Acidic ZSM-5 zeolite-coated long period fiber grating for optical sensing of ammonia

Abstract

A ZSM-5 zeolite thin film with Si/Al ratio of about 23 has been grown on a long period fiber grating (LPFG) for optical gas sensing by monitoring its resonance wavelength (λ_R) shift caused by molecular sorption into the zeolite cavity. The sensing selectivity, sensitivity and speed of response of the zeolite-coated LPFG (Z-LPFG) are determined by the adsorption equilibria and transport properties of the analyte molecules in the zeolite pores. The ZSM-5 zeolite was modified through ammonium ion exchange and subsequent calcination to form acidic ZSM-5 (H-ZSM-5). The surface acidified Z-LPFG (HZ-LPFG) achieved dramatically improved sensitivity and selectivity for detecting ammonia gas. Also, increasing operation temperature improves sensing selectivity for the strongly adsorbing ammonia over weakly adsorbing gases and enhances response speed but compromises detection sensitivity due to reduced amount of adsorption for ammonia.

---

Introduction

Zeolites are aluminosilicate crystals composed of TO₄ (T = Si⁴⁺, Al³⁺) tetrahedral primary units which are interconnected by oxygen atoms to form ordered micropores defined by O–T–O rings. Conventionally, zeolites have been used as gas adsorbents, membranes, and shape-selective catalysts for molecular separations and catalytic reactions. In recent years, research has revealed that zeolites also possess a number of sorbate-dependent optical properties, such as UV–vis–IR absorption spectra, fluorescence, refractive index, birefringence, and Raman spectra.^1–5 With the enormous surface-to-mass ratio for analyte concentration and excellent thermal and chemical stabilities, zeolites are well suited for sensing chemicals through monitoring their optical properties which vary upon loading of analyte species. However, the realization of zeolite-based optical chemical probes has been hindered by the challenges in integrating zeolites with optical devices and the lack of in depth understanding on the sorbate-dependent optical behaviors.

Recently, we successfully integrated zeolites with fiber optic devices by growing highly siliceous MFI-type zeolite thin films on the straight-cut fiber endface and the surface of long period fiber grating (LPFG).^6,7 These zeolite-fiber optical devices were demonstrated for sensing chemicals based on the change of zeolite optical refractive index (n_z) caused by adsorption of analyte molecules from the environment.^8,9 The LPFG is an index module created by inscribing periodic index perturbations into the core of an optical fiber using laser irradiation. As lights travel through the LPFG, some particular wavelengths, which is termed as resonant wavelength (λ_R), are coupled from lossless core mode to high-loss cladding modes, resulting in a series of attenuation bands in the transmission spectrum.^10–13 When the LPFG is coated with a thin film, its λ_R shifts when the refractive index of the overcoat (n_ov) changes. Thus optical chemical sensors can be constructed by coating thin films of chemically sensitive photonic materials on the LPFG.^6,14–16 The relationship between λ_R and n_ov is described by¹⁷

where n_cl,eff is the effective index of the cladding including the coated film. For a band in the normal cladding mode (n_ov < n_cl), increasing n_cl,eff decreases the coupling wavelength (dλ_R/dn_cl,eff < 0)^13,18 and λ_R exhibits a sharp blue shift as n_ov increases to approach n_cl, e.g. when ∼1.3 < n_ov < n_cl (n_cl ≈ 1.46 for fused silica fiber).¹⁹ The cladding mode LPFG are theoretically capable of detecting a n_ov variation of ∼10⁻⁹ when n_ov is in the close vicinity of n_cl.^15,18,19

The refractive indices of the microporous zeolite crystals (n_z) are smaller than that of the dense silica cladding, n_cl. The n_z increases with increasing the load of sorbate molecules in the zeolitic cavities.^5,8,20 Therefore, the zeolite coated LPFG (Z-LPFG) can measure the concentration of adsorbing gases by monitoring the shift of λ_R in the normal cladding mode. In our previous studies, the λ_R of the pure-silica MFI zeolite coated LPFG shifted ∼1.45 nm and ∼ 0.7 nm when contacting 5.5 ppmv isopropanol vapor and 0.22 ppmv toluene vapor, respectively.⁶ The magnitude of these responses in λ_R promises high detection sensitivity because wavelength displacement as small as 0.001 nm can be measured by commercial optical spectrum analyzers or tunable lasers.

