以理論計算與模擬探討石墨烯的儲鋰物理機制以及其吸附之鋰離子的擴散與成核動力學行為

Computational Study of the Origins of the Enhanced Li Storage Capacity and the Kinetic Behaviors of Li Ions on Reduced Graphene Oxide

林昆翰(Kun-Han Lin)

指導教授 ： 郭錦龍

國立臺灣大學/工學院/材料科學與工程學研究所/碩士(2015年)

https://doi.org/10.6342/NTU.2015.00978

摘要

本研究運用第一原理計算來探討鋰離子電池中石墨烯負極材料的儲鋰物理機制，以及研究其與石墨相比下增加的儲鋰量成因。我們分別研究了不同濃度的官能基─包含修飾邊緣的官能基和存在於石墨烯平面上的官能基─以及不同鋰化程度下的石墨烯的儲鋰行為。此外，我們還進一步研究分析鋰離子在此石墨烯電極上的動力學行為模式。我們將這篇論文分成兩個部分：熱力學層面和動力學層面。
在第一部分的研究中，我們主要探討官能基在熱力學層面下對於儲鋰物理機制的影響。我們首先探討邊緣修飾之官能基的影響。我們的結果顯示，石墨烯以及以氫和羥基修飾邊緣的石墨烯奈米緞帶無法儲鋰。另一方面，羰基和羰基-環醚對的存在卻可以非常有效地增加鋰在石墨烯奈米緞帶上的吸附。此外，我們的研究發現這兩個官能基所修飾的石墨烯奈米緞帶在整個鋰化反應過程中最穩定的吸鋰位置並非是位於碳原子所構成的六方形中心，而是在邊緣修飾的官能基上。對於增加儲鋰量的能力，我們發現羰基要比羰基-環醚對來的好，顯示環醚的存在相對來說並無法有效增加儲鋰量。至於在石墨烯平面上的官能基─環氧基和羥基─它們在鋰化過程中扮演成核中心的角色。當鋰靠近這些官能基時，能夠形成氧化鋰和氫氧化鋰的小分子團簇，因而增加了石墨烯的儲鋰量。我們的結果發現，這些位於平面上的官能基對於增加儲鋰量的能力要比在邊緣修飾的官能基還要好。
在第二部分的研究中，我們探討官能基在動力學層面下對於鋰的擴散以及成核成長的物理機制造成的影響。我們的結果顯示，邊緣修飾的官能機種類和邊緣的形狀都會影響鋰的擴散行為。此外，當邊緣修飾的官能基是羰基和羰基-環醚對的時候，在石墨奈米緞帶邊緣附近的鋰能夠輕易地擴散到邊緣，最終吸附在官能基上。至於在石墨烯平面上的官能基對於鋰的動力學行為的影響是比較複雜的。當鋰遠離這些官能基時，鋰的擴散行為與在沒有官能基存在的石墨烯上差不多。然而，當鋰靠近環氧基/羥基時，它們會成核形成鋰-羰基對/氫氧化鋰分子。而每當有鋰靠近
這個已形成的核時，它將不斷擴大直到長成氧化四鋰/氫氧化三鋰分子團簇。從分子動態模擬的結果發現，這些分子團簇能夠持續吸附在石墨烯上，並沒有脫附而造成儲鋰量的流失。除此之外，我們更發現它們在石墨烯上的擴散能障比單顆鋰原子要來得小許多。由於它們在石墨烯上的擴散較為容易，因此有相當大的機率它們會互相碰撞。當碰撞發生時，我們的結果顯示它們會進一步形成更大的團簇分子，進而影響石墨烯鋰化過程的動力學行為。此外，我們發現當羥基存在於石墨烯表面時，其所催化形成的大分子團簇將有機會脫離石墨烯表面而造成儲鋰量的損失。從材料設計的觀點來看，在製造石墨烯負極材料時時應該盡量減少羥基的含量以避免儲鋰量的不可逆損失。

