Coupled split square resonator based metamaterial sensor for realtime monitoring of air bubbles in IV channels

Abstract

Air embolism caused by undetected air bubbles in intravenous (IV) lines poses a critical risk in medical treatments, necessitating accurate and reliable detection systems. This study addresses this challenge by developing a novel metamaterial-based sensor for real-time monitoring of air bubbles in IV channels. The sensor employs a Single Negative Complementary Split Square Resonator (CSSR) coupled with a polymer-based microfluidic channel (PMC) to detect dielectric changes caused by air bubbles in flowing saline. Unlike traditional optical or ultrasonic methods, the proposed sensor offers a compact, cost-effective, and fabrication-friendly solution while maintaining high sensitivity of 78% and precision. Experimental validation demonstrated its capability to detect even small air bubbles, with resonant frequency shifts observed between 6.6 GHz and 6.9 GHz. This innovation provides a robust tool for enhancing patient safety by preventing air embolisms, with potential applications in broader biomedical sensing scenarios. By addressing a critical medical need through advanced metamaterial technology, this work contributes significantly to both healthcare monitoring and sensor design.

Introduction

Intravenous (IV) therapy is a vital medical procedure, but undetected air bubbles in IV lines can pose severe risks to patient safety. Air embolisms caused by such bubbles can lead to life-threatening complications, underscoring the importance of real-time monitoring in medical settings. Conventional techniques for detecting air bubbles include optical and ultrasonic methods. Optical methods detect variations in transparency caused by bubbles, while ultrasonic methods rely on acoustic signals to identify changes in density. Although effective, these approaches often suffer from drawbacks, including complex setups, susceptibility to environmental interference, and high implementation costs. Additionally, integrating these systems into existing medical equipment presents significant challenges, particularly in ensuring scalability and robustness. To overcome these limitations, this study introduces a novel metamaterial (MTM)-based approach for detecting air bubbles in IV channels.

Metamaterial (MTM) is an artificially synthesized structure with non-natural material electromagnetic properties. MTMs have diverse applications as they can manipulate electromagnetic waves by tuning permittivity, permeability, and refractive index. The MTM electromagnetic properties can be tuned by modifying the dimension and geometry of the unit cell structure^1,2. MTMs are also appropriate for applications such as sensing³, invisible clocking⁴, camouflage technology⁵, lensing⁵, antennae⁶, SAR⁷, filters⁸, physical property detection^9,10, etc. The diversity of MTM applications has attracted much research attention. MTM can be applied in various sensing applications like solid material sensing^10,11, gas sensing¹², liquid sensing^13,14, earth material sensing¹³, etc. A microwave sensor employing non-identical double-split ring resonators¹⁵ demonstrated high sensitivity and multi-band operation for liquid permittivity detection. However, its design involves complex power divider branches and focuses primarily on general liquid characterization, lacking the specificity required for real-time monitoring of dynamic phenomena such as air bubble detection in IV channels. Another sensor¹⁶ operating at 9 GHz achieved a high Q-factor of 240 for non-invasive glucose monitoring but relies on a substrate-integrated waveguide cavity, adding significant fabrication complexity and limiting adaptability to other biomedical sensing needs. Similarly, a meandered microstrip sensor¹⁷ achieved impressive sensitivity (0.64%) and a Q-factor of 506 for liquid permittivity measurement but operates in a narrow frequency band (6.21 GHz), making it unsuitable for applications requiring real-time responsiveness to rapid dielectric changes.

