3.1. Absorption and emission spectra
Figure 2a shows an image depicting Fl (yellow-green) and seven rhodamine derivatives: (rhodamine 110 (R110), rhodamine 123 (R123), rhodamine 590 (R590), rhodamine 610 (R610 or RhB), kiton red 620 (K620), rhodamine 640 (R640), and sulforhodamine 640 (S640)) under ambient light conditions at 21°C. Fl, a popular fluoropore for use in LIF, was chosen as a baseline for comparison purposes. The color of the rhodamine derivatives ranges from yellow-green (R123) to dark pink (S640), indicating that their emissions range from ~ 500 nm to ~ 600nm. Although R123, R110, and R6G are discernable in the figure, RhB, K620, R640, and S640 are difficult to distinguish by color due to the similarity of their emission spectra.
Figure 2b and 2c show absorption and emission spectra for Fl and seven rhodamine derivatives at a 10 µM concentration, while their maximum absorption and emission values are tabulated in Table 1. For the rhodamine derivatives, the seven dyes combined cover a wavelength range from 420nm to 650nm. It is noteworthy that four of these fluorophores, R610, K620, R640, and S640, have a second absorption peak at a shorter wavelength, making their absorption range broader than other similar dyes. This second absorption peak characteristic also results in a second peak of fluorescence at longer wavelengths (Fig. 2c). These broadening effects are beneficial, especially when these fluorophores are employed in LIF techniques because 1) there is more flexibility in the selection of the excitation sources (e.g., Ar + laser at 488, 514 nm, Nd:YAG laser at 532 nm), and 2) various combinations of fluorophores are possible in dual-tracer applications by minimizing crosstalk between two wavelength bands [1, 7].
Table 1
Absorption and emission maximum for Fl and seven rhodamine dyes
Dye | Fl | R110 | R123 | R590 | RhB | K620 | R640 | S640 |
Wavelength at absorption maximum (nm) | 490 | 496 | 500 | 526 | 554 | 565 | 586 | 586 |
Wavelength at emission maximum (nm) | 515 | 521 | 534 | 554 | 580 | 588 | 609 | 610 |
3.2. Temperature and pH dependence
LIF provides a practical, non-invasive approach to measuring liquid temperatures that has been extensively used in different applications [1, 7, 35]. The temperature-dependent characteristics on the quantum yield of some dye molecules allow for thermometry by monitoring the fluorescence intensity changes with temperature. To accurately measure temperatures using LIF, it is essential to carefully choose the appropriate temperature-sensitive tracers. Figure 3 shows the temperature-dependent fluorescence spectra for Fl and the seven rhodamine derivatives picked in this study. All spectra were obtained in a temperature-controlled spectrometer (SpectraMax, USA) at 27, 35, 45 and 55°C. It is well known that Fl is a good temperature tracer with a strong temperature-sensitive emission [1], as seen in Fig. 3 (a). The temperature sensitivity of the rhodamine dyes varies depending on the specific fluorophore. The emission of Rhodamine 110 shows no temperature dependence, suggesting that this dye serves as a good reference dye when no temperature sensitivity is desired. On the other hand, all the other derivatives show temperature-dependent emissions, with a uniform decrease in intensity across the emission spectrum but without distortion or shift on its wavelength profile.
Table 2 shows quantitative results of changes in both the emission maximum and the total amount of emission as temperatures change from 26°C to 55°C. There are very slight differences in the exact percentage change between the peak fluorescence intensity and the sum of the total emission intensities as a function of temperature, mostly due to experimental error, but the overall characteristics of change are essentially the same. Fl shows about 31% change of emission, resulting in a temperature sensitivity of -1.08%/°C. Among the rhodamine derivatives, the dyes that show the largest changes in their emissions are R610 and K620, showing a temperature sensitivity of -1.55 and − 1.53%/°C, respectively. These sensitivities are in agreement with the literature [1, 7]. If a dual fluorescence LIF approach is to be used, a tracer pair combination of a temperature-sensitive dye and a temperature-insensitive dye is typically required. By taking the ratio of two emissions, the source of errors due to fluctuation of excitation intensity can be minimized. As long as both dyes are excitable by the same illumination source, it is apparent that the combination of R610-R110 shows the highest temperature sensitivity and, therefore, provides the most accurate temperature measuring system.
