Mixed-matrix composite integrated fiber optic CO2 sensor

Abstract

Novel chemical sensors that improve detection and quantification of CO.sub.2 are critical to ensuring safe and cost-effective monitoring of carbon storage sites. Fiber optic (FO) based chemical sensor systems are promising field-deployable systems for real-time monitoring of CO.sub.2 in geological formations for long-range distributed sensing. In this work, a mixed-matrix composite integrated FO sensor system was developed that reliably operates as a detector for gas-phase and dissolved CO.sub.2. A mixed-matrix composite sensor coating on the FO sensor comprising plasmonic nanocrystals and zeolite embedded in a polymer matrix. The mixed-matrix composite FO sensor showed excellent reversibility/stability in a high humidity environment and sensitivity to gas-phase CO.sub.2 over a large concentration range. The sensor exhibited the ability to sense CO.sub.2 in the presence of other geologically relevant gases, which is of importance for applications in geological formations. A prototype FO sensor configuration which possesses a robust sensing capability for monitoring dissolved CO.sub.2 in natural water was demonstrated. Reproducibility was confirmed over many cycles, both in a laboratory setting and in the field.

Claims

1. An optical fiber comprising a coating composition disposed on the exterior of a glass fiber; and wherein the coating composition comprises an optical response enhancer and sorbent particles disposed in a polymer.

2. The optical fiber of claim 1 wherein the glass fiber is at least 90 wt % silica, or at least 99 wt % silica.

3. The optical fiber of claim 1 wherein the optical response enhancer is a plasmonic nanocrystal.

3. (canceled)

4. The optical fiber of claim 1 wherein the optical response enhancer comprises ITO.

5. (canceled)

6. The optical fiber of any of the above claims wherein the coating has a refractive index within the range of 1.30 to 1.70 or 1.30 to 1.60.

7. The optical fiber of claim 1 wherein the optical response enhancer is a light absorbing material that, when present in the polymer matrix exhibits at least two times greater absorption, or at least 5 times, or in the range of 2 to ten times or five to ten times greater absorption of the light reflected from the sensor coating at the measurement wavelength as compared to the absorption from the coating without enhancer and as compared to the coating in the absence of CO.sub.2 and CH.sub.4.

8. The optical fiber of claim 1 wherein the optical response enhancer comprises plasmonic nanocrystals having at least one dimension in the size range of 1 nm to 30 nm or 2 nm to 20 nm, or 5 to 20 nm, or 20 to 200 nm based on the smallest diameter of the particles.

9. (canceled)

10. The optical fiber of claim 1 wherein the length of the coating is 5 to 10 cm.

11. The optical fiber of claim 1 wherein the sorbent particles comprise a zeolite.

12. (canceled)

13. The optical fiber of claim 1 wherein the optical enhancer comprises plasmonic nanocrystals and wherein the sorbent mass average particle size of the sorbent is larger than the mass average particle size of the plasmonic nanocrystals.

14. (canceled)

15. The optical fiber of claim 1 wherein the polymer comprises a polysiloxane.

16. The optical fiber of claim 1 wherein the coating composition comprises 0.5-20 wt % or 1% to 20 wt %, or 2 to 10 wt % plasmonic nanocrystals; at least 5 wt % adsorbent, or 5 to 80 wt %, or 10 to 70 wt %, or 40 to 80 wt % adsorbent; and at least 10 wt % polymer, or 10 to 90 wt %, or 20 to 90 wt %, or 20 to 70 wt %, or 30 to 70 wt % polymer.

17. The optical fiber of claim 1 wherein the optical enhancer comprises plasmonic nanocrystals and wherein at least 50 mass % or at least 80 mass % of the nanocrystals have sizes in the range of 5 to 40 nm, or 5 to 25 nm, or 6 to 20 nm.

18. (canceled)

19. A coating composition comprising: plasmonic nanocrystals and sorbent particles disposed in a polymer.

20. The coating composition of claim 19 wherein the wherein the sorbent particles comprise a zeolite.

21. The coating composition of claim 19 wherein optical response enhancer is a light absorbing material that, when present in the polymer matrix which coats an optical fiber to a thickness of 5 μm, exhibits at least two times greater absorption, or at least 5 times, or in the range of 2 to ten times or five to ten times greater absorption of the light absorbed from the sensor coating at the measurement wavelength as compared to the absorption from the coating without enhancer and as compared to the coating in the absence of CO.sub.2 and CH.sub.4.

22. A sensor, comprising: a glass fiber coated with the coating composition of claim 19.

23. The sensor of claim 22 wherein the glass fiber is disposed in a metal or plastic tube having a plurality of holes formed along the length of the tube.

24-26. (canceled)

27. A method of making a composite, comprising: mixing a suspension of polymer, optical enhancer and absorbent particles; applying the suspension to a substrate, and curing or setting the polymer.

28. The method of claim 27 wherein the optical enhancer comprises plasmonic nanocrystals, and wherein the substrate is a glass fiber.

29. (canceled)

30. (canceled)

31. A method of measuring an amount of a molecule of interest, comprising: exposing the sensor of any of the above claims to the molecule of interest, and measuring light transmission through the fiber.

32-33. (canceled)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0026] FIG. 1 illustrates a processing scheme for the preparation of mixed-matrix NC.sub.PZ.sub.HPP.sub.CL composites on FO sensor platform (NC.sub.P: Plasmonic nanocrystals, Z.sub.HP: Hydrophobic zeolites, P.sub.CL: Cross-linked polymer).

[0027] FIG. 2. (a) TEM image of mono-dispersed NC.sub.P synthesized by hot-injection method. SEM images of (b) Z.sub.HP and (c) NC.sub.PZ.sub.HPP.sub.CL mixed-matrix film on FO sensor platform. (d) XRD patterns of NC.sub.P, Z.sub.HP and NC.sub.PZ.sub.HPP.sub.CL composites.

[0028] FIG. 3. Schematic diagram of the mixed-matrix composite FO sensor system with double jacket polytetrafluoroethylene (PTFE) membrane.

[0029] FIG. 4. NC.sub.PZ.sub.HPP.sub.CL QCM sensors. (a) Frequency changes for variable CO.sub.2 concentrations at different % RH conditions. Arrows in 4a indicate injection of water vapor into the gas chamber. (b) Responses curves as a function of CO.sub.2 concentration constructed from ΔFrequency in 4a (solid lines indicate Langmuir-Freundlich fitting). (c) CO.sub.2 concentration dependent sensitivity at different % RH (% relative humidity) conditions relative to the dry condition. (d) Dynamic frequency changes over 7 cycles at 6 different CO.sub.2 concentrations in 95% RH condition.

[0030] FIG. 5. Optical responses of the NC.sub.PZ.sub.HPP.sub.CL FO sensors in dry condition. (a) Transmission spectra after exposure to different concentration of CO.sub.2. (b) ΔT with different NC.sub.PZ.sub.HPP.sub.CL thickness to 100% CO.sub.2. (c) Time-resolved % T response with different sensing layers to 6 different CO.sub.2 concentration.

