Open Access
Add to BibSonomy
Abstract
The use of fluorescent antennas in optical wireless communications (OWC) has been demonstrated previously, and it has been shown that it is an efficient method for enhancing receiver performance, providing both signal gain and a wide field of view (FoV). To achieve a high concentration gain at the receiver output, the selected fluorophores should have a high photoluminescence quantum yield (PLQY), limited overlap between their absorption and emission spectra, and emit light that can be efficiently detected. In addition, to support a high modulation bandwidth, the photoluminescence (PL) lifetime of the fluorophore needs to be short. In this paper, we propose a new fluorescent antenna architecture based on Förster resonance energy transfer (FRET). Our results show that, due to the photophysical interactions between the energy donor and energy acceptor, the use of FRET simultaneously increases PLQY and reduces PL lifetime. Additionally, employing FRET leads to an increased Stokes shift, ensuring that the emitted light has longer wavelengths, thus reducing self-absorption. This shift can also increase the efficiency with which the fluorescence is detected by a typical silicon (Si) photodetector. Consequently, our OWC results show that a new FRET-based antenna can achieve a significantly higher concentration gain and a wider transmission bandwidth than a conventional non-FRET antenna, leading to much higher data rates.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Optical wireless communication (OWC) is expected to play a crucial role in the future 6G network, augmenting the existing radio frequency (RF) system [1]. In OWC, the photodetector needs to have a small area so that its capacitance is low enough to support a high modulation bandwidth. However, a small photodetector can significantly reduce the intensity of the detected signal. To increase the receiver’s output signal, the conventional approach is to place a lens or a compound parabolic concentrator in front of the photodetector so that more light can be concentrated onto the photodetector. However, these methods conserve étendue and so they limit the field of view (FoV) of the receiver, [2]. Recent studies show that overcoming the étendue limit can be achieved through the utilization of fluorescent-doped optical concentrators, commonly referred to as fluorescent antennas, in OWC [3–10]. These antennas rely on fluorescence rather than reflection and/or refraction to concentrate light. Since étendue isn’t conserved by fluorescence, this enables them to achieve both a wide FoV and a high concentration gain [4]. However, when incorporating fluorescent antennas into OWC systems, certain crucial features must be considered, including: (i) the fluorophore should have high absorption of the light emitted from the transmitter but low absorption of ambient light; (ii) the fluorophore should have a high photoluminescence quantum yield (PLQY); (iii) the photoluminescence (PL) lifetime of the fluorophore should be short so that the antenna can react to fast intensity changes from the transmitter; (iv) a Stokes shift that is large enough to reduce self-absorption; and (v) the wavelength of the emitted light should be detected efficiently by a photodetector. In existing research, various fluorescent materials have been considered for building the antennas, including organic dyes [4,10], inorganic nanocrystals such as quantum dots [8,11], and commercially available fluorescent fibers [12–16]. To the best of our knowledge, the fluorescence concentrated by all these antennas is emitted by the fluorophore which absorbed light from the transmitter.
In this paper, we propose a novel approach to designing fluorescent antennas, particularly through the use of Förster resonance energy transfer (FRET). We demonstrate the ability of FRET to fulfill all the aforementioned performance criteria and the resulting significant improvement in the performance of an OWC system. In FRET, similar to a system with a single type of fluorophore, the absorption of a photon can move one of the electrons in the highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO). However, unlike a system with a single type of fluorophore, as shown in Fig. 1, the energy associated with the excited state of a donor fluorescent molecule can be transferred to an acceptor fluorescent molecule through nonradiative dipole-dipole interaction [17]. FRET is often utilized to tune the absorption and emission spectra of fluorophores for specific applications, including in luminescent solar concentrators (LSC) [18,19], and it has also been considered into a nanoscale molecular communication system [20]. Fluorescent antennas are superficially similar to LSC, however, the requirements for these applications are very different. For instance, an LSC should absorb a broad range of wavelengths from sunlight. In contrast, a fluorescent antenna needs to have a relatively narrow absorption band that primarily absorbs light from the transmitter. In addition, PL lifetime is irrelevant to an LSC, but a short PL lifetime is critically important for an antenna. Furthermore, the performance of LSC and antennas are assessed using very different criteria. It is these differences that motivated our investigation into the use of FRET in fluorescent antennas.
