Optimizing the Fabrication of All-Organic Solar Cells: A Kinetic Monte Carlo Study of Fullerene Adsorption in Phthalocyanine-Based Covalent Organic Frameworks

Optimizing the Fabrication of All-Organic Solar Cells: A Kinetic Monte Carlo Study of Fullerene Adsorption in Phthalocyanine-based Covalent Organic Frameworks

Brian Koo, Paulette Clancy

Chemical and Biomolecular Engineering, Cornell University

The class of materials known as covalent organic frameworks (COFs)¹ has gained traction recently in the field of photovoltaics. These two-dimensional polymers, originally studied extensively for small gas storage, have high surface area to volume ratios that rival the capacities of metal-organic frameworks and zeolites. Since the incorporation of chromophores into the framework, specifically phthalocyanine groups,² optoelectronic applications of COFs became viable. The advancement of synthesizing 2D COFs as thin films on graphene-covered substrates³ enabled the processing of COFs as a hole-conductor for organic photovoltaic cells. Indeed, recently, a solar cell was made from a combination of PCBM (electron carrier) and a thienothiophene-based COF (hole carrier) but displayed a disappointingly low solar efficiency.⁴  Our study explores how to improve this efficiency by studying the filling of thin film phthalocyanine COFs with fullerene molecules (C60) to create prototypical ordered heterojunction solar cells. COFs are a desirable material since they allow, in principle, the ability to transport free charge carriers perpendicular to its stacked layers, and its inherent ability to direct the structural characteristics of semiconducting guest molecules within its pores, such as fullerene.

Many complementary computational studies have probed the nature of stacking of 2D COF layers, given the importance of angstrom-sized offsets on the hole/electron mobility. Adjacent layers adopt a 1.0-1.7 angstrom offset⁵ in weakly interacting layers, leading to a stack of 2D layers. This, in turn, induces a large potential energy barrier on the pore surface, which limits molecular dynamics (MD) studies of the adsorption of guest molecules within the pores, since transport within micropores occurs purely by surface diffusion. In a previous study, we predicted that layers are offset with no preferred stacking direction; therefore layers are equally likely to adopt stacking patterns that only have local 3-layer persistence, e.g., helical, zigzag, and staircase stacking, each with locally defined lattice sites. Here, we calculate the rates of diffusion between these lattice sites from the free energy barriers extracted from steered molecular dynamics simulations, and find the location and stability of lattice sites using a minimization procedure. From these data, we use Kinetic Monte Carlo to access far longer timescales of adsorption and diffusion than are possible with MD to find the equilibrated packing structure of fullerene as prototypical n-type guest molecules. We also explore the role of solvent and the effect of pore size on the lattice mismatch between COF and C₆₀. Finally, we predict a theoretical maximum packing density and compare it to experiment⁴ to estimate the amount of improvement possible with this arrangement of solar cell component materials.

1.       Côté, A. P. et al. Porous, crystalline, covalent organic frameworks. Science 310, 1166–70 (2005).

2.       Spitler, E. L. & Dichtel, W. R. Lewis acid-catalysed formation of two-dimensional phthalocyanine covalent organic frameworks. Nat. Chem. 2, 672–677 (2010).

3.       Colson, J. W. et al. Oriented 2D Covalent Organic Framework Thin Films on Single-Layer Graphene. Science 332, 228–231 (2011).

4.       Dogru, M. et al. A Photoconductive Thienothiophene-Based Covalent Organic Framework Showing Charge Transfer Towards Included Fullerene. Angew. Chem. 125, 2992–2996 (2013).

5.       Koo, B. T., Dichtel, W. R. & Clancy, P. A classification scheme for the stacking of two-dimensional boronate ester-linked covalent organic frameworks. J. Mater. Chem. 22, 17460-17469 (2012).