View PDF Version
 
DOI: 10.1039/D5LC00390C (Paper) Lab Chip, 2025, Advance Article

High frequency capillary wave-enabled ultra-small droplets for inhaled drug delivery†

Haitao Zhanga, Zirui Zhaoa, Yangchao Zhoua, Zaoxian Meib and Xuexin Duan*a
aState Key Laboratory of Precision Measuring Technology & Instruments, Tianjin University, Tianjin 300072, China. E-mail: xduan@tju.edu.cn
bTianjin Hospital, Tianjin 300211, China

Received 23rd April 2025 , Accepted 6th June 2025

First published on 11th June 2025

Abstract

Benefiting from localization, targeting and rapid response, inhaled drug delivery has become an indispensable method for treating lung diseases. However, the efficacy of drug delivery is often compromised by the physical characteristics of the aerosol produced by current nebulization methods, i.e., large droplet size distributions, which are deposited in the upper airways. In this study, a portable, low-energy, and low-cost approach to nebulize drugs with appropriate size distribution is introduced using capillary wave breakup induced by gigahertz (GHz) acoustic waves. A delicately designed miniaturized nebulizer is developed by integrating a GHz bulk acoustic resonator with a semi-open microchannel to nebulize droplets of optimal size for pulmonary inhalation, with size distributions, in which 96% are smaller than 5 μm at low power, which surpasses existing methods. In addition, this technique facilitates the nebulization of liquids with viscosities of up to 5000 cP. Low-flux lung models achieve 88% drug delivery efficiency. Murine in vivo tests demonstrate the efficacy of the proposed nebulizer in lung-targeted delivery via autonomous inhalation, which is attributed to optimized droplet size and flux. The tunable sizes, broad range of nebulization viscosities, suitable fluxes, pumpless operation, and low cost highlight the potential for autonomous lung drug delivery and combination therapy targeting both the small airways and alveoli.

1. Introduction

Inhaled drug delivery (IDD) has revolutionized the treatment of respiratory diseases, such as asthma and chronic obstructive pulmonary disease (COPD), by enabling targeted therapy with rapid onset and minimized systemic side effects.1–3 Unlike oral or parenteral routes, IDD enables direct deposition of therapeutics at the site of airway inflammation, bypassing first-pass metabolism and reducing required dosages by 10- to 20-fold.4,5 However, the efficacy of IDD critically depends on the aerosol droplet size distribution: droplets of >5 μm predominantly deposit in the upper airways via inertial impaction, while those smaller than 5 μm can penetrate deeper into the small airways (<2 mm diameter), which are increasingly recognized as key contributors to asthma exacerbations and disease progression.6–8 Clinical studies demonstrate that monodisperse aerosols of 1.5–3 μm significantly enhance bronchodilator response compared to polydisperse formulations, underscoring the need for precise aerodynamic control.9

Current nebulization technologies—including jet, vibrating mesh, and ultrasonic systems—face inherent limitations in achieving optimal droplet size distributions.10,11 Jet nebulizers generate broad size distributions with a mass median aerodynamic diameter (MMAD) of 3–7 μm and a <15% fine particle fraction (FPF, <5 μm),12 while ultrasonic methods suffer from thermal degradation and limited viscosity tolerance (<50 cP).13 Although vibrating mesh devices improve FPF to 40–60%, their narrow orifice structures are prone to clogging with high-viscosity drugs (e.g., biologics or suspensions).14–20 Furthermore, all existing systems require patient coordination during forced inhalation or gas assistance, leading to <30% lung deposition efficiency even under ideal conditions. These limitations necessitate frequent dosing, increased costs, and compromised patient compliance—a critical barrier for pediatric and geriatric populations.10,11

Recent advances in acoustic-driven nebulization have emerged as promising alternatives.21,22 Capillary wave atomization, induced by MHz-range surface acoustic waves (SAW), can produce sub-5 μm droplets through Faraday instability.23–27 For instance, Qi et al. demonstrated the production of 2.8 μm aerosols using 20 MHz SAW devices, achieving a FPF of 65% for insulin delivery.28 Similarly, bulk acoustic wave (BAW) resonators operating at 1–10 MHz have shown improved viscosity tolerance (up to 175 cP) by leveraging acoustic streaming.29 Nevertheless, these systems still face challenges: (1) limited energy focusing restricts further high-viscosity nebulization and droplet size reduction due to viscous dissipation. (2) Polydisperse output geometric standard deviation (GSD > 2) results from uncontrolled wave interference. (3) High power consumption (>5 W) makes them unsuitable for portable applications.10

Here, we propose a GHz BAW resonator-based nebulizer that overcomes these limitations through three key innovations. First, the GHz-frequency operation (1–3 GHz) generates an ultra-short attenuation length (l = 10 μm), enabling consistent production of submicron to 5 μm aerosols (GSD = 1.57) via deterministic wave breakup on the thinner liquid film height (H = 30 μm). Second, the micro electro mechanical systems-fabricated semi-open microchannel design eliminates pumping requirements by harnessing GHz-driven acoustic streaming (velocity >1 m s−1), allowing the nebulization of high-viscosity formulations (up to 5000 cP). Third, the pumpless architecture operates at low power (<1 W), enabling battery-powered portability. Through coupled acoustic simulations and in vitro/in vivo testing, we demonstrate that this platform achieves 88% drug delivery efficiency and 49.3% alveoli diffusion for deeper lung penetration compared to commercial nebulizers. Our technology not only overcomes the viscosity–size trade-off inherent in existing systems but also enables autonomous inhalation therapy—a paradigm shift for treating small airway diseases.

