Pump Up the Jam: Granular Media as a Quasi‐Hydraulic Fluid for Independent Control Over Isometric and Isotonic Actuation

Abstract Elastomer‐granule composites have been used to switch between soft and stiff states by applying negative pressure differentials that cause the membrane to squeeze the internal grains, inducing dilation and jamming. Applications of this phenomenon have ranged from universal gripping to adaptive mobility. Previously, the combination of this jamming phenomenon with the ability to transport grains across multiple soft actuators for shape morphing has not yet been demonstrated. In this paper, the authors demonstrate the use of hollow glass spheres as granular media that functions as a jammable “quasi‐hydraulic” fluid in a fluidic elastomeric actuator that better mimics a key featur of animal musculature: independent control over i) isotonic actuation for motion; and ii) isometric actuation for stiffening without shape change. To best implement the quasi‐hydraulic fluid, the authors design and build a fluidic device. Leveraging this combination of physical properties creates a new option for fluidic actuation that allows higher specific stiffness actuators using lower volumetric flow rates in addition to independent control over shape and stiffness. These features are showcased in a robotic catcher's mitt by stiffening the fluid in the glove's open configuration for catching, unjamming the media, then pumping additional fluid to the mitt to inflate and grasp.


Assembly of printed components
We used a Carbon M1 3D printer to rapidly fabricate a range of structures throughout this project. The printer uses a projection stereolithography technique to draw three dimensional structures out of a vat of liquid resin. An image is projected through a transparent window on the bottom of the vat to photochemically solidify the resin. In the case of RPU 70 and SIL 30, photoirradiation produces a green state that requires a heat treatment to cure. Projection stereolithography returns short print times but can process only a single resin at a time. We printed our designs as a collection of smaller components and assembled them into composite structures following a technique previously described by the authors. [21] We washed uncured liquid resin off of printed components with isopropyl alcohol and dried them with compressed air. We assembled our fluidic elastomer actuators (FEA) when the printed components were still in the green state to chemically bond them to one another during the heat treatment. We coated seams with a layer of liquid SIL 30 resin and solidified them with an ultraviolet light source prior to the heat treatment to enhance bonding and to ensure a seal.

Cyclically Syringe Pumping Hollow Glass Spheres
The main text of this paper focuses on the use of a peristaltic pump to transport granular fluids through narrow channels into and out of QH-FEAs. Here, we use a syringe pump to demonstrate an alternative mechanism to pump the hollow glass spheres (HGS) as a hydraulic fluid. These tests were conducted using the method described in "4.1.3 Syringe Pump" and "2.2.1 Fluidic Device Design," with the exceptions of: (i) Attaching a 3D printed nozzle over the cut off end of an off-the-shelf syringe barrel ( Figure S1). Using an off-the-shelf syringe ensured a seal between the plunger and barrel. (ii) A rubber latex balloon was stretched over the nozzle of the syringe to act as a FEA. (iii) The syringe pump drove the plunger back and forth at its maximum speed of 1 mm·s -1 over 5 cycles and (iv) the syringe pump was sat on a level surface. Figure S1A is a series of time lapse photographs of the HGSs inflating a balloon as they are extruded out of a syringe, and then being withdrawn back into the syringe barrel once the pumping direction was changed after time = 5 s. We visually observed the syringe filling with granular fluid, as the syringe barrel is transparent. The displacement, Δs, of the plunger as a function of time as the pump inflates and deflates the balloon with HGSs over the 5 pumping cycles is shown in Figure S1B. The plunger achieves the maximum displacement (Δs = 40 mm) after each filling event and returns to Δs = 0 mm after each emptying event for syringes with nozzle diameters of di =10 mm and di =6.4 mm. The motor drives the syringe plunger at a constant rate throughout operation which indicates that the HGSs do not jam when extruded through the narrow channel and demonstrates that flowability of the granules is unaffected by the pumping technique.

Scanning Electron Microscopy
These tests were conducted using the method described in "4.2 Scanning Electron Microscopy". Figure S2A is a SEM image of undeformed HGSs that we imaged as they were received. This image reveals spheres with a disperse range of sizes with diameters up to ~70 µm. The image also shows that there is a presence of some fragments of damaged spheres within the granular fluid. We also imaged the solid glass spheres as they were received ( Fig. S2B and confirmed that the SGSs contain a high proportion of complete =70 µm spheres. We captured SEM images of pumped HGSs to determine the effects of each of the mechanisms on the granules. Figure S2C is a SEM image of HGSs that had been pumped back and forth through the peristaltic pump five times prior to imaging. This image shows an absence of the largest spheres (~70 µm) that were present in the undeformed sample and showed an increased presence of glass shards. These findings indicate that the rotating barrel mechanism crushes the hollow glass spheres during pumping. We compensated for this degradation to the HGSs in our further experiments by regularly replacing the pumped granular fluid from the reservoir with undeformed material. HGSs collected from the syringe pump after five pumping cycles ( Figure   S2D) contained a high prevalence of ~70 µm spheres and very few shards of broken granules which indicates that the syringe pump transports the fluid without crushing the spheres.

Rheology of Crushed Hollow Glass Spheres
This test was conducted using the method described in "4.1.1 Rheology." Figure S3A

Fluidic Elastomer Actuators with Cyclically Pumped Hollow Glass Spheres
This test was conducted using the method described in "2.2.3 QH-FEA Characterization and Performance" in "Compression." We measured the mechanical performance of a cylindrical QH-FEA as it was filled with the HGSs and jammed (∆ = -14 kPa) by the peristaltic pump over a series of five cycles. We reused the same batch of HGSs throughout all cycling tests to observe the effect of crushing HGSs with the peristaltic pump on the mechanical properties of the filled QH-FEA. Figure S3B shows that the degradation to HGSs from cyclic pumping has limited effect on the compressive modulus, ! and tangent modulus, ".$ of the FEA. After five pumping cycles, the jammed QH-FEA had a ! of 210 ± 10 kPa and a ".$ of 700 ± 100 kPa which represents a minor increase in stiffness compared to QH-FEAs inflated with undeformed HGSs ( ! = 200 ± 30 kPa and ".$ = 600 ± 100 kPa).

Attaching a Vacuum Pump to the Fluidic Device
We attached a dedicated vacuum pump (GAST, Model DOA-P704-A4) to our fluidic device which enhanced the effect of jamming on the QH-FEAs. Figure S5 is a flow schematic of the fluidic device with the attached vacuum pump in configuration to jam a QH-FEA. Attaching the vacuum pump enabled us to increase the magnitude of the pressure differential within the QH-FEAs from ∆P= -14 kPa to ∆P= -70 kPa.