In this work, we explore the use of FRET in fluorescent antennas using two types of organic dyes. Specifically, Tris(8-hydroxyquinolinato)aluminium (Alq3) is selected as the energy donor, while 4-dicyanomethylene-2-methyl-6-p-dimethylaminostyryl-4 H-pyran (DCM) is chosen as the energy acceptor. The efficiency of the energy transfer in FRET depends strongly on the distance between the donor and the acceptor molecules. It therefore depends upon the relative concentrations of the donor and acceptor. The initial focus of the work on FRET-based antennas was therefore the optimization of the ratio of concentrations of Alq3 and DCM. Following this optimization, the results demonstrate that employing FRET increases the PLQY by a factor of more than 3 to 0.53. Additionally, the use of FRET reduces the PL lifetime from 9.3 ns to only 3.5 ns. Subsequently, we evaluated the antenna’s performance in an OWC system. The results show that the use of the new FRET antenna resulted in a concentration gain of 6.5, compared with a gain of 1.3 achieved by an antenna that only contains Alq3. Moreover, the new FRET antenna, with its shorter PL lifetime, supports a higher modulation bandwidth compared to the non-FRET antenna. As a result of these two improvements, the new antenna increases the transmission data rate by a factor of more than 7 compared to the case without an antenna and by a factor of 3.5 compared to a conventional non-FRET antenna.
2. Material measurements
2.1 Selected fluorophores
To implement FRET in an OWC antenna, we selected two fluorophores: Alq3 and DCM. The molecular structures of Alq3 and DCM are shown in Figs. 2(a) and (b). DCM has a high PLQY, a short PL lifetime, and emits red light, which is efficiently detected by a typical Si photodetector. These attributes mean that it has been incorporated into fluorescent antennas previously [10,21]. They also mean that it has been selected as the energy acceptor in the new FRET-based antenna. Within the antenna, it has been paired with Alq3 because the significant overlap between the emission spectrum of Alq3 and the absorption spectrum of DCM means that energy transfer between these molecules during FRET will be efficient [22].
2.2 Absorption spectra measurements
We first measured the absorption spectra of samples with different concentrations of DCM in Alq3. In this measurement, solution-based samples were prepared by dissolving materials into chloroform. The concentration of Alq3 in chloroform was set at 0.1 mg/mL, corresponding to a molar concentration of $0.218\times 10^{-5}$ mol/L. Then, different samples were prepared by varying the quantity of DCM added to the Alq3, resulting in different weight ratios between DCM and Alq3. These ratios include 1:20, 1:50, 1:100, 1:150, 1:200, 1:300, and 1:400, corresponding to the mole fraction of DCM in Alq3 as 7.04${\% }$, 2.93${\% }$ 1.49${\% }$ 1.00${\% }$, 0.75${\% }$ 0.50${\% }$, and 0.38${\% }$, respectively. As shown in Fig. 3(a), the absorption spectra were primarily influenced by Alq3, given its much higher concentration compared to DCM. However, at the higher concentrations, the DCM leads to the absorption of wavelengths longer than 450 nm. Once the ratio of DCM in Alq3 drops below 1:150, this additional absorbance from DCM becomes almost negligible.
Fig. 3. (a) The measured absorbance of the samples with different concentrations of DCM in Alq3, (b) the measured emission spectra of the thin film samples with different concentrations of DCM in Alq3, (c) a photograph of the thin film samples under illumination.
2.3 Emission spectra and PLQY measurements
To determine the impact of energy transfer, the emission spectra of the samples were also measured. Since energy transfer occurs effectively when the acceptor and donor molecules are closely spaced within the range of the Förster radius ($R_{0}$), we prepared thin film samples composed of Alq3 and DCM for this measurement. In the preparation of the thin film samples, chloroform was used as the solvent, and the concentration of Alq3 in chloroform was fixed at 10 mg/ml. Then the amount of DCM added to each sample was varied to achieve the same weight ratios as those used for absorption measurements, listed above. For reference, we also prepared a pure sample of Alq3. Subsequently, each of these samples was dropped onto a quartz substrate for spin coating. The spin coating process was carried out at a speed of 2000 RPM for 30 seconds. After spin coating, the samples were left on the holder for several minutes to allow the chloroform to evaporate. Figure 3(c) shows a photograph of the prepared thin film samples after spin coating under UV light.