2. Materials and methods

2.1 Reagents

Ethanol was purchased from Tianjin Yuanli Chemical Technology. Octanol, glycol, olive oil, glycerol and rhodamine B were purchased from Heowns. Salbutamol was purchased from Shanghai Macklin. Nanoparticles (500 nm, green fluorescence) were purchased from Beijing Zhongkeleiming Daojin Technology and diluted with octanol. The solution was sonicated for 3 min to improve the monodispersity. Glycerol and deionized water were used to prepare glycerin solutions at different percentages.

2.2 Acoustic resonator fabrication

The GHz-acoustic resonator was designed and fabricated by a standard IC process.30 Silicon dioxide and molybdenum were deposited on the silicon as a Bragg reflection layer, above which was a bottom electrode–piezoelectric layer–top electrode sandwich structure, the materials of which were gold, aluminum nitride and gold, respectively. A silicon dioxide thin film with a thickness of 200 nm was deposited as the passive layer to protect the resonator. More details are provided in Fig. S1.† A miniaturized circuit was designed, and a portable power source was used for to provide energy to the whole nebulization system. The portable nebulizer is shown in Fig. S2.†

2.3 Aerosol characterization

The droplet size distributions were measured using laser diffraction (Spraytec, Malvern Instruments, UK), with droplets ranging between 0.1 and 2000 μm, covering the desired 0.1–5 μm range. The nebulization flow rate was determined according to a high-precision Harvard pump. In addition, a high-speed camera was used for the visualization of the whole nebulization process and to record the breaking process of the free capillary wave at the liquid surface on the side.

2.4 Lung model

A glass single-cut-point twin-stage impinger (TSI) was used as an in vitro model of the human pulmonary system, following established pharmaceutical standards.28,31,32 The impinger stages were filled with ethanol to collect and dissolve the salbutamol–octanol droplets, and an airflow of 60 L min−1 was drawn through the impinger during nebulization (Fig. S3†). At this flow rate, droplets with aerodynamic diameters >5.8 μm are expected to be deposited in the upper respiratory airway wall regions A and B—due to the dominance of inertia in their motion. In contrast, droplets with aerodynamic diameters of less than 5.8 μm are deposited in the lower respiratory airways, represented by subregions C1 and C2, primarily due to diffusion-driven molecular collision with lung tissue or gravitational sedimentation.33

The theoretical dose refers to the total volume of the drug solution delivered to the nebulizer. Droplets deposited in part A contribute to the upper respiratory dose, while droplets remaining in subzones C1 and C2 contribute to the lower respiratory dose. Consequently, the theoretical dose and the lower respiratory dose are the critical metrics in this study. They not only assess the efficiency of the nebulization drug delivery, but also gauge the potential of the device for pulmonary administration. In this experiment, a UV spectrophotometer (UV-2600i, Shimadzu Corporation) was employed to measure the doses for the upper and lower respiratory tract. The absorption peak values of salbutamol at various concentrations in ethanol solution were calibrated through pre-sample treatment (Fig. S3†). The absorption peaks were observed at 227 nm and 278 nm, and the absorption peak value is linear with the concentration of salbutamol. By fitting the curve, the salbutamol concentration in the upper and lower respiratory tracts can be determined based on the absorption peak values. For accuracy, absorbance measurements at 227 nm were preferred due to the low concentrations involved. In addition, to account for residual drug, samples were collected by cleaning different parts of the impinger with ethanol solution, which was then combined with the receiving solution and diluted to 20 mL in a measuring cylinder for the final absorbance measurement.

2.5 Mouse lung model

Inbred strain C57BL/6 (6–8 weeks) mice were purchased from Guosheng Zhengyuan Technology (Tianjin), and were used for studying drug deposition in the lung based on the proposed nebulizer. The proposed nebulizer has a suitable nebulization flow rate and does not require passive air flow to promote the absorption of droplets. Thus, the mice could autonomously inhale atomized droplets, which could reflect the effect of the drug more realistically. A mixture of rhodamine B and octanol was nebulized, and the mice were fixed on a board during the process. The mice were euthanized by isoflurane overdose at the indicated time point. The lungs of the mice were dissected and placed in a fully automated fluorescence imaging analysis system (ChampChemi) for tissue imaging. Subsequently, the lungs of the mice were removed and quick-frozen to prevent postmortem tissue changes that would affect the later staining results. OCT embedding agent was used to immerse the tissue, and the embedded tissue was placed in a small cup filled with liquid nitrogen or dry ice, and then fixed after it was frozen. A layer of OCT embedding adhesive was applied to the sample holder, the tissue was placed on it, and the refrigerator was pre-cooled at 4 °C for 5–10 min, so that the OCT adhesive could penetrate the tissue. Another layer of OCT glue was added on top of the tissue to completely cover the tissue. A thermostatic cryotome (LEICA CM1900, Leica Instruments LTD) was used to prepare frozen sections. The resulting slices were placed on a slide and fixed with 4% paraformaldehyde (Tianjin Guosheng Zhongyuan Technology, GSZY-S0008) for 5–10 min. PBS was used to clean the sample three times for 5 min each time, and then the excess liquid was removed. Drops of anti-fluorescence quenching sealing tablets (Beijing Soleibo Technology, S2100) were used to slow the quenching of the fluorescent dyes, and a Panoramic Scan instrument (3DHISTECH, Panoramic Scan) was used for scanning.

2.6 Ethical statement

All animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals of Tianjin Medical Experimental Animal Care and approved by the Animal Ethics Committee of Yi Shengyuan Gene Technology (Tianjin), Ltd. (protocol number YSY-DWLL-2024609).