[0031] FIG. 6. Sensitivity and selectivity of the NC.sub.PZ.sub.HPP.sub.CL FO sensors to CO.sub.2 in dry and humid conditions. (a) Time-resolved % T response to 6 different CO.sub.2 concentration. (b) Responses curves as a function of CO.sub.2 concentration constructed from ΔT in FIG. 6a (solid lines indicate the Langmuir-Freundlich fitting). (c) CO.sub.2 concentration dependent sensitivity at different % RH conditions against the dry condition. (d) Dynamic responses to 5 cycles of 6 different CO.sub.2 concentration in 95% RH condition. (e) Time-resolved % T response to different gas mixtures.

[0032] FIG. 7. (a) Time-resolved % T of the NC.sub.PZ.sub.HPP.sub.CL FO sensor for 6 subsequent cycles to different CO.sub.2 concentration dissolved in DI water. Power changes of the NC.sub.PZ.sub.HPP.sub.CL FO sensor to different CO.sub.2 concentration (b) dissolved in DI water and (c) dissolved in synthetic AMD water.

[0033] FIG. 8. (a) SEM image and (b) XRD patterns of the NC.sub.PZ.sub.HPP.sub.CL FO sensors after field testing. Inset shows an enlarged XRD patterns of the NC.sub.PZ.sub.HPP.sub.CL FO sensors after field testing indicated by the dotted square box in FIG. 8b. FWHM values of the peak around 2θ=7.96° for fresh and used samples are identical as 0.0836. (d) Dynamic % T responses of the NC.sub.PZ.sub.HPP.sub.CL FO sensor after field testing to 86 cycles of 20% CO.sub.2 for 8 days in 95% RH condition.

[0034] FIG. 9. Long-term test of the mixed-matrix fiber optic CO.sub.2 sensor. Each data points were obtained from the maximum changes in optical power to 20% CO.sub.2 exposure in 45% RH conditions. The sensor testing was performed from July/2021 through March/2022. The sensing tests were performed without pretreatments such as heating and/or evacuating under vacuum.

DESCRIPTION OF PREFERRED EMBODIMENTS/EXAMPLES

[0035] Synthesis of indium-tin oxide (ITO) nanocrystals (NCs). ITO NCs were synthesized by using a hot-injection method reported elsewhere..sup.45 Briefly, a solution of oleylamine (10 mmol) and octadecene (5 mL) was injected into a solution containing of In(Ac).sub.3 (1.08 mmol), Tin(II) 2-ethylhexanoate (0.12 mmol), 2-ethylhexanoic acid (3.6 mmol), and octadecene (10 mL) in a three-neck flask at 240° C., and then heated to 290° C. After 2 hours reaction, the ITO NCs were separated by centrifuging and then washed with toluene. The final product was redispersed in toluene (8 mL). An ITO content of 1.05 wt % in toluene was confirmed from Thermogravimetric Analyzer.

[0036] Hydrophobic ZSM-5 nanoparticles. Hydrophobic ZSM-5 (MZ5-1500p, Si/Al=1500) was purchased from ACS Material. Before mixing with PDMS solution, the surface of the ZSM-5 particles was modified with isobutyl siloxane to help suspend the particles in hexane. ZSM-5 (2.5 g) and isobutyltriethoxysilane (100 mg) were added into heptane (20 mL) and then refluxed overnight. The NCs were separated by centrifuging and then washed sequentially with heptane and ethanol.

[0037] Cross-linked polydimethylsiloxane (PDMS) solution. The Sylgard-184 silicone elastomer and curing agent were mixed together in a 10:1 ratio. This mixture was stirred and then placed on a 100° C. hot plate for 20 minutes. Afterwards, n-hexane was added to the cooled mixture to achieve a 2 wt % PDMS solution and the solution was stirred for an hour.

[0038] Mixed-matrix precursor solution. Precursor suspension was prepared by mixing ITO, ZSM-5 and PDMS solution. ZSM-5 (26 mg) was firstly added into the 2 wt % cross-linked PDMS solution in hexane (2 mL), and then 0.6 mL ITO solution was added. When this suspension is cast as a film, it should be 6% ITO, 50% ZSM-5 and 44% PDMS (wt %). Before application of the coating, the precursor suspension was sonicated for 30 minutes to ensure a good dispersion of suspended particles.

[0039] Synthetic acid mine drainage (AMD) water. Al.sub.2(SO.sub.4).sub.3.18H.sub.2O (0.58 g L.sup.−1), CaSO.sub.4.2H.sub.2O (0.75 g L.sup.−1), MnSO.sub.4.H.sub.2O (0.07 g L.sup.−1), MgCl.6H.sub.2O (0.67 g L.sup.−1), NaCl (0.28 g L.sup.−1), NaHCO.sub.3 (0.15 g L.sup.−1), KCl (0.06 g L 1), and FeCl.sub.3.6H.sub.2O (0.01 g L.sup.−1) were first dissolved in DI water (500 mL) and then pH was adjusted to 3.5 using H.sub.2SO.sub.4, followed by bringing the solution to 1 L final volume with DI water.

[0040] Films on substrates. Quartz crystal microbalance (QCM) substrate: QCM substrate was obtained from INFICON (MAXTEK 5 MHz polished Ti/Au, one-inch diameter). QCM substrate was first cleaned by successive sonications in acetone, methanol, and DI water. A mixed matrix solution was dropped on the center of the quartz disc and then spin-coated at 2000 rpm for 10 seconds. The film was finally cured at 120° C. for 2 hours in air. Fiber optic (FO) substrate: Optical multimode fibers with 105 m core diameter were used (FG105LCA, 0.22 NA, Thorlabs, Inc.). Before applying the NC.sub.PZ.sub.HPP.sub.CL coating, the polymer jacket of the FO was removed, and the fiber was etched (˜5 cm in length) in a buffered oxide hydrofluoric acid etchant solution for 90 minutes. It was then cleaned by immersing in DI water. Precursor solution was coated on an etched FO by dragging upward (1 cm s.sup.−1) a generated droplet at the tip of the micropipette, and drying at room temperature. The thickness of the nanocomposite films was controlled by repeating the above process. The film was finally cured at 120° C. for 2 hours in air.

[0041] Sensor testing of QCM and FO sensors in dry and wet conditions. Sensor measurements were performed in a gas cell at room temperature and atmospheric pressure. The gas cell inlet was connected to an automated gas delivery system with a total flow rate of 100 mL min.sup.−1. Different CO.sub.2 gas concentrations were obtained by controlling the flow rates of pure N.sub.2 and CO.sub.2 gas. Before measurement, the gas cell was first purged with pure N.sub.2 gas for 60 minutes. In order to achieve different humidity levels (˜45% or ˜95% RH) in the gas chamber, the dry N.sub.2 gas stream was run through a bubbler filled with saturated potassium carbonate or DI water at atmospheric condition. A selectivity test of the NC.sub.PZ.sub.HPP.sub.CL FO sensor was performed with 20% doses of dry CO.sub.2, O.sub.2, H.sub.2, and CH.sub.4 diluted in N.sub.2 with 100% N.sub.2 flowed between each exposure. Note that all sensing tests were performed without pretreatments such as heating and/or evacuating under vacuum.

