Mesoporous supraparticles with a tailored solid–liquid–gas interface for visual indication of H₂ gas and NH₃ vapours

Abstract

Dual-gasochromic supraparticles that undergo rapid eye-readable and gas-specific colour changes upon reaction with hydrogen or ammonia are reported. This functionality is achieved by tailoring the solid–liquid–gas interface within the mesoporous framework of supraparticles via spray-drying.

---

Gasochromic materials that visually indicate the presence of toxic, flammable, or explosive gases through a change in their absorbance become increasingly relevant. Their optical readout facilitates remote and spatially resolved gas detection without a power supply. In addition, optical signal transduction makes an operation in explosive atmospheres possible by eliminating electrical contacts, i.e., the risk of sparks.¹ Hydrogen (H₂) gas and ammonia (NH₃) vapours are currently relevant as targets for gasochromic detectors. They are considered promising energy carriers to foster the decarbonisation of our energy system. However, both gases pose danger to humans due to the high flammability range of H₂/air mixtures¹ and the toxicity of NH₃.^2,3 Therefore, any unintended release during the production, transport, or usage of both gases must be immediately detected and the leakage precisely localized.

A gasochromic particle that can be incorporated as an additive into diverse materials and indicate the presence of both gases through rapid, eye-readable, and gas-specific colour changes at every point of use, i.e., a dual-gasochromic particle, has the potential to meet this demand.

Despite the large variety of gasochromic materials available for H₂ gas^4,5 and NH₃ vapours,^6–8 respectively, no dual-gasochromic particle for these two species has been reported to the best of our knowledge. It is important to mention that electrical dual-gas detectors for H₂ and NH₃⁹ or multi-gas detector arrays^7,10 were established but not in the form of one flexibly applicable, gasochromic particle. Most materials for visual H₂ detection are based on the activation of H₂ molecules at catalytically active noble metal surfaces and the subsequent reduction of organic¹¹ or inorganic chromophores.^4,12 NH₃ vapours are typically detected by impregnating organic pH indicator dyes into porous host materials and exploiting the basic nature of NH₃ as a proton acceptor.⁸ Both types of gasochromic materials incorporate functional chromophores into porous support materials. Recently, we introduced supraparticles (SPs) as a new class of (meso)porous support materials for gasochromic indicators.¹³ SPs are hierarchical assemblies of nanoparticle (NP) building blocks with a consistent structural motif,¹⁴ offering a highly customizable material design.^13,15–17 The established gasochromic SPs exploit (ir)reversible colour changes of incorporated Reichardt's betaine dye or resazurin (RES) for the detection of NH₃ vapours¹³ and H₂ gas,^15–18 respectively. However, the NH₃-indicator SPs cannot detect H₂, and vice versa.¹⁸ Thus, the development of a SP for the visual detection of both gases is still an open challenge.

Herein, we report a pioneering example of a dual-gasochromic SP for H₂ gas and NH₃ vapours. Aiming for such a SP, we elaborated a suitable material design that involves the selection of appropriate building blocks and the adjustment of the chemical microenvironment within the solid–liquid–gas interface, provided by the porous framework of the SP. We found that a suitable dye species must undergo distinguishable colour change reactions upon reduction and deprotonation, respectively. Furthermore, we demonstrate that a suitable microenvironment can be created by spray-drying of a mixed dispersion containing SiO₂ NPs, Pt NPs, the chosen dye molecules, and an acid (H₂SO₄), functioning as an additional key component. The resulting SPs achieve the targeted functionality through the interaction of H₂ with the incorporated Pt NPs and NH₃ with the acidic liquid phase within the gas-accessible pores of the SP. Both reactions at the solid–gas and liquid–gas interface induce a gas-specific colour change of the comprised dye molecules via reduction or deprotonation, respectively. From a broader perspective, the confined space within such mesoporous SPs provides a reaction environment similar to solution chemistry with the following advantages: exotic precursor combinations can be united almost independently of their surface-chemistry using forced assembly. The resulting SP powders can be easily incorporated as active material into sensor devices¹⁹ or as pigment in coatings and clothing.¹⁸

