Viewing entries in
LIFA-X

Oxygen transport across slippery and curved gas-liquid interfaces using phosphorescence lifetime imaging

The transport phenomena at interfaces often determine or limit the overall performance of processes. Direct investigations of interfacial transport of momentum, mass and heat at the interfaces in micron scale are highly appreciable for further optimization of various micro and macroscale technologies. The Soft Matter Group at the University of Twente (The Netherlands), lead by Prof. Dr. Rob Lammertink, aims at gaining a better understanding of transport phenomena near boundaries, so that various processes such as desalination, separation of species and (photo)catalytic reactions can be improved.

Microfluidics offer an ideal platform allowing for the integration of 'controllable' surfaces and direct measurements of transport phenomena near them. Elif Karatay used a microfluidic bubble mattress during her PhD studies at the University of Twente, fabricating one of the microchannel walls as a superhydrophobic surface consisting of alternating solid walls and micro-bubbles (figure 1). She experimentally measured and numerically estimated the dynamic mass transfer of gas absorption at stable gas-liquid interfaces for short contacting times.

 Figure 1. Microfluidic bubble mattress (a). Numerical simulations of dissolved oxygen in water with identical settings in FLIM experiments (b), the color bar shows the oxygen concentration. Lifetime field resolved by FLIM superimposed on the bright-field microscopy image showing bubbles protruding into the water (c), the color bar indicates fluorescence lifetime in nanoseconds.

Figure 1. Microfluidic bubble mattress (a). Numerical simulations of dissolved oxygen in water with identical settings in FLIM experiments (b), the color bar shows the oxygen concentration. Lifetime field resolved by FLIM superimposed on the bright-field microscopy image showing bubbles protruding into the water (c), the color bar indicates fluorescence lifetime in nanoseconds.

The rate of gas absorption into water was studied by in situ measurements of dissolved oxygen concentration profiles in aqueous solutions flowing over oxygen bubbles by frequency-domain fluorescence-lifetime imaging (FD-FLIM) microscopy.  FLIM was used to image the oxygen concentration using an oxygen sensitive luminescent dye, ruthenium tris(2,20 -dipyridyl) dichloride hexahydrate (RTDP), obeying a mono-decay function as quenched by oxygen. For the FLIM experiments, an eXtended Lambert Instruments FLIM Attachment (LIFA-X) system was used on a Zeiss Axio Observer inverted microscope. The LIFA-X, consisting of a LED light source and a Lambert Instruments intensified CCD camera (TRiCAM), was operated in gated mode to obtain phosphorescence lifetimes.

For calibration measurements, lifetimes of the RTDP were measured in oxygen-free (N2 saturated), aerated and oxygen-saturated aqueous solutions, where the micro-bubbles were established by nitrogen, air and oxygen gases, respectively. During the mass transfer experiments, oxygen gas micro-bubbles were established at the boundary of the microchannels and deoxygenated RTDP aqueous solution was the working liquid flowing past the transversely aligned oxygen bubbles (figures 1c and 2).

The lifetime of RTDP across the liquid side microchannel height was measured at different axial locations (figure 2). The bubble interface profiles and locations were experimentally determined by locating the minimum lifetime data measured near the hybrid wall. Figure 2} shows the successive lifetime fields measured with FLIM at different axial locations along the same microchannel embedded with micro-bubbles where the increasing boundary layer thickness along the downstream flow can be observed. Here, in figure 2c, the thickness of the diffusion boundary layer is 23% of the microchannel height H, and does not extend further into the microchannel due to a relatively large Reynolds number Re of 7.5.

 Figure 2. Successive lifetime fields in axial position x. Quantitative visualization of the increasing boundary layer thickness along downstream flow (45 μl/min). The flow direction is from left to right. The color bar refers to the lifetime, which is given in nanoseconds. The dashed arrows indicate the axial positions at which the local oxygen concentration profiles across the microchannel height are obtained.