[0042] QCM sensor test: The QCM device was held in an Inficon (Maxtek) crystal holder with pogo pins, which was connected to an Inficon (Maxtek) RQCM-QCM research system. The frequency change was recorded with MaxTek RQCM data-logger software.

[0043] FO sensor test: Optical transmittance was recorded by connecting one end of the fiber to a spectrometer (JAZ, Ocean Optics). The other end of the fiber was connected to a broad-band tungsten halogen light source (DH-2000-BAL, Ocean Optics). 100% transmission as reference spectrum was collected with pure N.sub.2 gas flow in a gas cell.

[0044] Sensor testing in aqueous solutions using double jacket membrane system. A schematic diagram of the experimental set-up for testing the NC.sub.PZ.sub.HPP.sub.CL FO sensor in aqueous solution is shown in Figure S1. A broad-band tungsten halogen lamp was used as light source (LS-1, Ocean Optics). Optical power change was recorded by a power meter (PM 100D, Thorlabs) equipped with a photodiode (S151C, λ=400-1100 nm, P.sub.max=20 mW, Thorlabs, Inc.), which was connected to a computer for logging the data. Dry ice was used as a CO.sub.2 source to control the total concentration of dissolved CO.sub.2 in testing waters. First, testing water (either DI water or synthetic AMD water) was fed into the commercial IR CO.sub.2 sensor (GM70 Handheld CO.sub.2 Meter, Vaisala) in a flow-through cell, and then into a flow-through cell with the FO sensor. The IR sensor probe and FO sensors were encased within a liquid water impermeable double jacket polytetrafluoroethylene (PTFE) membrane (density: 1.13 g cm.sup.3, outer diameter: 6.7 mm, wall thickness: 0.28 mm, International Polymer Engineering). A perforated stainless-steel tube (⅛ inch) was used to support the FO sensor assembly. The testing water was delivered by a peristaltic pump (Barnant L/S portable sampler) at a flow rate of 500 mL min.sup.−1.

[0045] Sensor testing in the field. The configuration of the FO sensor system described above was modified to run on battery power for field testing (Figure S2). For the practical application, a portable FO sensor was tested with shallow groundwater from a monitoring well at the NETL Pittsburgh site, Pennsylvania (lat. 40° 18′ 25.0″ N, long. 790 58′ 40.2″ W) and with coal mine drainage at Lambert Run, Clarksburg, W. Va. (lat. 390 19′ 43.0″ N, long. 800 21′ 59.3″ W). The testing waters were introduced directly into the flow-through system without any purification for real-time monitoring of dissolved CO.sub.2. The sensor testing at the field sites were performed in October-November 2020.

[0046] On-line monitoring in the field. To realize the real-time online monitoring functions in the field, the FO sensor system was modified without a pumping part for low-power operation. The two FO cables were inserted into high-density PFTE tubing (⅛ inch), and the ends of the FO cables were spliced with 5 cm lengths of coreless fiber (125 m core diameter, FG125LA, Thorlabs, Inc.). The coreless FO part was coated with the NC.sub.PZ.sub.HPP.sub.CL composite, which was then encased by a perforated stainless-steel tube (⅛ inch) within a PTFE membrane. The FO sensor together with a commercial IR sensor were lowered into the NETL Pittsburgh site well (480 feet depth). The optical power change was recorded by a power meter equipped with a photodiode, which was connected to a data logger (X2 Environmental Data Loggers, NexSens Technology, Inc.). The data logger was operated by a rechargeable battery powered with solar panels and the data was transferred through cellular lines from the field to the website. The experimental set-up for on-line monitoring with a wireless communication system is shown in Figure S3. The sensor testing at the NETL Pittsburgh well was performed in March 2022.

[0047] Results and Discussion

[0048] Preparation and Characterization of Mixed-Matrix NC.sub.PZ.sub.HPP.sub.CL Composites on FO Sensor Platform. The first step in the process toward mixed-matrix NC.sub.PZ.sub.HPP.sub.CL composites for CO.sub.2 sensing is the colloidal synthesis of the plasmonic NCs (NC.sub.P). Mono-dispersed ITO NCs were synthesized, yielding an average size of 12 nm in diameter (FIG. 2a). TEM and Fourier transform image (Inset of FIG. 2a) show single-crystalline ITO with an atomic lattice fringe of 0.29 nm corresponding to interplanar spacing of (222) and 0.25 nm for (400) lattice planes, respectively, of the cubic bixbyite structure of In.sub.2O.sub.3 (JCPDS file 06-0416).sup.46. In the next step, hydrophobic ZSM-5 zeolite (Z.sub.HP) particles (FIG. 2b) were added into cross-linked PDMS (P.sub.CL) solution, and the suspension was subsequently mixed with the NC.sub.P solution. This precursor suspension was coated on FO sensor platforms and then cured in order to complete the cross-linking of PDMS oligomers. Note that the total time to integrate the NC.sub.PZ.sub.HPP.sub.CL composites layers onto the FO takes only a few tens of seconds. This rapid application is well-suited for the production of long lengths of FO sensors which are required for distributed sensors. SEM imaging indicates a dense NC.sub.PZ.sub.HPP.sub.CL layer on the surface of FO (FIG. 2c), in which NC.sub.P and Z.sub.HP were well inter-connected by PDMS binder without serious particle aggregation. Multiple SEM images taken at the different locations showed an excellent uniformity of the NC.sub.PZ.sub.HPP.sub.CL layer on FO. A cross-section SEM image also shows a uniform coating of the NC.sub.PZ.sub.HPP.sub.CL layer on FO. Furthermore, inspection of XRD patterns of the NC.sub.PZ.sub.HPP.sub.CL layer reveals that their crystalline nature remains intact compared with the respective individual components (FIG. 2d).

[0049] Water Vapor Mitigation of NC.sub.PZ.sub.HPP.sub.CL Composite. For real-world applications, the sensors should be able to detect CO.sub.2 and remain stable at high humidity levels. With this application in mind, the ability of the mixed-matrix NC.sub.PZ.sub.HPP.sub.CL sensor material to efficiently mitigate the effects of water vapor was investigated. QCM sensors, a highly sensitive and simple transduction platform, detects mass variations per unit area by measuring the change in frequency of a quartz crystal resonator..sup.47 The QCM sensing platform is therefore useful to evaluate the selectivity of CO.sub.2 gas in humid conditions. The precursor solution was spin-coated onto a QCM substrate and then cured to form a mixed-matrix NC.sub.PZ.sub.HPP.sub.CL composite layer. The response of the frequency was then tracked during different gas cycles at room temperature to study the CO.sub.2 concentration dependence in dry and humid conditions. The gas cycles were switched alternatively between pure N.sub.2 and a mixture of CO.sub.2 in N.sub.2 while maintaining a constant total flow rate. FIG. 4a shows the QCM frequency changes during the gas cycling, which alternated every 10 minutes between pure N.sub.2 and a mixture of N.sub.2 and CO.sub.2. The NC.sub.PZ.sub.HPP.sub.CL coated QCM sensor showed a rapid decrease in frequency and reached equilibrium within 12 seconds in dry condition, indicating a fast diffusion of CO.sub.2 molecules into NC.sub.PZ.sub.HPP.sub.CL composite layer. In humid conditions (45% RH and 95% RH), the sensitivity remains despite a slightly smaller and slower response/recovery times (˜2 minutes) as compared to dry conditions. This increase in response/recovery times can be attributed to slightly slower CO.sub.2 adsorption kinetics within the NC.sub.PZ.sub.HPP.sub.CL layer in humid conditions because CO.sub.2 molecules compete with water molecules for internal pore volume of the Z.sub.HP. In order to demonstrate the hydrophobicity of Z.sub.HP, gravimetric gas adsorption and in situ Fourier transform IR analysis were then investigated. According to gravimetric gas adsorption, the total weight change for CO.sub.2 uptake is 10 times higher than for H.sub.2O uptake on the Z.sub.HP material at 25° C. and 1 bar. The FT-IR results displayed spectral features consistent with physisorption of CO.sub.2 at around 2330 cm.sup.−1 with no evidence of bicarbonate formation at around 1362 cm.sup.−1. The CO.sub.2:SiO.sub.2 ratio of the FT-IR absorbance peaks is the same in dry CO.sub.2 and humid CO.sub.2 condition, indicating that the amount of adsorbed CO.sub.2 in the Z.sub.HP material remains unchanged even at high humidity (95% RH).