It has been also observed that the response time of the Z-LPFG in chemical sensing coincides with the equilibrating time for the molecular adsorption,^7,21 which is controlled by the sorption and desorption kinetics as well as the molecular diffusivity in the zeolite pores. Because the adsorption-induced change of zeolite refractive index cannot reveal the identity of adsorbed molecules, the sensing selectivity/specificity of the Z-LPFG relies on the gas adsorption selectivity in the zeolite film. In this research we synthesize a ZSM-5 zeolite film on the LPFG and investigate the effects of zeolite surface acidification and operating temperature on its performance in ammonia gas sensing.

Experimental

Zeolite film synthesis

The LPFG was obtained using point-by-point CO₂-laser irradiation on the Corning fused silica SMF-28 single mode fiber (refractive index 1.4682).¹¹ The grating section was 5-cm long with a 520-μm grating period (Λ). The ZSM-5 zeolite thin film was grown on the 125 μm-diameter LPFG fiber surface by in situ hydrothermal crystallization in a precursor solution with a nominal Si/Al atomic ratio of 15. The synthesis solution was a clear solution derived by mixing 11.3 ml of tetrapropylammonium hydroxide (TPAOH, 1M, Sigma-Aldrich), 19.2 ml of tetraethyl orthosilicate (TEOS, 98%, Acros), 60 ml of DI water and 0.458 g of NaAlO₂ (Sigma-Aldrich). TPAOH is the structure directing agent (SDA). The apparatus for zeolite film synthesis on the LPFG has been described in a previous publication.⁹ The hydrothermal synthesis was conducted at 180 °C under autogeneous pressure for 4 h. After the hydrothermal reaction, the zeolite-coated LPFG was rinsed thoroughly with DI water and then dried and calcined in air at 500 °C for 2 h to remove the SDA from the zeolite channels. The entire process of zeolite film growth was monitored online by continuously scanning the LPFG transmission spectrum† to ensure that the optical function of the zeolite-coated LPFG is preserved and the zeolite film is thick enough to avoid thickness influence.

Zeolite modification and characterization

The resultant ZSM-5 film on the fiber surface was examined by field emission scanning electron microscopy (FESEM) and the actual elemental composition of the zeolite thin film was determined by energy dispersive spectroscopy (EDS). Adsorption isotherms of NH₃, CO₂, and CH₄ were measured by a Micromeritics ASAP 2020 unit for the ZSM-5 zeolite particles collected from the residual liquid in the synthesis vessel. Both the zeolite-coated LPFG and ZSM-5 particles were treated in a 0.1 M NH₄Cl solution at 80 °C for 60 min after SDA removal to exchange the extraframework Na⁺ ions with NH₄⁺. The ion exchanged ZSM-5 film and particles were then dried and calcined in air at 350 °C to convert the [≡Si–O⁻(NH⁺₄)] to [≡Si–O⁻(H⁺)] acidic sites in the zeolite structure.

Optical gas sensing

The transmission spectrum and λ_R of the Z-LPFG were measured by an apparatus same as that reported in previous works.⁶ The zeolite-coated LPFG segment of the fiber was mounted in a 4-mm-ID stainless steel tube (total chamber volume ∼9.6 cm³) which was placed in a temperature-programmable tubular furnace. The transmission spectrum covers a near IR wavelength range of 1510–1640 nm. The source light was provided by a tunable laser equipped with a laser power detector (Agilent 8164A) and a computer data acquisition system. In operation, one end of the Z-LPFG connects to the tunable laser (light source) and the other end (output) connects to the laser power meter. The output light intensity was measured point-by-point by the power meter to obtain the transmission spectrum and the λ_R value. A wavelength increment of 1 nm and a dwelling time of 1 s were used in the gas sensing experiments. In all measurements, the total flow rate of the sample gas was kept at a small value of 10 cm³ (STP)/min for maintaining temperature stability and avoiding mechanical disturbance to the suspended fiber sensor.