Figure 2. Successive lifetime fields in axial position x. Quantitative visualization of the increasing boundary layer thickness along downstream flow (45 μl/min). The flow direction is from left to right. The color bar refers to the lifetime, which is given in nanoseconds. The dashed arrows indicate the axial positions at which the local oxygen concentration profiles across the microchannel height are obtained.

The local profiles of oxygen flux from the gaseous phase to the liquid phase were obtained using the local concentration gradients measured by FLIM. Furthermore the space-averaged total flux of oxygen absorption was calculated from these local flux profiles.

The experimental results obtained by FLIM revealed an additional mass transfer resistance to gas dissolution on slippery micro-bubbles at short contact times of gas and liquid. This mass transfer resistance results in slower gas absorption than predicted by the conventionally accepted equilibrium interface model, Henry’s Law. Whereas the experimental results are in good agreement with the numerical results obtained from simulations considering non-equilibrium conditions. The results indicate that the phase equilibrium state may not be established at short contacting times.

Acknowledgements

Graphs courtesy of Dr. Elif Karatay, Stanford University, USA

References

E. Karatay, A.P. Tsai and R.G.H. Lammertink, Rate of gas absorption on a slippery bubble mattress, Soft Matter, 2013, 9, 11098-11106

Frequency-Domain FLIM: From nanoseconds to milliseconds

In modern day science Fluorescence Lifetime Imaging Microscopy (FLIM) is more and more recognized as a powerful diagnostic tool. The lifetime of fluorescence is the decay time of the emission after excitation stopped.

Fluorescence lifetime can be affected by a number of effectors including pH changes and changes in oxygen concentration. One very potent application of FLIM is Forster Resonance Energy Transfer (FRET) analysis for studying protein interactions. In case of FRET, the emission energy is transferred to an acceptor molecule instead of being released as a photon (emission). FRET effectively reduces the decay time of the donor emission.

Fluorescence lifetimes fall typically within the range of 1–10 ns. Lambert Instruments offers a user friendly robust FLIM system, the Lambert Instruments FLIM Attachment, for resolving lifetimes within the range of 0 to 300 ns, with an accuracy of 80 ps. Now this lifetime resolving capacity range has been extended to 1 ms with their new product, the LIFA-X.

Frequency-Domain FLIM

 Figure 1. Modulated excitation will result in modulated emission with an identical repetition frequency. The phase shift and decrease in modulation depth are caused by the decay of the fluorescent molecule and can be used for deducing the lifetime information.

Figure 1. Modulated excitation will result in modulated emission with an identical repetition frequency. The phase shift and decrease in modulation depth are caused by the decay of the fluorescent molecule and can be used for deducing the lifetime information.

Frequency-domain FLIM is a method to acquire wide-field FLIM images. For all pixels in the image the decay times of fluorophores are collected simultaneously. In order to achieve this, a homodyne detection scheme is applied in which the intensity of the excitation light is modulated in a sinusoidal fashion. Consequently, the emission signal will have a modulated intensity with an identical repetition frequency. However, the phase of the emission signal will be delayed (= phase shift) because of the decay of the fluorophore. The decay of the emission also results in a lower amplitude of the emission signal (= decrease in modulation depth). Therefore, the emission signal contains two parameters from which the fluorescence lifetime can be deduced: the phase shift and decrease in modulation depth (figure 1).

The lifetime of a fluorescent material can be determined from the phase-shift by the following relation:
\begin{equation}
    \tau_\phi = \frac{1}{\omega} \tan \phi,
    \label{eq:PhaseLifetime}
\end{equation}
where \(\omega\) is the angular frequency of the modulation and \(\phi\) is the phase-shift.

Equally, the lifetime can be determined from the decrease in modulation depth \(m\). The decrease in modulation depth \(m\) is the relative modulation depth of the emission signal as compared to the excitation:
\begin{equation}
    \tau_m = \frac{1}{\omega} \sqrt{\frac{1}{m^2} - 1}.
    \label{ModulationLifetime}
\end{equation}
In case of a single exponential decay it can be shown that \(\tau_\phi\) is equal to \(\tau_m\). A significant difference between these two calculated values indicates the presence of more components having different lifetimes. For further analysis of multi-exponential decay the software offers the Multi-Frequency method and the Polar Plot method.

