3-1 Time-Resolved Photoluminescence (TRPL) System
Nowadays, fluorescence spectroscopic investigations are very common, and the temporal evolution of the laser induced fluorescence can provide various important information on dynamics of radiative relaxation. Time-correlated single photon counting (TCSPC) technique has been widely used to study the radiative lifetime of laser induced fluorescence. Although time-resolved fluorescence spectroscopy is a powerful analysis tool in sciences, it is a challenge to record its fast decay time which lasts hundred picoseconds to tens of nanoseconds. In order to recover fluorescence lifetimes as short as 300 ps, one must be able to resolve the exponential decay by tens of sampling steps. This means the transient recorder required to sample at 30 ps time steps. And it is a hard task to achieve with ordinary electronic transient recorder. In addition, the fluorescence intensity can be too weak to sample an analog temporal decay. One of the solutions for these problems is TCSPC technique. Since with periodic excitation, it is possible to extend the data collection over multiple cycle and one can reconstruct the single cycle decay profile from single photon events collected over many cycles. By setting the count rate small enough than the repetitive excitation source from the laser, TCSPC measures the time difference between the single pulses from the trigger that is synchronized with excitation signals in the single photoelectron state as shown in Fig. 3-3. In a single cycle, the timing electronics will record the single photon and the time delay with the trigger pulse. After multiple cycles, we can get the probability distribution of photons related to delay time as shown in Fig 3-1. For TRPL measurement, a picosecond pulsed diode laser which emits light pulses as short as 70 ps at repetition rates from single shot up to 80 MHz with peak powers up to 1 Watt was used.
18
3-1.1 The principle of Time-Correlated Single Photon Counting
The working principle of TCSPC is based on the repetitive recording a fluorescence photon when it was excited by the laser pulse. Fig. 3-1 shows how the histogram is formed over many cycles. We should use the pulsed laser to excite a fluorescence photon and measure the time difference between excitation pulse and emission photon precisely. After many cycles of counts, we can get the histogram by accumulating the photons.
In order to register only one photon or no photon at every excitation, it is necessary to attenuate the excitation power. If there are more than one photon to be registered, the decay time will be shorten and make it inaccurate. In fact, if the single photon probability is met, there will be no photons at all in many cycles. So it is important to meet the single photon probability condition.
Fig. 3-1 TCSPC histogram.
19
3-1.2 Principles behind PicoHarp 300
Our time-resolved PL measurement was performed using the commercial TCSPC system, so called, PicoHarp 300. Figure 3-2 is the building blocks of PicoHarp 300, in which the signal is fed into the constant fraction discriminator (CFD). CFD has some important function in PicoHarp 300: (1) It can extract precise timing information from the electrical detector pulses that may vary in amplitude. (2) The overall system Instrument Response Function (IRF) may be tuned to become narrower for improving the resolution and some of the random background signal can be suppressed. The same could not be achieved with a simple comparator. (3) Pulses originating from random electrons generated at the dynode of the PMT can be suppressed. Therefore, CFD can recognize the laser pulses as well as photon signal pulse and extract precise timing information. Similar to the detector signal, the sync signal must be made available to the timing circuitry. Since the sync pulses are usually of well-defined amplitude and shape, a simple settable level trigger (comparator) is enough to adapt to different sync sources. The signals from CFD and sync trigger are fed to a Time to Amplitude Converter (TAC). The result is a voltage proportional to the time difference between the two signals. The voltage obtained from TAC is then fed to an Analog to Digital Converter (ADC) which provides the digital timing value to histogrammer.
20
Fig. 3-2 Building blocks of PicoHarp 300.
3-1.3 Experimental Setup for TPRL measurement with TCSPC
Figure 3-3 shows the schematic arrangement of TRPL measurement system with TCSPC. The central wavelength of the laser head is 405nm and the repetition rate and pulse width are 2.5 MHz and 70 ps, respectively. The pump laser pulses are reflected by the dichroic mirror and excite the sample through an optical microscope with a 60X objective lens. The luminescence signal of the sample goes through the dichroic mirror and is detected by monochromator. Before the photon signal goes into a photomultiplier tube (PMT), we should tune the grating in the monochromator to the desired wavelength, and then PMT can turn the photon signal into electrical signal.
