Fluorescent Quantum Dots Make Feasible Long ... - ACS Publications

Aug 2, 2017 - Few seminal examples already showed the feasibility of molecular bits;(8-14) however, as the transfer of the messengers occurs via relea...
0 downloads 0 Views 2MB Size
Letter pubs.acs.org/JPCL

Fluorescent Quantum Dots Make Feasible Long-Range Transmission of Molecular Bits Nunzio Tuccitto, Giovanni Li-Destri,* Grazia M. L. Messina, and Giovanni Marletta Laboratory for Molecular Surfaces and Nanotechnology (LAMSUN), Department of Chemical Science, University of Catania and CSGI, viale A. Doria 6, 95125 Catania, Italy S Supporting Information *

ABSTRACT: The modeling and realization of an effective communication platform for long-range information transfer is reported. Messages are encrypted in molecular bits by concentration pulses of fluorescent carbon quantum dots having self-quenching emission that dynamically depends on the concentration pulses. Messages are transferred along longer paths when received and decoded by means of dynamical emission response with respect to the ones encoded by absorbance scaling linearly with messenger concentration. These results represent a significant breakthrough in view of the futuristic development of a nonspecific molecular communication platform to encode and transfer information in multiple fluid environments, ranging from physiological to industrial ones.

occurs via release and transport in a fluid, they are subject along their path to hydrodynamic dispersion, which causes an exponential decrease of the transmission efficiency with the distance from the source.15 This unavoidable effect has so far limited diffusion-based and flow-based nonspecific molecular communication to the short and medium ranges,16 for example, to travel paths no longer than a few millimeters, while attempts to reach a communication range of meters are still based on the use of specific molecular messengers.17 Very recently, it has been shown that the transmission efficiency can be greatly increased by employing a nonlinear molecular signal generated by chemical reactions.14 However, in ref 14, the signal decoding requires complex machine-learning algorithms, taking into account the fact that the intensity of each bit is affected by the previous ones. The present paper, instead, for the first time modeled and developed an experimental platform for an innovative and simple long-range communication protocol, based on the use of concentration pulses of fluorescent carbon quantum dots (CQDs) to encode messages in binary code. Given the characteristic self-quenching behavior of CQDs, we prove that fluorescence intensity does not scale linearly with concentration and, as a consequence, leads to lower bit broadening and weakening with travel distance. This nonlinear molecular signal allows one to store and transfer the information along longer distances without the need for specific decoding algorithms. Encoding, emission, propagation, and decoding of the concentration pulses are the keys determining the efficiency

he storage and transport of information through fluids is at the basis of the processes controlling biochemical cycles1 and several functions of cells, tissues, and organs,2 which are generally activated by the many chemical and electrical communication processes that nature has developed. A similar approach can be used in theranostics as the ability reached to miniaturize devices down to the micro- and nanoscale may allow development of miniaturized implanted biosensors for physiological status monitor of patients,3 implanted biochips used to retrieve medical information from the body,4 and microactuators allowing in situ controlled drug delivery.5 This requires a proper communication platform. Accordingly, several chemical messengers have been synthesized and shown to efficiently and rapidly store and transfer information.6,7 However, in this strategy, efficient communication is only possible with highly specific messengers and receptors. In other words, current messenger/receptor couples are only able to transfer one piece of information per couple, that is, the mere presence or absence of the messenger itself and the many pieces of information required for the proper functioning of a living body are stored and transferred by thousands of chemical messengers developed by nature. Alternatively, by drawing inspiration from well-established electronic communication protocols, information can be more efficiently transmitted by means of nonspecific messenger concentration pulses, decoded by the receptor as 1-bit, the presence (high concentration) of the messenger, or 0-bit, the absence (low concentration), that is, a “molecular bit” approach. This approach, if n bits are used, allows one to encode and transfer 2n different pieces of information (see Scheme 1). Few seminal examples already showed the feasibility of molecular bits;8−14 however, as the transfer of the messengers