LIFA

The Lambert Instruments FLIM Attachment, or LIFA, is a complete, easy to use lifetime imaging system. It can be mounted onto any wide-field fluorescence microscope. The detector is an intensified CCD (ICCD) camera that uses an image intensifier as prime detector that is fiber optically coupled to a CCD. This ICCD allows detection of low-light levels, a typical feature of fluorescence. The high sensitivity of the detector allows the usage of a low excitation intensity, minimizing photo-bleaching of the sample. Modulated LEDs are used as the standard source of light in the LIFA system.

 Figure 2. The LIFA system can be mounted onto any wide-field fluorescence microscope.

Figure 2. The LIFA system can be mounted onto any wide-field fluorescence microscope.

The image intensifier of the ICCD is modulated at the same frequency as the excitation signal. Here, modulation means modulation of detector sensitivity. The phase of modulation is altered through 360 phase degrees with dedicated steps. All pixels are detected simultaneously. When the detector is most sensitive at the peak of the emission signal, the measured intensity per pixel will be high; i.e. the detector modulation is in phase with the emission signal. If the detector modulation is out-of-phase with the emission signal, the detector will be least sensitive when the emission signal is most intense. The measured intensity per pixel will be low. The measured intensities per pixel are then plotted versus the detector phase and the fitted sine wave contains both the phase and the modulation depth information (figure 3).

By using a uniform material with a known lifetime (e.g. rhodamine 6G) as a reference, the unknown lifetime of a sample can be distilled when comparing the two fitted sine waves. This calculation of the lifetime is done for every pixel of the image simultaneously, which is the fundament for wide-field FLIM.

The LIFA, a set-up with modulation of both the excitation light and detector sensitivity, can be used for modulation frequencies from 100 kHz to 80 MHz with LED as a light source. Moreover, modulation frequencies up to 120 MHz are feasible with the use of modulated laser diodes. This frequency range is adequate for the measurement of lifetimes from 0.1 ns to 300 ns with a temporal resolution of 80 ps.

 Figure 3. The fit through the data points when plotting single-pixel intensities versus detector phase contains information about both the phase of the modulated emission signal as well as the modulation depth in that pixel.

Figure 3. The fit through the data points when plotting single-pixel intensities versus detector phase contains information about both the phase of the modulated emission signal as well as the modulation depth in that pixel.


The theoretically optimal frequency for resolving fluorescence lifetimes is determined using Equations 3 and 4:
\begin{align}
    f & = \frac{1}{2\pi\tau}\text{ for phase lifetime,} \label{eq:PhaseFrequency} \\
    f & = \frac{\sqrt{2}}{2\pi\tau}\text{ for modulation lifetime,} \label{eq:ModulationFrequency}
\end{align}
with \(\tau\) = lifetime (s) and \(f\) = frequency (Hz). This means that for a lifetime of e.g. 4 ns the optimal frequency is approximately 40 MHz. 

For much longer lifetimes like 10 \(\mu\)s the optimal frequency is approximately 15 kHz. For these long lifetimes a much lower frequency of the excitation is needed, else the modulation of the emitted signal from the sample is too low for accurate measurements. For coverage of the low frequency range Lambert Instruments developed the LIFA-X.

The new LIFA-X

 Figure 4. The new LIFA-X ICCD camera with integrated pulse generator.

Figure 4. The new LIFA-X ICCD camera with integrated pulse generator.

The LIFA-X is Lambert Instruments' latest addition to their FLIM product range with the 'X' indicating the extended range of lifetimes this product can resolve. The LIFA-X can resolve lifetimes from 0.1 ns up to 1 ms (phosphorescence) and is well suited, e.g., for measuring oxygen concentrations using optodes or reactive oxygen species in cells.

