It was 3:15 a.m. on a cold night in Urbana, Illinois about 25 years ago. I had been on my feet since the previous morning, preparing the experiment. Professor Enrico Gratton had instructed me to machine a brass piece to replace a window in the cryostat. I finished the piece on a lathe, drilled a hole in the center, soldered a 1/4-in. copper tube and sealed the end outside the cryostat with a copper cap. I bent the open end so that it was a couple of millimeters from a calcium fluoride window, taking care not to obstruct the infrared beam from the interferometer. I loaded some acetanilide powder in the copper tube—it would serve as a crude but effective oven. The first nervous test passed when my clumsy solder job actually held vacuum. The whole procedure had taken a long time because of my inexperience. When the cryostat was finally cold, I took a heat gun and gently warmed the tube on the outside. A few micrograms of the material sublimated and got deposited into an amorphous film on the cold window. Prof Gratton was looking at the emerging infrared spectrum as the film developed. There was the amide I band, near 1,667 cm − 1, familiar from many past experiments on crystals. But the soliton band at 1,650 cm − 1was missing. That was it! We had confirmed Al Scott’s interpretation of the anomalous infrared band in acetanilide. But even as we celebrated, I knew in my heart that we had just had doomed prospects for Scott–Davydov solitons in proteins. As a biological physicist in training, I would never really work on this wonderful problem again. It was then, and still remains, beautiful physics. Years later, when my colleague Bob Austin attempted to lure me back to do some more experiments, I resisted. Along the way, I had the privilege of meeting brilliant physicists like Al Scott and mavericks like Giorgio Careri. My admiration remains for superb theorists like Denise Alexander, who deserves to be remembered for her memorable paper in a tragically short career. Peter Hamm’s beautiful experiments provided not only confirmation but also greatly expanded on the original idea. For all that, the lesson learned on a cold Midwestern night was permanently etched in my heart: beautiful physics is sometimes irrelevant to biology. My advisor, Prof Hans Frauenfelder, had taught me that. My interaction with Al Scott and his ideas on the Davydov soliton started with Enrico Gratton, who was a superbly gifted mentor to me. I was a beginning graduate student, struggling with all aspects of physics. Prof Gratton came to my desk and showed me some papers with complicated mathematics—papers from Al Scott and Davydov among others. You are Indian, you must be good at mathematics. Clearly, Professor Gratton had forgotten the oral examination in which he asked me to derive Ohm’s law from first principles. I had struggled mightily, butchering the expression for the drift velocity. And that was a linear problem! He had nevertheless passed me in an act of grace and kindness. But he needed an assistant in infrared spectroscopy. I had just learned how to take an infrared spectrum without breaking too many expensive CaF2 windows. So, Ohm’s law or not, I got the job. A clumsy experimentalist with indifferent mathematical skills—that was my rather unpromising entry into the heady world of self-trapped states. I began to learn about the great problem in energy transfer that Davydov had set out to solve. I read Scott’s infrared-active extension of the audacious proposal for a particular type of a soliton. And Enrico taught me about Giorgio Careri’s ideas on acetanilide as a model system. We started a systematic series of