Olá, Carlos, obrigado pela pergunta. O método é um pouco mais pesado que o do Le Roy, por exemplo, e as previsões são mais ou menos as mesmas. Mas veja: é um algoritmo novo, que foi aplicado a alguns sistemas e ainda será aplicado em outros. O escopo aqui é esse, entende? Hoje, o grupo está aplicando o método no método HFR e de correlação eletrônica, principalmente. Mas vejo com bons olhos aplicar na espectroscopia vibracional, uma vez que os cálculos freq do gaussian, por exemplo, gera espectros através de métodos da física clássica. Ou seja, ainda tem bastante água pra rolar embaixo dessa ponte.

Muito obrigado, Carlos, espero ter respondido sua questão.

Quem por acaso quiser, estou nessa sala a disposição para conversarmos hoje.

If you would like to talk with me today, I'm in this link.

Hello  Annika. Could we talk a little bit in a meet? I am with my daughter and write in English now is quite impossible. If you don't mind, I'm here, meet.google.com/zfq-qfyr-cgi

Of course, if you can't talk with me now, I answer you as soon as possible.

Hello again, Annika. It's a pleasure to talk with you, I have already read some of your articles. I am having some hardware issues right now, so your first question was in presentation but it is not audible. We use VQMC method. The variational princicle is almost the only imposition of the algorithm. 2. No, our method is, nowadays, only for calculation of ro-vibrational wavefunctions and energies. It means (if I understood your question) that we use some potential curve generated from another place. In this work, Morse Potential. But I do wish that we publish an article soon that we apply our method from MRCI, CC and/or composite vibrational curve. There are some people using something different in way that you told about, but it is not the scope of our attention right now. 3. I don't think I understood your question. Are you interested in the error from the experimental results? Or the error from our calculation instead? In the case of the first one, our errors are quite small, specially if you see the results on another calculations like that. People uses grid from 100 to 1000 point, we are using only 50 and the results are very close to the experimental ones. About the error of the method, well. I don't know, I confess. Once the variational principle is used, the method intend to be exact, if I'm not completely wrong. So, I don't think the error bar could be applied in this case. But I am definetly not sure about this. As soon as possible I refresh this third question.

Anyway, thank you so much for your questions, it is a pleasure talk with you. I have classes on thursday in your lecture, but I will watch it later for sure. Thank you again.

Grande Regis, saudades de você! Grande abraço.