[0050] The CO.sub.2 concentration dependent frequency change is clearly observed both in dry and humid conditions (FIG. 4b). The ΔFrequency of the NC.sub.PZ.sub.HPP.sub.CL QCM sensor shows a logarithmic response which fits very well to a Langmuir-Freundlich model, indicative of a physical adsorption process occurring within the zeolite micropores. This response curves in FIG. 4a were used to construct the sensitivity change of the NC.sub.PZ.sub.HPP.sub.CL QCM sensor in humid conditions and the sensitivity change was calculated from the equation as follow:

[00001]$\Delta \mathrm{Sensitivity}=\left(\frac{\Delta {\mathrm{Frequency}}_{\mathrm{humid}}-\Delta {\mathrm{Frequency}}_{\mathrm{dry}}}{\Delta {\mathrm{Frequency}}_{\mathrm{dry}}}\right)\times 100$

[0051] The CO.sub.2 sensitivity of the NC.sub.PZ.sub.HPP.sub.CL QCM sensor is slightly reduced in humid conditions compared to dry condition: approximately 7% at 45% RH and 16% at 95% RH of the frequency change for 100% CO.sub.2 exposure (FIG. 4c). In addition, the sensing experiment was repeated to demonstrate the long-term stability of the NC.sub.PZ.sub.HPP.sub.CL in 95% RH (FIG. 4d). Remarkably, the response to CO.sub.2 is virtually unchanged over the cycling test. This stability in sensing response can be attributed to the hydrophobicity and chemical inertness of the zeolite and PDMS.

[0052] Optical Responses of NC.sub.PZ.sub.HPP.sub.CL FO Sensor in Dry and Humid Conditions. To evaluate the CO.sub.2 sensing function of FO integrated with NC.sub.PZ.sub.HPP.sub.CL, a commercial multimode fiber with 105 m SiO.sub.2 core and F-doped SiO.sub.2 cladding was used, wherein the cladding was removed before applying the NC.sub.PZ.sub.HPP.sub.CL coating on the surface of the FO platform (see details above). In this way, the RI of the NC.sub.PZ.sub.HPP.sub.CL sensing layer on the FO sensing platform changes upon CO.sub.2 gas adsorption, which in turn, gives rise to a wavelength-dependent variation of the transmitted light intensity (% T), according to the Lorenz-Lorentz law when CO.sub.2 replaces N.sub.2. In order to identify the wavelength range corresponding to the flank of the “peak”, which has been shown to work as a sensor and yield the highest sensitivity, the wavelength-resolved optical transmission of the NC.sub.PZ.sub.HPP.sub.CL FO sensor was measured when the gas environment was switched from pure N.sub.2 to various concentrations of CO.sub.2. The spectra revealed a distinct decrease in transmittance for the wavelength range of 600-650 nm when CO.sub.2 is introduced (FIG. 5a). Note that the drop in transmission in the 600-650 nm range is due to the total reflection in the optical modes guided by the sensor film.

[0053] We hereafter focused on the monochromatic signal at 635 nm (T.sub.max) as the readout, which can easily be applied to a commercial power meter equipped with a photodiode. Next, the thickness dependent optical response of the NC.sub.PZ.sub.HPP.sub.CL FO sensor was investigated. The sensitivity to 100% CO.sub.2 (ΔT.sub.max response) was plotted as a function of NC.sub.PZ.sub.HPP.sub.CL thickness, as shown in FIG. 5b. Thickness information was obtained from SEM analysis, and the average film thickness measured at 5 different positions on the film. The ΔT.sub.max response was observed at around 5 μm with no enhancement observed with the application of additional coating cycles. This indicates a long-range diffusivity with negligible film thickness dependency on the NC.sub.PZ.sub.HPP.sub.CL sensing layer. A series of FO sensors with different sensor components were also prepared and their optical responses were measured to determine how each component in the NC.sub.PZ.sub.HPP.sub.CL affects the sensitivity. FIG. 5c shows the time-resolved % T response of NC.sub.P, NC.sub.PP.sub.CL, Z.sub.HPP.sub.CL and NC.sub.PZ.sub.HPP.sub.CL FO sensors to different concentration of CO.sub.2. No response for NC.sub.P FO sensors was observed, and only a small increase in % T for NC.sub.PP.sub.CL was observed. The results show the importance of porous materials to effectively change the RI of the sensing layer. Surprisingly, the incorporation of NC.sub.P into Z.sub.HPP.sub.CL matrix showed a 7-fold improvement in sensitivity, clearly indicating the plasmonic enhancement in the optical sensing response associated with gas adsorption in the sensing layer. In addition, the changes in % T when cycling between pure N.sub.2 and pure CO.sub.2 were reproducible at room temperature without the need for vacuum or added heat to regenerate the NC.sub.PZ.sub.HPP.sub.CL FO sensor.