T

© 2017 American Chemical Society

Received: July 4, 2017 Accepted: August 2, 2017 Published: August 2, 2017 3861

DOI: 10.1021/acs.jpclett.7b01713 J. Phys. Chem. Lett. 2017, 8, 3861−3866

Letter

The Journal of Physical Chemistry Letters Scheme 1. Schematic Representation of the Molecular Communication Approach

messenger concentration the signal must be low, undergoing amplification at intermediate dilutions, and after reaching a maximum at a critical concentration, it must exponentially decay with further dilution. This approach fulfills both requirements of an optimal box-shaped pulse maintaining and “first hitting” detection protocol. The simulated signal profile of a single pulse of this molecular messenger is reported in Figure 1b (blue trace). It is evident that, by using this nonlinear kernel, at a given dispersion-induced pulse broadening, the profile of the signal originating from the nonlinear response is closer to the initial box shape of the pulse. As a result, binary signals carried out by messengers having nonlinear behavior show, at a given detection distance, a higher ratio between 1 and 0 pulses and, therefore, are expected to be detectable at longer paths, as shown in Figure 1d. The ratio between maxima and minima as a function of the travel path is reported in Figure 1e for both linear and nonlinear responses. In both cases, an exponential decrease of the signal is observed; however, if we impose, as a detection threshold, a typical acceptable experimental ratio (vide infra) between maxima and minima (e.g., 0.2), we observe that the range of detectability increases by 1 order of magnitude in the case of molecular messengers having nonlinear response. The second parameter affecting the efficient decoding of the signal with distance is the fwhm of the pulses. Indeed, in the case of 1-bit pulses separated by 0-bit ones, the broadening with time causes their overlap when the fwhm exceeds half of the width of the 0-bits. Moreover, the fwhm determines the temporal and spatial resolution of the process. Again, the nonlinear approach here employed allows one to face the problem, as long as the fwhm at increasing detection distances is kept constant, as shown in Figure 1f, where the fwhm as a function of the time is reported for molecular messengers having linear and nonlinear response. In particular, it can be seen that, although an initial stronger broadening of the pulses occurs for nonlinear response, the saturation of the signals at intermediate concentration causes a reasonably long travel path with constant fwhm, that is, a significantly smoother increase with respect to the linear response. It should be noted that the initial fwhm broadening can be easily overcome with the appropriate choice of the width of the 0-bit pulse. In addition, interestingly, both the initial fwhm increase and the length of the plateau are strongly influenced by the dispersion coefficients of the molecular messenger. Figure S2 shows that by increasing the hydrodynamic diffusion coefficient D, for example, the fluid turbulence, we observe larger broadening of the pulses and shorter plateaus. Thus, it is evident that molecular messengers having nonlinear response provide, especially in the case of long-range communication, more efficiently detected pulses, thanks to the higher amplitude and the reduced fwhm broadening. However, detailed character-

of data transfer; thus, a rigorous and quantitative description of their mechanism is essential to properly tailor the properties of each component and obtain the required functions. The partial differential equation describing the evolution of the concentration profile of a molecular messenger C(x,t) along the traveling distance x over the time t is ∂C /∂t = ∇·(D∇C) − ∇·(vC)

(1)

where v is the speed of the carrier fluid and D (corresponding to the hydrodynamic dispersion coefficient) is a function describing the dispersion of the chemical signals along the path.18 In the case of a sequence of binary pulses traveling at constant speed through a fluid from the source to receivers located at various downstream sites, eq 1 allows one to describe the intensity and shape evolution with distance of the pulses. When moving away from the source, the initial box-shaped pulse traveling toward the receivers readily adopts a Gaussian shape as a consequence of the hydrodynamic dispersion phenomena (proportional to D). The pulse broadening causes fast lowering of the signal intensity with an exponential decreasing trend (Figure S1). Typically, detectors engineered to decode chemical signal, for example, resistive, electrochemical, and optical detectors, show a linear response with concentration of the chemical messenger, as shown in Figure 1a (green trace), and saturation above a threshold concentration. As the single pulse broadens during dispersion through the liquid medium, the detected signal broadens accordingly (Figure 1b green and red curves). As a consequence, if we simulated a multibit sequence (1010101) propagating on a confined pathway (travel velocity: 1 mm/s; hydrodynamic dispersion coefficient: 0.15 mm2/s mimicking blood fluid19,20), we confirmed that the recorded signal gets broader the longer the travel path, and more importantly, we determined that the intensity difference between maxima and minima, for example, between 1- and 0-bits, weakens (Figure 1c). Thus, there will be a threshold path length above which the decoding of the binary information is no longer possible. To solve this problem, that is, to increase the range of data transfer, the decodable signal generated by molecular messengers must not scale linearly with concentration but, preferably, follow a nonlinear trend characterized by a maximum of intensity at low concentrations. In this way, the box-like shape of the signal pulse can be preserved for longer ranges. In addition, in the case of a closedloop flow, as in the body fluids, after the detection step, the multiple detection must be avoided, which is known as the “first hitting method”,8 so that the signal must be removed from the circuit. We have chosen to simulate the data transfer by means of a molecular messenger providing a decodable signal with a nonlinear response to concentration (Figure 1a red trace). In particular, our approach forecasts that at high molecular 3862

