Full interview

——Congratulations on your publication in Nature!

Dr. Naoya Aizawa(Right),
Dr. Tatsuo Nakagawa[UNISOKU](Left)

UNISOKU :First of all, congratulations on the publication of your research in Nature. This is truly an outstanding achievement, and we are delighted that our picoTAS was used in your research. Moreover, this is first time which picoTAS contributes to a Nature’s publication. Regarding the title of your paper, I was already amazed by the reversal of energy between Singlet and Triplet states, disobeying the Hund's rule, however, I'm curious, why is it titled "Delayed fluorescence from inverted..."? Could you provide some insight into this?

Dr. Aizawa :Thank you very much. The reason we added "Delayed Fluorescence" to the title is that simply having the reversal of energy between Singlet and Triplet states isn't unique, as there are other examples. To claim it as a world-first, we needed to emphasize the aspect of delayed fluorescence. By indicating that we have reversed the energy difference (ΔE_ST) between the excited Singlet state (S₁) and Triplet state (T₁) and achieved delayed fluorescence, experts will recognize that this is something truly groundbreaking and unprecedented.

UNISOKU :Was there a special reason for submitting to Nature instead of Science? Looking at the paper, I noticed that the submission date was April 29, 2021, and it was accepted on July 21, 2022, which is a period of 1 year and 3 months. What took so long for the review process?

Dr. Aizawa :Personally, I feel more delighted about the publication in Nature. In the field of chemistry, Nature has a higher impact compared to Science, in my opinion. As for the lengthy review process, it simply came down to the fact that some reviewers took a long time to respond (laughs). Additionally, the review itself was quite rigorous. While some reviewers were quick to provide feedback, others took their time...

——How has the response been?

UNISOKU :How has the response been after the publication in Nature? Also, what are your thoughts on the future implications and applications of your research?

Dr. Aizawa :We have received positive feedback from everyone, expressing their fascination with the fact that we have broken the Hund's rule which is taught in textbooks. Our research was even nominated for the 2022 C&EN's molecules of the year. We have also seen an increase in opportunities for collaborative research and received inquiries from companies interested in practical applications.
The study of organic light-emitting diodes (OLEDs) is truly an applied research field that requires knowledge and expertise from various disciplines such as organic chemistry, photochemistry, quantum chemistry, and semiconductor physics. Interestingly, even in applied research like this, we often gain insights into fundamental science from it. This study, which disobeys the Hund's rule by achieving a negative ΔE_ST, serves as a great example of that.
Moving forward, we aim to discover new materials that have potential for practical applications, such as those used in smartphone displays. For instance, in the field of delayed fluorescence, we have proposed a new type called "H-type" as the heptazine molecule was first synthesized for this purpose. We want to further investigate the properties of this new type of delayed fluorescence molecule with a negative ΔE_ST to advance the fundamental photochemical science.

UNISOKU : How has the response been in Yamagata University and other collaborators? I noticed that Yamagata University issued a press release as well. I found many scientists involved in your research. Could you please tell us about the division of responsibilities if you don’t mind to?

Dr. Aizawa : Yes, Dr. Pu, RIKEN, who is the second author, is also affiliated with Yamagata University, so they have issued a press release, too. I primarily handled experiments and calculations related to emission spectra and other spectroscopic aspects, while organic synthesis was primarily led by Dr. Miyajima, who works in RIKEN.

——About inspiration for you research

UNISOKU :There is a concept called serendipity, where unexpected phenomena or results are discovered first, and theories or objectives are developed afterward. How about your research in that regard?

Dr. Aizawa :This research had a theoretical foundation first, then we aimed to design materials that exhibit a negative ΔE_ST based on that. The possibility of a reversed ΔE_ST in heptazine derivatives had been discussed in theoretical studies since the 1980s, but there have been no experimental examples. As an experimentalist, I had the idea around 2020 to synthesize and validate these theories. I collaborated with Prof. Miyajima and others who excel in synthesis, with the intention that even if the ΔE_ST didn't turn out to be negative, we would still create molecules having a small ΔE_ST. By the end of that year, we had started to collect core data, but when we discovered that the synthesized molecules indeed had a negative ΔE_ST, I was quite surprised.

