Direct Measurement of the Triplet Exciton Diffusion Length in Organic Semiconductors

We present a new method to measure the triplet exciton diffusion length in organic semiconductors. $N$,$N^{\prime }$-di-[(1-naphthyl)-$N$,$N^{\prime }$-diphenyl]-1,$1^{\prime }$-biphenyl)-4,$4^{\prime }$-diamine (NPD) has been used as a model system. Triplet excitons are injected into a thin film of NPD by a phosphorescent thin film, which is optically excited and forms a sharp interface with the NPD layer. The penetration profile of the triplet excitons density is recorded by measuring the emission intensity of another phosphorescent material (detector), which is doped into the NPD film at variable distances from the injecting interface. From the obtained triplet penetration profile we extracted a triplet exciton diffusion length of $87\pm 2.7\mathrm{nm}$. For excitation power densities $>1\mathrm{mW}/{\mathrm{mm}}^2$ triplet-triplet annihilation processes can significantly limit the triplet penetration depth into organic semiconductor. The proposed sample structure can be further used to study excitonic spin degree of freedom.

Figure 1

Sample structure for triplet exciton diffusion length measurements (a); energy levels of triplet excited states (b); and (c) chemical structures of iridium(III)bis(2-(4 6-difluorephenyl)pyridinato-$N$,$C^2$) (FIrPic), $N$,$N^{\prime }$-di-[(1-naphthyl)-$N$,$N^{\prime }$-diphenyl]-1,$1^{\prime }$-biphenyl)-4,$4^{\prime }$-diamine (NPD), and meso-tetratolylporphyrin-Pd (PdTPP).

Figure 2

(a) Solid lines are photoluminescence spectra of Reference and FIrPic sides of a typical sample used for triplet exciton diffusion measurements. The arithmetical difference between these spectra is also presented. (b) Typical dependence of the phosphorescence emission of PdTPP (detected at 700 nm) depending on the sample spatial position. The average intensities of $I_0$ and $I_1$ correspond to the Reference and FIrPic sides of the sample, respectively.

Figure 3

Normalized variation of the phosphorescence intensity at 700 nm ($\Delta i$) vs NPD layer thickness $L$. Samples were excited at 466 nm; various excitation power densities are presented. The dashed lines serve as guides to the eye; the solid line is a fit with Eq. (4).

Figure 4

The dependence of the emission intensity on the incident power density in a NPD-PdTPP-NPD heterostructure at 700 nm (circles) and at 550 nm (triangles); in PdTPP-polystyrene blend at 700 nm (squares). Laser excitation was at 466 nm.