DOI: 10.1021/acs.jpclett.7b01713 J. Phys. Chem. Lett. 2017, 8, 3861−3866

Letter

The Journal of Physical Chemistry Letters

Figure 1. (a) Simulated linear and nonlinear detector responses. (b) Simulated broadening of box-shaped signal induced by diffusivity detected by using linear and nonlinear responses. (c) Simulated advection−diffusion multibit sequence (1010101) detected by linear response. (D) Simulated advection−diffusion multibit sequence (1010101) detected by nonlinear response. (e) Trend of maximum to minimum amplitude as a function of traveling distance referred to (c) and (d) simulations. (f) The fwhm as a function of traveling distance referred to (c) and (d) simulations.

ization of the dispersion coefficients and flow velocity is required to optimize the communication system, especially with respect to the width of the 0-bit detection and the efficient transmission distance. To convert into reality the requirements obtained from simulations, a molecular messenger providing a detectable signal with a tailored nonlinear response with concentration is required. In this view, fluorescence self-quenching systems, providing a characteristic nonlinear emission intensity with concentration, are ideal candidates. In particular, CQDs21,22 show high quantum yield fluorescence emission in the visible region if excited in the ultraviolet region of the spectrum23 and

also self-quenching due to particle aggregation leading to nonradiative decay processes.24,25 Accordingly, we have successfully synthesized and characterized CQDs obtained by decomposition through hydrothermal treatment of citric acid (see the Supporting Information for a detailed description of the synthesis and characterization). Then, to foster the aggregation, and in turn the self-quenching, we selected the fraction with higher size (average diameter of 4 nm) as this provides the highest interfacial area and, in turn, the strongest interparticle attraction, that is, the tendency to aggregate. In Figure 2a, the dependence of fluorescence intensity from CQD concentration is reported; as expected, the highest intensity 3863

DOI: 10.1021/acs.jpclett.7b01713 J. Phys. Chem. Lett. 2017, 8, 3861−3866

Letter

The Journal of Physical Chemistry Letters

Figure 2. (a) Fluorescence spectra (λecc = 350 nm) of CQDs acquired at several percentages of dilutions. (b) Emission intensity (λemi = 450 nm) and absorbance at 350 nm as a function of CQD concentration.

consequence, making it difficult to decode already at 30 cm. On the contrary, the box-shaped bits generated by the nonlinear response are characterized, at any travel path, by a higher signal-to-noise ratio, which easily allows detection even at the longest tested distances. It is also worth observing that at intermediate travel paths the absorbance signal greatly weakens while the emission increases. This is due to the characteristic self-quenching of CQDs due to the obvious dilution effect, ensuring higher emission yields at intermediate dilutions (Figure 3b). Also, the fwhm of the experimental absorbance and emission follows the simulation prediction. Although the detection of absorbance and fluorescence at 0 cm was hampered by the microvalves, the time control on the pulse release permits a reasonably assumption that the initial fwhm is 4 ± 1 s. After the 10 cm path, the fwhm increases to 14 ± 1 s for absorbance peaks and to 31 ± 8 s for emission ones. At longer distances, the fwhm of absorbance pulses increases up to 21 ± 1 s, while the one for emission is constant. This agrees perfectly with the simulation previously described (Figure 1f), where an exponential increase of the fwhm is expected for responses linear to concentration while a steeper increase followed by a long plateau occurs in the case of nonlinear behavior. This steeper increase can hamper the bit decoding if the distance between two consecutive maxima, for example, the 0-bit signal, is not high enough to prevent overlapping. Thus, we have optimized the communication setup by adopting a 0-bit width of 40 s, which, with the highest fwhm being around 30 s, allows the decoding at any travel distance. In conclusion, we have demonstrated a novel communication platform tuned to transfer complex binary codes along the cm− dm scale in flowing fluids, taking profit of the peculiar nonlinear response of on purpose synthesized self-quenching CQDs, whose nonlinear fluorescence properties are dynamically optimized with concentration. Accordingly, we have realized a prototypal molecular communication system and compared its information transfer efficiency respectively for the linear (adsorbance) and nonlinear (fluorescence) detection of CQD pulses. The results showed that nonlinear response allows communication along significantly longer paths. The reported preliminary results suggest that the present approach is very promising with respect to nonspecific molecular communication processes in very different flowing environments, where efficient communication between net-