3. Results and discussion

3.1 Theory and system design

The GHz-acoustic waves are generated by the vibration of the bulk acoustic resonator. As a beam of acoustic waves propagates through a fluid medium, it is attenuated due to viscous dissipation, relaxation effects, and turbulent fluctuations within the fluid.34 The attenuation coefficient is as follows:
 
image file: d5lc00390c-t1.tif(1)
where ω is the angular frequency of the acoustic waves, ρ is the liquid density, cL is the acoustic speed in the liquid, and μ and μb are the viscosity and bulk viscosity of the fluid, respectively. As the attenuation coefficient scales with the square of acoustic frequency, the GHz-acoustic beam generated by the resonator attenuates within a very short distance (l = 10 μm).35 When the GHz-frequency acoustic waves act as an actuation source on a fluid film, the high-frequency vibrations induce fluid motion that excites capillary waves at the free surface,23,36 with the frequency of the free surface capillary waves being fc ∼ γ/μH.

The wavelength λ of capillary waves is governed by the Kelvin equation:21–23

 
image file: d5lc00390c-t2.tif(2)
where γ is the surface tension of the liquid, ρ is the liquid density, μ is the liquid viscosity, and H is the height of the liquid film. During this process, acoustic energy is converted into surface-tension-dominated capillary forces, driving continuous elongation and progressive thinning of the fluid interface.37,38 As the fluid velocity increases, the neck region undergoes dynamic destabilization under capillary force perturbations—a mechanism analogous to the Rayleigh–Plateau instability. Specifically, when the perturbation wavelength significantly exceeds the radial dimension (typically several times the radius of the cylindrical fluid column), the dominant balance between surface tension and inertial forces is disrupted, leading to periodic neck contraction and ultimately determining the diameter of the ruptured droplets:39,40
 
image file: d5lc00390c-t3.tif(3)
 
U ∝ f (4)
where U is the velocity of the jet at the interface, and f is the frequency of the acoustic waves. Although the initial fluid velocity can reach 1 m s−1, viscous dissipation substantially depletes its inertial kinetic energy. Ultimately, the neck undergoes pinch-off, releasing micron-sized droplets with an ejection velocity attenuated to approximately 0.1 m s−1. The short attenuation length of GHz-acoustic waves enables nebulization at a reduced liquid film height H.

According to eqn (3), the decrease in the liquid height creates conditions in which smaller droplets can be effectively generated by the breakup of capillary waves. This property provides a valuable opportunity for achieving smaller-size droplets–thinner height of liquid. Crucially, using the GHz-acoustic waves as the external energy, the attenuation length of the acoustic waves within the liquid is at 10 μm.35,41 Taking advantage of this characteristic, the GHz-acoustic waves focus the energy over the attenuation length, causing the liquid to form a thinner liquid height, generating a smaller droplet size distribution and thus providing a novel idea for IDD.

The proposed nebulizer synergizes a BAW resonator with capillary-driven microfluidics (Fig. 1). Here, the BAW generates high-frequency acoustic waves under the excitation of the signal generator, which induces the acoustic fluids, causing instability of the liquid surface and local capillary waves. Under constant disturbance, the capillary wave breaks, resulting in nebulization. To enable continuous nebulization, a semi-open microchannel is used here to deliver the liquid and maintain the height of the liquid film. This is achieved by guiding the liquid to the edge of the acoustic resonator at a specific angle (108°) along the microchannel. Driven by gravity, surface tension and the capillary force, the liquid tends to move along the edge. Due to the structure parallel to the edge of the resonator (108°), the area of the liquid interacting with the acoustic waves is increased. This design ensures continuous liquid replenishment and consistent near-monodisperse nebulized droplets, which are crucial for pulmonary therapy.


image file: d5lc00390c-f1.tif
Fig. 1 The nebulizer platform. The semi-open microchannel and GHz-acoustic waves deliver the nebulized liquid containing the drug to the action region of the acoustic resonator, ensuring the stable formation of a thin liquid height through interaction with the resonator. At the outlet of the microchannel, the opening parallel to the edge of the acoustic wave resonator is designed. Under the action of gravity, surface tension and the capillary force, the liquid tends to move along the edge of the solid. By the structure parallel to the edge of the resonator, the area of the liquid interacting with the acoustic wave is increased. This design ensures continuous liquid replenishment and consistent nebulized near-monodisperse droplets, which are crucial for pulmonary therapy.

Efficient nebulization is ideal for the delivery of next-generation therapeutics (e.g., exosomes, nanomaterials and oligonucleotides) that require microdosing inhalation due to limited production or prohibitively high costs.42–44 Microdosing high-efficiency IDD enables targeted lung delivery of subcellular components and biomolecules in low doses, addressing inefficiencies and waste in current pulmonary delivery systems. The capillary-driven nebulizer is employed for microdosing (μL min−1), portability and battery-power consumption.

To ensure that the liquids flowing from the microchannel just reach the area of action of the acoustic resonator and avoid sticking to the side walls, the motion of the liquid at different microchannel heights was simulated (Fig. 2). In the proposed nebulizer, the distance between the BAW and the outlet of the microchannel is 200 μm. When the height of the microchannel is small, the liquid outflow experiences significant adhesion to the side walls before reaching the area affected by the acoustic waves. As the height increases, the thickness of the liquid within the acoustic wave region gradually increases, which hinders nebulization.


image file: d5lc00390c-f2.tif
Fig. 2 Motion of the liquid at different heights of the microchannel outlet. When the height of the microfluidic channel is 50 μm, the liquid exiting the outlet does not reach the action region of the GHz-acoustic waves. As the height increases to 150 μm, the liquid gradually meets the action region. Further increasing the microchannel height results in a thicker liquid level in the action region, which hinders nebulization.