[0054] The NC.sub.PZ.sub.HPP.sub.CL FO sensor was further tested in humid conditions (45% RH and 95% RH) to see the effect of water vapor on the sensitivity and response/recovery times (FIG. 6a). Similar to the QCM sensor, the response/recovery time also increased slightly in humid conditions due to the competitive nature of the CO.sub.2 and H.sub.2O diffusion process. The optical response of the NC.sub.PZ.sub.HPP.sub.CL FO sensor shows a logarithmic response between partial pressure of CO.sub.2 gas and the overall drop in % T, indicative of a guest molecule-induced optical response modification (FIG. 6b). Compared to dry conditions, the NC.sub.PZ.sub.HPP.sub.CL FO sensor showed a decrease in ΔT response by approximately 23% for 100% CO.sub.2 in 95% RH (FIG. 6c). As expected, the effect of water vapor on the FO sensor is different from the QCM sensor results, which showed a decrease of 16% in sensitivity in 95% RH. This variation results from the different sensing mechanisms of the QCM and FO sensors. Upon CO.sub.2 exposure, the QCM sensor measures the mass variation calculated according to the Sauerbrey equation, whereas the response of the FO sensor is based on the RI changes that follows the density variation of sensor layer. Since the molecular weight of H.sub.2O is smaller than CO.sub.2, H.sub.2O will have a smaller impact on the mass-based QCM sensor. In other words, if the same amounts of H.sub.2O and CO.sub.2 are adsorbed in the QCM and FO sensor coatings, a larger impact from water would be expected on the NC.sub.PZ.sub.HPP.sub.CL FO sensor. Therefore, the sensitivity is more affected at low concentration of CO.sub.2 in humid condition on the NC.sub.PZ.sub.HPP.sub.CL FO sensor, as shown in FIG. 6c. The exceptional dynamic response of the NC.sub.PZ.sub.HPP.sub.CL FO sensor clearly displays reproducible results over 5 cycles for 6 different CO.sub.2 concentrations from 100% to 10% CO.sub.2 in 95% RH (FIG. 6d). The selectivity of the NC.sub.PZ.sub.HPP.sub.CL FO sensor toward CO.sub.2 as compared to other gases such as O.sub.2, H.sub.2, and CH.sub.4 was also investigated (FIG. 6e). The NC.sub.PZ.sub.HPP.sub.CL FO sensor was exposed to 20% of O.sub.2, H.sub.2, and CH.sub.4 diluted in N.sub.2. The NC.sub.PZ.sub.HPP.sub.CL FO sensor displayed a selective response to CO.sub.2, with a much smaller response for CH.sub.4 and essentially no response for O.sub.2 and H.sub.2 in both dry and 95% RH conditions.

[0055] Sensor Testing of NC.sub.PZ.sub.HPP.sub.CL FO Sensor in Aqueous Solution. With excellent sensitivity to CO.sub.2 and repeatability in humid conditions, the NC.sub.PZ.sub.HPP.sub.CL FO sensor was further tested with water solutions to observe its robustness in a more complex environment. A prototype FO sensor configuration was demonstrated for testing in aqueous solution. The NC.sub.PZ.sub.HPP.sub.CL FO sensor is encased within a double jacket PTFE polymer membrane system. This system protects the FO sensor against liquid water agitation while simultaneously allowing only gas phase CO.sub.2 from aqueous dissolved CO.sub.2 to diffuse into the FO sensor and interact with the NC.sub.PZ.sub.HPP.sub.CL sensing layer. Diffusion kinetics of dissolved CO.sub.2 in water and water vapor through a PTFE polymer membrane system were investigated. The responses from different concentrations of acidified bicarbonate solution were monitored by an IR CO.sub.2 sensor. Note that the IR sensor probe was also enveloped within a PFTE membrane to guard against liquid water agitation. As a result, the amount of dissolved CO.sub.2 that permeates through the PTFE membrane showed a linear relationship to the concentration of acidified bicarbonate solution with a correlation coefficient (R.sup.2) of 0.9995. RH increases to as high as 95% by diffusion of water vapor through the PTFE membrane.

[0056] For the FO sensor experiments in aqueous medias, the aqueous CO.sub.2 solution was composed of either DI water or synthetic AMD water. The CO.sub.2 solution was first pumped through the commercial IR sensor chamber, and then through the cell containing the FO sensor enclosed within the double jacket PTFE membrane. The changes in optical response were monitored at 635 nm by using either a spectrometer or power meter with the data logged by computer. In this experiment, dry ice was employed as the CO.sub.2 source in water, where a linear relationship between the CO.sub.2 concentration and the added amount of dry ice in DI water was observed. Cyclic sensing of the NC.sub.PZ.sub.HPP.sub.CL FO sensor to different CO.sub.2 concentrations dissolved in DI water was performed. The CO.sub.2 concentration in the testing water was varied by adding dry ice every 20 minutes and purging with DI water between steps. After 3 hours of testing, the sensor was left under DI water flow overnight, and then the CO.sub.2 concentration was adjusted again for the next cycle. FIG. 7a shows that the time-resolved % T response is reversible, and sensitivity and stability of the NC.sub.PZ.sub.HPP.sub.CL FO sensor are nearly identical over the 6 cycles. The calibration curve (CO.sub.2% vs AT) was constructed by combining the result from the FIGS. 6b and 7a tested in 95% RH condition. The curve shows a logarithmic response and the responses fit very well to Langmuir-Freundlich model (R.sup.2=0.9999). The linear working range and the linearity were extracted from the calibration curve. A perfect linearity between AT and CO.sub.2 concentration in a range of −3% to 20% CO.sub.2 (R.sup.2=0.9999) with the sensitivity of 0.118 was observed.

[0057] To enable the sensor system for field application, a monochromatic readout using inexpensive components such as light emitting diodes and photodiode detectors that consume low power was utilized in the NC.sub.PZ.sub.HPP.sub.CL FO sensor system. This was further corroborated by an extended cycling test in DI water, in which the FO sensor system provided a stable response in agreement with the commercial CO.sub.2 IR spectrometer (FIG. 7b). Therefore, it is expected that the FO sensor system studied in this work can be readily transferred to field applications. The NC.sub.PZ.sub.HPP.sub.CL FO sensor was further tested with synthetic AMD water to see how the sensor would response to a more complex water solution matrix. This time, the CO.sub.2 concentration in the synthetic AMD water was gradually increased at the beginning of the testing cycle and then randomly varied throughout the remainder of the tests. (FIG. 7c). As can be seen, the sensor's response/recovery was consistent in all cycles. Thus, it can be concluded that the NC.sub.PZ.sub.HPP.sub.CL FO sensor system exhibits an excellent reproducibility for dissolved CO.sub.2 detection in both pure water and synthetic AMD water, with no degradation of the signal amplitude over time. By using power meter as the readout, the NC.sub.PZ.sub.HPP.sub.CL FO sensor shows the LOD of 0.218% in DI water and 0.409% in AMD water, respectively. Increasing the length of the sensor coating would increase sensitivity to lower CO.sub.2 concentrations.