However, these works underscore important innovations but lack the focus on detecting localized, transient changes in dielectric properties, such as air bubbles in IV lines, a critical need in biomedical applications, yet has received comparatively less focus in MTM research. This gap motivates the current study, which leverages the ability of MTMs to achieve high precision and reliability through resonance-based sensing. The proposed MTM sensor employs a Single Negative Coupled Split Square Resonator (CSSR) integrated with a polymer-based microfluidic channel (PMC). It achieves real-time monitoring by detecting shifts in resonant frequency, which occur due to the low dielectric constant of air bubbles in flowing saline. This design offers significant advantages over traditional methods, including compactness, ease of fabrication, cost-effectiveness, and suitability for integration into existing medical systems. Transmission line-based MTM sensors combine the unique properties of metamaterials with established transmission line theory, enabling the development of compact and efficient sensors. These sensors have been applied in material characterization, moisture content detection, and biomedical sensing. Air bubble detection in saline lines represents a particularly urgent challenge due to the potential for air embolisms. Traditional methods, such as optical and ultrasonic techniques, while effective, often lack the robustness and simplicity offered by MTM-based solutions. Most conventional MTM designs used for sensing rely on complex multilayered structures, which limit their practical applicability due to fabrication challenges and high costs. However, polarization-insensitive and angle-insensitive behaviours make MTM designs highly appealing. In recent years, researchers have developed many sensing techniques using MTM, including periodic cross-resonators¹⁸, symmetrical Double Split Ring Resonator (DSRR)¹⁹, circular spiral resonators²⁰, three-layer sandwich models²¹, etc. While these approaches have expanded the scope of MTM applications, they often depend on intricate fabrication processes and the specific structure of the test object.

Consequently, there is a growing need for MTM solutions that provide a broad sensing spectrum, simplified production processes, and cost-efficiency for diverse applications. Another challenge in MTM sensors is achieving a narrow transmission band, as multiple or broad transmission bands can interfere with each other, leading to non-linear results. This issue is particularly critical when different test materials are placed on the resonator. By precisely controlling the dielectric permittivity and permeability of the structure, an isolated single transmission band can be achieved. A carefully optimized geometrical design can match the free-space impedance and input impedance of the MTM, minimizing reflected power at specific frequencies and enhancing sensing accuracy.

In this context, the study focuses on addressing the unmet need for a accurate air bubble detection system in IV channels. This research demonstrates the design and validation of a metamaterial-based sensor for real-time monitoring of air bubbles in IV channels. The sensor utilizes a Single Negative Coupled Split Square Resonator (CSSR), with a narrow transmission band. The final design exhibited a strong electric field distribution and a low Half Power Bandwidth (HPBW) of 0.1 GHz, indicating high sensitivity to the presence of air bubbles. Experimental results from the sensing application confirmed the sensor’s ability to detect air bubbles by monitoring shifts in resonant frequency, validating its potential for precise bubble detection in medical settings. By combining compact design, ease of fabrication, and high precision, the proposed sensor addresses critical challenges in biomedical sensing and contributes significantly to the advancement of MTM technology in healthcare.

MTM sensor design

The behavior of an MTM’s electromagnetic response to a wave’s incidence is controlled by two essential factors: permeability and permittivity. These are the intrinsic qualities of a material. Each MTM has a different electromagnetic response due to different atomic arrangements¹⁸. The MTM unit cell acts as an electric dipole to generate a unique electromagnetic response like negative permeability, permittivity, and refractive index, which results in a change in the reflection coefficients (S₁₁) or transmission coefficients (S₂₁)²². This phenomenon can be utilized to develop metamaterial-based sensing technology. Frequency-dependent permittivity and permeability are given in Eqs. (1) and (2), where ω_pc, ω_pm and ω_mr respectively electric plasma frequency, magnetic plasma frequency and magnetic resonance frequency. Γ_e and Γ_m are the electric and magnetic damping factors²².

1

2

Unit-cell design

The conceptualization of the proposed sensor revolves around the transmission line principle, as depicted in Fig. 1. The sensor uses a modified transmission line featuring an MTM unit cell on top of it. A 1.67 mm thick FR-4 lossy substrate material has been used with a dielectric constant of 4.3, and loss tangent of 0.025. The FR4 material was chosen for its favorable dielectric properties and ease of fabrication, making it a common substrate in high-frequency applications. The transmission line and the MTM are made of copper with a thickness of 0.035 mm. The overall size of the sensor is 40 mm x 10 mm and the unit cell size is 6.6 mm x 9.20 mm. The design parameters are given in Fig. 1; Table 1.

Fig. 1.

(a) MTM-based sensor, (b) MTM unit cell, and (c) The sensor associated with the equivalent circuit model.