Table 2
Changes in emissions of Fl and rhodamine derivatives with temperatures and their average temperature sensitivity in the range of 27–55°C.
| Fl | R110 | R123 | R590 | R610 | K620 | R640 | S640 |
% change of emission maximum* | 33.1 | 0.4 | 10.6 | 14.5 | 46.5 | 46.4 | 20.3 | 18.3 |
% change in total emission (based on area)* | 31.5 | 3.3 | 9.2 | 14.3 | 45.0 | 44.6 | 18.8 | 16.1 |
Temperature Sensitivity (%/°C )** | -1.08 | -0.11 | -0.31 | -0.49 | -1.55 | -1.53 | -0.64 | -0.55 |
*Temperature changes from 27 to 55°C |
**When curve-fitted to a linear |
Temperature-dependent fluorescence occurs because either the quantum yield or the fluorophore's absorption spectrum is sensitive to temperature changes. Fl is a well-known fluorophore with different temperature sensitivities when excited at different wavelengths due to its temperature-sensitive absorption characteristics [1]. For example, Fl has a temperature sensitivity of -0.16%/°C when excited at 488 nm, while it has 2.43%/°C at 514nm excitation. To investigate the effect of temperature on the fluorescence of the rhodamine derivatives, their absorption spectra were measured at temperatures of 27, 35, 45, and 55°C. Figure 4 shows representative absorption spectra for R610 and S640 dyes. It is apparent from Fig. 4 that the absorption spectra of all derivatives are insensitive to temperature. These results support the notion that the temperature sensitivity of their emissions is unaffected by the excitation wavelength. Rather, the emissions change with temperature, apparently due to variations in quantum yield with temperature.
Another important aspect to consider when using organic fluorescent molecules as tracers for measurements of physical properties is their sensitivity to solution pH. Changes in pH cause variations in emission intensities due to the pH-dependent absorptivity of the dye molecules. Figure 5 shows the emission spectra for the representative rhodamine dyes R123 and S640 that exhibit emission characteristics independent of solution pH over the 4.5, 6.5, and 8.5 values. Although rhodamine dyes are a good choice as tracers for temperature measurements in environments with variable pH conditions, other organic fluorophores, such as Fl, are better choices for pH measurements.
As one of the representative demonstrations for pH measurements with these dyes, we measured pH changes in DI water with fluorescein and rhodamine B upon continuous CO2 injection into test solutions. Because of cascading hydrogen ion production from CO2-water reactions, the carbonated solution by CO2 becomes acidic, lowering the solution pH. Figures 6(a) and (b) show changes in emission intensities of both fluorescein and rhodamine B dyes at 6 different time steps along the microchannel when the solution was pressurized by a CO2 gas. In the case of fluorescein dye, the emission changed from the meniscus (not shown in the figure) of the CO2-water interface. Since a CO2 gas continuously dissolves into the test solution (from the righthand side of the figure), the solution pH becomes acidic. Figure 6(a) shows an intensity decrease of fluorescein’s emission over time due to its pH sensitivity. On the other hand, emission intensities of rhodamine B remained constant over time upon CO2 injection, as expected, because of the pH-insensitivity of its emission intensity.
3.3. Fluorescence lifetime
Fluorescence lifetime (τ), defined as the average time that a molecule resides in the excited state before returning to its ground state, is an intrinsic property of the fluorophore. Due to the high sensitivity of the fluorescence lifetime to environmental conditions, measurements of the fluorophore’s lifetime can be used to evaluate changes in pH [39–42], temperature [43–45], and ion concentration[46, 47]. For example, Sanders et al.[39] used the ratio of fluorescence lifetime measurements in two different time windows to map pH values over Chinese hamster ovary cells with a carboxy SNAFL-1 whose fluorescence lifetime is pH-dependent. Similarly, Nakabayashi et al. [42] used fluorescence lifetime imaging microscopy (FLIM) to image the intracellular pH of HeLa cells by using the pH-dependent ionic equilibrium of the chromophore from enhanced green fluorescent proteins. In order to measure intercellular temperature distributions in a living COS7 cell, Okabe et al. [44] employed the temperature-dependent fluorescent lifetime of a fluorescent polymeric thermometer that diffuses throughout the entire cell, combined with a time-correlated single photon counting system-based FLIM. In addition, Tremier et al. [48] utilized FLIM to determine fluorescence resonance energy transfer (FRET) in living cells as a powerful method for visualizing protein-protein interactions and biochemical reactions. They studied the impact of photobleaching on the measured fluorescent lifetime from fluorescent proteins such as CFP, YFP, and GFP in FRET applications. Lastly, Bopp et al. [49] used fluorescence lifetime measurement to estimate the photobleaching time of one bacteriochlorophyll molecule by using a confocal fluorescence microscope with picosecond time resolution.
