Department of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, MD, USA.
Department of Mechanical Engineering, Johns Hopkins University, Baltimore, MD, USA.
Nat Commun. 2024 Sep 5;15(1):7773. doi: 10.1038/s41467-024-51198-9.
The ability to transform matter between numerous physical states or shapes without wires or external devices is a major challenge for robotics and materials design. Organisms can transform their shapes using biomolecules carrying specific information and localize at sites where transitions occur. Here, we introduce gel automata, which likewise can transform between a large number of prescribed shapes in response to a combinatorial library of biomolecular instructions. Gel automata are centimeter-scale materials consisting of multiple micro-segments. A library of DNA activator sequences can each reversibly grow or shrink different micro-segments by polymerizing or depolymerizing within them. We develop DNA activator designs that maximize the extent of growth and shrinking, and a photolithography process for precisely fabricating gel automata with elaborate segmentation patterns. Guided by simulations of shape change and neural networks that evaluate gel automata designs, we create gel automata that reversibly transform between multiple, wholly distinct shapes: four different letters and every even or every odd numeral. The sequential and repeated metamorphosis of gel automata demonstrates how soft materials and robots can be digitally programmed and reprogrammed with information-bearing chemical signals.
将物质在多种物理状态或形状之间转换而无需电线或外部设备，这是机器人技术和材料设计的主要挑战。生物体可以使用携带特定信息的生物分子来改变形状，并在发生转变的部位定位。在这里，我们介绍了凝胶自动机，它同样可以响应生物分子指令的组合文库在大量规定的形状之间进行转换。凝胶自动机是由多个微段组成的厘米级材料。一个 DNA 激活序列库可以通过在其中聚合或解聚来可逆地生长或收缩不同的微段。我们开发了最大化生长和收缩程度的 DNA 激活设计，以及一种用于精确制造具有精细分段模式的凝胶自动机的光刻工艺。通过形状变化的模拟和评估凝胶自动机设计的神经网络的指导，我们创建了可以在多个完全不同的形状之间可逆转换的凝胶自动机：四个不同的字母和每个偶数或奇数数字。凝胶自动机的顺序和重复的变形演示了如何使用带有信息的化学信号对软材料和机器人进行数字编程和重新编程。