Figure 3(b) shows the measured emission spectra of the thin film samples when the excitation light has a wavelength of 365 nm, which is also the peak wavelength of the light-emitting diode (LED) used in our OWC transmission measurements. It can be seen that an increasing concentration of DCM in Alq3 results in longer emission wavelengths. The results also show that the intensity of the light emitted from the thin film sample varies between samples. Since the intensity of the emitted light is directly related to the PLQY, we also measured the PLQY of the thin film samples to obtain the results shown in Fig. 4. It can be seen that the concentration of 1:200 (0.75 mol ${\% }$) gives the highest PLQY. This is consistent with the emission spectra measurement shown in Fig. 3(b), confirming that a higher PLQY corresponds to a higher intensity peak. As shown in these results, the PLQY initially increases as the DCM concentration increases, before decreasing at higher concentrations.
The initial increase of PLQY as the DCM concentration increases is consistent with an increase in the probability of energy transfer as the average distance between DCM and Alq3 molecules decreases. The decrease in PLQY at higher concentrations is then consistent with DCM molecules quenching each other as they get closer together, a phenomenon known as concentration quenching (CQ) [23]. To confirm that it is these processes that lead to the observed behaviour, we calculated various distances between DCM molecules and also between DCM and Alq3 molecules in the prepared thin film samples. The results, summarized in Table 1, indicate that, as the mole fraction of DCM changes from 7.04${\% }$ to 0.38${\% }$, the average distance between DCM molecules increases from 2.73 nm to 7.23 nm. Due to the uniform distribution of Alq3 molecules among DCM molecules, this means that the distance between DCM and Alq3 molecules changes from 1.37 nm to 3.62 nm. The CQ of DCM has been previously investigated in an inert matrix of poly(methyl methacrylate) (PMMA) [23]. The fluorescence decay rate ($k_{\text {fl}}$), the sum of non-radiative ($k_{\text {nr}}$) and concentration quenching rates ($k_{\text {CQ}}$) as a function of the DCM-DCM distance showed that in this system, the non-radiative decay of DCM significantly increases at DCM-DCM distances less than 7 nm [23]. This distance is close to that of the sample with the weight ratio of 1:300. This comparison confirms that the decrease in PLQY at higher DCM concentrations is due to CQ (see Fig. S4 in Supplement 1).
Table 1. Distances between DCM-DCM and Alq3-DCM molecules in thin films with different DCM concentrations, along with measured PLQY and PL lifetime.
Another quantitative comparison is to compare the average Alq3-DCM distance and the Förster transfer radius ($R_{0}$). The Förster transfer radius is defined as the average intermolecular distance at which the donor exciton (Alq3) has an equal probability of either decaying on the donor or non-radiatively transferring to the nearest acceptor (DCM) molecule. The $R_{0}$ of the Alq3-DCM system has been reported as 3.3 nm [24]. Since the average Alq3-DCM distance of the sample with the weight ratio of 1:300 coincides with $R_{0}$, a higher DCM ratio of more than 1:200 would be beneficial for efficient energy transfer. The maximum PLQY, therefore, occurs from a combination of increasing the efficiency of transfer from Alq3 to DCM molecules on one hand and CQ on the other hand.
Considering the case when the concentration of DCM in Alq3 is 1:200, Fig. 5 shows the normalized absorption and emission spectra. This is compared to both the Alq3 thin film sample and the DCM solution sample (0.1 mg/ml in chloroform). In the case of DCM, we considered a solution rather than a thin film. This is because DCM has a strong CQ effect, and consequently a film of DCM will not emit light. The results in Fig. 5 show that the absorption spectra of Alq3 and Alq3+DCM are nearly identical, suggesting light, including the 365 nm excitation wavelength used in other experiments, is mainly absorbed by Alq3. However, their emission spectra are markedly different. In the Alq3+DCM sample, the emission is predominantly from DCM. This shows that energy transfer has occurred. However, the Alq3+DCM emission spectrum isn’t the same as the DCM emission spectra because the Alq3+DCM sample is a thin film, whilst the DCM sample had to be a solution. The difference in matrix polarity of the two samples then results in different emission spectra [25].