The LIFA-X includes a modified ICCD camera in which a dual pulse generator is built in. It can be used in modulated mode in which it can be operated like the LIFA, resolving short lifetimes (< 300 ns). Additionally it can be operated in pulsed mode for the measurement of long lifetimes (> 300 ns). The design allows retro fitting onto existing LIFA systems. When the system is operated in ‘long lifetime mode’ – i.e. the image intensifier is gated - the pulse generator controls the gating of the image intensifier. The pulse generator also controls the LED or laser excitation source. The use of a pulsed signal creates a block wave modulation pattern (figure 5). Especially at low frequencies a block wave is more effective than a sinusoidal wave in relation to an efficient use of photons.

 Figure 5.&nbsp;Frequency domain FLIM using a pulsed lightsource results in a block wave excitation and emission pattern.

Figure 5. Frequency domain FLIM using a pulsed lightsource results in a block wave excitation and emission pattern.

In gated mode frequencies of 1 kHz to 100 kHz are supported. Data analysis comprises the same statistic strategies as with modulated excitation and detection and can be done with the same LI-FLIM software package as for the LIFA.

Software

For data acquisition the LIFA and LIFA-X are supplied with the software program LI-FLIM. LI-FLIM is a user-friendly software package that controls the hardware components, collects phase images, and processes the intensity information per pixel. These data are analyzed instantaneously allowing visual representation of the lifetime data in a matter of seconds. 

 Figure 6. Screenshot of the LI-FLIM acquisition software. This software is used for analyzing the FLIM data. In the recorded sample shown, CHO cells are transfected with either GFP (donor-only, ROI with long lifetimes – red pixels) or GFP-RFP fusion protein (widely used FRET pair) of which the donor protein has a significantly shorter fluorescence lifetime due to FRET (ROI with short lifetimes – blue pixels). An overlay of the sample intensity image with lifetime data is shown on the right. The left plane of the screen shot displays the statistics window with analyzed lifetime data of the drawn regions of interest.

Figure 6. Screenshot of the LI-FLIM acquisition software. This software is used for analyzing the FLIM data. In the recorded sample shown, CHO cells are transfected with either GFP (donor-only, ROI with long lifetimes – red pixels) or GFP-RFP fusion protein (widely used FRET pair) of which the donor protein has a significantly shorter fluorescence lifetime due to FRET (ROI with short lifetimes – blue pixels). An overlay of the sample intensity image with lifetime data is shown on the right. The left plane of the screen shot displays the statistics window with analyzed lifetime data of the drawn regions of interest.

The data can be presented in an overlay of lifetime values over an intensity image of the sample. Polar plot presentation is also supported. Regions of interest can be selected for statistical analysis (figure 6). Other features include time-lapse recordings (minutes to days) and multi-frequency measurements for unmixing multiple lifetime components.


Glossary

Image intensifier

A second and third generation image intensifier consists of a photocathode that emits electrons when illuminated.  The emitted electrons are accelerated in the direction of the anode screen. Between cathode and anode a micro-channel plate (MCP) is positioned. A MCP is the micro-channel equivalent of a photon multiplier tube (PMT) and multiplies the number of primary electrons in each channel by secondary emission from the channel wall. The gain of the intensifier can be controlled by the voltage across the MCP and by the cathode voltage. 

When the electrons hit the anode screen, their energy is converted into light which is emitted through the output window of the intensifier. The wide dynamic range allows imaging of dim and bright features within one image. The spatial resolution is determined by a number of parameters such as the individual channel diameter in the MCP and the applied voltages to cathode and anode.

 Figure 6.&nbsp;Schematic representation of an image intensifier. p, photon.

Figure 6. Schematic representation of an image intensifier. p, photon.

The photocathode needs to be negatively charged in order to realize sensitivity. Namely, if positive the released electrons will remain at the photocathode; i.e. the tube is closed. Modulation/gating of the image intensifier is achieved by (rapidly) changing the voltage at the photocathode.