In the gas sensing test, N₂ was used as the carrier gas and the calibrating baseline of the sensor's optical response was measured in pure N₂ in a temperature range of 25–350 °C. The baseline λ_R,N2 is given as a function of temperature, (λ_R,N2)_T = λ⁰_R,N2 + ζT, because the λ_R of a LPFG is known to exhibit excellent linear dependence on temperature. The ζ values were typically around 0.11 nm/° C with slight variations among different Z-LPFGs. Thus, the gas sensing signal at a specific temperature T, which is the shift of λ_R in response to switching from N₂ to an analyte gas “i” (Δλ_R,i)_T, is given by:

(Δλ_R,i)_T = (λ_R,i)_T − (λ_R,N2)_T (2)

where (λ_R,i)_T is the resonant wavelength in gas i at temperature T. (λ_R,i)_T is determined based on the transmission spectrum by a simple computer program and (λ_R,N2)_T is calculated by the predetermined calibrating baseline equation at T. The response speed of the Z-LPFG in gas sensing was studied by measuring the single wavelength intensity as a function of time. The response time is represented by the time required for stabilizing the output light intensity at the fixed wavelength. The specific single wavelength is chosen at each temperature for high sensitivity.†

The Δλ_R of the Z-LPFG was measured for N₂-carried single gases including NH₃ and CO₂ to evaluate the sensing selectivity for NH₃ over CO₂. Other gases such as H₂, CH₄, CO, H₂S and H₂O were also tested as they are common interfering molecules in potential applications like biomass-derived syngas and environments of food storage and animal farms. The ideal sensing selectivity of NH₃ against a gas i, α_NH3/i, is defined as

where Δλ_R,NH3 and Δλ_R,i are measured at same temperature and same volume concentration under atmospheric pressure. The zeolites' gas adsorption selectivity of NH₃ over CO₂ and CH₄, β_NH3/i (i = CO₂ and CH₄), was obtained by measuring the adsorption isotherms using a Micromeritics ASAP™ 2020 BET/Chemisorption Analyzer. β_NH3/i is calculated based on the amount of gas adsorption (ads_i) at same partial pressure:

Results and discussion

Morphology and chemistry of the zeolite film

The zeolite film had a typical morphology of ZSM-5 films with low Si/Al ratios in which the individual particles do not have smooth surface and well-defined orientation planes. Although the outer surface of the ZSM-5 films is microscopically rough, a 3–4 μm thick dense layer is clearly formed at the fiber surface as shown by the SEM images in Fig. 1. This film thickness ensures that the cladding modes are confined in the cladding so that the influence of film thickness variation on λ_R is negligible for different sensors.

Fig. 1 SEM images of fractured cross-section of a ZSM-5 zeolite coated LPFG fiber (Insert: high magnification showing zeolite/fiber interface).

The EDS test confirmed that both the zeolite film and particles collected from residual liquid had similar average Si/Al ratios of ∼22.8 (±1.5) and a Na⁺/Al³⁺ atomic ratios of ∼0.74 (±0.1). The Na⁺/Al³⁺ atomic ratio in the zeolite decreased to ∼0.1 after the NH₄⁺ ion exchange and firing. The MFI zeolite unit cell composition is M^m+_x/m[Si_96−xAl_xO₁₉₂] where M^m+ is the extraframework cation compensators, which are Na⁺ and H⁺ in the current ZSM-5 zeolites. Thus, in each unit cell, the as-synthesized ZSM-5 contained ∼3.0 [≡Si–O⁻(Na⁺)] sites and ∼1.0 [≡Si–O⁻(H⁺)] site while the modified HZSM-5 contained ∼0.4 [≡Si–O⁻(Na⁺)] sites and ∼3.6 [≡Si–O⁻(H⁺)] sites. The modified ZSM-5 film on LPFG and ZSM-5 zeolite particles had greatly increased numbers of acidic sites and hence are denoted hereafter as HZ-LPFG and HZSM-5, respectively.

