Limits...
An Analog Filter Approach to Frequency Domain Fluorescence Spectroscopy.

Trainham R, O'Neill M, McKenna IJ - J Fluoresc (2015)

Bottom Line: The rate equations found in frequency domain fluorescence spectroscopy are the same as those found in electronics under analog filter theory.The techniques described here can be used to separate signals from fast and slow fluorophores emitting into the same spectral band, and data collection can be greatly accelerated by means of a frequency comb driver waveform and appropriate signal processing of the response.The simplification of the analysis mathematics, and the ability to model the entire detection chain, make it possible to develop more compact instruments for remote sensing applications.

View Article: PubMed Central - PubMed

Affiliation: Special Technologies Laboratory, National Security Technologies, LLC, 5520 Ekwill Street, Santa Barbara, CA, 93111, USA. trainhcp@nv.doe.gov.

ABSTRACT
The rate equations found in frequency domain fluorescence spectroscopy are the same as those found in electronics under analog filter theory. Laplace transform methods are a natural way to solve the equations, and the methods can provide solutions for arbitrary excitation functions. The fluorescence terms can be modelled as circuit components and cascaded with drive and detection electronics to produce a global transfer function. Electronics design tools such as SPICE can be used to model fluorescence problems. In applications, such as remote sensing, where detection electronics are operated at high gain and limited bandwidth, a global modelling of the entire system is important, since the filter terms of the drive and detection electronics affect the measured response of the fluorescence signals. The techniques described here can be used to separate signals from fast and slow fluorophores emitting into the same spectral band, and data collection can be greatly accelerated by means of a frequency comb driver waveform and appropriate signal processing of the response. The simplification of the analysis mathematics, and the ability to model the entire detection chain, make it possible to develop more compact instruments for remote sensing applications.

No MeSH data available.


This family of curves was generated using the same uranyl fluoride and coumarin fluorophores, the same laser, and the same photodiode as were used for Fig. 6, These curves were constructed using only three drive frequencies: 10 Hz, 250 Hz, and 2.5 kHz. The driver signal in this measurement was a square wave, and the data were collected by a fast digitizer. The data were processed for phase responses of the harmonic components up to 50 harmonics
© Copyright Policy - OpenAccess
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4644193&req=5

Fig9: This family of curves was generated using the same uranyl fluoride and coumarin fluorophores, the same laser, and the same photodiode as were used for Fig. 6, These curves were constructed using only three drive frequencies: 10 Hz, 250 Hz, and 2.5 kHz. The driver signal in this measurement was a square wave, and the data were collected by a fast digitizer. The data were processed for phase responses of the harmonic components up to 50 harmonics

Mentions: Figure 9 shows experimental results from data collected with only three drive frequencies of the square wave, and exploitation of the phase shifts of harmonics up to harmonic 50. By comparison, the data in Fig. 6 were collected using a pure sine wave and 80 frequency samples for each curve. The experimental setup for Fig. 9 was identical to the setup used to collect the data shown in Fig. 6, except that the lock-in detector was replaced by a National Instruments USB-6251 fast waveform digitizer. The time domain signals from the driver and response waveforms were each averaged over a fixed number of cycles, typically 100, and then transformed into the frequency domain by Fast Fourier Transform (FFT). The FFT arrays were then averaged for mean and standard deviations of all complex frequency values of the harmonics. The phase delays between the response and driver signals of harmonics up to 50 were computed, and these are shown in Fig. 9. The error bars in the plot are the standard deviations of the computed phase shifts. The three frequencies used to reconstruct the family of phase shift curves were 10 Hz, 250 Hz, and 2.5 kHz. The various curves in the figure are for different mixtures of fast and slow lifetime fluorophores. The data collection time for each curve in Fig. 9, using three drive frequencies, was only a few seconds. By contrast, the data collection time in Fig. 6 required 6 minutes for each curve.Fig. 9


An Analog Filter Approach to Frequency Domain Fluorescence Spectroscopy.

Trainham R, O'Neill M, McKenna IJ - J Fluoresc (2015)

This family of curves was generated using the same uranyl fluoride and coumarin fluorophores, the same laser, and the same photodiode as were used for Fig. 6, These curves were constructed using only three drive frequencies: 10 Hz, 250 Hz, and 2.5 kHz. The driver signal in this measurement was a square wave, and the data were collected by a fast digitizer. The data were processed for phase responses of the harmonic components up to 50 harmonics
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4644193&req=5

Fig9: This family of curves was generated using the same uranyl fluoride and coumarin fluorophores, the same laser, and the same photodiode as were used for Fig. 6, These curves were constructed using only three drive frequencies: 10 Hz, 250 Hz, and 2.5 kHz. The driver signal in this measurement was a square wave, and the data were collected by a fast digitizer. The data were processed for phase responses of the harmonic components up to 50 harmonics
Mentions: Figure 9 shows experimental results from data collected with only three drive frequencies of the square wave, and exploitation of the phase shifts of harmonics up to harmonic 50. By comparison, the data in Fig. 6 were collected using a pure sine wave and 80 frequency samples for each curve. The experimental setup for Fig. 9 was identical to the setup used to collect the data shown in Fig. 6, except that the lock-in detector was replaced by a National Instruments USB-6251 fast waveform digitizer. The time domain signals from the driver and response waveforms were each averaged over a fixed number of cycles, typically 100, and then transformed into the frequency domain by Fast Fourier Transform (FFT). The FFT arrays were then averaged for mean and standard deviations of all complex frequency values of the harmonics. The phase delays between the response and driver signals of harmonics up to 50 were computed, and these are shown in Fig. 9. The error bars in the plot are the standard deviations of the computed phase shifts. The three frequencies used to reconstruct the family of phase shift curves were 10 Hz, 250 Hz, and 2.5 kHz. The various curves in the figure are for different mixtures of fast and slow lifetime fluorophores. The data collection time for each curve in Fig. 9, using three drive frequencies, was only a few seconds. By contrast, the data collection time in Fig. 6 required 6 minutes for each curve.Fig. 9

Bottom Line: The rate equations found in frequency domain fluorescence spectroscopy are the same as those found in electronics under analog filter theory.The techniques described here can be used to separate signals from fast and slow fluorophores emitting into the same spectral band, and data collection can be greatly accelerated by means of a frequency comb driver waveform and appropriate signal processing of the response.The simplification of the analysis mathematics, and the ability to model the entire detection chain, make it possible to develop more compact instruments for remote sensing applications.

View Article: PubMed Central - PubMed

Affiliation: Special Technologies Laboratory, National Security Technologies, LLC, 5520 Ekwill Street, Santa Barbara, CA, 93111, USA. trainhcp@nv.doe.gov.

ABSTRACT
The rate equations found in frequency domain fluorescence spectroscopy are the same as those found in electronics under analog filter theory. Laplace transform methods are a natural way to solve the equations, and the methods can provide solutions for arbitrary excitation functions. The fluorescence terms can be modelled as circuit components and cascaded with drive and detection electronics to produce a global transfer function. Electronics design tools such as SPICE can be used to model fluorescence problems. In applications, such as remote sensing, where detection electronics are operated at high gain and limited bandwidth, a global modelling of the entire system is important, since the filter terms of the drive and detection electronics affect the measured response of the fluorescence signals. The techniques described here can be used to separate signals from fast and slow fluorophores emitting into the same spectral band, and data collection can be greatly accelerated by means of a frequency comb driver waveform and appropriate signal processing of the response. The simplification of the analysis mathematics, and the ability to model the entire detection chain, make it possible to develop more compact instruments for remote sensing applications.

No MeSH data available.