Table 1.

Design parameters.

Parameters	Value (mm)	Parameters	Value (mm)
L	40	Mw	6.6
W	10	Mt	0.9
T	2	Mc	0.3
C	10	Mg	0.8
Ml	4.2

Design evolution

Figure 2 shows the design evolution of the sensor and Fig. 3 shows the Simulated and Measured S₂₁ pattern of the final sensor design. Figure 4 represents the electric (E) field distribution for every design step. Figure 5 shows the measured S₂₁ pattern in the VNA. Initially, the structure is built with a single copper transmission line. The width of the transmission line is 2 mm.

Fig. 2.

(a–d) Design evolution of the MTM sensor.

Fig. 3.

Simulated and measured S₂₁ pattern of the final sensor design.

Fig. 4.

Electric field distribution for (a) first design, (b) second design, (c) third design, and (d) final design.

Fig. 5.

Measured S₂₁ pattern in the VNA.

The transmission line width is set to 2 mm. Where w is the width of the microstrip line, Z₀ is the characteristic impedance, h is the height of the substrate, t is the thickness of the conductive layer, f is the operating frequency, and L is the length of the microstrip line.

The initial design featured only a transmission line, which did not exhibit any resonance in the desired frequency range. Introducing a slotted transmission line also resulted in no noticeable resonances. However, when a single metamaterial structure was added on top of the slotted transmission line, a resonant frequency appeared at 3.25 GHz with a peak of − 15 dB in the S₂₁ parameter, demonstrating successful coupling. Finally, when the metamaterial was modified into a coupled resonator design, with two resonators facing each other, two resonant bands emerged at 4.94 GHz and 6.62 GHz, with peaks of − 17.55 dB and nearly − 27 dB, respectively, indicating enhanced frequency response.

Equivalent circuit model

The resonance frequency of an CSSR is directly dependent on its structural parameters, including the size of the resonator, distance between resonators, split gap, length of split gap extension, and resonator width. These parameters significantly affect the capacitance and inductance of the resonator, collectively determining its resonance frequency. The equivalent circuit is presented in Fig. 6 is to provide a basic understanding of the capacitance and inductance in the Coupled split-square resonators (CSSRs). It helps explain how the sensor operates and where the capacitive and inductive elements are derived from, offering insight into the sensor’s resonance behavior.

3

4

Fig. 6.

Equivalent circuit model associated with CSSR-loaded transmission line.

Equation (3)¹⁵ shows the relation of the RLC components and the resonating frequency (f_r). In this equation, ‘L_r’ and ‘C_r’ denote the inductance and capacitance of the SSR respectively, while ‘n’ represents individual SSR. Additionally, in (4), ‘L_t’ represents the inductance of the microstrip line, and ‘Z₀’ denotes the characteristic impedance of the microstrip line, which is adjusted to 50 Ω. The term ‘R_r’ refers to the resistance of the SSR. The Q factor of the sensor is 75.5, as described in Eq. (4)²². Where HPBW (half power bandwidth) is 0.084 GHz.

Characteristic analysis of the CSSR resonator-based MTM

The CSSR resonator-based Metamaterial (MTM) unit cell has been extensively examined across various domains to showcase its unique metamaterial characteristics. In Fig. 7a, The surface Current distribution of the MTM and the coupling between the transmission line and the resonator are illustrated. Figure 7b shows the Permittivity, and Permeability of the resonator, respectively. Permittivity (ε) and permeability (µ) are intrinsic characteristics of materials that determine how they interact with electromagnetic waves.

Fig. 7.

(a) Surface Current distribution in the MTM. (b) Permittivity and Permeability of the MTM.

The proposed metamaterial is referred to as a “Single Negative Coupled Split Ring Resonator” (CSSR) due to the unique electromagnetic behavior of the metamaterial used in its design. Metamaterials are classified as double negative when both permittivity and permeability are negative, and single negative when only one of these parameters exhibits negativity. In this work, the real part of the permeability (µ) becomes negative near 7 GHz, which is the operating frequency for the sensing application. This single negative property contributes to the strong metamaterial characteristics of the resonator, enabling it to achieve high sensitivity and precise detection of dielectric variations in the test medium.