Therefore, a microchannel with a height of 150 microns was used for nebulization. The semi-open microchannels utilize the synergistic effects of gravity, surface tension, capillary forces and the pumping ability of the GHz-acoustic waves themselves to spread the liquid in the chamber along the edges of the resonator, forming a thin liquid film (20–30 μm). The semi-open microfluidic structure eliminates the need for an additional pump to replenish the liquid, and avoids pulse fluctuations, ensuring liquid film height and stability of the nebulization, resulting in near-monodisperse droplets. This design also enables portable applications. The outlet of the microfluidic channel is coordinated with the position of the resonator, and thus, the liquid exits and arrives at the nebulized area, nebulizing the liquid more efficiently.

3.2 Nebulization characteristics

The unique capillary wave dynamics induced by GHz-acoustic waves were captured through synchronized high-speed imaging and nebulization characterization. As shown in Fig. 3A, at high liquid film height (∼300 μm), long-wavelength capillary waves (λ > 50 μm) dominated the liquid interface, generating larger-size droplets (D ≈ 100 μm) through classical Plateau–Rayleigh instability. When the liquid film was thinner (∼30 μm), the short capillary wave ruptures under the strong acoustic pressure induced by the rapid attenuation of the high-frequency acoustic waves, resulting in the release of small droplets (Fig. 3B). This is consistent with the previous surmise–thinner liquid height resulting in smaller droplet sizes. The velocity of the jet is fast at the beginning of its formation. However, as the liquid thread is pinched off to form a droplet, most of its inertial energy is lost owing to viscous dissipation, resulting in a slower droplet velocity. Consequently, the generated droplets have a very low momentum.
image file: d5lc00390c-f3.tif
Fig. 3 (A) Long-wavelength capillary waves are broken at high liquid film height, leading to the formation of large droplets. Scale bar: 100 μm. (B) As the film becomes thinner, capillary waves are broken more readily, resulting in smaller droplets. Scale bar: 100 μm.

In addition, we observed that the proposed nebulizer features in situ droplet size adjustment owing to the variation in the thickness of the liquid film at different acoustic pressures. As shown in Fig. 4A, at low applied power (0.5 W), no droplets were generated. When the power was increased to 0.6 W, small droplets (Dv50 = 2.7 μm) began to form. With increasing the power to 0.8 W, the liquid film was thinned to a critical height (∼20 μm), producing near-monodisperse aerosols (Dv50 = 1.997 μm, GSD = 1.57), which is ideal for deep lung targeting. With further increasing the power (1.25 W), smaller-size aerosols were generated (Dv50 = 0.424 μm), which are suitable for alveolar transfer. This transition aligns with our theoretical analysis, in which reduced liquid thickness directly governs droplet diameter (D ∝ H1.5), as shown in Fig. 4B. Notably, the rapid decay of the velocity of the nebulized droplets (<0.1 m s−1 within 2 ms) minimizes inertial impaction in the upper airways—a key limitation of conventional jet nebulizers for inhalation.


image file: d5lc00390c-f4.tif
Fig. 4 (A) State of the liquid film being pushed away at different vibration amplitudes with increasing power from (i) to (iv). In state (i), no droplets are generated owing to the low power (0.5 W). As the applied power is increased, the nebulization rate steadily increases. Once the power exceeds a certain threshold, as seen in state (iv), the liquid is pushed too far, causing the nebulization rate to decrease instead. Scale bar: 100 μm. (B) Droplet size distribution for salbutamol/octanol at different power levels. As the power is increased, the droplet size decreases. Two parameters, Dv50 and Dv90, were chosen to characterize the aerosol size distribution for the salbutamol/octanol system. The parameters represent the volume equivalent diameter below which 50% and 90% of the droplets fall, respectively. The Dv50 values are 2.712 μm, 1.997 μm, 1.401 μm, and 0.424 μm at 0.6 W, 0.8 W, 1 W and 1.25 W, respectively. (C) Relationship between nebulization flow rate and applied power. (D) Schematic illustration of the broad viscosity range of liquids nebulized by the proposed nebulizer, which spans nearly four orders of magnitude, and corresponding images of the nebulization of different liquids using this method. Scale bar: 100 μm.

Power-dependent optimization revealed a self-regulating mechanism between acoustic force and capillary replenishment. Fig. 4C illustrates the relationship between applied power and the nebulization flow rate, with the flux reaching 10 μL min−1 at 1 W (Video S1†). This flux level is optimal for spontaneous inhalation delivery—too low a flux diminishes effectiveness, while too high a flux results in waste and potential respiratory issues. While increasing the power enhanced nebulization flux up to 10 μL min−1 at 1.0 W (Dv50 = 1.4 μm), exceeding 1.25 W displaced liquid from the action region of the resonator, resulted in coarsened droplets with two peaks (1.5 W, Fig. S4†). This non-monotonic response underscores the importance of balancing energy input with microfluidic autonomy—a feat achieved through our semi-open channel design (150 μm height × 200 μm width). The capillary-driven flow ensured continuous liquid supply without external pumps, while the low power consumption of the GHz resonator (<1 W) enabled sustained operation on batteries.