We can get the time-resolved PL result after the electrical signal is fed to the TCSPC electronics (PicoHarp 300) via a preamplifier. In addition, the driver directly provides the electric sync signal needed for the photon arrival time measurement.
Fig. 3-3 Experiment setup for TPRL measurement with TCSPC.
21
3-1.4 Experimental Setup for micro-PL measurement
The experimental setup for micro-PL measurement is shown in Fig. 3-4. It is similar to the schematic experimental arrangement of TPRL measurement with TCSPC, but here the time-integrated PL is measured by a spectrometer. For micro-PL measurement, either continuous (CW) or pulsed light source can be used as excitation source. Since the organic materials used in this study are highly reactive with air under intense laser excitation, all the measurements were performed under the vacuum environment. The luminescence signal of each sample goes through the dichroic mirror and is fed into a spectrometer to get the PL spectrum.
Fig. 3-4 Experiment setup for PL measurement.
22
3-2 Organic Dyes
3-2.1 DCM
4-(Dicyanomethylene)-2-methyl-6-(4-dimethylaminostyryl)-4H-pyran (DCM), shown in Fig. 3-5 is known as one of the best laser dyes in the red emission field. It is often used as a dopant in the host-guest system. It has a great effectiveness in 600~700nm region and is widely used in laser spectroscopy. Its emission efficiency is in great relation with the doping concentration, and its transition from the ground state to the excited state will have an increase in dipole moment.
Fig. 3-5 Molecule structure of DCM.
3-2.2 Alq3
Tris-(8-hydroxyquinolinato) aluminum (III) (Alq3) shown in Fig. 3-6 is one of the metal chelates. Alq3 has high thermal stability owing to its high symmetry of molecule structure and high glass transition temperature (Tg), besides its morphology is very stable and it can emit light. So it has been widely used as emission and electron transport layer due to these significant properties.
Fig. 3-6 Molecule structure of Alq3.
23
3-2.3 BAlq
Bis (2-methyl-8-quinolinato)-4-phenylphenolate aluminum (III) (BAlq) shown in Fig. 3-7 is usually used as the exciton-blocking layer between the emissive layer in phosphorescent light-emitting devices to prevent the long lifetime excitons in the triplet excited state diffuse to the electron transport layer.
Fig. 3-7 Molecule structure of BAlq.
3-2.4 Znq
Bis (8-hydroxyquinoline) zinc (II) (Znq) is one of the metal chelates shown in Fig. 3-8 and is used as emission and electron transport layer. It is a highly luminescent and electroluminescent material and has shown advantages over Alq3 in electron transport and shown higher quantum yields in device performance. So it has lower operating voltages than Alq3.
Fig. 3-8 Molecule structure of Znq.
24
3-3 Sample preparation
For this study, three different dyes of Znq:DCM (2.5%), BAlq:DCM (2.5%), and Alq3:DCM (2.5%) grown on mica, and Alq3:DCM grown on SOG which covered the gold nanoplates were prepared. Znq, BAlq, Alq3 are the host emitter while DCM is
Table 3-1 Symbol assignment of four samples.
3-3.1 Samples on mica
Sample A, B, and C were grown on mica substrate and all of them had the same structures as shown in Fig. 3-9. As for the organic layers, they were deposited by co-doping technique in a thermal evaporator. The gold and SiO2 layer were deposited by e-beam evaporation. The organic layer is 150nm of thickness. The gold film is 50nm of thickness, and SiO2 buffer layer between the organic dye and SiO2 protect layer over the organic dye are 5nm of thickness.
Fig. 3-9 Structure of the samples on mica.
25
3-3.2 Sample on SOG
The sample with the dye of Alq3:DCM(2.5%) was positioned on the SOG, which is 5nm of thickness and covered the gold nanoplates, as shown in Fig. 3-10(b). The organic layer was deposited by co-evaporation technique. The average length of the gold nanoplates is about 10μm (Fig. 3-10(a)). The organic layer is 170nm of thickness.
The protect layer SiO2 over the organic dye is 20nm of thickness. The three silver bars for orientation are 50nm of thickness.
(a)
(b)
Fig. 3-10 (a)OM image of gold nanoplates. (b) Structure of the sample on SOG.
26