Sensing application

The final sensor design was evaluated through both simulation and measurement to assess its performance. The measurement setup is shown in the Fig. 8a. In the simulation, two resonating frequency bands were observed at 4.94 GHz and 6.62 GHz. However, when the sensor was measured experimentally, a slight shift in the resonant frequencies was noted. The first resonant frequency band shifted to 5.02 GHz, while the second band moved to around 6.95 GHz. Which is shown in the Fig. 8b. These shifts between the simulated and measured responses are likely due to fabrication tolerances, material imperfections, or external environmental factors that influenced the sensor’s electromagnetic behavior in practice.

Fig. 8.

(a) Measurement setup and (b) simulated and measured S₂₁ of the fabricated MTM sensor.

Sensing principle

The capacitive area of the sensor is where the electric field is most concentrated, and this region plays a critical role in detecting changes in dielectric properties. To accurately measure the dielectric characteristics of a substance, it must be placed near the sensor’s high electric field zone. When a dielectric material is introduced, the electric field interacts with the sample, causing a change in the device’s capacitance. This change is dependent on the material’s complex permittivity and results in a shift in the sensor’s resonant frequency. Thus, variations in dielectric properties are detectable through the resonance frequency shift caused by capacitance changes. Furthermore, the lossy characteristics of the dielectric material can be quantified by observing changes in the Q factor response. In contrast, the 6.9 GHz frequency band demonstrates strong metamaterial behavior, with the real part of the permeability (μ) showing a negative value. This characteristic indicates that the metamaterial properties are dominant at this frequency, making the sensor highly responsive and sensitive. During testing (referring to subsection B), the 6.9 GHz band exhibits substantial frequency shifts in response to changes in the dielectric properties of the sample, confirming its suitability for sensing applications.

Since the dielectric materials tested in this experiment are liquids, such as saline, a medium is required to guide the liquid across the sensor’s high electric field regions. A polymer-based microfluidic channel (PMC) is used for this purpose, transporting the liquid over the sensitive regions of the CSSR cells. When the microstrip line excites the CSSR, high electric fields are generated in the splitting gaps, making this area responsive to dielectric changes at resonance, which leads to a measurable resonance frequency shift. Figure 9a shows the illustrated measurement setup for air bubbles in IV channel.

Fig. 9.

The measurement setup for air bubbles in IV channel monitoring (a) is the illustration, and (b) is the actual fabricated sensor with PMC pipe line.

In the first experimental setup, saline flowed through the PMC and water was flowed through the PMC in the second experiment. Later, an air bubble is introduced into the flow to act as an insoluble particle. As the insoluble particles (air bubbles) passed through the capacitive region of the CSSR, the corresponding frequency band in the S₂₁ parameter shifted. This frequency shift occurred due to changes in the dielectric constant in the capacitive region. Figure 8a and b show the simulation and measurement setups, respectively, with the PMC placed directly on top of the SSR splitting gaps for accurate sensing.

Monitoring air bubbles in IV channels

The sensing principle behind the metamaterial-based sensor tags relies on their sensitivity to changes in the dielectric properties of the medium in close proximity to the resonator structure. The sensor was connected to a Vector Network Analyzer (VNA), where the transmission coefficient (S₂₁) was measured, to monitor the resonant response. The resonators were designed with capacitive gaps within their conducting layers. These gaps create a localized electric field, making the resonators highly sensitive to changes in the surrounding dielectric environment. When the dielectric properties near the resonator change, such as when a material with a different dielectric constant moves into proximity the resonant frequency of the sensor shifts. This shift occurs because the capacitance in the resonator’s structure is altered by the presence of a different dielectric medium, such as air or saline. This principle was utilized to detect the presence of air bubbles within a fluid flowing through a saline tube.