Current drug development and delivery methods rely on existing nebulization technology, but the viscosity range of liquids that can be nebulized using current technologies is small, limiting the development of new drugs. Due to viscous loss, the remaining energy produced in current technology is not sufficient to break up the surface tension of high-viscosity liquids. A hallmark of the proposed technology is its unprecedented viscosity tolerance, spanning four orders of magnitude (0.7–5000 cP). As shown in Fig. 4D, low-viscosity fluids such as octanol (η = 7.5 cP) formed 1–3 μm droplets via inertial thinning, whereas high-viscosity honey (η = 5000 cP) required localized acoustic pressures >5 MPa to overcome viscous dissipation. This capability far exceeds the limits of current vibrating mesh nebulizers (typically <50 cP), thus addressing a critical barrier for nebulizing next-generation biologics, such as lipid nanoparticles or polymer-based gene therapies.44,45

3.3 In vitro and in vivo evaluation of targeted delivery

Nebulization, as a method of inhalation therapy, delivers the drug directly to the lungs, avoiding the time delays associated with intravenous injection. In addition, a suitable nebulization flow rate allows patients to inhale at their own pace and depth, improving comfort and drug delivery efficiency.42,46 To validate clinical relevance, we engineered salbutamol/octanol (η = 7.5 cP) droplets with tunable sizes from 0.4–2.7 μm by modulating the power (Fig. 3B). At 0.8 W, the Dv50 of 1.997 μm matched the optimal range for small airway deposition (FPF < 5 μm = 96%), while 1.25 W produced 0.4 μm droplets suitable for alveolar delivery. The mass output rate (8 μg salbutamol per min for 3 min) aligns with emergency dosing protocols without requiring forced inhalation—a transformative feature for pediatric or dyspneic patients. The Peltier effect was applied to the nebulizer for cooling, and thermal characterization confirmed biocompatibility, with a core temperature <37 °C during operation (Fig. S5†), which is well below thresholds for protein denaturation or patient discomfort.
In vitro testing. To establish the clinical translatability of GHz-acoustic nebulization, we employed a pharmacopeia-grade TSI model that replicates human respiratory dynamics under tidal breathing conditions (30–60 L min−1). The selection of salbutamol/octanol (1 mg mL−1) as a model formulation was guided by its dual aerodynamic and biopharmaceutical advantages. The salbutamol/octanol droplets were then inhaled through the TSI using a flow pump set to 60 L min−1 (Fig. 5A). Samples were collected from each position. The absorption values were measured using a UV spectrophotometer and the delivery efficiency was calculated using the fitting formula obtained. As shown in Fig. 5B, when nebulization was conducted at 0.8 W (8 μL min−1 flux), the system achieved 88 ± 2.1% deep lung deposition (part C)—a 4–8 fold improvement over commercial jet nebulizers (10–20%).47 This exceptional targeting efficiency stems from the convergence of near-monodisperse droplets and breath-synchronized delivery, where droplet velocities decay below 0.1 m s−1 within 2 ms, effectively eliminating oropharyngeal impaction. This performance not only rivals that of ultra-high-efficiency dry powder inhalers (DPIs) but also crucially eliminates their reliance on patient inhalation effort—a transformative advancement for pediatric and geriatric populations with compromised respiratory coordination.
image file: d5lc00390c-f5.tif
Fig. 5 (A) Salbutamol inhalation lung model. (B) Dose rate of salbutamol/octanol deposited in different parts of the lung. Nearly 90% was deposited in the lower respiratory tract (n = 3).
In vivo testing. Having confirmed the potential of the proposed nebulizer for drug delivery, the translational promise of this platform was further validated in spontaneously breathing C57BL/6 mice, whose airway dimensions (trachea: 1.2 mm; small airways: 0.3–0.5 mm) mirror human pediatric anatomy. Previous results indicated that the nebulized droplet velocity is relatively low due to energy losses, suggesting that the proposed nebulizer has a gentle inhalation characteristic. This means that at low-flux levels, the nebulizer relies on the natural breathing of mice to deliver the drug to the small airways, rather than depending on external gases or intubation. To facilitate observation of lung deposition, we used rhodamine B, a fluorescent live cell dye, instead of salbutamol. The experimental group received an octanol solution containing rhodamine B, which was nebulized for 30 and 60 min, respectively. The control group was administered octanol alone. The process is illustrated in Fig. 6A. Autonomous inhalation of rhodamine B/octanol aerosols (Dv50 = 1.997 μm) resulted in time-linear lung accumulation—1.8 fold higher at 60 min compared to 30 min, with >90% of fluorescence localized in the deep lung and alveolar regions (Fig. 6B). This biodistribution profile underscores the capacity of the system for lung-specific targeting while circumventing the systemic exposure risks inherent to intravenous routes. These results demonstrate that the proposed nebulizer provides an appropriate droplet size distribution through autonomous breathing, and thus holds promise as a therapeutic approach for treating lung diseases, particularly targeting small airways and alveoli.
image file: d5lc00390c-f6.tif
Fig. 6 Lung distribution of rhodamine B in mice. (A) Schematic of nebulized drug delivery and tissue processing procedures. (B) Representative images of mouse lungs showing rhodamine B distribution following inhalation through the proposed nebulizer. The control group inhaled octanol for 30 minutes. The experimental group was divided into two groups. One inhaled rhodamine B/octanol for 30 minutes, while the other inhaled the same mixture for 60 minutes (n = 3 per group). Representative images of mouse lungs after inhalation ((i) and (ii)). The color scale indicates rhodamine B density. (iii) Representative images of lung sections. Red fluorescence represents the location of inhaled rhodamine B.

