Experimental Section
Gripper Fabrication: The silicone elastomer and carbon
black-filled elastomer composite were fabricated out of conductive
nanoparticles (AkzoNobel, Ketjenblack EC-300J) and a liquid silicone
elastomer (Smooth-On, Ecoflex 00–30) in a planetary centrifugal mixer
(Thinky, ARE-250) following already existing manufacturing
approach.[35] The fabrication of the gripper
consists of six steps. At the first step, the silicone layer for
electroadhesive (EA) pad was prepared. The liquid silicone was poured
onto a PET sheet and spread using an applicator coater (Zehntner,
ZUA2000) and a variable gap applicator (Zehntner, ZAA2300) with a gap of
100 μm and a drawing speed of 5 mm/sec. The elastomer layer was then
cured in an oven for 30 min at 80 °C. Then, in the second step, the
conductive silicone layer was formed by pouring conductive silicone
composite onto the silicon layer and blade casting it using a variable
gap applicator with the gap of 25 μm and a drawing speed of 15 mm/sec.
The multilayer sample was then placed place in the oven for 60 min at 80
°C to form a thickness of conductive layer around 20 μm. In the third
step, the electrode layer was ablated using a laser engraver (Trotec,
Speedy 300) to obtain the interdigitated electrode structure (Figure
2a-ii), and then the sample surface was cleaned with a solvent
(isopropyl alcohol) to remove etched conductive particles. Uncleaned
regions with carbon particles will lead to a drastic decrease in the
breakdown voltage of the pads. Conductive tape (3M, 9713) was placed
onto the electrodes from both sides to connect the device with the power
supply after assembly. In the fourth step, the first step was repeated
to encapsulate the interdigitated electrode structure. A variable gap
applicator with a gap of 1000 μm and a drawing speed of 5 mm/sec was
used. In the step five, liquid silicone was poured into the mold with
the pad screwed to the bottom side of the mold. Then the silicone was
left at least two hours at room temperature to cure. Leaving it at room
temperature also has the advantage to leave more time for the trapped
air to emerge to the surface. In the sixth step, the coffee granules,
which was already used in conventional GJ-based
grippers,[5,11] with the diameter of 0.2 mm were
poured into already fabricated silicone bag.
Grasping Parameters and Grasping Force Characterization: All the
experiment the objects were 3D printed using a printer (Utimaker, S5)
and polylactic acid (PLA) material (Ultimaker, Tough PLA) with the
highest layer thickness of 0.4 mm (instead of a usual 0.06 mm). Porous
hemisphere was made by drilling multiple through-holes to prevent
airtight seal between the gripper and the manipulated object. All the
objects had a screw hole on the bottom to ensure the connection between
the object and the surface under the gripper in the linear stretcher
(Instron, 5965). The gripper moved down with the speed of 1 mm/sec and
stopped when the gripper’s surface is 10 mm lower than the highest point
of the manipulated object to ensure immersion of the object into
granules from all the sides. After moving the gripper down, the force
data form the load cell (Instron, EX2580-500N), which corresponds to
applied force, was recorded. Then, the gripper was activated and moved
up, while the force data was collected from the load cell of the linear
stretcher. The GJ mode was activated using vacuum pump (Thomas VTE 3),
which performance was measured using a manual pressure sensor. The pump
was able to generate a pressure drop of 84 KPa, thus an absolute vacuum
of 16 kPa considering the ambient pressure to be 100 kPa. This is close
to the datasheet value of 15 kPa for an absolute vacuum. The maximum
generated force in each of the experiments was depicted and used to
represent grasping force results. The grasping angle data was collected
using a camcorder of the phone (IPhone, 6s). Then, the data was manually
processed. In the electroadhesion (EA) characterization, a flat sheet of
paper was mounted on the bottom of the linear stretcher instead of the
target object. The paper was chosen because of it does not stick to the
silicone bottom layer of the gripper.[1] A high
voltage supply (Stanford Research Systems, PS350/5000V-25W) was used to
activate the devices. A low capacity load cell (Instron, EX2580-10N) was
used to measure the grasping forces.