Fig. 5. The measured absorption and emission spectra using the Alq3 sample and using the Alq3+DCM sample.
2.4 PL lifetime measurements
Another important parameter for a fluorescent antenna is its PL lifetime, $\tau$, which is directly related to the 3 dB bandwidth of the antenna, $f_{\text {3dB}}$, in particular [26]
Figure 6 shows the PL decay data of different samples, measured using a streak camera system (Hamamatsu, C4780). The corresponding PL lifetimes are shown in Fig. 4 and Fig. 6. It can be seen that the PL lifetime of Alq3 is 9.31 ns, corresponding to a 3 dB bandwidth of 17 MHz. We can also see that the PL lifetime is reduced when DCM is added to Alq3. Furthermore, the results indicate a direct correlation between the PL lifetime and the quantity of DCM mixed with Alq3. A higher concentration of DCM leads to a shorter PL lifetime. At lower concentrations of DCM, this is because DCM has a shorter PL lifetime than Alq3. When energy transfer occurs, the light is emitted by DCM, and consequently, the PL lifetime becomes shorter. However, as discussed previously, as the concentration of DCM increases CQ becomes important. This reduces the PL lifetime further, but at the cost of also reducing the PLQY. Since the concentration ratio of 1:200 yields the highest PLQY, antennas were made using this weight ratio. The results in Fig. 6, show that this means that these antennas have a PL lifetime is 3.52 ns, and therefore, a 3 dB bandwidth of 45 MHz.3. Fluorescent antenna and the transmission system
3.1 Fluorescent antenna structure
Figure 7 illustrates the structure of the fluorescent antenna, which includes both the waveguide layer and the fluorescent layer. In this work, we used a glass substrate with a length of 75 mm, a width of 8 mm and a thickness of 1.1 mm to form the waveguide layer. Then, to create the fluorescent layer, the prepared solution with a weight ratio of 1:200 between DCM and Alq3 was applied to the surface of the glass substrate through drop casting. After the solvent evaporated this resulted in the creation of a thin film fluorescent layer. Also, we have coated both sides of the glass substrate to form two fluorescent layers so that the light which is not absorbed by the top layer can be absorbed by the bottom layer. As illustrated in Fig. 7, when incident light arrives at the fluorescent layer, if its wavelength falls within the absorption range of Alq3, it can be absorbed. A short while later some of this absorbed energy results in the emission of photons from DCM with longer wavelengths. Using this structure, although many photons would be emitted into the air due to the random angles of photon emission, a large number of photons are still being trapped within the glass and guided towards the antenna edge, where a photodetector is positioned. This antenna is relatively simple to manufacture. However, since the antenna’s perimeter is much larger than the photodetector only a fraction of the guided light is detected. Despite this inefficiency, the use of the fluorescent antenna can increase the intensity of light reaching the photodetector.
3.2 OWC transmission system
Figure 8 shows the block diagram of our experimental setup also shown in the photograph in Fig. 9(a). A photograph of the two fabricated fluorescent antennas is shown in Fig. 9(b). In this experimental setup, an arbitrary waveform generator (AWG, Siglent SDG2082X) was used to output the transmitted signal, which was subsequently amplified through a power amplifier (Mini-circuits, ZHL-32A-S). The amplified signal was superimposed onto a DC current via a bias-T (Mini-circuits, ZFBT-4R2GW) to drive a 365 nm LED transmitter (Thorlabs, M365D2). A detailed discussion on the selection of this wavelength is provided in Supplement 1. At the receiver side, our fluorescent antenna collected the light and guided it to a commercial avalanche photodiode (APD, Thorlabs, APD130A). The APD output was then captured using a digital storage oscilloscope (DSO, LeCroy, 204Xi-A).
Fig. 9. (a) A photograph of the experiment setup. The blue arrow indicates the direction of the incident light. (b) A photograph of both the non-FRET antenna, made of only Alq3, and the new FRET antenna, made of Alq3+DCM.