[0058] Field Deployment of NC.sub.PZ.sub.HPP.sub.CL FO Sensor in Real-World Setting. To assess the robustness of the NC.sub.PZ.sub.HPP.sub.CL FO sensor toward natural water, the same experimental protocol as used in the laboratory was conducted for field testing. Note that the configuration of the sensor system was modified to run on batteries for the field testing. The field sites were at NETL Pittsburgh, Pa. and Lambert's Run, W. Va. The testing water was introduced into the sensor system and switched to tap water for baseline. When the signal was stable, the tap water was switched to testing water, and the power dropped about 0.0364 μW, which is associated to 3.1% CO.sub.2, as observed by the IR sensor. A slight change of baseline level that occurs at approximately 40 min was attributed to a small movement of the FO by wind at the field site. To further assess the NC.sub.PZ.sub.HPP.sub.CL FO sensor in more realistic conditions, coal mine drainage water at Lambert Run field was tested, wherein commercial Aquafina® bottled drinking water was used for the baseline. When the testing water was introduced at 20 min, a 0.1044 W power drop corresponding to 12% CO.sub.2 was observed, which is fully recovered by replacing with Aquafina® water. To benchmark the NC.sub.PZ.sub.HPP.sub.CL FO sensor, the response/recovery times was investigated by comparing with a commercial IR sensor. The NC.sub.PZ.sub.HPP.sub.CL FO sensor system shows response/recovery times 4 times faster than the commercial IR sensor. For example, in the ground water test, the measured response/recovery times of the NC.sub.PZ.sub.HPP.sub.CL FO sensor were 5 min and 3 min, respectively, while 30 min and 12 min were observed for IR sensor. Similarly, the response/recovery times with the coal mine drainage water were 3 min and 2 min, respectively. These results were further corroborated by a second introduction of testing water at around 45 min, for which very clear and rapid responses were observed for the NC.sub.PZ.sub.HPP.sub.CL FO sensor. The origin of this accelerating effect in FO sensor is due to the fast diffusion kinetics of dissolved CO.sub.2 from the testing water to the sensing area of the FO surface. In contrast, the IR sensor requires a relatively large sensing volume to be sensed which increases the CO.sub.2 diffusion time. The difference in the response/recovery times between the sites can be assigned to the different flow rates: 200 mL min.sup.−1 and 500 mL min.sup.−1 were flowed for NETL Pittsburgh site and for Lambert Run site, respectively. This indicates that the response/recovery times are associated with how quickly the water sample is flowing onto the NC.sub.PZ.sub.HPP.sub.CL FO sensor. Note that when operating the FO sensor system in the field under battery power, a compromise such as flow rate must be made to determine the optimum system conditions to balance the operational requirements with analytical performance.

[0059] Having successfully assessed the NC.sub.PZ.sub.HPP.sub.CL FO sensor performance, it is important to put the obtained sensing metrics into a wider perspective. To this end, a scatterplot of all of the NC.sub.PZ.sub.HPP.sub.CL FO sensor readouts under controlled laboratory conditions and the CO.sub.2 concentrations measured by the IR sensor, as well as natural water samples were compiled. The combined results indicate that there is a linear relation between the FO sensor and the IR sensor readouts, and that the responses from the two test sites are more in line with the DI water test than with the synthetic AMD water tests. The accuracy of the NC.sub.PZ.sub.HPP.sub.CL FO sensor was evaluated to determine its efficacy when exposed to diverse sources for realistic scenarios. Linear regressions were used to determine the best-fit line between the IR sensor and the corresponding the NC.sub.PZ.sub.HPP.sub.CL FO sensor readout in a range of 0.5˜15% of CO.sub.2. The accuracy of the NC.sub.PZ.sub.HPP.sub.CL FO sensor was then calculated based on the equation as follow:

[00002]$\mathrm{Accuracy}=100-.Math.\frac{\mathrm{reference}-{\mathrm{readout}}_{\mathrm{Fo}\mathrm{sensor}}}{\mathrm{reference}}.Math.100$

[0060] References were obtained from linear regressions for each measurement result in DI water or synthetic AMD water. For example, a power drop of 0.02644 μW in DI water and 0.01411 μW in synthetic AMD water corresponding to 3.1% CO.sub.2 were calculated from linear regressions and used as references. The power drop corresponding to 3.1% CO.sub.2 on the FO sensor observed at the NETL Pittsburgh site was 0.0364 μW. As a result, the NC.sub.PZ.sub.HPP.sub.CL FO sensor calibrated with DI water handled higher degrees of accuracy (62% for NETL Pittsburgh site and 98% for Lambert's Run site), but the NC.sub.PZ.sub.HPP.sub.CL FO sensor showed a very low accuracy when it was calibrated with the synthetic AMD water. This implies that the NC.sub.PZ.sub.HPP.sub.CL FO sensor can quantitatively detect dissolved CO.sub.2 in natural waters (e.g., shallow ground water) when calibrated under appropriate conditions. The potential influence of different testing waters on the measurement was further investigated. The observed power readouts corresponding to tap water (2.548 W) or Aquafina® drinking water (2.597 W) for the baseline were closer to DI water (2.629 W) than observed for synthetic AMD water (1.935 W). This is possibly due to the competitive effects of sulfur-bearing gases present in synthetic AMD water, such as H.sub.2S or SO.sub.2, although the diffusion of H.sub.2S through fluorinated polymers such as PTFE is much less favorable than CO.sub.2. Such components cannot be ruled out as potential interference factors in the sulfidic deep waters or eutrophic waters where high concentrations of competitive gases could reduce the sensitivity of CO.sub.2 detection. The overall response and sensitivity to dissolved CO.sub.2 in real water samples remains constant throughout the experiment.

[0061] In addition, a wireless telemetry system was used to transfer the data in real time through cellular lines to provide on-line monitoring capabilities. These concepts were demonstrated with NexSens X2 Environmental Data Loggers (see the details in Materials and Methods section). The optical response of the NC.sub.PZ.sub.HPP.sub.CL FO sensor system to shallow ground water at the NETL Pittsburgh site well was monitored with the data transferred from the field to the website for 3.5 days. As expected, the temperature of the subterranean water remained unchanged with time due to the depth of the test. The on-line monitoring system provided a stable response with 0.0028 μW of power variation, which corresponded to 0.3% CO.sub.2, as calculated from the linear curve. This is in good agreement with the commercial CO.sub.2 IR sensor showing 0.1% variation. The trend in optical power changes is consistent with the concentration of CO.sub.2 measured by the commercial IR sensor.

[0062] The structural stability of the NC.sub.PZ.sub.HPP.sub.CL FO sensors was investigated via SEM and XRD analysis after field testing. The results reveal that the NC.sub.PZ.sub.HPP.sub.CL layer on FO exhibits remarkably constant surface morphology (FIG. 8a) and no degradation in crystallinity occurs after the entire testing (FIG. 8b). This observable structural stability agrees well with the stable and reversible CO.sub.2 sensing response displayed over the long-term. The NC.sub.PZ.sub.HPP.sub.CL FO sensor was subjected to 86 cycles of 20% CO.sub.2 for 8 days in 95% RH condition and it consistently responded with the same amplitude during the entire test (FIG. 8c). Over the 8-day testing period, the sensor exhibited a ΔT=2.78±0.23% (average±standard deviation) and 95% confidence level is between 2.73 and 2.83%. The response of fresh NC.sub.PZ.sub.HPP.sub.CL FO sensor to 20% CO.sub.2 in 95% RH condition was ΔT=2.74%. This value is also within 95% confidence level, indicating that the original sensor performance was maintained. Stability testing over eight months under real world conditions is shown in FIG. 9.