occurs at intermediate CQD concentrations, while higher concentrations induce self-quenching. Figure 2b shows a comparison between the absorbance, which follows the characteristic linear Lambert−Beer trend, and emission responses of various aqueous solutions of CQDs. From this comparison, it is evident that the fluorescence response to concentration is very close to the one described in the simulation. This confirms that CQDs are good candidates for molecular messengers in long-range systems. It must be noted that CQD fluorescence emission decreases when using aqueous salt solutions and, therefore, CQDs would require specific functionalization to be employed as molecular messengers in real fluid environments. However, this goes beyond the scope of this Letter, which is to show that a tailored nonlinear molecular response greatly increases the range of applicability of molecular messengers. The prototypal molecular communication system herein employed is outlined in Scheme 2. It consists of in vitro Scheme 2. Schematic Representation of the Proposed Communication Platforma

a

Messenger pulses are injected during the encoding, transmitted by the flowing fluid, detected downstream by a proper detector, and finally erased by further dilution.

phantom vascular flow with optical detectors placed downstream at several distances from the source (10, 20, and 50 cm) to detect the binary code as a function of absorbance and fluorescence emission (Figure 3). The shapes of these bits closely resemble the simulated ones, being Gaussian in the case of (linear) absorbance and box-shaped with a local minimum in the case of (nonlinear) emission. In both cases, the time distance between the two 1-bit signals is constant (41 ± 4 s) and unaltered by the traveled path. As expected, the intensity of absorbance rapidly decreases with the travel path, causing a strong increase of the signal-to-noise ratio and, as a 3864

DOI: 10.1021/acs.jpclett.7b01713 J. Phys. Chem. Lett. 2017, 8, 3861−3866

Letter

The Journal of Physical Chemistry Letters

Figure 3. Experimental molecular communication performed by using linear (a) and nonlinear (b) response detectors as a function of travel distance. Dashed lines are for eye guidance.



worked devices can pave the way to real-time targeted information retrieval. Furthermore, due to the high versatility of the proposed protocol, not requiring any specific messengers/receiver coupling, that is, not requiring any specific encoding limitation, we can forecast fruitful applications in as different fields as smart drug delivery processes and targeted industrial fluid treatments.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.7b01713. Materials and methods details, supplementary results and discussion, including figures of simulated broadening of the box-shaped signal induced by diffusivity, the fwhm as a function of the travel distance, and spectroscopic and AFM characterization, and additional references (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Nunzio Tuccitto: 0000-0003-4129-0406 Giovanni Li-Destri: 0000-0001-6195-659X Notes

The authors declare no competing financial interest.



REFERENCES

(1) Youk, H.; Lim, W. A. Secreting and Sensing the Same Molecule Allows Cells to Achieve Versatile Social Behaviors. Science 2014, 343, 1242782. (2) Malak, D.; Akan, O. B. Molecular Communication Nanonetworks Inside Human Body. Nano Commun. Netw. 2012, 3, 19−35. (3) Schwiebert, L.; Gupta, S. K. S.; Weinmann, J. Research Challenges in Wireless Networks of Biomedical Sensors. Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM 2001, 151−165. (4) Umay, I.; Fidan, B.; Barshan, B. Localization and Tracking of Implantable Biomedical Sensors. Sensors 2017, 17, 583. (5) Saltzman, W. M.; Olbricht, W. L. Building Drug Delivery into Tissue Engineering. Nature Rev. Drug. Discovery 2002, 1, 177−186. (6) Giménez, C.; Climent, E.; Aznar, E.; Martínez-Máñez, R.; Sancenón, F.; Marcos, M. D.; Amorós, P.; Rurack, K. Towards Chemical Communication between Gated Nanoparticles. Angew. Chem., Int. Ed. 2014, 53, 12629−12633. (7) Pramanik, S.; De, S.; Schmittel, M. Bidirectional Chemical Communication between Nanomechanical Switches. Angew. Chem., Int. Ed. 2014, 53, 4709−4713. (8) Farsad, N.; Yilmaz, H. B.; Eckford, A.; Chae, C. B.; Guo, W. A Comprehensive Survey of Recent Advancements in Molecular Communication. IEEE Commun. Surv. Tutorials 2016, 18, 1887−1919. (9) Parcerisa Giné, L.; Akyildiz, I. F. Molecular Communication Options for Long Range Nanonetworks. Comput. Netw. 2009, 53, 2753−2766. (10) Nakano, T.; Moore, M. J.; Wei, F.; Vasilakos, A. V.; Shuai, J. Molecular Communication and Networking: Opportunities and Challenges. IEEE T. Nanobiosci. 2012, 11, 135−148. (11) Hiyama, S.; Moritani, Y. Molecular Communication: Harnessing Biochemical Materials to Engineer Biomimetic Communication Systems. Nano Commun. Netw. 2010, 1, 20−30. (12) Kuran, M. S.; Yilmaz, H. B.; Tugcu, T.; Ö zerman, B. Energy Model for Communication via Diffusion in Nanonetworks. Nano Commun. Netw. 2010, 1, 86−95. (13) Farsad, N.; Guo, W.; Eckford, A. W. Tabletop Molecular Communication: Text Messages through Chemical Signals. PLoS One 2013, 8, e82935.

