Design and Characterization of a Flexible Self-Inflating Mechanical Structure

Soft and reconfigurable structures are widely used in space structures [1–3], smart systems [4,5], and artistic applications [6,7] because of their tunable physical properties such as shape, color, surface morphology, and stiffness. These property variations are usually obtained through the alteration of the form, size, and strain state of the construct. Different techniques such as origami-inspired structural design [8,9] have been used to create such systems with tunable properties; however, inflation of a soft structure is a superior technique in terms of versatility, ease of integration with other soft systems, and cost of fabrication [1,10,11]. In terrestrial device applications, inflatable soft structures offer a wide range of shapes and properties for different purposes such as soft robotics [12–14] and wearable devices [15,16]. In the past, several inflatable structures have been developed to perform various engineering tasks. Rafsanjani et al. introduced a biomimetic crawling soft robot, where the inflation of a rubbery core structure changes frictional properties of the robot exterior surfaces and enables the robot to crawl forward [17]. In addition, Konishi and Kosawa developed an inflation-driven micro-actuator using a conversion mechanism to create high-output bending motion [18]. In another example, Gorissen et al. presented a spherical structure capable of harvesting low inflation rates to activate a snap-through instability-driven motion. This sudden release of energy can be used to make the structure jump [19]. These novel engineered systems operate with an external source of pressurized fluid, which is usually made from non-flexible and bulky stand-alone components. Integrating a rigid fluid pump with a soft structure may sacrifice the overall flexibility of the system and mitigate some of the main advantages of the soft structures. Our thorough literature survey did not result in any papers addressing design of flexible pumps and its incorporation in an integrated system to control stiffness of device. Our novel design allows such an integrated device which can be incorporated into soft robotic applications as well as other devices. There has been a lot of research in developing pumps in biomechanics applications. Cacucciolo et al. [20] designed a class of soft-matter bidirectional pumps based on charge-injection electrohydrodynamics. They showed that these solid-state pumps are flexible, stretchable, modular, scalable, quiet, and rapid. Furthermore, they integrated the designed pump into a glove and demonstrated wearable active thermal management. Sinn et al. [21] introduced an adaptive structure called the self-inflating adaptive membrane. They showed in a single structure that smart components can be incorporated by changing its shapes and flexibility to different mission requirements. The idea behind this adaptive design is taken from nature’s heliotropism principle. It is shown that through this bio-inspired concept, a high ratio of adaptability of the membrane can be obtained. Diteesawat et al. [22] introduced the electro-pneumatic pump, a flexible, high-performance pneumatic pump driven by dielectrophoretic liquid zipping actuation that overcomes the limitations of conventional electromagnetic pneumatic power sources, generating higher air flowrate and consuming less power. Yamaguchi et al. [23] developed a soft pneumatic balloon operated by a soft pump which applies the appropriate pressure over a tactile sensor onto the radial artery of the wrist to detect detailed heart pulse waves (P1 and P2). Stergiopulos et al. [24] designed and manufactured a monolithic high-pressure diaphragm pump made entirely of soft elastomer material and driven by a combustion chamber incorporated within the soft pump structure. The pump can deliver pressures up to 60 kPa and can reach output flows up to 40 ml/min. Methane (CH₄) combustion is used as the actuation source. It is worth noting that there are currently no force induced for self-inflating a soft pump and its incorporation in soft robotics. The proposed self-inflating device overcomes this deficiency. The proposed pump system can be used to control the stiffness of the structure with many applications in soft robotics which relies only to external loading applied to the device.

In this paper, we propose a novel structural solution for the development of a self-inflating mechanical structure with flexible embedded fluid pump. The proposed structure operates independent of any external pressurized fluid source and offers a fully flexible self-inflating structure with easy integration with other soft systems.

The proposed structure inhales and stores the ambient air internally when it experiences an external compressive load. The residual internal pressure increases as a function of the intensity of applied compressive loads and the initial internal air pressure inside the system. Repeating this loading cycle results in a gradual increase in the internal air pressure until it reaches a steady-state pressure equilibrium after several loading cycles. This family of self-inflating structures can be widely used in medical devices and automotive industries because of their high reliability and low fabrication cost stemming from their simple working principle and design. This self-inflating mechanical structure comprises of two internal air chambers connected through multiple air channels, see Fig. 2(a). The first chamber is located at the top of the structure resembling a dome construct and acts as the embedded flexible fluid pump (shown as pump). The second chamber is attached to the bottom of the pump chamber and stores the pressurized air (shown as capsule).

We first study the baseline compressive performance of pressurized elastomeric capsule. We will subsequently compare the results to those from the self-inflating structure (integrated pump and capsule) to validate the performance and effectiveness of the proposed self-inflating system.