In bandwidth measurements, sinusoidal waves with varying frequencies were generated by the AWG, and the peak-to-peak voltage detected by the DSO was measured. In data transmission measurements, we considered orthogonal frequency division multiplexing (OFDM) modulation. This signal was generated offline using MATLAB, with the main signal processing steps shown in Fig. 8. To create an OFDM signal suitable for intensity modulation, we initially mapped the binary data sequence onto complex 4-quadrature amplitude modulation (QAM) symbols. These QAM symbols were then constrained to have Hermitian symmetry, ensuring that the output time domain signal sequence from the inverse fast Fourier transform (IFFT) contained only real numbers suitable for directly driving the LED transmitter. Subsequently, the signal sequence underwent parallel-to-serial (P/S) conversion, a cyclic prefix (CP) was added, and the signal was clipped at both its top and bottom to address the high peak-to-average power ratio (PAPR) common in OFDM signals with a Gaussian distribution. In our case, the clipping level was fixed at 10 dB, which means the clipping level was 3.16 times larger than the standard deviation of the Gaussian distribution. This allowed the removal of obvious signal peaks while maintaining negligible distortion in the clipped signal [27]. At the receiver side, to recover the transmitted data, we performed OFDM demodulation. This involved transforming the received signal into the frequency domain using a fast Fourier transform (FFT) and applying single-tap equalizers for individual subcarriers. After equalization, the received QAM symbols were converted to binary bits based on the maximum likelihood (ML) detection principle. Finally, we compared the transmitted bit sequence with the received bit sequence to obtain bit error rates (BERs) for evaluating the performance of the transmission link.
4. Communication measurement
4.1 Bandwidth and concentration gain
Figure 10(a) shows the measured frequency response of the system with and without using the fluorescent antenna. These results show that the 3 dB bandwidth of the system without an antenna is 7.5 MHz. Since the 3 dB bandwidth of the APD is 50 MHz, this means that, like other LED-based OWC systems, the bandwidth of this system is limited by the LED. The results also show that, since both antennas have a bandwidth that is much higher than 7.5 MHz, they don’t reduce the system’s 3 dB bandwidth. However, at frequencies higher than the 3 dB frequency of the system the antennas slightly reduce the systems response.
Fig. 10. The measured frequency response without antenna and with two different types of antenna, (a) with the y-axis being the attenuation (in dB), (b) with the y-axis being the measured peak-to-peak voltage (in mV).
Figure 10(b) shows raw data used to generate Fig. 10(a). Since this data is the measured peak-to-peak voltage it can be used to determine the signal gain obtained by using either antenna. In particular, this data shows that the use of the non-FRET fluorescent antenna increases the signal by 1.3 times. However, this is significantly less than the gain obtained when using the FRET-based antenna, which is 6.5. Most of the increase in gain obtained by using FRET arises from the increase in PLQY. However, the change in the spectrum of the emitted light also both reduces self-absorption losses and means that the APD detects the emitted light more efficiently. Together these two processes contribute approximately 33${\% }$ of the increased gain obtained by using FRET.
4.2 Data transmission measurements
Figure 11(a) show the measured BERs when the transmission distance is 20 cm, without an antenna, and with both the non-FRET antenna and the FRET-based antenna. In this measurement, the transmission data rate was varied from low to high by adjusting the transmission sampling rate. It can be seen that the BER increases as the data rate increases. By considering the forward error correction (FEC) limit of $3.8 \times 10^{-3}$, when no antenna is used, the best transmission data rate is only 7 Mbps. This can be increased to 14 Mbps using the non-FRET antenna and 50 Mbps using the FRET-based antenna. This significant improvement achieved using the FRET-based antenna is attributed to its higher gain. Figures 11(b)-(d) shows the received QAM constellations when the transmission data rate is 20 Mbps. It can be seen that the use of the new FRET-based antenna results in a much clearer constellation distribution, which is associated with a BER of $0$.
Fig. 11. (a) The measured BERs when the transmission distance is 20 cm, (b) the received QAM constellations without antenna when the transmission data rate is 20 Mbps, (c) the received QAM constellations using the non-FRET antenna when the transmission data rate is 20 Mbps, (d) the received QAM constellations using the FRET-based antenna when the transmission data rate is 20 Mbps.