關鍵字

鋰離子電池 ； 鋰電池 ； 石墨烯 ； 負極材料 ； 第一原理計算

並列摘要

We employed first-principles density functional theory calculations to investigate the lithiation mechanism of the reduced graphene oxides (RGO) and explore the origins of their enhanced storage capacity for the anode of Li-ion batteries. Here we have investigated the Li storage and kinetic behaviors of various types of functional groups located at the edge as well as those lying on the basal plane within different levels of lithiation and functionalization of RGO.
In the first part of the thesis, we investigated the effect of functional groups on the lithaition behaviors of graphene narrow ribbons (GNRs) in terms of the thermodynamic viewpoints. Our result shows that lithiation is almost unlikely to happen in pristine graphene and GNRs terminated with OH and H groups. For functional groups terminating the edge, only ketone and K-E pair can effectively enhance Li adsorption on GNRs, and the most favorable sites for Li adsorption turn out to be these edged-oxidized groups rather than the hollow sites on the basal plane. Furthermore, as the ketone-terminated GNRs were fully lithiated, the Li/O atomic ratio was found to be mostly around 1.0~1.25 depending on the concentration of ketone and types of edge, while that for the K-E pair was found to be 0.5 irrespective of the K-E pair concentration and the orientations of graphene edge. This indicates that these edge-oxidized functional groups can indeed enhance Li storage capacity of GNRs. As for the functional groups located on the basal plane, they can serve as the nucleation centers for Li clustering, thereby enhancing the Li storage capacity of GNRs. The Li/O atomic ratio was found to be 4 for the epoxy and 3 (2 for armchair GNR and graphene) for the hydroxyl group. These results indicate that the epoxy and hydroxyl groups on the basal plane tend to be more effective in enhancing the Li storage capacity than the edge functional groups.
In the second part of the thesis, we studied the effect of functional groups on the kinetic and dynamic behaviors of the lithaition process in RGO. Our results show that as Li atom is not located in the vicinity of the edge functional groups, the migration energy barriers of Li on the basal plane are mostly comparable to that on pristine graphene. As a Li atom was located near the edge ketone /K-E pair, it was found to diffuse easily toward the edge sites and then adsorb onto the edge functional groups without any sizable energy barrier. As for the functional groups on the basal plane, their effect on the kinetics and dynamics of the lithiation process is much more complicated. The migration energy barrier for Li away from the epoxy/hydroxyl group is nearly identical to that on pristine graphene. However, as Li is in the vicinity of the epoxy/hydroxyl group, Li can diffuse readily towards the functional group and form Li-O pair/Li(OH) cluster on the basal plane without any sizeable energy barrier. Furthermore, these small clusters can grow bigger through the interaction with other Li atoms at the cost of a very low energy barrier. Very interestingly, theses Li clusters can undergo diffusion with a much smaller energy barrier than that for Li diffusion on pristine graphene. Accordingly, these LinO/Lin(OH) clusters are very likely to diffuse towards the neighboring sites of other nanoclusters and then coalesce into an even bigger one as evidenced in our ab initio MD simulations. On the other hand, our MD simulations also show that some of these nanoclusters containing OH groups may tend to desorb from the RGO surface, which can thus lead to the irreversible loss of the Li storage capacity for RGO.

並列關鍵字

Li-ion battery ； anode materials ； reduced graphene oxide ； first-principles calculations ； kinetics behavior

參考文獻

3. Dahn, J. R.; Zheng, T.; Liu, Y.; Xue, J. S. Science 1995, 270, (5236), 590-593.

23. Uthaisar, C.; Barone, V. Nano Lett. 2010, 10, (8), 2838-2842.

24. Hohenberg, P. Phys. Rev. 1964, 136, (3B), B864-B871.

27. Thomas, L. H. Mathematical Proceedings of the Cambridge Philosophical Society 1927, 23, (05), 542.

28. Dirac, P. A. M. Mathematical Proceedings of the Cambridge Philosophical Society 1930, 26, (03), 376.