A closed cylindrical capsule with radius of 60 mm, height of 60 mm, and wall thickness of 5 mm is fabricated through multi-step molding of silicone rubber as shown in Fig. 1(a). The top face and the sidewall are molded together using a multi-component separable 3D printed mold (please see Section 3). Silicone rubbers are curable elastomers with excellent properties such as thermal stability and enhanced elongation [25], which make them a promising candidate for inflatable systems experiencing large deformation. Two rigid 3D printed plates (please see Section 3) were glued to the top and bottom of the molded rubber capsule to avoid bulging of the top and bottom faces during the inflation. The bottom rigid plate has two embedded air channels, which connect the capsule to a pressure sensor to monitor the air pressure and a syringe to adjust the internal air pressure. A one-way valve was placed at the outlet of the syringe to prevent the backflow and maintain the desired air pressure in the capsule. A sectioned schematic view of the rubbery construct is shown in Fig. 1(b).

Compressive testing was subsequently carried out on manufactured capsule by axially compressing the silicon rubber capsule with different initial (before compression) air pressure values using a single-column Instron universal testing machine. In addition, an open capsule configuration was tested where the syringe and the one-way valve were detached from the system (constant atmospheric pressure inside the capsule). In this testing, the cylindrical capsule was axially compressed for 10 mm and the axial load was measured during compression. Figure 1(c) shows compressive force versus capsule axial deformation. All the eight tested samples showed an initial linear relation between load and axial deformation (approx. 5 mm displacement). In this region, all sample responses were almost identical and independent of the internal pressure. However, their behavior in the nonlinear region diverged. In this region, the stiffness of the structure dropped depending on the initial internal air pressure. The lowest softening was observed for a capsule with the maximum pressure of 9.7 kPa. The softening was more pronounced for lower initial pressures. The open capsule configuration showed the most pronounced softening (negative slope of the force versus displacement curve) in the second half of the compression test (see Fig. 1(c)). We are now in a position to understand the effect of addition of a pump to the capsule.

Experimental study is carried on the integrated capsule-dome system to evaluate evolution of capsule internal pressure when the dome is subjected to cyclic axial compression and verify our numerical results. For this, a sample of the proposed self-inflating structure was fabricated. Both capsule and pump were fabricated separately using a multi-step molding technique discussed in Section 3. These two components were glued to each other with a rigid 3D printed plate in between. The plate contains embedded air channels to transfer air between the pump, capsule, ambient environment, and pressure sensor, while passing them through one-way valves to avoid backflow. Please see Fig. 2(a) for schematic of the internal air channels and one-way valves. Figure 3(a) shows the experimental results of the capsule internal pressure with increase of cyclic compression of the fabricated sample. In this testing, the structure is subjected to 15 consecutive cycles of compressive loading when the force fluctuates between 0 and 25 N. The residual capsule pressure is initially zero and increases gradually until reaching a constant pressure of 10.5 kPa after 13 cycles. The presented results validate the self-inflating functionality of the fabricated sample and shows an excellent agreement with the numerical result presented in Fig. 2. Figure 3(a) shows the inflatable structure prior to loading, after four cycles, and after 15 cycles. In Fig. 3(b), the stiffening effect is quantified through measurement of the force versus axial displacement of the fabricated sample under the cyclic compression. Each curve in this plot corresponds to a compression cycle. The compressive force is initially related linearly to the displacement, and then it follows by a sudden drop in the stiffness (slope of the curve). The observed drop happens because of valve opening when the trapped air inside the pump opens its way to the capsule chamber. As the number of cycle increases (higher residual pressure in the capsule), the effect of valve opening diminishes and eventually an almost linear force versus displacement curve can be observed after 10 cycles. In general, the overall stiffness of the device after valve opening increases because of structural inflation. Note, the same quality of the stiffening effect captured in the capsule inflation test, presented in Fig. 1(c), is also observed in the self-inflating structure. This similarity verifies the performance of the embedded deformable pump as an alternative low-cost method for the external air pump.