——How did you choose the two molecules?

UNISOKU :The term "Computational Screening" is mentioned in the paper. How did you go about screening over 35,000 compounds and selecting the approximately 1,000 molecules? And regarding the two molecules, HzTFEX₂ and HzPipX₂, how did you choose them finally for this study? And I’m curious about the quantum chemical calculations of ΔE_ST and how much time was required for each calculation?

Dr. Aizawa :As for the selected molecules, our criteria included ease of synthesis and, yes, a touch of intuition (laughs). We first synthesized HzPipX₂, which had a borderline positive ΔE_ST and thought it might work. Then, when we synthesized HzTFEX₂ as the second molecule, which exhibited a negative ΔE_ST. It was quite lucky. From a more practical perspective, there may be high-performance molecules to explore. Considering not only the emission properties but also the carrier injection and overall quantum yield, I believe there are still many molecules that can exhibit negative ΔE_ST. Furthermore, since it is a novel area in photochemistry, there are aspects that are yet to be understood, such as the effect of increasing the magnitude of negative ΔE_ST. I would like to investigate these fundamental scientific aspects as well.
Regarding computational time, it took about two months to finish a single calculation using EOM-CCSD method* for each molecule. While it may be possible to optimize the calculations for faster speed, it is still faster to synthesize molecules and verify its properties experimentally (laughs). Computational time remains a challenge. *EOM-CCSD: A theoretical calculation method called Equation-of-Motion Coupled Cluster with Single and Double excitations.

——How did our picoTAS and CoolSpek contribute for the publication?

UNISOKU :What role did our products, picoTAS and CoolSpeK, play in the process leading up to the publication of the paper? How did picoTAS contribute during the peer review process? At what point did you gain confidence? Additionally, in the paper, the transient absorption results are presented in the Extended Figure rather than the main figure. Was this due to limitations on the number of figures in Nature?

Dr. Aizawa :Through transient absorption measurements using picoTAS, we were able to obtain decisive evidence indicating reversible intersystem crossing between S₁ and T₁ states (Extended Data Fig. 1a-c). Transient fluorescence alone cannot definitively demonstrate the involvement of the T₁ state in emission. During the peer review process, we were asked to provide low-temperature transient absorption data, and thanks to the CoolSpeK-SLIM specifically designed for picoTAS, we were able to fulfill this request (Extended Data Fig. 3a-c). In this study, we also obtained another crucial piece of data using CoolSpeK. It involved measuring the emission spectra of the molecule at low temperatures. For typical TADF molecules, the emission lifetime becomes longer as the temperature decreases, but in the case of heptazine molecules, we confirmed that the emission lifetime becomes shorter (Fig. 3-d). At that moment, I distinctly remember feeling a sense of excitement and thinking, "This is it!"
Regarding the figures, you're correct. Nature imposes limitations on the number of figures that can be included in the main article. However, Extended Figures hold a position closer to the main figures than the supporting information.

Transient absorption data of HzTFEX₂ (from Extended Data Fig. 1a-c):
a. Transient absorption of HzTFEX₂ as a function of wavelength and time in a deaerated toluene solution.
b. Integrated transient absorption spectra of HzTFEX₂ over 0-500 ns in deaerated and aerated toluene solutions.
c. Transient absorption decays of S₁ and T₁ monitored at 700 nm and 1,600 nm, respectively.

Rate constants k_ISC and k_RISC of HzTFEX₂ and HzPipX₂ (from Extended Data Fig. 3a-c):
a. Temperature dependence of the rate constants k_ISC and k_RISC of HzTFEX₂ in deaerated toluene.
b. Temperature dependence of the rate constants k_ISC and k_RISC of HzPipX₂ in deaerated toluene.
c. Schematic diagram of the potential energy surfaces of S₁ and T₁ and the activation energies of ISC and RISC.

——The data can be obtained only with picoTAS!

Temperature dependence of
the time constant of delayed fluorescence τ_DF of
HzTFEX₂ and HzPipX₂ (From Fig. 3-d).