The first step to create a dual-gasochromic SP for H₂ gas and NH₃ vapours was the identification of suitable chromophore systems and reaction conditions. Therefore, we probed the colour response of various indicator dye solutions by mixing them with NH₃ solution or H₂ gas (Fig. S1, ESI†). The main findings can be summarized as follows: Pt NPs with a size of roughly 5–10 nm (dynamic light scattering, DLS, Fig. S2, ESI†) are required for the H₂-induced reduction of dye molecules by providing active hydrogen at their surface.¹¹ Many dyes require an acid (e.g., the presence of H₂SO₄) to realize a strong colour change upon reaction with NH₃ solution. An indicator dye species suitable for the envisaged SPs must undergo distinguishable colour responses upon deprotonation, induced by an increase in the pH value of its surrounding medium, and additionally also due to the reduction by active hydrogen. Among all tested indicator dyes (see ESI†), resazurin (RES) and methyl red (MR) are most suitable for the visual indication of both H₂ and NH₃.

In the second step, we created a SP that provides the required chemical microenvironment by forced assembly of the identified key elements, i.e., Pt NPs, dye molecules, H₂SO₄, and framework-building SiO₂ NPs (≈5–15 nm, DLS, Fig. S3, ESI†) via spray-drying (Fig. 1A1 and A2). In this process, the different building blocks are united in a mixed dispersion, fed through a nozzle and atomized into small droplets. These droplets enter a hot chamber, which induces the evaporation of the solvent thereby forcing the assembly of all components at the liquid–gas interface, yielding a SP entity. Spray-drying is a continuous, high-throughput and industrially established process, which makes its scalability possible.²⁰ Herein, we achieved a SP production capacity of 12 g h⁻¹ using a lab-scale spray dryer.

Fig. 1 Synthesis of SPs via spray-drying without (A1) and with the addition of H₂SO₄ (B1) and related SEM images of SPs without H₂SO₄ (A2-A4) and with H₂SO₄ (B2-B4): overview images (A2, B2), images of SP cross-sections using backscattered electron detection at different magnifications (A3-4, B3-4). Bright spots partially highlighted with coloured circles indicate the Pt NPs. N₂ ad-/desorption isotherms at 77 K (C). Full symbols indicate adsorption branch and empty symbols indicate desorption branch. Non-local density functional theory (NLDFT) pore size distribution (D). H₂O ad-/desorption isotherms at standard temperature (298 K) and pressure (STP) (E).

For a proof-of-concept study, we exemplary used RES as an indicator dye as it is well studied by others¹¹ and us.^15–18 We focused in our study on how the addition of H₂SO₄ to the mixed dispersion prior to spray-drying influences the morphology, structure, texture and functionality of the resulting SPs. The mixed dispersion of the reference SP system containing SiO₂ NPs, Pt NPs and RES had a pH ≈ 10 due to the NaOH-stabilisation of the SiO₂ NP dispersion. The other mixed dispersion had a pH ≈ 2 due to the addition of H₂SO₄. The performed spray-drying processes of the two mixed dispersions yielded homogenous, free-flowing powders with a purple and orange colour, respectively (Fig. 1A1 and B1). These colours correspond to the deprotonated (purple) and protonated species of RES (RES-H⁺, orange), respectively,²¹ matching the colours and UV-vis spectra of the aqueous solutions of RES (Fig. S1, ESI†). Scanning electron microscopy (SEM) imaging (Fig. 1A2 and B2) and laser diffraction measurements (Fig. S4, ESI†) revealed that neither the morphology nor the size distribution of the SPs are significantly altered by the addition of H₂SO₄. Both types of SPs show a spherical or doughnut-like morphology and a size distribution ranging from ≈2 to ≈8 μm. As the Pt NPs are essential for the detection of H₂,¹⁶ we studied their distribution within the SP framework via SEM imaging of SP cross-sections using back-scattered electron detecting to obtain elemental contrast. Without the addition of H₂SO₄, the Pt NPs are well dispersed throughout the entire SP (Fig. 1A3) and partially form small agglomerates (Fig. 1A4). In contrast, the addition of H₂SO₄ to the mixed dispersion results in the formation of larger Pt NPs agglomerates within the SP that, in turn, are distributed throughout the entire SP (Fig. 1B3 and B4). The enhanced agglomeration of the citric acid stabilised Pt NPs is attributed to the H₂SO₄-induced destabilisation of the colloidal dispersion.²²

