By taking a holistic approach, researchers at the University of Michigan in Ann Arbor have developed a multicolor fluorescence detection scheme that uses a fast, broadband white light to excite several types of fluorophores. A time-resolved detector captures the whole-spectrum fluorescence, without the use of excitation or emission filters. Team member Jing Yong Ye noted that the combination of supercontinuum generation and time-resolving detection is novel. “It has not been done before — to use unfiltered supercontinuum as a unique excitation source for fluorescence measurements,” he said. The fluorescence measurement of a concentrated mixture of 6-Tamra at 570 nm and Deep Red at 670 nm, seen here, shows how multiple signals can be differentiated in time. The short-duration, broad-spectrum supercontinuum that excites the fluorescence appears in the detection window as the spike near 0 ns (left) and the broad line near 0 ns (right). Adjusting the delay moves it out, enabling the detection of much weaker signals. Courtesy of Jing Yong Ye, University of Michigan, Ann Arbor. The separation of fluorescence from excitation traditionally has been done in the frequency, or wavelength, domain. The excitation light is at one frequency and often is generated with a laser because of its narrow-wavelength output. The fluorescence signal is at a different wavelength, some spectral distance away. If multiple fluorophores are used, all emissions must be spectrally distinct from each other and from the excitation light. Furthermore, for this scheme to work, the excitation and emission should not significantly overlap, which can be hard to achieve. Another drawback is that many fluorophores absorb at specific wavelengths, so having multiple types present means using various excitation lasers and different detectors, making the setup challenging. The Michigan researchers, therefore, chose to differentiate between excitation and emission in time. To do so, they needed a fast light source and a detector capable of capturing the resulting fluorescence. To avoid using multiple lasers, they required a supercontinuum source that produces light across a wide spectral range, enabling the excitation of multiple fluorophores at once, without resorting to filters and dichroic mirrors. They used a laser from Coherent Inc. of Santa Clara, Calif., that fired 50-fs pulses at 800 nm into a nonlinear photonic crystal from Crystal Fibre A/S of Birkerød, Denmark. The output was a subpicosecond supercontinuum burst that stretched from 460 to >850 nm. The investigators focused this light into a solution containing a mixture of fluorescent dyes. They collected the fluorescence and the scattered excitation light using lenses located at right angles to the incoming beam. With a spectrometer from Horiba Jobin Yvon of Edison, N.J., they created overlapping spectra. To distinguish between the multiple fluorescence signals and the scattered excitation light, they used a Hamamatsu streak camera, synchronizing it with the start of the pulse. By adjusting the delay, they avoided the scattered excitation light, which appeared in the first 10 ps after the pulse, and captured only the fluorescence, which had a lifetime of hundreds of picoseconds to tens of nanoseconds. The researchers sent the output of the streak camera to either a cooled CCD camera from Princeton Instruments of Trenton, N.J., or to a photon-counting photomultiplier tube made by Hamamatsu. The former recorded the time-resolved fluorescence spectrum, while the latter detected rapidly changing signals. With this setup, the group imaged samples with two fluorescent dyes, Deep Red and 6-Tamra, in both a cuvette and in a flow cytometer. The emissions of the dyes could be differentiated clearly in a plot of fluorescence versus wavelength and in a plot of signal intensity versus time, a consequence of different spectral outputs and fluorescence lifetimes. As for applications of the method, Ye said that these include flow cytometry, fluorescence microscopy and endoscopy. “We believe the development of this technology will lead to significant improvements in many different kinds of fluorescence-based instruments,” he said. Optics Express, Aug. 6, 2007, pp.10439-10445.