The discovery of a new material is often accidental. In this small article, I would like to introduce the most memorable failures in the author's research career in the development of new materials, and also the success story that were the other side of those failures. Specifically, I will introduce the story about the quantum kagome antiferromagnet Ca-Kapellasite, which was born into the world at the moment of the failure during the flux removal in a crystal growth of the high-T_c cuprate superconductor. Failure and success are indeed two sides of the same coin, and I hope if I could tell even a little of the dynamism of success born from failure.

  When the author was a master’s course student, she experienced a failure in molecular beam epitaxial (MBE) growth of one of the dilute bismide (Bi-) III-V compound semiconductors, GaAs_1-xBi_x. This failure was caused by her cutting corners when she cleaned the surface of the substrate by chemical etching before MBE growth. However, this experience later led to the realization of the first laser oscillation of GaAs_1-xBi_x. The failures leading to the success in MBE growth of GaAs_1-xBi_x are introduced in this manuscript to discuss what the crystals of GaAs_1-xBi_x have taught the author through her experience.

  Polycrystalline Cu(In,Ga)Se₂ (poly-CIGS) solar cells are widely studied for their application in thin-film photovoltaics. Single-crystalline CIGS solar cells can potentially address the issues of poly-CIGS solar cells, and we fabricated epitaxial CIGS (epi-CIGS) layers on GaAs substrates using techniques developed for high-efficiency poly-CIGS solar cells such as Ga grading, Na doping, and heat-light soaking. Scanning transmission electron microscopy was used to compare the structural characteristics of poly-CIGS and epi-CIGS solar cells with conversion efficiencies of more than 21%. Both types of solar cells had similar Ga gradient profiles. These observations suggest that grain boundaries in poly-CIGS layers are not the main factor limiting the conversion efficiency of poly-CIGS solar cells.

  Solution-mediated transformation is crucial in a wide variety of fields such as the formation of materials and minerals. However, no one knows the universal strategy for the growth of crystals formed by solution-mediated transformation. Here, we present the growth of norsethite [BaMg(CO₃)₂] by adding NH₄NO₃, and discuss the key to overcoming the difficulties of the solution-mediated transformation. Norsethite is poorly-soluble, and appears through witherite [BaCO₃]. The solubilities of norsethite and witherite in NH₄NO₃ solutions were ~35 and ~45 times higher than those in the absence of NH₄NO₃, respectively. At 90℃, the growth rate of norsethite in NH₄NO₃ solutions was ~17 times higher than that in NH₄NO₃-free solutions. Finally, we succeeded in synthesizing bulky norsethite of 150 μm size only in one month, utilizing 1M NH₄NO₃ solutions of 90℃. We explained the appearance of large norsethite by adding NH₄NO₃, considering that norsethite dominantly consumes the solute compared with witherite.

  Many inspirations and insights in research are gained in the laboratory. A plan is made before starting an experiment in the laboratory, and a simulation is carried out in the mind. You may encounter unexpected failures or phenomena while working with an image of what will go well. Such unplanned situations are the seeds of new research. In this paper, I would like to introduce four episodes I have experienced with my students.