Z-LPFG and HZ-LPFG responses to NH₃ and CO₂

The Δλ_R of the zeolite-coated LPFG was tested before and after the modification for their optical responses to NH₃ and CO₂ at room temperature (23 °C) and atmospheric total pressure. Fig. 2 (a) presents Δλ_R,NH3 andΔλ_R,CO2 as a function of the gas concentration in N₂ and Fig. 2 (b) presents the NH₃/CO₂ sensing selectivity, α_NH3/CO2, and zeolite gas adsorption selectivity, β_NH3/CO2. The refractive index of the porous MFI zeolite crystal is about 1.34 (<n_cl ≈ 1.46) in the non-adsorbing N₂ gas and increases with increasing the amount of gas adsorption.⁸ Thus, blue shifts of λ_R, i.e.Δλ_R < 0, occur when the zeolite film adsorbs NH₃ and CO₂.

Fig. 2 Effect of zeolite acidification on (a) Δλ_R in response to NH₃ and CO₂ and (b) NH₃/CO₂ sensing selectivity at 23 °C and adsorption selectivity at 35 °C under a total pressure of p_o = 101.3 kPa.

At same volume concentration, Δλ_R,NH3 of the Z-LPFG was greater than Δλ_R,CO2 because the adsorbance of NH₃ is larger than that of CO₂. In Fig. 2 (b), the sensing selectivity α_NH3/CO2 followed the trend of the adsorption selectivity β_NH3/CO2, which decreased with increasing gas concentration. Both α_NH3/CO2 and β_NH3/CO2 increase sharply as the gas concentrations approach the lower end. At low concentration, the NH₃ adsorption is dominated by strong chemisorption, [≡Si–O⁻(H⁺)]+NH₃ → [≡Si–O⁻(NH⁺₄)], and the NH₃ adsorbance is much greater than that of CO₂ which is physically adsorbed in the zeolites. The HZSM-5 zeolite particles and HZ-LPFG have larger populations of acidic sites and therefore exhibit higher β_NH3/CO2 and α_NH3/CO2 than the unmodified ZSM-5 particles and Z-LPFG. The β_NH3/CO2 and α_NH3/CO2 decrease with increasing NH₃ and CO₂ gas concentrations because of the increases in the less selective physisorption with increasing adsorbate partial pressure.

Effect of temperature on sensor performance

The temperature-dependence of gas adsorption in zeolites varies with the adsorbing strength. Adsorption isotherms were measured for NH₃ and CO₂ in a temperature range of 35 to 200 °C.†Fig. 3 presents the amounts of adsorption of NH₃ and CO₂ and adsorption selectivity β_NH3/CO2 in ZSM-5 and HZSM-5 zeolites as functions of temperature at a fixed pressure of ∼0.05 bar for both NH₃ and CO₂.

Fig. 3 Temperature dependences of amount of adsorption (ads) and β_NH3/CO2 for NH₃ and CO₂ in the ZSM-5 and HZSM-5 zeolites.

Distinct effects of zeolite modification were obtained on the adsorption of NH₃ and CO₂. At low temperature, physical adsorptions of NH₃ is significant in both the ZSM-5 and HZSM-5 zeolites and the total amounts of NH₃ adsorption in ZSM-5 and HZSM-5 are almost the same. As temperature increases, the NH₃ physisorption diminishes and the NH₃ chemisorption remains strong. Hence the HZSM-5 zeolite adsorbs a larger amount of NH₃ than the ZSM-5 zeolite at elevated temperature. In contrast, the extraframework alkali metal ions enhance the adsorption of CO₂ gas. Thus, replacing the Na⁺ with H⁺ in the zeolite reduces the adsorption of the acidic CO₂. The relatively weak adsorption of CO₂ diminished to nearly zero at 110–150 °C in the HZSM-5 and at ∼200 °C in the ZSM-5 zeolite, respectively. Meanwhile, the chemisorption of NH₃ remained significant in both zeolites at 200 °C, namely 13.4 cm³(STP)/g in ZSM-5 and 18.32 cm³(STP)/g in HZSM-5.

Therefore, the modification of zeolite enhances the NH₃ chemisorption while reducing the CO₂ adsorption that, in combination, dramatically increased the β_NH3/CO2 in the HZSM-5, especially at elevated temperatures. As shown in Fig. 3, β_NH3/CO2 values are 4.1 in ZSM-5 and 9.1 in HZSM-5 at 35 °C which increased to 9.1 in ZSM-5 and 156.2 in HZSM-5 at 150 °C. The gas adsorption data indicate that the sensing selectivity for NH₃ over CO₂ can be significantly improved by the surface modification together with proper selection of operation temperature.

