Determination of Fluorescence Lifetime

Determination of the Fluorescence Lifetime
 
After the absorption of light, a molecule remains in the electronically excited state for a certain period of time before it releases the absorbed energy. Photochemical processes of different probability compete with each other. In addition to non-radiative deactivations, light can also be emitted, e.g. as fluorescence.
The time after which the number of excited molecules (n1) has dropped to a factor of 1/e (approx. 37%) is called the fluorescence decay time (τfl):

n1(t) = n1(0) e(- t / τfl)

The fluorescence decay time of an organic dye can be used to identify it. However, the fluorescence lifetime is not a constant value, but is influenced by the environment of the dye, i.e. solvent and temperature.
The fluorescence lifetime of the ATTO dyes is typically in the range of nanoseconds (10-9 s). For our catalogue, the values for the carboxy derivative in aqueous solution (PBS, pH 7.4) at 22 °C were determined using the method of "time-correlated single photon counting" described below. The TemPro lifetime measuring instrument from HORIBA Jobin Yvon was used for this purpose.
 
Time correlated single photon counting (TCSPC)
The emission of a fluorescence photon after excitation is a statistical process. Thus, the duration of a single molecule remaining in the excited state is also a statistical quantity. However, if one considers an ensemble of many identical molecules, a well-defined decay statistic results.
In the TCSPC method, the method most frequently used today for measuring fluorescence lifetimes, this statistic is very well time-resolved - up to the picosecond range (10-12 s) - and measured with high detection sensitivity. For this purpose, a sample is irradiated with periodic, extremely short light pulses. After excitation, the sample emits fluorescence photons. The exact time span is measured until the first photon reaches the detector. These individual events are recorded, i.e. registered in a histogram as "count". The exponential decay curve of the sample is finally obtained by adding up the counts of a large number of single experiments.
 

License: public domain
https://commons.wikimedia.org/wiki/File:TCSPC_Histogram.gif?

The measuring electronics has to be reset after each laser pulse and counting process. During the dead time no photon can be registered. The conditions are choosen that only about 1 photon per 100 excitation pulses is detected. At a higher rate, the short times in the histogram would be overestimated and the decay curve distorted ("pile-up"). In practice, the measurement is often carried out in the so-called "time-inverted" mode, i.e. the fluorescence photon starts the time measurement while the excitation laser pulse registered by another diode stops the time measurement.
 
Before measuring a dye sample, the following information of the dye should be available:

  • Absorption and fluorescence maximum, e.g. from the measurement of an absorption and fluorescence spectrum of the solution or from literature.
  • approximate fluorescence decay time, e.g. estimated from previous measurements, measurements of comparable systems or from literature.
The spectral data determines the selection of the excitation light source - for which intensive laser or semiconductor diodes are commonly used today - and the choice of an appropriate cut-off filter. This filter should prevent scattered excitation light from reaching the detector.
 
In addition to the selection of the excitation light source and the cut-off filter, some other experimental prerequisites must also be considered:
  • The sample solution should have an optical density < 0.1 in the 1 cm cell at and above the excitation wavelength to avoid reabsorption. Diluted solutions also prevent problems caused by e.g. aggregation and concentration quenching.
  • During data analysis, scattering solutions can show additional unexpectedly short lifetimes. This can be avoided, e.g. by µ-filtration of the sample solution.
  • Transparent liquids are usually measured in the 90 ° arrangement in 1 cm fluorescence cells.
  • The detection time frame should be long enough (at least 10000 counts).
For fluorescence lifetime measurements using the TCSPC principle, it is also necessary to ensure that the count rate does not exceed 2% of the repetition rate of the pulsed excitation light source. This avoids - as explained above - the so-called "pile-up".
In order to reduce the counting rate, the following can be done:
  • The sample solution can be diluted.
  • A neutral glass filter is inserted into the excitation channel reducing the intensity of the excitation light.
  • A filter with longer wavelength cut-off can be inserted into the emission channel reducing the spectral range of the detected fluorescence.
Reverse actions can be applied to increase the count rate up to 2 %.
To determine the instrument function, i.e. the response of the electronics, a scattering solution - e.g. 0.01 % Ludox AS40 colloidal silica in water - should be measured. The result of this reference measurement can be taken into account during later data evaluation.
The data is analyzed by the supplied software. The measured data points are fitted by a curve. The theoretical course of the fluorescence decay curve, like the radioactive decay, corresponds to a 1st order reaction and is therefore mono-exponential. However, a higher exponential curve can also result as a better fit for the measuring points. This may be the case, for example, if different species with different fluorescence lifetimes are present in the solution.