experiments to test whether Careri’s work could be married to Scott’s innovation. First, we repeated earlier cryogenic studies on acetanilide in pressed KBr pellets. The temperature data could be reasonably well fit with Scottt’s model. But the amide peaks were evidently being distorted in the KBr pellets. So I learned from Professor Dana Dlott how to zone refine the material, how to grow crystals large enough to study high harmonics, and how to deuterate the samples. I learned how to grow extremely large crystals for high harmonic studies and how to grow extremely thin single crystals for polarization studies. One particular experiment stands out in my mind. I had mounted a thin ~1 μ m ACN crystal on a cryostat using an Al foil as holder. As I cooled the sample below 77 K, the all important soliton band grew as predicted by Scott. But when I cooled to 10 K, all of a sudden, the peak disappeared and I got a 300 K spectrum. It took a few anxious moments to realize what had happened: the crystal had separated slightly from the Al foil, which was in fact at 10 K. The infrared beam from spectrometer had then heated the thermally insulated thin crystal all the way to near room temperature. More careful mounting of the crystal solved this problem. Single crystal studies allowed us to make polarization measurements, providing further support to Scott’s interpretation. Eventually, the studies led up to one final prediction of the Scott model: the soliton band should disappear in the amorphous phase. That led to the memorable later night experiment on amorphous ACN. We published a number of papers on our results. I am proud to say that our experimental results have held up robustly. Over the years, the model has been criticized on theoretical grounds and alternative interpretations proposed. Scientists far more qualified than I will talk about this work elsewhere in this issue. For me, the most appealing interpretation came from a brilliant young theory graduate student at Cornell, Denise Alexander. She had been working with Jim Krumhansl’s group and had learnt from that great master to be as comfortable with nonlinear phenomena as I had been intimidated by them. She proposed a remarkable model that was inspired by the Holstein small polaron, but with the role of the electrons played by the optical phonon coupled to a bath of acoustic phonons. It fit our data well. Denise was sufficiently enthused by her success that she decided to switch to biological physics. Tragically, she was killed in a car accident in New York City where she had moved for post-doctoral research—a huge loss to the field. Someone at the time remarked that singularly bad luck seemed to pursue all the great theorists working in the field. I think she would have been very happy with Peter Hamm’s ultrafast studies, and her insights would have been invaluable. How can I best explain my mixed emotions on that Illinois night? Our experiments showed me that solitons as proposed were simply too delicate to survive in biology. Prof Frauenfelder taught us that proteins in fact existed in a great number of conformational substates. Even proteins with apparently well-defined crystal structures were in fact dynamic entities, with fluctuations that revealed themselves in the Debye–Waller factor and in non-exponential binding kinetics. A universal scheme for transferring energy in biological processes needs to be robust because of the messy and arbitrary nature of fluctuations and mutations. A delicate phenomenon that needed a precise and subtle crystal structure simply could not survive the rigors of evolution. That was what our experiment