Science is premised on being a self-correcting enterprise, building upon past discoveries to reach new heights. Reported literature is subject to vetting by others to confirm the validity of reported results. But what happens when this process is disrupted?

We rely on scientists to uphold rigorous standards, ensuring their results are accurate and reproducible. Unfortunately, there's not always much incentive to double-check other researchers' work. This is because, in practice, credit is given for reporting new advances, not verifying the results of others. In turn, a wild chase of an “exciting new development” often leads to poorly vetted, “sloppy” results being published quickly. This has led to a "reproducibility crisis"[i] in some fields, where major studies can't be replicated, casting doubt on their findings, and sometimes even the validity of the field, at large.

While many prominent displays of the scientific process working as it should do exist, examples of Hendrik Schön[ii] or, more recently Ranga Dias[iii], cast immediate doubt on the integrity of the process as a whole. False claims are risky to public health; they short the stock of the public’s trust in the scientific process, and they also divert funding from valid research while also running the risk of damaging the careers of involved but unsuspecting scientists.

The desire to fix the situation drives the need for scientists to conduct experiments with the highest standards. We conjecture that one efficient way to assist the realization of this lofty goal is to formalize standards for acceptable practices and minimal experimental demonstrations, which are required to confirm long-sought-after goals. In practice, this entails developing common but field-specific guidelines or checklists[iv] for the reporting of results. Arguably, such practice would minimize low-quality (incorrect or irreproducible) reporting and, in tandem, will provide clear, testable measures for reproducibility – all while improving the efficiency of the scientific enterprise. It may even prevent future controversies!

The field of condensed-phase laser refrigeration is one field where standards are needed. The first 1995 report of a 0.3-K cooling of a rare-earth-doped glass [v] quickly motivated a vibrant community around the idea of using coherent light to cool solids. Today, laser cooling to 91 K is possible using Yb³⁺-doped crystals. Unfortunately, the existence of a minimum achievable temperature with these materials means that going to lower temperatures requires a shift to a different material system. A race has therefore begun to use light to cool semiconductors, promising cooling floors as low as 10 K due to favorable Fermi-Dirac statistics.

Currently, there are several reports of observed laser cooling in semiconductors.[vi],[vii],[viii] Unfortunately, there has been no standardization of reported results, causing large variability in the quality of such important claims. For instance, this has led to controversy as some reported results did not conform to established physical principles.[ix] At a broader level, claims of semiconductor optical cooling are generally hampered by incomplete experimental details. Key performance metrics are often implied (or assumed) rather than directly measured.

Our Expert Recommendation article in Nature Reviews Physics is therefore born out of a need to standardize the reporting of condensed phase laser cooling results.[x] We therefore revisit the fundamental tenets of optical refrigeration to establish these rules. Four key principles result in guiding the verification and reporting of new cooling results: (1) the reporting of clear and consistent optical cooling metrics, specifically sample emission quantum yields and absorption/up-conversion efficiencies, (2) explicit demonstrations of heating and cooling regimes, depending on excitation frequency, (3) establishing the thermodynamic consistency of cooling timescales and final temperatures, and (4) a reporting of the temperature measurement technique, calibration procedure, and time resolution. These principles are intended to aid the evaluation of new claims as well as provide a framework for critically reviewing existing literature in the field.

It is our sincere hope that these Expert Recommendations help advance progress in this nascent field. More broadly, it is time to recognize that demanding research, especially the type which relies on the absolute performance characteristics of advanced materials - be it condensed phase optical refrigeration, room-temperature superconductivity, light transistors, Majorana fermions, etc…. all require such rigorous standardization. Beyond impacting our optical refrigeration field, we hope that setup of such guides can be adopted by other fields. Finally, it is important to note that the presented tenets are based on fundamental arguments, which are unlikely to change even if new conceptual developments occur in the field. Nonetheless, the guidelines themselves are subject to a public debate. And that’s a good thing. For everyone.

*The article is contributed by Yang Ding, Zhuoming Zhang, Denis V. Seletskiy, Peter J. Pauzauskie, Masaru Kuno.

[i] Reflections on scientific fraud. Nature 419, 417 (2002).

[ii] Physicist found guilty of misconduct. Nature (2002). https://doi.org/10.1038/news020923-9

[iii] Garisto, D. Superconductivity researcher who committed misconduct exits university. Nature 635, 791–792 (2024).

[iv] Checklists work to improve science. Nature 556, 273-274 (2018).

[v] Epstein, R. I., Buchwald, M. I., Edwards, B. C., Gosnell, T. R. & Mungan, C. E. Observation of laser-induced fluorescent cooling of a solid. Nature 377, 500-503 (1995).

[vi] Zhang, J., Li, D., Chen, R. & Xiong, Q. Laser cooling of a semiconductor by 40 kelvin. Nature 493, 504-508 (2013).

[vii] Ha, S.-T., Shen, C., Zhang, J. & Xiong, Q. Laser cooling of organic–inorganic lead halide perovskites. Nat. Photonics 10, 115-121 (2016).

[viii] Roman, B. J., Villegas, N. M., Lytle, K. & Sheldon, M. Optically cooling cesium lead tribromide nanocrystals. Nano Lett. 20, 8874-8879 (2020).

[ix] Morozov, Y. V. et al. Can lasers really refrigerate CdS nanobelts? Nature 570, E60-E61 (2019).

[x] Zhang, Z., Ding, Y., Pauzauskie, P.J. et al. Principles for demonstrating condensed phase optical refrigeration. Nat. Rev. Phys. (2025).