Photograph of the instrument.
Constant delay timing histograms of channel 1 vs. sync input at different repetition rates for assessment of timing jitter. The x-axis offsets of the individual histograms are arbitrarily chosen for clarity of display only. Channel 2 shows visually identical behavior. See Table I for numerical details.
(a) Fluorescence lifetime measurement of Coumarin 6 in ethanol. A theoretical model which takes the instrument response function (IRF) into account is fitted to the data. A lifetime of 2.52 ns is measured. The lower panel (b) depicts the fit residuals.
Experimental setup: Photon pairs are generated inside a PPKTP crystal, which is placed inside a linear cavity. The cavity length is stabilized to a resonance of the pump laser, which is produced by second harmonic generation (SHG). After spectral filtering signal and idler photons are separated at a polarizing beam splitter (PBS) and coupled into single-mode fibers. Signal photons are split by a 50/50 fiber beam splitter (FBS). Three avalanche photo diodes (APD) are used to detect photons. Their arrival times are measured with the TCSPC electronics in time tagging mode T2.
Measured (symbols) and theoretical (lines) normalized second order cross-correlation function for the three different pump powers ( , , and ). For low pump powers strong signal-idler bunching can be observed. If left as a free parameter in a fit of Eq. (1) , the pair creation rate R 0 can be estimated from this data. The values are in good agreement with the creation rates obtained from the detection efficiency and the APD counts rates.
Measured (symbols) normalized second-order auto-correlation function of the signal photons for the three different pump powers ( , , and ). The deviation of from 2 is caused by imperfect signal/idler-separation in measurement (a) and by uncorrelated background photons, which are not blocked by the filters behind the OPO in measurements (b) and (c). The solid lines are theoretical fits to the model given in Eq. (5) .
Measured triple coincidence probability P ssi (t i − t 1, t i − t 2) for different pump powers ( , , and ). The histograms have a bin width of 4.5 ns. The data points are normalized such that P ssi (0, ∞) = P ssi (∞, 0) = 1. In all three cases the same color bar is used.
Measured (symbols) conditioned auto-correlation function for the three different pump powers ( , , and ). A binning of 4.5 ns is chosen. As a guide to the eye a fit (lines) to the theoretical model given in Ref. 33 is added.
Channel 1 and 2 rms jitter in ps against the sync input for different repetition rates.
Article metrics loading...
Full text loading...