ACKNOWLEDGMENTS

The authors gratefully acknowledge financial support from the Program FIR 2014 (University of Catania, Italy) and the Project FIRB “Accordi di Programma” (MIUR, Rome, Italy), Contract No. Rbap11zjfa_002. 3865

DOI: 10.1021/acs.jpclett.7b01713 J. Phys. Chem. Lett. 2017, 8, 3861−3866

Letter

The Journal of Physical Chemistry Letters (14) Farsad, N.; Pan, D.; Goldsmith, A. A Novel Experimental Platform for In-Vessel Multi-Chemical Molecular Communications. arXiv:1704.04810, 2017. (15) Bicen, A. O.; Akyildiz, I. F. System-Theoretic Analysis and Least-Squares Design of Microfluidic Channels for Flow-Induced Molecular Communication. IEEE T. Signal Proces. 2013, 61, 5000− 5013. (16) Unluturk, B. D.; Akyildiz, I. F. An End-to-End Model of Plant Pheromone Channel for Long Range Molecular Communication. IEEE T. Nanobiosci. 2017, 16, 11−20. (17) Bresler, E. Simultaneous Transport of Solutes and Water under Transient Unsaturated Flow Conditions. Water Resour. Res. 1973, 9, 975−986. (18) Marieb, E. N.; Hoehn, K. The Cardiovascular System: Blood Vessels. Human Anatomy & Physiology, 9th ed.; Pearson Education, 2013. (19) Liu, M.; Nicholson, J. K.; Parkinson, J. A.; Lindon, J. C. Measurement of Biomolecular Diffusion Coefficients in Blood Plasma Using Two-Dimensional 1H-1H Diffusion-Edited Total-Correlation NMR Spectroscopy. Anal. Chem. 1997, 69, 1504−1509. (20) Xu, X.; Ray, R.; Gu, Y.; Ploehn, H. J.; Gearheart, L.; Raker, K.; Scrivens, W. A. Electrophoretic Analysis and Purification of Fluorescent Single-Walled Carbon Nanotube Fragments. J. Am. Chem. Soc. 2004, 126, 12736−12737. (21) Baker, S. N.; Baker, G. A. Luminescent Carbon Nanodots: Emergent Nanolights. Angew. Chem., Int. Ed. 2010, 49, 6726−6744. (22) Du, F.; Jin, X.; Chen, J.; Hua, Y.; Cao, M.; Zhang, L.; Li, J.; Zhang, L.; Jin, J.; Wu, C.; Gong, A.; Xu, W.; Shao, Q.; Zhang, M. Nitrogen-doped Carbon Dots as Multifunctional Fluorescent Probes. J. Nanopart. Res. 2014, 16, 2720. (23) Kwon, W.; Do, S.; Lee, J.; Hwang, S.; Kim, J. K.; Rhee, S. W. Freestanding Luminescent Films of Nitrogen-Rich Carbon Nanodots toward Large-Scale Phosphor-Based White-Light-Emitting Devices. Chem. Mater. 2013, 25, 1893−1899. (24) Sharma, A.; Gadly, T.; Neogy, S.; Ghosh, S. K.; Kumbhakar, M. Molecular Origin and Self-Assembly of Fluorescent Carbon Nanodots in Polar Solvents. J. Phys. Chem. Lett. 2017, 8, 1044−1052. (25) Sharma, T.; Gadly, A.; Gupta, A.; Ballal, A.; Ghosh, S. K.; Kumbhakar, M. Origin of Excitation Dependent Fluorescence in Carbon Nanodots. J. Phys. Chem. Lett. 2016, 7, 3695−3702.

3866

DOI: 10.1021/acs.jpclett.7b01713 J. Phys. Chem. Lett. 2017, 8, 3861−3866