The proposed self-inflating structure can be used in various applications, where the on-demand inflation or stiffening of a mechanical construct is desired. An important application of such a device is a self-inflating tire, where the rotation of the tire results in a cyclic loading and provides the energy source for the inflation. In this application, the tire inflation continues until it reaches the desired pressure. The desired pressure can be tuned using an adjustable pressure regulator and the internal tire pressure can be changed based on the road quality, temperature, or driving style [28,29]. Figure 4(a) shows schematic of the sectioned self-inflating tire. The structure of the tire is made by revolving the shown cross section, similar to the cross section of the self-inflating structure studied earlier in this paper; see Fig. 2(a). In addition, four walls radially divide the internal structure of the tire to four equal sections (walls are placed every 90 deg). The internal structure of the tire includes the pump and the capsule components interacting through a pneumatic unit. The pneumatic unit is located in the rim of the wheel for easier access and fabrication. The applied cyclic loading on the tire causes the pressure increase in the pump component and eventually pumps the pressurized air to the capsule. As discussed in Fig. 3, the capsule pressure increases gradually until it reaches a plateau. Figure 4(b) shows two images of a fabricated self-inflating tire prior to loading (depressurized) and after five full rotations (pressurized). See Section 3 for more information on the fabrication of the self-inflating tire. To validate the expected performance of the fabricated self-inflating tire, an experimental testing is performed. In this testing, a vertical downward load is applied on the wheel axle and the wheel is located on a moving surface with adjustable speed. In the first part of the testing, three different force values of 8.5 N, 17 N, and 22.5 N are applied on the wheel axle, while the wheel is rotating at 60 rpm. Figure 4(c) shows that the tire pressure increases as a function of time. The initial tire air pressure depends on the axial force and as the tire rotates, the pressure gradually increases until it reaches a plateau. The final tire pressure is higher for greater axial load. In the second part of the testing, the rotational speed of tire has been set to three different values of 30 rpm, 60 rpm, and 100 rpm while the axial load is maintained constant at 17 N for all three cases. Figure 4(c) bottom shows the capsule pressure as a function of time, where the higher rotational speed causes a faster pressure increase in the tire and the final steady-state pressure is almost independent of the rotational speed.

Self-inflating cylinder: This structure is made through multi-step molding of Dragon Skin FX-Pro (Smooth-On Inc., Macungie, PA). Dragon Skin FX-Pro is a soft and durable platinum silicone rubber with a core hardness of 2 A and break elongation of 763%, which makes it a proper choice for deformable structures. Dragon skin FX-Pro comes in two different containers of A and B which should be mixed with a ratio of 1:1 (volume or weight), poured to the mold and left to dry for at least 40 min in room temperature. All molds are 3D printed using VeroWhite material in an Eden 260 V Polyjet 3D printer (Stratasys Ltd.). As shown in Fig. 1(a), the cylinder is made of two separate components: (1) sidewall and top face, which is made using a three-piece mold to ease the mold release process, and (2) bottom face, which is made using a one-piece mold. These two components were later glued together using Sil-Poxy (Smooth-On Inc., Macungie, PA) to form a fully enclosed construct. In addition, the internal channels are 3D printed and attached to the cylinder. The structure shown in Figs. 2 and 3 includes an additional dome-shaped top layer, which is molded separately and glued to the cylinder top face to form the final self-inflating structure.

This construct includes a rigid rim with embedded air channels to enable air transfer between the ambient air and tire. The rim is 3D printed using VeroWhite material in an Eden 260 V Polyjet 3D printer (Stratasys Ltd.). Tire, including embedded air channels, is 3D printed in two separate parts using TangoBlackPlus material in the same 3D printer. TangoBlackPlus offers properties similar to rubber (tensile strength of 0.8–2.4 MPa and 170–220% elongation at break). Two 3D printed components of the tire are glued together and later glued to the rim using Sil-Poxy (Smooth-On Inc., Macungie, PA) to create a sealed connection. Finally, eight one-way air valves are assembled externally to the rim to control the airflow direction as described in the paper.

Pressure measurement: A Honeywell HSCDRRN005PD2A5 pressure sensor was used to measure the internal air pressure of the self-inflating cylinder and tire. An Arduino Uno is used as the controller board to process and save the pressure sensor data.

Inflatable structures are commonly used in a variety of engineering applications such as robotics, space structures, medical devices, and automotive safety devices. A novel design for a self-inflating structure with an embedded soft fluid pump is introduced. The performance of this conceptual device is fully characterized and validated by the finite element method and experimentation. It is shown with this design a self-inflating silicon rubber cylinder is capable to increase its internal air pressure from 0 to 10 kPa when the dome is subjected to 15 cyclic compressions of 0–25 N. The self-inflating cylinder design was used as a foundation in the design of new generation of self-inflating structures such as tires, which harvests the applied cyclic loading on the tire, as a result of tire rotation and weight of the vehicle, to pressurize the tire without any external air pump. It is shown again that the tire can be pressurized using this conceptual design. The speed of inflation depends on the tire speed and the compressive load to the axel of the tire.

This work was partially supported by the National Science Foundation and Department of Mechanical and Industrial Engineering of Northeastern University.

There are no conflicts of interest.

The authors attest that all data for this study are included in the paper.