Similarly, the interstitial pores between the assembled NPs of the SP framework are essential for the targeted functionality.¹⁵ The mesopores are intended to provide the required micro-environment within the solid–liquid–gas interface that consists of solid NPs, liquid pore water that is adsorbed in humid atmospheres¹⁵ and the surrounding gas atmosphere. Therefore, we investigate the effect of the addition of H₂SO₄ to the mixed dispersion prior to spray-drying on the textural properties and the water adsorption capacity of the resulting SPs. Details of the conducted N₂ ad-/desorption at 77 K and H₂O ad-/desorption measurements at 298 K are provided in the ESI† (Fig. S5), while the main findings can be summarized as follows: both SPs are mesoporous, which is indicated by their almost identically shaped type IV N₂ isotherms (Fig. 1C).²³ Their non-local density functional theory (NLDFT) pore size distribution is narrow (≈2–6 nm) with minor differences likely associated with the colloidal destabilization induced by H₂SO₄ (Fig. 1D). Additionally, the presence of H₂SO₄ in the SPs increases the total water adsorption capacity and causes a small shift of the relative pressure range at which water condensation occurs to smaller values (Fig. 1E). We attribute these changes mainly to the hygroscopic nature of H₂SO₄.²⁴

Next, we probed the gasochromic functionality of the two SP powder samples by exposing the SP powders to H₂ gas for ≈ 20 s and NH₃ vapours for ≈3 s, respectively – optionally, also after ex situ H₂O dosing in a climate chamber (1 h, 30 °C, 98% r.h.) to increase the amount of water in the pores.¹⁵ The SP powder without H₂SO₄ showed the previously reported rapid, eye-readable, two-step (ir)reversible colour change upon H₂ dosing (Fig. 2A1, after ex situ H₂O dosing).^15–17 First exposure to H₂ leads to the irreversible reduction of deprotonated RES (purple) to deprotonated resorufin (RF, pink), followed by a reversible reduction to hydroresorufin (hRF, colourless) upon further H₂ dosing (Fig. 2A2). The condensed water in the mesopores grants the mobility of the incorporated dye molecules. This dye mobility enables their migration to the surface of the Pt NPs, where they can be reduced by active hydrogen when H₂ gas is present.¹⁵ SP powder without ex situ H₂O dosing did not show a complete conversion to colourless hRF due to a lower dye mobility (Fig. S6, ESI†). However, as the RES molecules in SPs without H₂SO₄ are deprotonated, no colour change is observed upon dosing NH₃ vapours (Fig. 2A3).

In contrast, SP powder containing Pt NPs, SiO₂ NPs, RES and H₂SO₄ showed rapid, eye-readable, gas-specific colour change reactions upon exposure to H₂ or NH₃ vapours (Fig. 2B, more snapshots provided in Fig. S7, ESI†). First reaction with H₂ irreversibly reduced the orange RES-H⁺ to the slightly brighter RF-H⁺ (Fig. 2B1–B3). This is confirmed by distinct changes in their absorption and fluorescence spectra (Fig. S8, ESI†). Further H₂ dosing led to blue powder due to the formation of a scarcely reported stable radical state of hRF (hRF, Fig. 2B3).²¹ When the dye mobility was high enough, hRF· was further reduced by H₂ dosing to colourless hRF (Fig. 2B2). After stopping H₂ dosing, the colourless SP powder (hRF) turned blue (hRF), before reaching a final orange colour (RF-H⁺). It is important to mention that the colouration of these SPs occurred much slower (up to 180 s) compared to their reduction (≈5–10 s) and the colouration of SPs without H₂SO₄ (≈5–10 s). We attribute this slow reversibility to the stability of the blue hRF· radical in air.