To investigate how the bandwidth affects the transmission performance the distance between the transmitter and receiver was varied and the received signal strength was quantified using the measured peak-to-peak voltage when a 1 MHz sine was transmitted. Figure 12 shows the data rate corresponding to a BER of $3.8 \times 10^{-3}$, for various received signal strengths. The results in this figure show that as expected the achievable data rate increases as the signal strength increases. However, as the signal strength becomes relatively high, increasing it only results in a slight increase in the data rate. Moreover, it can be seen that when the signal strength is the same, the case when no antenna is used supports the best data rate. This is because the use of a fluorescent antenna reduces the bandwidth of the system due to the PL lifetime of the fluorophore. However, when comparing the non-FRET antenna with the FRET-based antenna, the FRET-based antenna supports a higher data rate. This is because, as shown in Fig. 6, the FRET-based sample has a shorter PL lifetime due to the use of FRET. In Fig. 12, we also show the obtained result when no antenna is used and the transmission distance is decreased to the point where the peak-to-peak voltage of the APD output reaches 1800 mV, which corresponds to the saturation voltage of the considered APD. At this voltage level, the associated data rate is 68 Mbps, which is considered as the best transmission data rate that can be achieved using the considered LED.
Fig. 12. The data rate at which the BER equals the FEC limit for different received signal strengths.
More importantly, the results in Fig. 12 show that the gain of the FRET antenna means that it achieves the same receiver output voltages at significantly longer transmission distances than the other two receivers. For example, when the voltage is approximately 600 mV, without an antenna or with the non-FRET antenna, the transmission distance is only 10 cm. In contrast, the transmission distance when using the FRET-based antenna is 30 cm. This means that in a particular scenario, the system that incorporates the FRET-based antenna will perform significantly better than the other two systems.
5. Conclusion
In this paper, we have reported the results of an investigation into the use of FRET-based fluorescent antennas in OWC. The results demonstrate that the use of FRET can enhance all aspects of the performance of a fluorescent antenna. In particular, when Alq3 is used as the energy donor and DCM as the energy acceptor, FRET reduces the PL lifetime from 9.3 ns to 3.5 ns. This means that FRET significantly increases the transmission bandwidth of the fluorescent antenna. In addition, the use of FRET increases the PLQY from 0.16 to 0.53. Moreover, a shift in the peak of the emission spectrum of the antenna from 525 nm to 610 nm has two effects: it reduces the losses due to self-absorption and increases the efficiency with which the emitted light is detected by a Si photodetector. The combination of a high PLQY and much longer emission wavelengths results in an antenna concentration gain improving from only 1.3 to 6.5. The result is a significant improvement in the SNR at the output of an OWC receiver. Results have been presented which show that the improvements arising from using a FRET-based antenna significantly increase the transmission data rate from only 7 Mbps to 50 Mbps.
Funding
This work was supported by Japan Society for the Promotion of Science (JSPS) KAKENHI (JP 23K13332, JP 23KK0257).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
Supplemental document
See Supplement 1 for supporting content.
References
1. Z. Wei, Z. Wang, J. Zhang, et al., “Evolution of optical wireless communication for b5g/6g,” Prog. Quantum Electron. 83, 100398 (2022). [CrossRef]
2. D. O’Brien, S. Rajbhandari, and H. Chun, “Transmitter and receiver technologies for optical wireless,” Phil. Trans. R. Soc. A 378(2169), 20190182 (2020). [CrossRef]
3. S. Collins, D. C. O’Brien, and A. Watt, “High gain, wide field of view concentrator for optical communications,” Opt. Lett. 39(7), 1756–1759 (2014). [CrossRef]
4. P. P. Manousiadis, S. Rajbhandari, R. Mulyawan, et al., “Wide field-of-view fluorescent antenna for visible light communications beyond the étendue limit,” Optica 3(7), 702–706 (2016). [CrossRef]