We compared the parameters of nebulization liquids using various methods, as detailed in Table 1. The results highlight the superior performance of the proposed nebulizer. Crucially, this technology bridges critical gaps in pulmonary therapeutics. For pediatric asthma and neonatal RDS patients, the elimination of forced inhalation requirements (≥30 L min−1) enables therapy administration independent of respiratory effort. The unprecedented viscosity tolerance (5000 cP) of the platform opens avenues for biologic delivery, as demonstrated by successful nebulization of monoclonal antibody surrogates—a capability unattainable with current mesh nebulizers limited to 50 cP. While murine models confirmed targeting precision, future studies must address scaling challenges: rabbit or primate models with branching airway geometries closer to those of humans will better predict clinical performance, particularly for alveolar-targeted gene therapies requiring submicron aerosols. Long-term biocompatibility studies (>30 days exposure) will further solidify the suitability of the platform for chronic disease management, where sustained aerosol stability and mucosal tolerability are paramount.

Table 1 Nebulization parameters of different methods
Method Power (W) Dv50 (μm) Viscosity (cP)
Jet12,47 100–200 3–7 <6
Ultrasonic15,48,49 ∼50 >5 ∼175
Mesh14,16,17 1–5 1–5 ∼50
Hypersonic 0.6 0.4–2.8 5000

Conclusions

This paper introduces an innovative and portable nebulizer for more efficient drug delivery, leveraging the breakup of capillary waves induced by the combination of GHz-acoustic waves and a microchannel, which surpasses the limitations of traditional nebulization methods. We systematically investigated the effects of external conditions on nebulization parameters and validated the feasibility of the proposed nebulizer through experiments with lung models and murine inhalation models. The nebulizer generates droplets with suitable size distribution and flux, enhancing the efficiency of pulmonary drug delivery (up to 88%) and reducing the side effects. Due to the resonator being in direct contact with the liquid, all of the generated energy is utilized for nebulization, which improves the energy efficiency and allows more viscous liquids to be nebulized (5000 cP). This advancement is poised to facilitate combined treatments for both small airways and alveoli. Although our current studies focused on salbutamol and octanol, the platform is versatile. It can be adapted to other drugs and solvents with appropriate viscosities, making it applicable for treating conditions such as asthma and emphysema. For future studies, clinical performance studies will be conducted to develop disease-specific wearable devices with unique delivery models to solve the problems of inhalation therapy.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors gratefully acknowledge financial support from the National Natural Science Foundation of China (NSFC No. 22427807, No. 62174119). Quanning Li, Xuejiao Chen, Bohua Liu, and Chongling Sun are thanked for help with the device fabrication.