Fig. 2 Gasochromic functionalities of SP powders without (A) and with H₂SO₄ added (B): snapshots before, during and after H₂ dosing of SPs after ex situ H₂O dosing (A1) and related reaction network (A2). Snapshots of the same SPs before, during and after NH₃ dosing (A3). Snapshots before, during and after H₂ dosing of SPs containing H₂SO₄ before (B1) and after ex situ H₂O dosing (B2) and related reaction network (B3). Snapshots of the same SPs before (B4) and after ex situ H₂O dosing (B5) before, during and after NH₃ dosing and related reaction network (B6).

In contrast to H₂ dosing, NH₃ vapours induced a rapid, visual colour change of SPs containing H₂SO₄ from orange to purple (Fig. 2B4 and B5) due to deprotonation of RES-H⁺ to RES (Fig. 2B6). This reaction is caused by a pH increase of the liquid phase in the mesopores of the SPs that is likely induced by the adsorption and dissolution of gaseous NH₃ species. The proposed mechanisms are supported by a control SP sample containing SiO₂ NPs, RES, H₂SO₄ but no Pt NPs, which showed no response to H₂ but an identical colour change to NH₃ dosing (Fig. S9, ESI†). The NH₃-induced colour change of the SP powders is partially irreversible, as the initially orange colour related to RES-H⁺ was never fully recovered (Fig. 2B4 and B5). Comparing the reversibility of the colour change upon repeated NH₃/air dosing revealed a slightly higher degree of reversibility for SPs after ex situ H₂O dosing compared to pristine SPs (Fig. S10, ESI†). The enhanced amount of water after H₂O dosing increases the amount of NH₃ required to alter the pH of the liquid phase in the mesopores.

Besides testing separate H₂ and NH₃ dosing, the SPs containing Pt NPs, SiO₂ NPs, RES and H₂SO₄ were subjected to simultaneous exposure to both target species. Details of this investigation are found in the ESI† (Fig. S11), while the main finding of this investigation can be summarized as follows: NH₃ poisons the incorporated Pt NPs,²⁵ which results in impaired H₂-induced reduction of the dye species, when both NH₃ and H₂ are present. Therefore, upon dosing of H₂ and NH₃, either pink RF is formed, when H₂ reacts with the SPs first, or deprotonated purple RES, when the SPs react first with NH₃.

Next, we studied the adjustability of the conceptualised SPs and synthesized SPs containing MR as indicator dye. SPs with MR also showed eye-readable, gas-specific colour change reactions upon exposure to H₂ gas or NH₃ vapours (Fig. S12, ESI†). Preliminary quantitative in situ UV-vis measurements (Fig. S13, ESI†) demonstrate that dual-gasochromic SPs carrying RES or MR also undergo strong colour changes when they are exposed to ≈2 vol% H₂ in N₂ (i.e., below the lower flammability limit of 4 vol%¹) or ≈ 620 ppm NH₃ in N₂ (i.e., below the limit at which human health issues occur^2,3).

In conclusion, we synthesised and characterised the first dual-gasochromic SPs that undergo rapid, eye-readable, gas-specific colour changes upon reaction with H₂ gas or NH₃ vapours. We revealed the parameters that affect the visual response and characteristics of the SPs. We demonstrated that only SPs assembled from SiO₂ NPs, Pt NPs, suitable dye species (RES or MR) and H₂SO₄, achieve the desired functionality. Thereby, the addition of H₂SO₄ caused just minor changes to the morphology and texture of the SPs but improved the water adsorption capability and ultimately made the desired two gas-specific colour changes via scarcely reported reactions possible. The functionality of the SPs results from the synergistic interplay of all four types of building blocks within the solid–liquid–gas interface of their mesoporous framework (Fig. S14, ESI†). This interplay includes the adsorption and condensation of water in the mesopores, the dissociation of H₂ at the surface of Pt NPs and the mobility of incorporated indicator dye molecules in the liquid phase. The later move through and react with the surface of the solid framework, e.g., via hydrogenation. Furthermore, we herein demonstrated the possibility of adjusting the pH of the liquid phase in the mesopores during SP synthesis via H₂SO₄ addition as well as afterward through the interaction with NH₃ vapours. The dual-gasochromic SPs are therefore a pioneering example for the emergence of new functionalities by exploiting the customisation of the solid–liquid–gas interface within the mesopores of a micron-sized SP.