The following figure shows an example of the measurement of a batch of ATTO 488 carboxy with the TemPro device. The previously described parameters such as detection time window (peak count > 10000), reference measurement/prompt with the Ludox solution and the measurement of the sample can be seen. A bi-exponential decay with CHISQ = 1.1 and uniformly distributed standard deviation was chosen as the best fit. The fluorescence decay time of ATTO 488 carboxy in aqueous solution (PBS, pH 7.4) at 22 °C is 4.1 ns with a relative proportion of 96.78 % in this sample.


 
The quality of the fitted curve - and thus the correctness of the determined fluorescence lifetime - is indicated by χ2 (CHISQ, XSQ) and the weighted residuals in the software.
A good fit is recognized by:
  • χ2 < 1.2 and not significantly changing when adding another component (mono-exponential -> bi-exponential etc.)
  • an even distribution of the weighted residuals
  • Consistency with previous measurements or literature values
Providers of commercial devices provide comprehensive manuals and product documentation, which usually describe the correct operation of the device in question and the supplied software in detail. By considering the given instructions and measuring dyes with known fluorescence decay times at first, the necessary routine and experience can be obtained to measure "correct" lifetimes with the equipment used.
 
Literature about TCSPC:
D.V. O’Connor, D. Phillips, Time-correlated single photon counting, Academic Press, New York (1984).
J.R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd Edition, Springer Science+Business Media, New York (2006).

Literature employing ATTO dyes:
F. Koberling, M. Wahl, M. Patting, H.-J. Rahn, P. Kapusta, R. Erdmann, Two channel fluorescence lifetime microscope with two colour laser excitation, single molecule sensitivity and sub-micrometer resolution, Proceedings of SPIE 5143, 181 (2003). → ATTO 655
S. Raymond, D. Boas, B. Bacskai, A. Kumar, Lifetime-based tomographic multiplexing, Journal of Biomedical Optics 15, 46011 (2010). → ATTO 740
J. Bückers, D. Wildanger, G. Vicidomini, L. Kastrup, S. Hell, Simultaneous multi-lifetime multi-color STED imaging for colocalization analyses, Optics Express 19, 3130 (2011). → ATTO 590, ATTO 647N
P. Haus, M. Korbus, M. Schröder, F.-J. Meyer-Almes, Identification of Selective Class II Histone Deacetylase Inhibitors Using a Novel Dual-Parameter Binding Assay Based on Fluorescence Anisotropy and Lifetime, Journal of Biomolecular Screening 16, 1206 (2011). → ATTO 700
P.-Y. Lin, S.-S. Lee, C.-S. Chang, F.-J. Kao, Long working distance fluorescence lifetime imaging with stimulated emission and electronic time delay, Optics Express 20, 11445 (2012). → ATTO 647N
Q. Wang, W. Moerner, Lifetime and Spectrally Resolved Characterization of the Photodynamics of Single Fluorophores in Solution Using the Anti-Brownian Electrokinetic Trap, The Journal of Physical Chemistry B 117, 4641 (2013). → ATTO 647N
I. Ziomkiewicz, A. Loman, R. Klement, C. Fritsch, A. Klymchenko, G. Bunt, T. Jovin, D. Arndt-Jovin, Dynamic conformational transitions of the EGF receptor in living mammalian cells determined by FRET and fluorescence lifetime imaging microscopy, Cytometry Part A 83, 794 (2013). → ATTO 390
U. Bhattacharjee, C. Beck, A. Winter, C. Wells, J. Petrich, Tryptophan and ATTO 590, The Journal of Physical Chemistry B 118, 8471 (2014). → ATTO 590
U. Kaiser, N. Sabir, C. Carrillo-Carrion, P. del Pino, M. Bossi, W. Heimbrodt, W. Parak, Förster resonance energy transfer mediated enhancement of the fluorescence lifetime of organic fluorophores to the millisecond range by coupling to Mn-doped CdS/ZnS quantum dots, Nanotechnology 27, 55101 (2015). → ATTO 590, ATTO 633, ATTO 655
S. Isbaner, N. Karedla, D. Ruhlandt, S. Stein, A. Chizhik, I. Gregor, J. Enderlein, Dead-time correction of fluorescence lifetime measurements and fluorescence lifetime imaging, Optics Express 24, 9429 (2016). → ATTO 647N, ATTO 655