Fig. 4 shows the experimental Δλ_R for the Z-LPFG and HZ-LPFG in response to 5% CO₂ and 5% NH₃ measured at atmospheric total pressure as a function of temperature. The Z-LPFG response to CO₂ virtually disappeared (i.e.Δλ_R,CO2 ≈ 0 nm) at 140 °C where Δλ_R,NH3 was −0.35 nm. The HZ-LPFG reached a Δλ_R,CO2 of ∼0 nm at 50 °C where its Δλ_R,NH3 was as large as −2.9 nm. These demonstrate that the sensing selectivity α_NH3/CO2 is significantly enhanced in the HZ-LPFG as compared to the unmodified Z-LPFG. The absolute deviation of Δλ_R measurement was within ±0.05 nm based on three times of independent measurements. The HZ-LPFG optical response to the 5% NH₃ was found to disappear, i.e.Δλ_R,NH3 approaching zero, at around 300 °C. It should be noted that Δλ_R of CO₂ reaches zero at a temperature well below the temperature where CO₂ adsorption becomes negligible because the baseline λ_R was measured in N₂ but not in vacuum.

Fig. 4 The Z-LPFG and HZ-LPFG responses to 5vol% CO₂ and 5vol% NH₃ carried in N₂ as a function of temperature.

Sensor response to common small molecules

We also measured the adsorption isotherms of CH₄ in the ZSM-5 and HZSM-5 zeolites† with the results for P_CH4/P_o at 0.05 shown in Fig. 5. The zeolite modification had a minimal influence on CH₄ adsorption because the spherical and nonpolar CH₄ molecule is not sensitive to the surface acidity or extraframework ions in the temperature range of concern here. The adsorption selectivity of NH₃ over CH₄, β_NH3/CH4, is much greater than β_NH3/CO2 in the temperature range of 35–150 °C because the adsorbance of CH₄ in the ZSM-5 zeolites is smaller than that of CO₂. The β_NH3/CH4 value was 53.8 in ZSM-5 and 56.7 in HZSM-5 at 35 °C. The β_NH3/CH4 increased to 240.7 and to 373.4 in ZSM-5 and HZSM-5, respectively, at 150 °C.

Fig. 5 Temperature dependences of amount of NH₃ adsorption and β_NH3/CH4 in ZSM-5 and HZSM-5 zeolites.

This observation can be explained by the fact that NH₃ adsorption is similar in ZSM-5 and HZSM-5 at 35 °C where the total amounts of adsorption are both high due to physisorption; the NH₃ adsorbance becomes largely different in the HZSM-5 and ZSM-5 at 150 °C where the adsorbance is dominated by the highly selective chemisorption with negligible physisorption. Thus, after modification, the enhancement of β_NH3/CH4 is insignificant at low temperature but becomes significant at high temperature because the temperature-dependence of CH₄ adsorption is not affected by the modification. The large value of β_NH3/CH4 at 35 °C suggests the HZ-LPFG to be insensitive to CH₄ and other very weakly adsorbing small gases as compared to CO₂. In fact, the HZ-LPFG showed almost no response (i.e.Δλ_R = 0 ± 0.05nm) to H₂, CO, and CH₄ even at room temperature when concentrations of these gases are below 5% as shown in Fig. 6.

Fig. 6 Response of HZ-LPFG to H₂, CO, CH₄ and NH₃ at 23 °C and atmospheric pressure.

Additional optical sensing experiments were performed for the HZ-LPFG with 10% CO₂, 3% H₂S, and 3% H₂O in N₂ at atmospheric pressure. The results are presented in Fig. 7. The responses to the 10% CO₂ and 3% H₂S are similar, both reached the baseline (i.e.Δλ_R = 0 nm) at 50–65 °C. The response to the 3% H₂O vapor was found to be stronger than those to CO₂ and H₂S due to the hydrophilic zeolite surface but weaker than that to NH₃. The Δλ_R for the 3% H₂O vapor reached the baseline at ∼125 °C where the HZ-LPFG still exhibited a large Δλ_R,NH3 of ca. −1.0 nm to the 5% NH₃.