References

  1. F. Gagnadoux, J. Hureaux, L. Vecellio, T. Urban, A. Le Pape, I. Valo, J. Montharu, V. Leblond, M. Boisdron-Celle, S. Lerondel, C. Majoral, P. Diot, J. L. Racineux and E. Lemarie, Aerosolized Chemotherapy, J. Aerosol Med. Pulm. Drug Delivery, 2008, 21, 61–70 CrossRef CAS PubMed .
  2. Y. Han, Y. Zhu, H. A. Youngblood, S. Almuntashiri, T. W. Jones, X. Wang, Y. Liu, P. R. Somanath and D. Zhang, Nebulization of extracellular vesicles: A promising small RNA delivery approach for lung diseases, J. Controlled Release, 2022, 352, 556–569 CrossRef CAS PubMed .
  3. C. Li, W. Zhan, Z. Yang, C. Tu, G. Hu, X. Zhang, W. Song, S. Du, Y. Zhu, K. Huang, Y. Kong, M. Zhang, Q. Mao, X. Gu, Y. Zhang, Y. Xie, Q. Deng, Y. Song, Z. Chen, L. Lu, S. Jiang, Y. Wu, L. Sun and T. Ying, Broad neutralization of SARS-CoV-2 variants by an inhalable bispecific single-domain antibody, Cell, 2022, 185, 1389–1401.e1318 CrossRef CAS PubMed .
  4. C. Leach, G. L. Colice and A. Luskin, Particle size of inhaled corticosteroids: Does it matter?, J. Allergy Clin. Immunol., 2009, 124, S88–S93 Search PubMed .
  5. D. Hodgson, J. Anderson, C. Reynolds, G. Meakin, H. Bailey, I. Pavord, D. Shaw and T. Harrison, A randomised controlled trial of small particle inhaled steroids in refractory eosinophilic asthma (SPIRA), Thorax, 2015, 70, 559–565 CrossRef PubMed .
  6. T. C. Carvalho, J. I. Peters and R. O. Williams Iii, Influence of particle size on regional lung deposition – What evidence is there?, Int. J. Pharm., 2011, 406, 1–10 CrossRef CAS PubMed .
  7. T. C. Carvalho and J. T. McConville, The function and performance of aqueous aerosol devices for inhalation therapy, J. Pharm. Pharmacol., 2016, 68, 556–578 CrossRef CAS PubMed .
  8. R. J. Martin, Therapeutic significance of distal airway inflammation in asthma, J. Allergy Clin. Immunol., 2002, 109, S447–S460 CrossRef PubMed .
  9. M. van den Berge, N. H. T. ten Hacken, E. van der Wiel and D. S. Postma, Treatment of the bronchial tree from beginning to end: targeting small airway inflammation in asthma, Allergy, 2012, 68, 16–26 CrossRef PubMed .
  10. R. Ghanem, R. Youf, T. Haute, X. Buin, M. Riool, J. Pourchez and T. Montier, The (re)emergence of aerosol delivery: Treatment of pulmonary diseases and its clinical challenges, J. Controlled Release, 2025, 379, 421–439 CrossRef CAS PubMed .
  11. L. Yue, X. Zhang, C. Zhao, R. Chen, X. Chen and L. Rao, Inhaled drug delivery: Past, present, and future, Nano Today, 2023, 52, 101942 CrossRef CAS .
  12. B. L. Laube, D. L. Swift and G. K. Adams, Single-Breath Deposition of Jet-Nebulized Saline Aerosol, Aerosol Sci. Technol., 2007, 3, 97–102 CrossRef .
  13. J. C. Simon, O. A. Sapozhnikov, V. A. Khokhlova, Y. N. Wang, L. A. Crum and M. R. Bailey, Ultrasonic atomization of tissue and its role in tissue fractionation by high intensity focused ultrasound, Phys. Med. Biol., 2012, 57, 8061–8078 CrossRef PubMed .
  14. J. N. Pritchard, R. H. Hatley, J. Denyer and D. V. Hollen, Mesh nebulizers have become the first choice for new nebulized pharmaceutical drug developments, Ther. Delivery, 2018, 9, 121–136 CrossRef CAS PubMed .
  15. C. Sean, M. Ofer and F. James, Improving and Predicting Fluid Atomization via Hysteresis-Free Thickness Vibration of Lithium Niobate, Adv. Funct. Mater., 2018, 28, 1704359 CrossRef .
  16. P. Sharma, M. Quazi, I. R. Vazquez and N. Jackson, Investigation of droplet size distribution for vibrating mesh atomizers, J. Aerosol Sci., 2022, 166, 106072 CrossRef CAS .
  17. P. Sharma, I. R. Vazquez and N. Jackson, Atomization of High Viscous Liquids Using a MEMS Vibrating Mesh With Integrated Microheater, J. Microelectromech. Syst., 2023, 32, 362–370 CAS .
  18. X. J. Dai, W. J. Li, D. D. Xie, B. X. Liu, L. Gong and H. H. Han, Stimuli-Responsive Nano Drug Delivery Systems for the Treatment of Neurological Diseases, Small, 2025, 21, e2410030 Search PubMed .
  19. Y. Kohl, M. Muller, M. Fink, M. Mamier, S. Furtauer, R. Drexel, C. Herrmann, S. Dahnhardt-Pfeiffer, R. Hornberger, M. I. Arz, C. Metzger, S. Wagner, S. Sangerlaub, H. Briesen, F. Meier and T. Krebs, Development and Characterization of a 96-Well Exposure System for Safety Assessment of Nanomaterials, Small, 2023, 19, e2207207 CrossRef PubMed .
  20. W. Li, C. Wang, Y. Zhang and Y. Lu, Lipid Nanocarrier-Based mRNA Therapy: Challenges and Promise for Clinical Transformation, Small, 2024, 20, e2310531 CrossRef PubMed .
  21. G. Forde, J. Friend and T. Williamson, Straightforward biodegradable nanoparticle generation through megahertz-order ultrasonic atomization, Appl. Phys. Lett., 2006, 89, 064105 CrossRef .
  22. J. R. Friend, L. Y. Yeo, D. R. Arifin and A. Mechler, Evaporative self-assembly assisted synthesis of polymeric nanoparticles by surface acoustic wave atomization, Nanotechnology, 2008, 19, 145301 CrossRef PubMed .
  23. L. Y. Y. Aisha Qi and James R. Friend, Interfacial destabilization and atomization driven by surface acoustic waves, Phys. Fluids, 2008, 20, 074103 CrossRef .
  24. W. Connacher, A. Jain, M. Stambaugh, V. W. Leung and J. Friend, Portable acoustofluidic nebulizer, J. Acoust. Soc. Am., 2019, 146, 2816–2816 Search PubMed .
  25. O. Pourmehran, K. Zarei, J. Pourchez, S. Vreugde, A. Psaltis and P.-J. Wormald, Advancements in acoustic drug delivery for paranasal sinuses: A comprehensive review, Int. J. Pharm., 2023, 644, 123277 CrossRef CAS PubMed .
  26. L. Y. Yeo, J. R. Friend, M. P. McIntosh, E. N. Meeusen and D. A. Morton, Ultrasonic nebulization platforms for pulmonary drug delivery, Expert Opin. Drug Delivery, 2010, 7, 663–679 CrossRef CAS PubMed .