Fluorescence emission occurs when the excited fluorophore molecules return to their ground state releasing energy in the form of visible light during this process. This process and the corresponding fluorescence intensity follow an exponential decay over time that can be expressed as:
I(t) = I0 exp (-t/ τ),
where I0 is the maximum intensity of the fluorescent emission at the population inversion point. The fluorescence lifetime is defined as the time taken for this peak fluorescence intensity to decay to 1/e (37%) of its value. Although fluorescence derives from a stochastic excitation-relaxation process, its intensity initially increases as the number of molecules that populate the excited state increases until reaching its maximum intensity, Io, at the onset of the population inversion point before gradually decaying.
In the current study, we measured the fluorescence lifetime of the seven rhodamine derivatives under the same environmental conditions, specifically room temperature (23°C), atmospheric pressure, and pH 6.5. Figure 7 shows the fluorescence lifetime measurements for Fl and the seven rhodamine dyes in DI water. It is apparent that there are two distinct groups of fluorescence lifetimes for the rhodamine dyes: one for RhB and K620, and another one for the rest of the rhodamine derivatives (Table 3). These results indicate that the emissions of RhB and K620 have shorter decay times (less than 1.74 ns) than the other rhodamine dyes (decay times of 4.53 ns and more). Due to the large range in fluorescence lifetimes (1.71 ns to 5.08 ns), in terms of absolute lifetime values, these rhodamine derivatives seem comparable to the fluorophores used in the aforementioned applications above. For example, the lifetime of carboxy SNAFL-1 is 1.1 ns, while those of GFP and YFP are 3.5 and 4.5 ns, respectively. However, to ensure that these rhodamine dyes are competitive against these other tracers for the purposes of measuring environmental pH and temperature changes, further work on the dependence of their fluorescent lifetimes as a function of these variables is needed.
Table 3
Fluorescence lifetime for Fl and seven rhodamine dyes
Dye | Fl | RhB | K620 | R123 | R110 | R590 | S640 | R640 |
Fluorescence Lifetime (ns) | 3.74 | 1.71 | 1.74 | 4.53 | 4.66 | 4.80 | 4.92 | 5.08 |
3.4. Diffusion into PDMS
PDMS-based chips are well suited to LIF techniques due to their transparency and ease of fabrication [50, 51]. However, they have been shown to have issues with absorption and diffusion of small molecules into their surfaces and bulk [52, 53]. However, to date, no study has compared the quantitative degree of permeability of rhodamine derivatives into PDMS microchannels. To test their permeability into the PDMS material, each of the seven rhodamine derivatives was mixed in solution with DI water at a concentration of 10 µM and injected into a 100 µm-square PDMS microchannel. Two fluorescence images of the microchannel were taken at 0 and 60 minutes after the solution was injected at a flow rate of 10 µL/min. Figure 8 shows greyscale images of the fluorescence emission when RhB was injected into the chip at 0 min (Fig. 8a), depicting no initial absorption or diffusion of the dye solution into the PDMS and after 60 min (Fig. 8b). By continuing to expose the microchannel to the RhB solution, a large amount of RhB molecules start to permeate into the PDMS walls and subsequently penetrate into the PDMS bulk, producing a smoother fluorescence profile that gradually decays from the wall to the interior of the sample. After 60 min of injection, it is apparent that significant diffusion of RhB molecules into the PDMS material has occurred, as shown in Fig. 8b. To quantify and compare the degree of diffusion between the seven rhodamine molecules, the intensity profiles at t = 60 min were analyzed against the profile at t = 0 min as shown in Fig. 8c. As can be seen in the figure, the intensity profile of RhB is more pronounced towards the PDMS bulk material as compared to the others, suggesting that the diffusion of these molecules is more significant. Quantitatively, RhB penetrates into the PDMS material up to 30% of the microchannel width, while the others penetrate into the PDMS bulk by only 5% of its width.