5. Y. Dong, M. Shi, X. Yang, et al., “Nanopatterned luminescent concentrators for visible light communications,” Opt. Express 25(18), 21926–21934 (2017). [CrossRef]
6. M. A. Umair, M. Seminara, M. Meucci, et al., “Long-range optical wireless communication system based on a large-area, q-dots fluorescent antenna,” Laser Photonics Rev. 17(2), 2200575 (2023). [CrossRef]
7. M. Portnoi, P. A. Haigh, T. J. Macdonald, et al., “Bandwidth limits of luminescent solar concentrators as detectors in free-space optical communication systems,” Light: Sci. Appl. 10(1), 3–12 (2021). [CrossRef]
8. S. Alshaibani, O. Alkhazragi, I. Ashry, et al., “Wide-field-of-view optical detectors for deep ultraviolet light communication using all-inorganic CsPbBr3 perovskite nanocrystals,” Opt. Express 31(16), 25385–25397 (2023). [CrossRef]
9. Y. Wang, J.-X. Wang, O. Alkhazragi, et al., “Multifunctional difluoroboron β-diketonate-based luminescent receiver for a high-speed underwater wireless optical communication system,” Opt. Express 31(20), 32516–32528 (2023). [CrossRef]
10. C. He, S. Collins, and H. Murata, “Capillary-based fluorescent antenna for visible light communications,” Opt. Express 31(11), 17716–17730 (2023). [CrossRef]
11. C. H. Kang, O. Alkhazragi, L. Sinatra, et al., “All-inorganic halide-perovskite polymer-fiber-photodetector for high-speed optical wireless communication,” Opt. Express 30(6), 9823–9840 (2022). [CrossRef]
12. C. H. Kang, A. Trichili, O. Alkhazragi, et al., “Ultraviolet-to-blue color-converting scintillating-fibers photoreceiver for 375-nm laser-based underwater wireless optical communication,” Opt. Express 27(21), 30450–30461 (2019). [CrossRef]
13. M. Sait, A. Trichili, O. Alkhazragi, et al., “Dual-wavelength luminescent fibers receiver for wide field-of-view, gb/s underwater optical wireless communication,” Opt. Express 29(23), 38014–38026 (2021). [CrossRef]
14. C. He, Y. Lim, and H. Murata, “Study of using different colors of fluorescent fibers as optical antennas in white led based-visible light communications,” Opt. Express 31(3), 4015–4028 (2023). [CrossRef]
15. C. He and S. Collins, “A two-stage fluorescent antenna for visible light communication uplinks,” IEEE Photonics Technology Letters (2023).
16. C. He, Y. Tang, C. Chen, et al., “Wavelength division multiplexing in visible light communications using fluorescent fiber antennas,” Journal of Lightwave Technology (2024).
17. T. Förster, “Transfer mechanisms of electronic excitation energy,” Discuss. Faraday Soc. 27(0), 7–17 (1959). [CrossRef]
18. M. J. Currie, J. K. Mapel, T. D. Heidel, et al., “High-efficiency organic solar concentrators for photovoltaics,” Science 321(5886), 226–228 (2008). [CrossRef]
19. B. Zhang, G. Lyu, E. A. Kelly, et al., “Förster resonance energy transfer in luminescent solar concentrators,” Adv. Sci. 9(23), 2201160 (2022). [CrossRef]
20. M. Kuscu, A. Kiraz, and O. B. Akan, “Fluorescent molecules as transceiver nanoantennas: The first practical and high-rate information transfer over a nanoscale communication channel based on fret,” Sci. Rep. 5(1), 7831 (2015). [CrossRef]
21. P. P. Manousiadis, H. Chun, S. Rajbhandari, et al., “Optical antennas for wavelength division multiplexing in visible light communications beyond the étendue limit,” Adv. Opt. Mater. 8(4), 1901139 (2020). [CrossRef]
22. M. Pope and C. E. Swenberg, “Electronic processes in organic crystals,” Oxford University Press (1982).
23. A. P. Green and A. R. Buckley, “Solid state concentration quenching of organic fluorophores in PMMA,” Phys. Chem. Chem. Phys. 17(2), 1435–1440 (2015). [CrossRef]
24. Y. Wu, H. Wu, Y. Zhou, et al., “Excitation energy transfer between tris-(8-hydroxyquinoline) aluminum and a red dye,” Appl. Phys. Lett. 88(12), 123512 (2006). [CrossRef]
25. V. Bulović, A. Shoustikov, M. Baldo, et al., “Bright, saturated, red-to-yellow organic light-emitting devices based on polarization-induced spectral shifts,” Chem. Phys. Lett. 287(3-4), 455–460 (1998). [CrossRef]
26. T. Peyronel, K. Quirk, S. Wang, et al., “Luminescent detector for free-space optical communication,” Optica 3(7), 787–792 (2016). [CrossRef]
27. C. He and J. Armstrong, “Clipping noise mitigation in optical OFDM systems,” IEEE Commun. Lett. 21(3), 548–551 (2017). [CrossRef]