[0063] In summary, a mixed-matrix composite integrated FO sensor system was developed and the capability of the FO sensor to detect a wide range of gas-phase and dissolved CO.sub.2 in water for real-time monitoring of CO.sub.2 leakage for carbon storage application was examined. Specifically, a combination of plasmonic NCs and hydrophobic zeolite was employed to enhance the sensitivity of the sensor in humid environments. The FO sensor shows excellent stability and reversibility in the adsorption of CO.sub.2 molecules while effectively mitigating water vapor. The sensing response of the FO sensor remained unchanged even after an entire series of tests in the laboratory and in the field, showing a very robust performance in various environments. The sensor displays the sensitivity and long-term stability required for diverse application areas of CO.sub.2 sensors. While environmental conditions can vary depending on testing sites, the sensor in this work is targeted for monitoring CO.sub.2 migrations associated with sequestered CO.sub.2 where the background gas composition would be expected to remain relatively constant. The PTFE sleeve effectively blocks any interference on the sensor reading from dissolved ions since these ions will not diffuse through PTFE. The PTFE sleeve is also expected to significantly reduce interference from other dissolved gases, such as CH.sub.4 and H.sub.2S, due to their low solubility/diffusivity in the polymer..sup.50,53 As such, the PTFE sleeve not only protects the optical fiber sensor from physical damage, it also acts as a membrane to improve the selectivity of the sensor assembly. This further simplifies the detection of increasing CO.sub.2 concentrations by monitoring changes is equilibrium loading of the adsorbate in the sensor coating relative to the initial baseline. Since the sensor operates on a physisorption principle, temperature will have a measurable effect on the amount of CO.sub.2 adsorbed in the sensor coating, and hence on the sensor response. Temperature also affects the baseline transmission of light within the FO sensor. The use of a blank FO in tandem with a sensor FO can be used to compensate for modulations in transmitted light or power due to fluctuations in the source, detector, and temperature. The blank FO will also provide real time temperature data which can be used to calibrate the sensor response. Accounting for temperature variations at the sensor location will, of course, be more relevant for above ground measurements since subterranean temperatures change little with time below a certain depth. For dissolved aqueous CO.sub.2 monitoring, the correlation between the sensor reading of gas phase CO.sub.2 permeating the PTFE sleeve and the dissolved CO.sub.2 concentration will be affected by the pH of the aquifer. Since U.S. Environmental Protection Agency regulations will require monitoring of changes in the pH of the aquifer, these data will also be available for appropriate calibration of the sensor reading at the monitoring site.

[0064] The use of polymeric binder in the sensing material makes this system suitable for large scale “reel-to-reel” fabrication of mixed-matrix composite NC.sub.PZ.sub.HPP.sub.CL FO sensors since the coating can be deposited directly from suspension in a continuous process. The sensor scheme developed in this work is relatively simple in its design and could be further simplified by using a nearly monochromatic optical measurement system with a light emitting diode as the source and photodiode as the detector.

REFERENCES

[0065] 1. Celia, M. A. Geological storage of captured carbon dioxide as a large-scale carbon mitigation option. Water Resour. Res. 2017, 53, 3527-3533. [0066] 2. Tanaka, Y.; Sawada, Y.; Tanase, D.; Tanaka, J.; Shiomi, S.; Kasukawa, T.; Tomakomai CCS demonstration project of Japan, CO.sub.2 injection in process. Energy Procedia 2017, 114, 5836-5846. [0067] 3. https://www.netl.doe.gov/coal/carbon-storage/advanced-storage-r-d/monitoring-verification-accounting-and-assessment. March 2020. [0068] 4. Yang, C.; Jamison, K.; Xue, L.; Dai, Z.; Hovorka, S. D.; Fredin, L.; Trevino, R. H. Quantitative assessment of soil CO.sub.2 concentration and stable carbon isotope for leakage detection at geological carbon sequestration sites. Greenh. Gases: Sci. Technol. 2017, 7, 680-691. [0069] 5. Humez, P.; Lions, J.; Négrel, P.; Lagneau, V. CO.sub.2 intrusion in freshwater aquifers: Review of geochemical tracers and monitoring tools, classical uses and innovative approaches. Appl. Geochem. 2014, 46, 95-108. [0070] 6. Yang, C.; Delgado-Alonso, J.; Hovorka, S.; Mickler, P.; Trevino, R.; Phillips, S. Monitoring dissolved CO.sub.2 in groundwater for CO.sub.2 leakage detection in a shallow aquifer. Energy Procedia 2014, 63, 4209-4215. [0071] 7. Sun, Y.; Tong, C.; Trainor-Guitton, W. J.; Lu, C.; Mansoor, K.; Carroll, S. A. Global sampling for integrating physics-specific subsystems and quantifying uncertainties of CO.sub.2 geological sequestration. Int. J. Green Gas Con. 2013, 12, 108-123. [0072] 8. Carroll, S.; Hao, Y.; Aines, R. Geochemical detection of carbon dioxide in dilute aquifers. Geochem. Trans. 2009, 10, 4-22. [0073] 9. Athavale, R.; Pankratova, N.; Dinkel, C.; Bakker, E.; Wehrli, B.; Brand, A. Fast potentiometric CO.sub.2 sensor for high-resolution in situ measurements in fresh water systems. Environ. Sci. Technol. 2018, 52, 11259-11266. [0074] 10. Dunk, R. M.; Peltzer, E. T.; Walz, P. M.; Brewer, P. G. Seeing a deep ocean CO.sub.2 enrichment experiment in a new light: Laser Raman detection of dissolved CO.sub.2 in seawater. Environ. Sci. Technol. 2005, 39, 9630-9636. [0075] 11. Martin, P. E.; Barker, E. F. The infrared absorption spectrum of carbon dioxide. Phys. Rev. 1932, 41, 291-303. [0076] 12. Wang, Y.; Nakayama, M.; Watanabe, K.; Yagi, M.; Nishikawa, M.; Fukunaga, M. The NDIR CO.sub.2 monitor with smart interface for global networking. IEEE Trans. Instrum. Meas. 2005, 54, 1634-1639. [0077] 13. Rohwedder, J. J. R.; Pasquini, C.; Fortes, P. R.; Raimundo, I. M.; Wilk, A.; Mizaikoff, B. iHWG-μNIR: A miniaturised near-infrared gas sensor based on substrate-integrated hollow waveguides coupled to a micro-NIR-spectrophotometer. Analyst 2014, 139, 3572-3576. [0078] 14. Wang, X.-D. Wolfbeis, O. S. Fiber-optic chemical sensors and biosensors (2008-2012). Anal. Chem. 2013, 85, 487-508. [0079] 15. Jin, W.; Cao, Y.; Yang, F.; Ho, H. L. Ultra-sensitive all-fibre photothermal spectroscopy with large dynamic range. Nat. Commun. 2015, 6, 6767. [0080] 16. Caucheteur, C.; Guo, T.; Liu, F.; Guan, B.-O.; Albert, J. Ultrasensitive plasmonic sensing in air using optical fibre spectral combs. Nat. Commun. 2016, 7, 13371. [0081] 17. Wang, X.-D. Wolfbeis, O. S. Fiber-optic chemical sensors and biosensors (2013-2015). Anal. Chem. 2016, 88, 203-227. [0082] 18. Sun, Y.; Liu, J.; Xue, Z.; Li, Q.; Fan, C.; Zhang, X. A critical review of distributed fiber optic sensing for real-time monitoring geologic CO.sub.2 sequestration. J. Nat. Gas Sci. Eng. 2021, 88, 103751. [0083] 19. Sun, Y.; Xue, Z.; Park, H.; Hashimoto, T.; Zhang, Y. Optical sensing of CO.sub.2 geological storage using distributed fiber-optic sensor: From laboratory to field-scale demonstrations. Energy Fuels 2021, 35, 659-669. [0084] 20. Bao, B.; Melo, L.; Davies, B.; Fadaei, H.; Sinton, D.; Wild. P. Detecting supercritical CO.sub.2 in brine at sequestration pressure with an optical fiber sensor. Environ. Sci. Technol. 2013, 47, 306-313. [0085] 21. Joe, H.-E.; Zhou, F.; Yun, S.-T.; Jun, M. B. G. Detection and quantification of underground CO.sub.2 leakage into the soil using a fiber-optic sensor. Opt. Fiber Technol. 2020, 60 102375. [0086] 22. Kim, K.-J.; Lu, P.; Culp, J. T.; Ohodnicki, P. R. Metal-organic framework thin film coated optical fiber sensors: A novel waveguide-based chemical sensing platform. ACS Sens. 2018, 3, 386-394. [0087] 23. Chong, X.; Kim, K.-J.; Li, E.; Zhang, Y.; Ohodnicki, P. R.; Chang, C.-H.; Wang, A. X. Near-infrared absorption gas sensing with metal-organic framework on optical fibers. Sens. Actuators, B 2016, 232, 43-51. [0088] 24. Nazari, M.; Forouzandeh, M. A.; Divarathne, C. M.; Sidiroglou, F.; Martinez, M. R.; Konstas, K.; Muir, B. W.; Hill, A. J.; Duke, M. C.; Hill, M. R.; Collins, S. F. UiO-66 MOF end-face-coated optical fiber in aqueous contaminant detection. Opt. Lett. 2016, 41, 1696-1699. [0089] 25. Kim, K.-J.; Ellis, J. E.; Howard, B. H.; Ohodnicki, P. R. Centimeter-scale pillared-layer metal-organic framework thin films mediated by hydroxy double salt intermediates for CO.sub.2 sensor applications.

