Exciton emission in PTCDA thin films under uniaxial pressure

We study the strain dependent photoluminescence (PL) of a $90\text{−}\mathrm{nm}$-thick polycrystalline 3,4,9,10-perylene tetracarboxylic dianhydride film on Si(001) between 20 and $300\mathrm{K}$. Uniaxial pressure up to $\sim 5\mathrm{kbar}$ is applied along the molecular stacking direction using a specially designed pressure cell. With increasing pressure, we find a quenching of the integrated PL intensity, which is mainly attributed to the creation of defects. At low temperature, the charge transfer exciton emission (CT2) gains intensity relative to the Frenkel exciton emission. Furthermore, the CT2 transition reveals a shift to lower energies by up to $24\mathrm{meV}$. At room temperature, the PL is dominated by the excimer transition, which also shows a redshift of $\sim 24\mathrm{meV}$ at the highest uniaxial pressure. The relative increase of the CT2 transition at low temperature and the redshift of the emission bands are attributed to an increased exciton trapping probability and an enhanced binding energy with reduced distance between stacked molecules. This explanation is supported by a comparison with total energy calculations for pairs of adjacent molecules. Moreover, the calculated pressure-induced energy shifts of the CT2 and excimer transitions are in good agreement with the experimentally observed values.

Figure 1

PL spectra at ambient pressure of a $90\text{−}\mathrm{nm}$-thick PTCDA film on Si(001) excited at $3.05\mathrm{eV}$ at $20\mathrm{K}$. The experimental spectrum (open circles) and different emission channels according to model calculations (lines as labeled) are shown. The resulting calculated spectrum is given as a thick solid line.

Figure 2

Uniaxial pressure-dependent PL spectra of a $90\text{−}\mathrm{nm}$-thick PTCDA film on Si(001) excited at $3.05\mathrm{eV}$ and obtained at $20\mathrm{K}$. The applied uniaxial pressure increases from $P_0$ to $P_4$. The gray solid line $\left(\mathit{Re}\right)$ shows the PL spectrum after the strain was released.

Figure 3

Uniaxial pressure-dependent PL spectra of a $90\text{−}\mathrm{nm}$-thick PTCDA film on Si(001) excited at $3.05\mathrm{eV}$ and obtained at $40\mathrm{K}$. The applied uniaxial pressure increases from $P_0$ to $P_4$. The gray solid line $\left(\mathit{Re}\right)$ shows the PL spectrum after the strain was released.

Figure 4

Uniaxial pressure-dependent PL spectra of a $90\text{−}\mathrm{nm}$-thick PTCDA film on Si(001) excited at $3.05\mathrm{eV}$ and obtained at $80\mathrm{K}$. The applied uniaxial pressure increases from $P_0$ to $P_4$. The gray solid line $\left(\mathit{Re}\right)$ shows the PL spectrum after the strain was released.

Figure 5

PL spectra of a $90\text{−}\mathrm{nm}$-thick PTCDA film on Si(001) excited at $3.05\mathrm{eV}$ at 20, 40, and $60\mathrm{K}$ at an applied uniaxial pressure $P_3$. The experimental spectrum (open circles) and different emission channels according to model calculations (lines as labeled) are shown. The resulting calculated spectrum is given as a thick solid line.

Figure 6

Energies of the CT2 transition (zeroth vibronic subband) at 20, 40, and $80\mathrm{K}$ as a function of the estimated applied uniaxial pressure. The CT2 energies were obtained from model calculations fitted to the data as described in the text. Also shown are observed excimer transition energies at $300\mathrm{K}$ as a function of the estimated uniaxial pressure.

Figure 7

Uniaxial pressure-dependent PL spectra of a $90\text{−}\mathrm{nm}$-thick PTCDA film on Si(001) excited at $3.05\mathrm{eV}$ and obtained at $300\mathrm{K}$. The applied uniaxial pressure increases from $P_0$ to $P_3$. The gray solid line $\left(\mathit{Re}\right)$ shows the PL spectrum after the strain was released.

Figure 8

Photos under identical optical conditions showing the $\mathit{a}$-axis length change of an $\alpha$-PTCDA crystal between 1 and $55\mathrm{kbar}$ hydrostatic pressure in a diamond-anvil cell at $300\mathrm{K}$.

Figure 9

Length scaling of the PTCDA crystal parallel to the $\mathit{a}$ axis as a function of pressure. The solid line shows the best fit using Eq. (6) and the dashed line gives the linear pressure dependence of $a\left(p\right)∕a_0$ according to Eq. (4).