on a cold night in Illinois showed. I believe that the concept of self-trapped states in proteins is strong and has plenty of experimental and theoretical support. But the self-trapped states are not the solitons as envisioned in acetanilide. Bob Austin’s article in this volume makes the case rather convincing. Over the years, I have admired Al Scott and his brilliant insights. He was remarkably generous with his time and continued to maintain his amazingly positive outlook even after his debilitating accident. He will remain a bright light in my memory, and a model scientist. I am grateful for the opportunity of having met him and of doing experiments inspired by him.

中文翻译：

---

孤子，从下面

大约 25 年前，伊利诺伊州厄巴纳一个寒冷的夜晚，凌晨 3 点 15 分。从前一天早上起，我就站起来准备实验。Enrico Gratton 教授曾指示我加工一块黄铜片来替换低温恒温器中的一个窗口。我在车床上完成了这件作品，在中心钻了一个洞，焊接了一个 1/4 英寸。铜管，并用铜帽密封低温恒温器外的末端。我将开口端弯曲，使其距离氟化钙窗口几毫米，注意不要阻挡来自干涉仪的红外光束。我在铜管中装入了一些乙酰苯胺粉末——它可以作为一个粗糙但有效的烤箱。当我笨拙的焊接工作实际上保持真空时，第一次紧张的测试就通过了。由于我的经验不足，整个过程花费了很长时间。当低温恒温器终于冷却时，我拿了一把热风枪，轻轻地加热了外面的管子。几微克的材料升华并在冷窗上沉积成无定形薄膜。随着电影的发展，格拉顿教授正在研究新兴的红外光谱。在 1,667 cm - 1 附近有酰胺 I 带，这在过去的许多晶体实验中很常见。但是 1,650 cm − 1 处的孤子带丢失了。就是这样！我们已经确认了 Al Scott 对乙酰苯胺异常红外波段的解释。但即使在我们庆祝的时候，我心里也知道我们刚刚注定了蛋白质中 Scott-Davydov 孤子的前景。作为一名正在接受培训的生物物理学家，我再也不会真正研究这个奇妙的问题了。那时，而且仍然是美丽的物理学。多年后，当我的同事 Bob Austin 试图引诱我回去做更多实验时，我拒绝了。一路上，我有幸结识了像 Al Scott 这样杰出的物理学家和像 Giorgio Careri 这样的特立独行者。我仍然钦佩像丹尼斯·亚历山大这样出色的理论家，她在悲惨的短暂职业生涯中发表了令人难忘的论文，值得人们记住。彼得·哈姆 (Peter Hamm) 漂亮的实验不仅证实了这一点，而且大大扩展了最初的想法。尽管如此，在寒冷的中西部夜晚学到的教训永远铭刻在我的心中：美丽的物理学有时与生物学无关。我的导师 Hans Frauenfelder 教授教会了我这一点。我与 Al Scott 的互动以及他对 Davydov 孤子的想法始于 Enrico Gratton，他是我非常有天赋的导师。我是刚毕业的研究生，与物理学的各个方面作斗争。Gratton 教授来到我的办公桌前，向我展示了一些数学复杂的论文——Al Scott 和 Davydov 等人的论文。你是印度人，你一定擅长数学。显然，格拉顿教授忘记了他要求我从第一原理推导出欧姆定律的口试。我拼命地挣扎着，为漂移速度抹杀了表情。这是一个线性问题！尽管如此，他还是以一种优雅和善意的方式超越了我。但他需要一个红外光谱学助理。我刚刚学会了如何在不破坏太多昂贵的 CaF2 窗口的情况下获取红外光谱。所以，不管是不是欧姆定律，我得到了这份工作。一个笨手笨脚的实验者，数学技能冷漠——这是我进入自我陷害状态的令人兴奋的世界相当没有希望的入口。我开始了解达维多夫着手解决的能量转移中的重大问题。我阅读了 Scott 对特定类型孤子的大胆提议的红外主动扩展。Enrico 教会了我 Giorgio Careri 关于乙酰苯胺作为模型系统的想法。我们开始了一系列系统的实验来测试 Careri 的工作是否可以与 Scott 的创新相结合。首先，我们重复了早期对压制 KBr 颗粒中乙酰苯胺的低温研究。温度数据与 Scottt 的模型非常吻合。但是，KBr 颗粒中的酰胺峰明显扭曲。所以我从 Dana Dlott 教授那里学到了如何对材料进行区域精炼，如何生长足够大的晶体来研究高次谐波，以及如何对样品进行氘化。我学会了如何为高次谐波研究生长极大的晶体，以及如何为极化研究生长极薄的单晶。一个特别的实验在我的脑海中脱颖而出。我使用铝箔作为支架在低温恒温器上安装了一个约 1 μm 的 ACN 薄晶体。当我将样品冷却到 77 K 以下时，所有重要的孤子带都如 Scott 预测的那样增长。但是当我冷却到 10 K 时，突然间，峰值消失了，我得到了 300 K 的光谱。过了一会儿才意识到发生了什么：晶体与铝箔略微分离，实际上是 10 K。然后光谱仪的红外光束将绝热的薄晶体一直加热到接近室温. 更仔细地安装晶体解决了这个问题。单晶研究使我们能够进行偏振测量，进一步支持 Scott 的解释。最终，这些研究得出了斯科特模型的一个最终预测：孤子带应该在非晶相中消失。这导致了令人难忘的晚间关于无定形 ACN 的实验。我们发表了许多关于我们的结果的论文。我很自豪地说，我们的实验结果非常可靠。多年来，该模型因理论依据和提出的替代解释而受到批评。科学家们比我将在本期其他地方谈论这项工作更有资格。对我来说，最吸引人的解释来自康奈尔大学一位才华横溢的年轻理论研究生丹尼斯·亚历山大。她一直在与 Jim Krumhansl 的小组一起工作，从那位大师那里学到了如何适应非线性现象，就像我被他们吓倒一样。她提出了一个非凡的模型，该模型的灵感来自荷斯坦小极化子，但光学声子所扮演的电子的角色与声学声子浴耦合。它非常适合我们的数据。丹尼斯对她的成功充满热情，她决定转向生物物理学。可悲的是，她在纽约市的一场车祸中丧生，她搬到那里进行博士后研究——这是该领域的巨大损失。当时有人评论说，所有在该领域工作的伟大理论家似乎都遭遇了奇特的厄运。我认为她会对 Peter Hamm 的超快研究感到非常满意，而且她的见解将是无价的。在伊利诺伊州的那个夜晚，我该如何最好地解释我的复杂情绪？我们的实验表明，所提出的孤子太脆弱了，无法在生物学中生存。Frauenfelder 教授告诉我们，蛋白质实际上存在于大量的构象亚状态中。即使具有明显明确的晶体结构的蛋白质实际上也是动态实体，其波动在 Debye-Waller 因子和非指数结合动力学中显现出来。由于波动和突变的混乱和任意性质，在生物过程中转移能量的通用方案需要稳健。一种需要精确而微妙的晶体结构的微妙现象根本无法经受住进化的严酷考验。这就是我们在伊利诺伊州一个寒冷的夜晚进行的实验所表明的。我相信蛋白质中自陷状态的概念很强，并且有大量的实验和理论支持。但是自陷状态并不是乙酰苯胺中设想的孤子。鲍勃·奥斯汀 (Bob Austin) 在本卷中的文章使案例颇具说服力。多年来，我一直钦佩 Al Scott 和他出色的洞察力。他对自己的时间非常慷慨，即使在他虚弱的事故之后仍然保持着惊人的积极态度。他将是我记忆中的一盏明灯，一位模范科学家。我很感激有机会见到他并在他的启发下做实验。鲍勃·奥斯汀 (Bob Austin) 在本卷中的文章使案例颇具说服力。多年来，我一直钦佩 Al Scott 和他出色的洞察力。他对自己的时间非常慷慨，即使在他虚弱的事故之后仍然保持着惊人的积极态度。他将永远留在我的记忆中，成为一名模范科学家。我很感激有机会见到他并在他的启发下做实验。鲍勃·奥斯汀 (Bob Austin) 在本卷中的文章使案例颇具说服力。多年来，我一直钦佩 Al Scott 和他出色的洞察力。他对自己的时间非常慷慨，即使在他虚弱的事故之后仍然保持着惊人的积极态度。他将永远留在我的记忆中，成为一名模范科学家。我很感激有机会见到他并在他的启发下做实验。