[0090] ACS Appl. Mater. Interfaces 2021, 13, 2062-2071. [0091] 26. Yang, C.; Hovorka, S. D.; Delgado-Alonso, J.; Mickler, P. J.; Trevino, R. H.; Phillips, S. Field demonstration of CO.sub.2 leakage detection in potable aquifers with a pulselike CO.sub.2-release test. Environ. Sci.

[0092] Technol. 2014, 48, 14031-14040. [0093] 27. Zamora, M. L.; Xiong, F.; Gentner, D.; Kerkez, B.; Kohrman-Glaser, J.; Koehler, K. Field and laboratory evaluations of the low-cost plantower particulate matter sensor. Environ. Sci. Technol. 2019, 53, 838-849. [0094] 28. Smith, A. D.; Elgammal, K.; Fan, X.; Lemme, M. C.; Delin, A.; Risander, M.; Bergqvist, L.; Schröder, S.; Fischer, A. C.; Niklaus, F.; Ostling, M. Graphene-based CO.sub.2 sensing and its cross-sensitivity with humidity. RSC Adv. 2017, 7, 22329-22339. [0095] 29. Sahner, K.; Hagen, G.; Schönauer, D.; Reiß, S.; Moos, R. Zeolites—Versatile materials for gas sensors. Solid State Ion. 2008, 179, 2416-2423. [0096] 30. Xu, X.; Wang, J.; Long, Y. Zeolite-based materials for gas sensors. Sensors 2006, 6, 1751-1764. [0097] 31. Zhang, J.; Dong, J.; Luo, M.; Xiao, H.; Murad, S.; Normann, R. A. Zeolite-fiber integrated optical chemical sensors for detection of dissolved organics in water. Langmuir 2005, 21, 19, 8609-8612. [0098] 32. Xiao, H.; Zhang, J.; Dong, J.; Luo, M.; Lee, R.; Romero, V. Synthesis of MFI zeolite films on optical fibers for detection of chemical vapors. Opt. Lett. 2005, 30, 1270-1272. [0099] 33. Yang, R.; Xu, Z.; Zeng, S.; Jing, W.; Trontz, A.; Dong, J. A fiber optic interferometric sensor platform for determining gas diffusivity in zeolite films. Sensors 2018, 18, 1090. [0100] 34. Tang, X.; Provenzano, J.; Xu, Z.; Dong, J.; Duan, H.; Xiao, H. Acidic ZSM-5 zeolite-coated long period fiber grating for optical sensing of ammonia. J. Mater. Chem. 2011,21, 181-186. [0101] 35. Mayer, K. M.; Hafner, J. H. Localized surface plasmon resonance sensors. Chem. Rev. 2011, 111, 3828-3857. [0102] 36. Tittl, A.; Giessen, H.; Liu, N. Plasmonic gas and chemical sensing. Nanophotonics 2014, 3, 157-180. [0103] 37. Pohl, T.; Sterl, F.; Strohfeldt, N.; Giessen, H. Optical carbon dioxide detection in the visible down to the single digit ppm range using plasmonic perfect absorbers. ACS Sens. 2020, 5, 2628-2635. [0104] 38. Herkert, E.; Sterl, F.; Strohfeldt, N.; Walter, R.; Giessen, H. Low-cost hydrogen sensor in the ppm range with purely optical readout. ACS Sens. 2020, 5, 978-983. [0105] 39. Nugroho, F. A. A.; Darmadi, I.; Cusinato, L.; Susarrey-Arce, A.; Schreuders, H.; Bannenberg, L. J.; da Silva Fanta, A. B.; Kadkhodazadeh, S.; Wagner, J. B.; Antosiewicz, T. J.; Hellman, A.; Zhdanov, V. P.; Dam, B.; Langhammer, C. Metal-polymer hybrid nanomaterials for plasmonic ultrafast hydrogen detection. Nat. Mater. 2019, 18, 489-495. [0106] 40. Vandezande, W.; Janssen, K. P. F.; Delport, F.; Ameloot, R.; De Vos, D. E.; Lammertyn, J.; Roeffaers, M. B. J. Parts per million detection of alcohol vapors via metal organic framework functionalized surface plasmon resonance sensors. Anal. Chem. 2017, 89, 4480-4487. [0107] 41. Garcia, G.; Buonsanti, R.; Runnerstrom, E. L.; Mendelsberg, R. J.; Llordes, A.; Anders, A.; Richardson, T. J.; Milliron, D. J. Dynamically modulating the surface plasmon resonance of doped semiconductor nanocrystals. Nano Lett. 2011, 11, 4415-4420. [0108] 42. Kanehara, M.; Koike, H.; Yoshinaga, T.; Teranishi, T. Indium tin oxide nanoparticles with compositionally tunable surface plasmon resonance frequencies in the near-IR region. J. Am. Chem. Soc. 2009, 131, 17736-17737. [0109] 43. Lai, W.-C.; Chakravarty, S.; Wang, X.; Lin, C.; Chen, R. T. On-chip methane sensing by near-IR absorption signatures in a photonic crystal slot waveguide. Opt. Lett. 2011, 36, 984-986. [0110] 44. Robinson, J. T.; Chen, L.; Lipson, M. On-chip gas detection in silicon optical microcavities. Opt. Express, 2008, 16, 4296-4301.