UNISOKU :Upon examining the data, we can see that picoTAS has successfully captured remarkable features such as highly emissive samples, seamless data spanning from the visible to near-infrared range, and decay times ranging from nanoseconds to sub-microseconds. It seems that the referees did not specifically comment on these aspects, but we must emphasize here that these are intricately detailed measurements that would be challenging to obtain with other commercially available instruments. Without a doubt, the data obtained through picoTAS is extraordinary, and it is unlikely to be acquired using other devices. 　　　　　　　

Dr. Aizawa :Indeed, typically there are discontinuities or gaps in time-resolved absorption/emission spectra. The nanosecond rise portion is crucial as well.

——Had you ever measured transient absorption before picoTAS purchase？

UNISOKU :Was the consideration of purchasing picoTAS for transient absorption measurements in this study to obtain experimental evidence? Additionally, there are various techniques for transient absorption, such as the pump-probe method. Did you have any previous experience with transient absorption measurements?

Dr. Naoya Aizawa (left) and Dr. Kenichi Nakayama (right)
with picoTAS in the background

Dr. Aizawa :Yes, initially I purchased picoTAS solely for this research (laughs). The reason being, for known types of TADF molecules, there is a certain understanding that if the transient fluorescence takes a certain form, it then can be recognized as TADF. However, in this case, it is a completely new system, and the transient fluorescence alone is not convincing. To determine whether the triplet state contributes to delayed fluorescence over time, it is necessary to demonstrate through transient absorption that there is intersystem crossing from the triplet state and that the lifetime is extremely short. This could only be achieved with picoTAS.
I had no prior experience with transient absorption measurements myself, nor did I have anyone close to me who had such experience. I am grateful to UNISOKU for providing guidance. Moreover, I believe techniques like femtosecond pump-probe would have been challenging to handle, and since the region of interest for observation was in the nanosecond to sub-microsecond range, picoTAS perfectly fit in terms of temporal resolution and cost.

——Will you continue conducting transient absorption measurements?

UNISOKU :Will you continue conducting transient absorption measurements in the future?

Dr. Aizawa :Yes, definitely. Currently, with my students, we are trying to perform transient absorption measurements on the T₁ state, which has a small population.

UNISOKU :We are delighted to hear that you will continue with the measurements. By the way, in your paper, it seems that you used data at a wavelength of 1600 nm to track the triplet and data at 700 nm to track the singlet. So, you were able to use wavelengths that conveniently allowed tracking of either the triplet or singlet state. The transient absorption spectrum, which is a spectrum of the excited state, is not known in advance, right?

Dr. Aizawa :You are right, we could not know it before measuring. But, It is possible to calculate the transient absorption, and to some extent, you can predict the absorption wavelength in advance. It is a challenging task though. By performing calculations on approximately 50 higher-order S_n and T_n states, we can roughly reproduce the spectrum. However, there may be a slight discrepancy between the theoretical and experimental values.

UNISOKU :Recently, I attended an applied course on photochemistry focusing on quantum chemical calculations, but there was no mention of transient absorption. I wanted to ask if it is possible to calculate transient absorption, but I felt too inexperienced to ask. What are your thoughts on this matter?

Dr. Aizawa :I would appreciate if you asked. Calculating UV absorption and ground state properties is possible with Gaussian software. The difficulty lies in modeling the transient absorption between excited states and higher-order excited states. It involves not only energy but also oscillator strength, which determines the intensity of absorption. The calculation of the excited state dipole is completely different from that of the ground state dipole. Therefore, some ingenuity is required. Although Gaussian does not have a specific keyword for transient absorption calculations, it can be done by setting it up manually. It is not well-known, but if you search, you will find the relevant information. There are also research papers that calculate transient absorption using Gaussian. However, the accuracy is not that high, and there can be deviations of about 100 nm. Nevertheless, the overall trends tend to align for molecules in solvent where it can be treated as a single molecule.

*Gaussian: Software for quantum chemistry calculations

——Could you please share your past experiences and career background?

Dr. Aizawa

UNISOKU :This is an out of curiosity question, but could you please tell us about your past experiences, the timing and inspiration to become a researcher, and any experiences you had with overseas research?

