2. then I remove my electrode atoms from file (KEEPING THE LATTICE
>>>>>> VECTORS AS SAME)
>>>>>> 3. Then perform a CG relaxation run.
>>>>>> 4. Once the run completed then I modify my .fdf file
>>>>>>     (left electrode + RELAXED Device + right electrode)
>>>>>> 5. Then perfom the transiesta run.
>>>>>>
>>>>>>
>>>>>> Please tell me whether the above process is appropriate and correct
>>>>>> or not. I am really a bit confuse in this step. I really need help in 
>>>>>> this
>>>>>> please help me.
>>>>>>
>>>>>> Thank You
>>>>>>
>>>>>> With Regards
>>>>>> Rajan Singh
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Kind regards Nick
>>>>>
>>>>
>>>>
>>>
>>>
>>> --
>>> Kind regards Nick
>>>
>>
>>
>


-- 
Kind regards Nick

Responder a