氮掺杂石墨烯的结构调控—路在何方？

来源：知社学术圈

石墨烯具有优异的电学与光学性能、极高的电荷载流子迁移率、室温量子霍尔效应等，因而被认为在纳米电子学领域有着非常广泛的应用前景。然而，本征石墨烯零带隙的特点以及过低的载流子浓度大大限制了其在数字电路中的应用。

对石墨烯进行氮掺杂有望弥补本征石墨烯零带隙和载流子密度低的缺憾，实现其在电子器件领域的广泛应用。然而，目前氮掺杂石墨烯(NG)的合成主要面临两个问题：

氮掺杂石墨烯中的氮浓度普遍很低，除利用特殊的芳香烃分子合成的氮原子周期性分布的碳氮二维材料以外，目前报道的氮掺杂石墨烯中氮原子的掺杂浓度不高于20%，这使得NG中载流子浓度的调控受到很大限制。

NG中的掺杂氮原子通常以吡啶氮、吡啶氮氧化物、吡咯氮、石墨氮等多种形式共存，并且掺杂氮原子在石墨烯的面内排列无序，这也使得载流子在输运过程中遭遇更强的散射，导致载流子迁移率大大降低。

现阶段，如何可控地制备高浓度、高有序的NG是一个极富挑战性的难题。

理解氮原子掺杂浓度低、掺杂类型和位置不可控的本质是实现高浓度、高有序的NG制备的前提。来自华东师范大学精密光谱科学与技术国家重点实验室的博士生补赛玉（昆士兰大学交流学生）及其导师袁清红研究员与澳大利亚昆士兰大学澳大利亚生物工程及纳米科技研究所的Debra J. Searles教授等人，采用第一性原理计算与粒子群优化算法相结合的方法，分别对吡啶氮和石墨氮掺杂石墨烯的最稳定结构进行了预测。发现氮掺杂石墨烯的稳定结构与其中的氮原子浓度密切相关，低氮掺杂浓度下，NG中的石墨氮和吡啶氮具有相近的形成能，因而更易形成石墨氮和吡啶氮共掺杂的结构。随着氮原子掺杂浓度的增加，石墨氮掺杂石墨烯的形成能要高于吡啶氮掺杂石墨烯的形成能，因而更易形成吡啶氮掺杂的石墨烯结构。

特别是，当氮原子掺杂浓度高于0.25时，NG中以吡啶氮掺杂为主。此外，该研究还表明，低氮掺杂浓度的NG具有更低的形成能。这一系列研究结果解释了目前实验上NG中氮原子浓度低，以及多种类型的氮混合掺杂的实验现象。进一步，研究人员结合理论推导计算出含碳或氮的前驱体分子中的碳或氮的化学势随温度、压强的表达式，提出可通过控制NG合成过程中的前驱体类型、温度、压强实现碳和氮原子化学势的调控，从而实现NG中氮原子的类型和氮掺杂浓度的调控。该研究通过对不同氮浓度下的NG结构和能量进行研究，为实验上实现氮掺杂石墨烯的可控合成提供了理论依据。

该文近期发表于npj Computational Materials 6: 128 (2020)，英文标题与摘要如下。

Design of two-dimensional carbon-nitride structures by tuning the nitrogen concentration 

Saiyu Bu, Nan Yao, Michelle A. Hunter, Debra J. Searles & Qinghong Yuan 

Nitrogen-doped graphene (NG) has attracted increasing attention because its properties are significantly different to pristine graphene, making it useful for various applications in physics, chemistry, biology, and materials science. However, the NGs that can currently be fabricated using most experimental methods always have low N concentrations and a mixture of N dopants, which limits the desirable physical and chemical properties. In this work, first principles calculations combined with the local particle-swarm optimization algorithm method were applied to explore possible stable structures of 2D carbon nitrides (C1−xNx) with various C/N ratios. It is predicted that C1−xNx structures with low N-doping concentration contain both graphitic and pyridinic N based on their calculated formation energies, which explains the experimentally observed coexistence of graphitic and pyridinic N in NG. However, pyridinic N is predominant in C1−xNx when the N concentration is above 0.25. In addition, C1−xNx structures with low N-doping concentration were found to have considerably lower formation energies than those with a high N concentration, which means synthesized NGs with low N-doping concentration are favorable. Moreover, we found the restrictions of mixed doping and low N concentration can be circumvented by using different C and N feedstocks, and by growing NG at lower temperatures.