Dr. Aizawa :After obtaining my degree at Yamagata University under Prof. Kido, I moved to Kyushu University where I worked as a project assistant professor (2015-2018), and then I worked as a postdoctoral researcher in RIKEN (2019-2021). And now I am at Osaka University.
 I decided to pursue an academic career around the time of my master course. I went on some company visits before starting my job search, but I felt that they were not suitable for me. At that time, the research I was involved in became more interesting, and I also had a desire to continue pursuing it. During my first or second year of the doctoral program, I had the opportunity to study abroad at the Georgia Institute of Technology for six months. It was a memorable experience as it coincided with my decision to pursue a doctoral degree and the Great East Japan Earthquake.

UNISOKU :You have worked in Yamagata University, Kyushu University, RIKEN, and Osaka University. Have you noticed differences in the research environments at each institution?

Dr. Aizawa :Yes, there are differences between them. At RIKEN, I had more time. On the other hand, I found universities are also great because working with students is enjoyable. In RIKEN, the number of students is really small.

——Impression of UNISOKU, and relation to photochemistry

UNISOKU :What was your impression of our company before purchasing the picoTAS? How was your impression after visiting our company?

Dr. Aizawa :I didn't have any particular preconceptions about UNISOKU as I didn't know much about the company. I had an interest in transient absorption measurements since my time at Kyushu University, and I remember having a discussion with someone from UNISOKU who came to explain the product. I think it was around 2017-2018, and I believe it was you? At that time, I had the impression that the pump-probe measurement system was a bit expensive, around 50 million yen.
* Dr. Aizawa considered purchasing the picoTAS around early January 2020 and visited our company in mid-February of the same year.

UNISOKU :We have participated in photochemistry conferences in the past. Have you attended these conferences frequently?

Dr. Aizawa :No, I have mainly attended JSAP (Japan Society of Applied Physics), CSJ (Chemical Society of Japan), and organic EL-related fields conferences. I think this might be my first time attending the photochemistry conference. When I wanted to perform transient absorption measurements for this research, I searched and found the picoTAS. I purchased it using PRESTO funding.

——High-end models of picoTAS are available in the demo room

UNISOKU :Approximately two months after your visit, we have set up a demo room and started offering paid in-house experimental services using picoTAS, nano-second system, and other equipment. For example, although you purchased a picoTAS with an excitation wavelength of 355 nm, in the demo room, we have a picoTAS permanently set up that allows you to change the excitation wavelength from 410 nm to 680 nm. The temporal resolution is 100 ps, providing higher performance than yours. In your research, you may be interested in observing the molecular response when changing the incident wavelength, right?

Dr. Aizawa :When performing theoretical calculations, it is naturally attractive to be able to change the excitation wavelength since the absorption wavelength varies depending on the molecule and its electronic state. Currently, I am measuring transient absorption spectra by exciting higher-order S_n states. It would be interesting to see the direct excitation of the S₁ state by inputting light at 450 nm.

——Half-price fee for academic users

UNISOKU :Since you are a picoTAS user, you can use our in-house experimental service for free for up to two days per year. We would like you to take advantage of this opportunity, but from the third day onward, it becomes a paid service. The regular price is 80,000 yen per day. However, for academic users like you, the price is halved, so it would be 40,000 yen per day.

Dr. Aizawa :The price is very reasonable. Is the free in-house experiment limited to two visits per fiscal year or two visits per calendar year? (UNISOKU: It is limited to within the fiscal year.) Can we ask you to do experiments without visiting your company? (UNISOKU: I'm sorry, we don't offer that.) By the way, do you know any facilities that perform transient absorption experiments on behalf of customers?

UNISOKU :As far as we know, there probably aren't many facilities. For in-house experiments in the demo room, it is possible for students to conduct the experiments while the professor participates online.

Dr. Aizawa :Transient absorption measurements seem to be quite in demand. I often hear that facilities that have such equipment are busy.

UNISOKU :Finally, do you have any specific wishes or requests for UNISOKU or measurement instrument manufacturers?

Dr. Aizawa :It would be very convenient if picoTAS and CoolSpeK could be integrated to measure the temperature dependence of transient absorption automatically. Your customer care, such as technical support, has been excellent, and I hope you continue to provide that in the future.

*For more details on in-house experiments, please refer to this link.