  27. A. E. Rajapaksa, J. J. Ho, A. Qi, R. Bischof, T.-H. Nguyen, M. Tate, D. Piedrafita, M. P. McIntosh, L. Y. Yeo, E. Meeusen, R. L. Coppel and J. R. Friend, Effective pulmonary delivery of an aerosolized plasmid DNA vaccine via surface acoustic wave nebulization, Respir. Res., 2014, 15, 60 CrossRef PubMed .
  28. A. Qi, J. R. Friend, L. Y. Yeo, D. A. Morton, M. P. McIntosh and L. Spiccia, Miniature inhalation therapy platform using surface acoustic wave microfluidic atomization, Lab Chip, 2009, 9, 2184–2193 RSC .
  29. T. Xie, Y. Zeng, Z. Gui, M. Ma, Y. Huo, W. Zhang, T. Tan, T. Zou, F. Zhang and J. Zhang, Piezoelectric atomization of liquids with dynamic viscosities up to 175 cP at room temperature, Ultrason. Sonochem., 2023, 94, 106331 CrossRef CAS PubMed .
  30. H. Zhang, Y. Zhou, M. Zhang, W. Guo, C. Sun, X. Duan and W. Pang, presented in part at the 2023 IEEE 36th International Conference on Micro Electro Mechanical Systems (MEMS), 2023 Search PubMed.
  31. T. Srichana, G. P. Martin and C. Marriott, A Human Oral-throat Cast Integrated with a Twin-stage Impinger for Evaluation of Dry Powder Inhalers, J. Pharm. Pharmacol., 2010, 52, 771–778 Search PubMed .
  32. Q. Y. Huang, Y. Le, H. Hu, Z. J. Wan, J. Ning and J. L. Han, Experimental research on surface acoustic wave microfluidic atomization for drug delivery, Sci. Rep., 2022, 12, 7930 CrossRef CAS PubMed .
  33. G. Scheuch, M. J. Kohlhaeufl, P. Brand and R. Siekmeier, Clinical perspectives on pulmonary systemic and macromolecular delivery, Adv. Drug Delivery Rev., 2006, 58, 996–1008 CrossRef CAS PubMed .
  34. M. B. Dentry, L. Y. Yeo and J. R. Friend, Frequency effects on the scale and behavior of acoustic streaming, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2014, 89, 013203 CrossRef PubMed .
  35. H. Wu, Z. Tang, R. You, S. Pan, W. Liu, H. Zhang, T. Li, Y. Yang, C. Sun, W. Pang and X. Duan, Manipulations of micro/nanoparticles using gigahertz acoustic streaming tweezers, Nanotechnol. Precis. Eng., 2022, 5, 023011 Search PubMed .
  36. L. Y. Yeo, D. Lastochkin, S.-C. Wang and H.-C. Chang, A New ac Electrospray Mechanism by Maxwell-Wagner Polarization and Capillary Resonance, Phys. Rev. Lett., 2004, 92, 133902 CrossRef PubMed .
  37. Y. Zhou, M. He, H. Zhang, B. Liu, C. Sun, Z. Han and X. Duan, Pinch-off droplet generator using microscale gigahertz acoustics, Lab Chip, 2023, 23, 4860–4867 RSC .
  38. M. He, Y. Zhou, W. Cui, Y. Yang, H. Zhang, X. Chen, W. Pang and X. Duan, An on-demand femtoliter droplet dispensing system based on a gigahertz acoustic resonator, Lab Chip, 2018, 18, 2540–2546 RSC .
  39. D. J. Collins, O. Manor, A. Winkler, H. Schmidt, J. R. Friend and L. Y. Yeo, Atomization off thin water films generated by high-frequency substrate wave vibrations, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2012, 86, 056312 CrossRef PubMed .
  40. V. Daru, B. Vincent and M. Baudoin, High-speed and acceleration micrometric jets induced by GHz streaming: A numerical study with direct numerical simulations, J. Acoust. Soc. Am., 2024, 155, 2470–2481 CrossRef PubMed .
  41. Y. Yang, L. Zhang, K. Jin, M. He, W. Wei, X. Chen, Q. Yang, Y. Wang, W. Pang, X. Ren and X. Duan, Self-adaptive virtual microchannel for continuous enrichment and separation of nanoparticles, Sci. Adv., 2022, 8, eabn8440 CrossRef CAS PubMed .
  42. A. M. Gracioso Martins, D. B. Snider, K. D. Popowski, K. G. Schuchard, M. Tenorio, S. Akunuri, J. Wee, K. J. Peters, A. Jansson, R. Shirwaiker, K. Cheng, D. O. Freytes and G. P. Cruse, Low-dose intrapulmonary drug delivery device for studies on next-generation therapeutics in mice, J. Controlled Release, 2023, 359, 287–301 CrossRef CAS PubMed .
  43. K. P. Friis, S. Gracin, S. Oag, A. Leijon, E. Sand, B. Lindberg, E. Lazaro-Ibanez, J. Lindqvist, K. A. Whitehead and A. Bak, Spray dried lipid nanoparticle formulations enable intratracheal delivery of mRNA, J. Controlled Release, 2023, 363, 389–401 CrossRef CAS PubMed .
  44. D. Vanover, C. Zurla, H. E. Peck, N. Orr-Burks, J. Y. Joo, J. Murray, N. Holladay, R. A. Hobbs, Y. Jung, L. C. S. Chaves, L. Rotolo, A. W. Lifland, A. K. Olivier, D. Li, K. O. Saunders, G. D. Sempowski, J. E. Crowe, Jr., B. F. Haynes, E. R. Lafontaine, R. J. Hogan and P. J. Santangelo, Nebulized mRNA-Encoded Antibodies Protect Hamsters from SARS-CoV-2 Infection, Adv. Sci., 2022, 9, 2202771 CrossRef CAS PubMed .
  45. C. Horejs, Nebulized lipid nanoparticles, Nat. Rev. Mater., 2021, 6, 1077 CrossRef CAS PubMed .
  46. M. Zhang, H. Jiang, L. Wu, H. Lu, H. Bera, X. Zhao, X. Guo, X. Liu, D. Cun and M. Yang, Airway epithelial cell-specific delivery of lipid nanoparticles loading siRNA for asthma treatment, J. Controlled Release, 2022, 352, 422–437 CrossRef CAS PubMed .
  47. S. G. Devadason, M. L. Everard, J. M. Linto and P. N. Le Souef, Comparison of drug delivery from conventional versus “Venturi” nebulizers, Eur. Respir. J., 1997, 10, 2479–2483 CrossRef CAS PubMed .
  48. H. Naidu, O. Kahraman and H. Feng, Novel applications of ultrasonic atomization in the manufacturing of fine chemicals, pharmaceuticals, and medical devices, Ultrason. Sonochem., 2022, 86, 105984 CrossRef CAS PubMed .
  49. K. Yoshioka, Y. Kurashina, A. Ogawa and T. Asakura, Effect of the area of a lithium niobate transducer on the efficiency of ultrasonic atomization driven by resonance vibration, Ultrason. Sonochem., 2022, 86, 106019 CrossRef CAS PubMed .

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5lc00390c
This journal is © The Royal Society of Chemistry 2025
Click here to see how this site uses Cookies. View our privacy policy here.