The contribution of all authors according to the CRediT system is listed in the ESI.† All authors have approved the final version of the manuscript.

This work was financially supported by the BMBF (NanoMatFutur grant 03XP0149 and project IDcycLIB 03XP0393C). J. R. acknowledges his scholarship funding from the German Federal Environmental Foundation (DBU). The authors thank BÜCHI Labortechnik AG for providing the spray dryer equipment.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. T. Hübert, et al. , Sens. Actuators, B, 2011, 157, 329.
2. J. Pauluhn, Regul. Toxicol. Pharmacol., 2013, 66, 315.
3. L. Silverman, et al. , J. Ind. Hyg. Toxicol., 1949, 31, 74.
4. Y.-A. Lee, et al. , Sens. Actuators, B, 2017, 238, 111.
5. (a) H. G. Girma, et al. , ACS Sens., 2023, 8, 3004; (b) X. Sun, et al. , Appl. Surf. Sci., 2022, 599, 153878.
6. S. Sutthasupa, et al. , Food Chem., 2021, 362, 130151.
7. L. Engel, et al. , Sens. Actuators, B, 2021, 330, 129281.
8. A. T. Hoang, et al. , Sens. Actuators, B, 2016, 230, 250.
9. (a) L. Du, et al. , Sens. Actuators, B, 2023, 375, 132873; (b) D. D. Roy, et al. , Int. J. Hydrogen Energy, 2023, 48, 4931.
10. (a) S. H. Lim, et al. , Nat. Chem., 2009, 1, 562; (b) J. Chen, et al. , ACS Nano, 2018, 12, 6079; (c) J. Zhang, et al. , Sens. Actuators, B, 2021, 326, 128822; (d) Y. Zhang and L.-T. Lim, Sens. Actuators, B, 2018, 255, 3216.
11. M. E. Smith, et al. , Anal. Chem., 2020, 92, 10651.
12. S. Hwan Cho, et al. , Chem. Eng. J., 2022, 446, 136862.
13. S. Wintzheimer, et al. , Part. Part. Syst. Charact., 2019, 36, 1900254.
14. (a) S. Wintzheimer, et al. , ACS Nano, 2018, 12, 5093; (b) J. Reichstein, et al. , Adv. Mater., 2023, 35, e2306728.
15. J. Reichstein, et al. , Adv. Funct. Mater., 2022, 32, 2112379.
16. K. Zhang, et al. , Chem. Mater., 2023, 35, 6808.
17. K. Zhang, et al. , J. Chem. Phys., 2023, 158, 134722.
18. J. Reichstein, et al. , Adv. Mater. Technol., 2024, 2400441.
19. C. Pannek, et al. , Sens. Actuators, B, 2020, 306, 127572.
20. (a) D. P. Debecker, et al. , Chem. Soc. Rev., 2018, 47, 4112; (b) S. Wintzheimer, et al. , Adv. Mater., 2023, 35, e2306648.
21. S. Khazalpour and D. Nematollahi, RSC Adv., 2014, 4, 8431.
22. G. Marzun, et al. , Langmuir, 2014, 30, 11928.
23. M. Thommes, et al. , Pure Appl. Chem., 2015, 87, 1051.
24. K. B. Kiradjiev, et al. , Ind. Eng. Chem. Res., 2020, 59, 4802.
25. D. A. Finkelstein, et al. , J. Phys. Chem. C, 2015, 119, 9860.

---

Footnotes

† Electronic supplementary information (ESI) available: Details of experimental section, material characterization. See DOI: https://doi.org/10.1039/d4cc01247j

‡ These authors contributed equally to this work.

---