Fig. 7 The HZ-LPFG responses to 10vol% CO₂, 3vol% H₂S, 3vol% H₂O and 5vol% NH₃ in N₂ as function of temperature at atmospheric pressure.

Effect of temperature on response speed and sensitivity

The above results demonstrate that the NH₃ sensing selectivity by the HZ-LPFG can be greatly improved when operating at elevated temperatures. The operating temperature will also affect the speed of response and the detection sensitivity. The speed of response in gas sensing is expected to be faster at higher temperature because the endothermic desorption as well as diffusion of NH₃ molecules in the microporous zeolite are facilitated at high temperature. Meanwhile, the NH₃ adsorption in the zeolite is reduced at elevated temperature that decreases the magnitude of Δλ_R,NH3 to lower the sensitivity.

Single wavelength sensing experiments were performed to investigate the HZ-LPFG temporal response to switching sample gas flow between pure N₂ and a mixture of 612 ppmv NH₃ carried in N₂. The dynamic response of transmission light intensity to the switch of gas composition was monitored at a fixed wavelength. The fixed single wavelength was specifically selected for each temperature such that the variation of the transmission light intensity would be the same if same amount of Δλ_R occurs.† Therefore, the peak height of the transmission light intensity can qualitatively reflect the detection sensitivity. The experimental results are shown in Fig. 8.

Fig. 8 Temporal response of the HZ-LPFG when switching between pure N₂ and 612 ppmv NH₃ in N₂ at various temperatures.

Below 110 °C, the response time for switching from N₂ to the 612 ppmv NH₃, which involves exothermic NH₃ adsorption, is much shorter than the reverse operation, which involves the endothermic NH₃ desorption. At 22 °C, the output signal, i.e. the transmission light intensity, stabilized within 2 min after switching the sample gas from pure N₂ to the NH₃/N₂ mixture but the signal stabilization took ∼120 min for the reverse process, i.e. switching back to pure N₂ for regeneration. As temperature increased to 150 °C, the stabilization time had no obvious change for the NH₃ sensing (adsorbing) step but was dramatically shortened to ∼2.5 min for the regeneration (desorption) step. It should be noted that the response time includes nearly 2 min needed for the sample gas (flow rate 10 cm³/min) to purge the 9.6-cm³ tube hosting the fiber. The stabilization time is determined by the time needed for the zeolite film to equilibrate the gas adsorption/desorption in the new gas environment. The results indicate that, at high temperature, the shortening of response time for regeneration is due primarily to the facilitated desorption of NH₃ rather than the enhancement of NH₃ diffusion in the microporosity because the kinetic size of NH₃ (∼3Å) is much smaller than the MFI zeolite pore diameter (5.6Å).

It is also shown in Fig. 8 that the sensitivity of NH₃ detection measured by the difference in output light intensities between 612 ppmv NH₃ and pure N₂, decreases with increasing temperature which is consistent with results in Fig. 4. This decrease of sensitivity is a result of the reduced amount of NH₃ adsorption at elevated temperatures. Therefore, the enhancements of NH₃ sensing selectivity and response speed at high temperature are achieved at a cost of lowered detection sensitivity.

Conclusion

ZSM-5 zeolite films with a low Si/Al ratio of ∼23 have been synthesized on LPFG optical fibers. The zeolite-coated LPFG (Z-LPFG) offers high sensitivity for gas sensing by monitoring the shift of the resonant wavelength upon molecular adsorption in the zeolite. The sensing selectivity of the Z-LPFG relies on the selectivity of gas adsorption in the zeolite pores. Zeolite surface modification combined with temperature control provides an effective way to enhance the gas sensing selectivity and response speed. However, operation at high temperature compromises the sensitivity. While the sensing performance in multicomponent mixtures is yet to be investigated, this study shows the feasibility of using zeolites as effective optical chemical sensing materials for construction of fiber optic sensors. The present work demonstrates that the optical response of HZ-LPFG to NH₃ (Δλ_R,NH3) has excellent quantitative correlations with NH₃ concentration and temperature. Thus with temperature and concentration (C_NH3) calibration data, a mathematic model of Δλ_R,NH3 = f(T, C_NH3) can be readily established and built into the computer program for quantitative measurement. There exist a large number of zeolitic materials with distinct crystallographic structures, pore sizes, material chemistry, and capacities for surface modification. This offers an excellent opportunity to develop a new class of zeolite-enabled fiber optic chemical sensors for applications in many important areas such as energy production processes, environmental monitoring, and chemical and biological threat detection. The zeolite-fiber integrated device is also potentially useful for studying molecular sorption and diffusion mechanisms in zeolites because of its high sensitivity and temporal resolution.