The experimental setup was designed to validate the performance of the fabricated metamaterial sensor in detecting air bubbles within a saline flow system. The sensor was equipped with two SMA ports, positioned on either side, and connected to an N5227A Performance Network Analyzer (PNA) for transmission coefficient (S₂₁) measurements. A microfluidic channel (PMC), commonly used in biomedical applications, was utilized as the to guide the liquid flow over the sensor. The pipe was securely positioned atop the sensor using a glue gun to ensure consistent alignment and minimize movement during the experiment. Two types of liquids were used during the tests: plain tap water and a medical-grade saline solution. These liquids were introduced into the PMC to simulate realistic conditions. The flow of the liquid across the sensor was carefully maintained, allowing the S21 parameter to be measured under steady-state conditions. The baseline response was first established using each liquid to understand the sensor’s behavior without air bubbles. Subsequently, air bubbles were systematically introduced into the flow. The bubbles interacted with the sensitive region of the sensor, where the dielectric contrast between the air and the liquid caused significant changes in the local electromagnetic environment. These changes triggered resonance at specific frequencies, which were recorded and analyzed using the PNA. Figure 9b shows the actual measurement setup with fabricated sensor and PMC pipe line for air bubbles in IV channel.

Figure 10 presents the different positions of the air on the MTM, and Fig. 11 presents the comparison of the sensor results for different positions of the air in saline on the MTM. The results demonstrated that in the absence of air bubbles, the metamaterial sensor did not resonate at their designated frequencies due to the high dielectric constant of the liquid flowing through the saline tube, which effectively suppressed their resonant response. However, when an air bubble, characterized by its much lower dielectric constant, passed through the liquid, a significant change was observed. The sensor began to resonate at its expected frequency once the air bubble covered the sensitive area of the sensor. The sensor exhibited two distinct frequency peaks, with one prominent peak at 6.975 GHz. When the pipeline contained only air (empty pipe), the resonant frequency remained stable at this value. However, the presence of saline in the pipeline caused a noticeable shift in the resonant frequency. Specifically, if the CSSR is completely covered by saline, the peak frequency is moved to 6.6 GHz.

Fig. 10.

Illustration of different positions of the air.

Fig. 11.

Measured S₂₁ response of the MTM for different positions of the air.

The resonating frequency band shifted in-between 6.6 GHz and 6.9 GHz as the air bubbles moved at positions A and B. In positions A and B, the CSSR was partially covered by Air bubbles. For those positions, the resonant frequency peak was observed at 6.7 GHz.

This frequency shift indicates the sensor’s sensitivity to the presence of air bubbles, with the change in resonant frequency directly correlating to the detection of bubbles. The observed shift from 6.6 to 6.7 GHz, and 6.9 GHz demonstrates the sensor’s capability to identify air bubble within the saline, and its position, highlighting its potential for precise bubble detection in medical applications.

The ability of the sensor to detect and localize the air bubbles was confirmed by the clear and distinct changes in their resonant frequencies, validating their effectiveness for real-time monitoring in biomedical applications.

The Eq. (5) defines the sensitivity () of the metamaterial sensor by quantifying the relationship between the frequency shift percentage and the area coverage percentage of the air bubble on the sensor. The numerator represents the percentage shift in the sensor’s resonant frequency, calculated as

5

Here, is the resonant frequency when water fully covers the sensor, and is the resonant frequency when air fully covers the sensor. This shift is considered 100% when the sensor is completely covered by the air bubble. When the sensor is fully covered by air, this is considered 100%. The sensitivity () is thus a ratio of the frequency shift percentage to the area coverage percentage. If the air bubble covers half of the sensor area (in Position A and B) and the corresponding frequency shift is 50% of the total shift.

The sensitivity of the sensor is given in Table 2. Sensitivity is presented in GHz/ε (frequency shift in GHz for 1 unit change in permittivity. The average sensitivity of the MTM for all materials is 78%.

Table 2.

Sensitivity of the sensor.