Acknowledgements

This material is based upon work supported by the National Science Foundation (Grant CBET-0854203) and the U.S. Department of Energy NETL University Coal Research program (Grant No. DE-NT0008062).

References

1. K. Hoffmann and F. Marlow, In Handbook of Zeolite Science and Technology; S. M. Auerbach, K. A. Carrado, P. K. Dutta, ed.; Marcel Dekker: NewYork, 2003, p 921–949.
2. J. L. Meinershagen and T. J. Bein, J. Am. Chem. Soc., 1999, 121, 448.
3. M. E. Davis, Nature, 2002, 417, 813.
4. P. Kortunov, L. Heinke, M. Arnold, Y. Nedellec, D. J. Jones, J. Caro and J. Kärger, J. Am. Chem. Soc., 2007, 129, 8041.
5. R. J. Kavanagh, K. K. Iu and J. K. Thomas, Langmuir, 1992, 8, 3008.
6. J. Zhang, X. Tang, J. Dong, T. Wei and H. Xiao, Opt. Express, 2008, 16, 8317.
7. J. Zhang, J. Dong, M. Luo, H. Xiao, S. Murad and R. A. Normann, Langmuir, 2005, 21, 8609.
8. J. Zhang, M. Luo, H. Xiao and J. Dong, Chem. Mater., 2006, 18, 4.
9. J. Zhang, X. Tang, J. Dong, T. Wei and H. Xiao, Sens. Actuators, B, 2009, 135, 420.
10. A. M. Vengasarkar, P. J. Lemaire, J. B. Judkins, V. Bhatia, T. Erdogan and J. E. Sipe, J. Lightwave Technol., 1996, 14, 58.
11. Y. Li, T. Wei, J. A. Montoya, S. V. Saini, X. Lan, X. Tang, J. Dong and H. Xiao, Appl. Opt., 2008, 47, 5296.
12. T. Erdogan, J. Lightwave Technol., 1997, 15, 12774.
13. N. D. Rees, S. W. James, R. P. Tatam and G. J. Ashwell, Opt. Lett., 2002, 27, 686.
14. A. Cusano, P. Pilla, L. Contessa, A. Iadicicco, S. Campopiano, A. Cutolo, M. Giordano and G. Guerra, Appl. Phys. Lett., 2005, 87, 234105.
15. Z. Gu, Y. Xu and K. Gao, Opt. Lett., 2006, 31, 2405.
16. X. Tang, K. Remmel, X. Lan, J. Deng, H. Xiao and J. Dong, Anal. Chem., 2009, 81, 7844.
17. V. Bhatia, Opt. Express, 1999, 4, 457.
18. R. Hou, Z. Hassemlooy, A. Hassan, C. Lu and K. P. Dowker, Meas. Sci. Technol., 2001, 12, 1709.
19. Y. J. Rao, Opt. Lasers Eng., 1999, 31, 297.
20. U. Schemmert, J. Karger, C. Krause, R. A. Rakoczy and J. Weitkamp, Europhys. Lett., 1999, 46, 204.
21. U. Schemmert, J. Karger and J. Weitkamp, Microporous Mesoporous Mater., 1999, 32, 101.

---

Footnote

† Electronic supplementary information (ESI) available: Schematic illustration of the zeolite film coated LPFG. Experimental setup Z-LPFG test. LPFG transmission spectra and resonant wavelength as functions of film synthesis time. Figure showing selection of single wavelength for the test of sensing response. Adsorption isotherms of CO₂ and NH₃ in ZSM-5 and HZSM-5 at different temperatures. Adsorption isotherms for CH₄ and NH₃ in ZSM-5 and HZSM-5 zeolites at various temperatures. See DOI: 10.1039/c0jm02523b

---