					(%)
100%	6.605	6.97	0.365	0.365	93
50%	6.605	6.72	0.365	0.115	63

The Table 2 shows that the size of the air bubble has a significant impact on the resonant frequency and sensitivity of the sensor. When the air bubble fully covers the area of the coupled resonators ( =100%), the resonant frequency shifts closer to the value observed in the absence of water on the sensor. Conversely, when the bubble covers only ( = 50% of the sensor area, the resonant frequency lies between the initial frequency (completely water-covered) and the final frequency (completely air-covered). This behavior is further validated by the sensitivity data in Table 2, where full coverage by the air bubble yields a sensitivity of 93%, while 50% coverage reduces the sensitivity to 63%. These results demonstrate that the sensor’s sensitivity decreases with a reduction in bubble size, highlighting the importance of bubble size in the sensing performance.

The proposed CSSR-based sensor exhibits competitive performance compared to existing designs, as summarized in Table 3. Operating in the 5–7 GHz frequency range, the sensor achieves a compact size of 40 × 10 × 1.6 mm, making it more compact than most designs, including the Quad SSR sensor in¹⁴ (70 × 20 × 1.6 mm) and the MTM-based sensor in²³ (40 × 40 × 1.6 mm). Additionally, the use of an FR-4 substrate balances cost and fabrication simplicity, offering a practical alternative to the higher-cost Rogers substrates used in^14,24. In terms of sensitivity, the proposed sensor achieves a remarkable 78%, which is significantly higher than the 0.75 GHz/ε reported in¹⁴ and the 0.73 GHz/ε in¹⁶, and far surpasses the 0.00767 (GHz/unit) in²⁴ and 0.000005 (GHz/unit) in²³. This exceptional sensitivity highlights the effectiveness of the proposed sensor for precise air bubble detection in biomedical applications. Overall, the sensor offers a superior combination of compact design, high sensitivity, and practical fabrication, making it a competitive solution for real-time sensing applications.

Table 3.

Comparison table.

Ref	Freq. GHz	Sensor structure	Size (mm)	Substrate	Sensitivity
24	1–2	ACS feed monopole	20 × 17.5 × 1.6	FR-4	0.000005 (GHz/unit)
16	9	Waveguide resonator	42 × 18 × 0.5	Rogers RO4003	0.00767 (GHz/unit)
23	13–14	MTM with microstrip feed line	40 × 40 × 1.6	FR-4	73 (GHz/ε)
14	3–5	Quad SSR	70 × 20 × 1.6	Rogers RT5880	75 (%)
Proposed	5–7	CSSR with microstrip feed line	40 × 10 × 1.6	FR-4	78 (%)

Conclusion

This study has successfully demonstrated the design, and validation of a metamaterial-based sensor for real-time monitoring of air bubbles in IV channels. The sensor’s design utilizes a Single Negative Coupled Split Square Resonator (CSSR) coupled with a slotted transmission line, with a polymer-based microfluidic channel (PMC) positioned directly over the resonator gaps to facilitate the flow of liquids. Through both simulation and experimental validation, the sensor has proven its ability to detect the presence of air bubbles with high precision by monitoring shifts in the resonant frequency of the metamaterial (MTM). In the experimental setup, air bubbles passing through the capacitive area of the resonator caused the resonant frequency to shift between 5 GHz and 7 GHz, demonstrating the sensor’s sensitivity to dielectric changes within the saline tube. The final design, featuring Coupled resonators, produced two distinct resonant frequency bands at 6.6 and 6.9 GHz. The second resonant frequency band shows stronger sensitivity. This band was used for air bubble detection due to its higher response in the electromagnetic field distribution, making it more effective for sensing applications. The sensor’s ability to detect even small air bubbles highlights its potential as a reliable tool for real-time medical monitoring. The integration of this MTM-based sensor into IV lines offers significant potential for improving patient safety by preventing air embolisms during intravenous therapy. The development of this highly sensitive (78%) and efficient device contributes to the advancement of MTM technology in biomedical engineering, opening new avenues for enhanced healthcare monitoring and patient safety.

Author contributions

A.M. conceptualized the work, simulated, experimented, prepared methodology and visualization and wrote the main manuscript , M.T.I. and T.A. did formal analysis, provided the lab facilities. Funding acquisition was done by S.A., A.A., A.S.S., H.A., M.S.S., and M.S.I. All authors have reviewed the manuscript and agreed to the published version of the manuscript.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.