Cell Manipulation 
A. Force Feedback Interface for Cell Injection
Motivation: Manual pronuclei injection and intracytoplasmic sperm injection (ICSI) requires long training and has low success rates primarily due to poor control over the injection force. Consequently, there is a need for quantification of forces during biological cell injection and for an automated cell injection system, which can provide force feedback to the operator improving the success rate of the injection task.
Materials and Methods: Figure 1 shows the overall setup on the biomanipulation side, which consists of a Nanomanipulator (Model: MP-285, manufactured by Sutter, Inc.) holding the PVDF force sensor and the pipette holding apparatus. The Nanomanipulator has three degrees of freedom in x, y and z direction and an additional fourth degree of freedom for the diagonal advancement of the pipette in the XZ plane (refer to Figure 1). The travel range is 25mm in all three axes. The lowest resolution is 0.02 µm/step and highest resolution is 40nm/step. The glass micropipette is integrated to the PVDF film.(Thickness: 28µm, Model: LDT1-028K of MSI, Inc.) with the help of a connector as shown in the Figure 2. This setup allows the easy removal and replacement of the micropipette if the tip of it gets damaged during micromanipulation (5µm ID). To test our force feedback interface for cell injection, we performed several cell injections on two different types of egg cells, namely, salmon fish egg cell and flying fish egg cell.A plastic micropipette holds the egg cells. The holding micropipette is mounted on to the MMJR, manual manipulator (manufactured by WPI, Inc.) which has three degrees of freedom in x, y, and z-direction. The travel range is 37mm in X-axis, 20 mm in Y-axis and 25mm in Z-axis. The resolution is 0.1 mm in all three axes. The cell manipulation system has a total of six degrees of freedom (3 DOF for holding pipette and 3 DOF freedom for the injection pipette) as shown in Figure 3. A magnetic base is used for the positioning and holding of the manual manipulator. The whole set up is mounted on a vibration isolation table and covered by an enclosure (not shown in figure 3). The egg cells are sucked inside the holding pipette, which facilitate the penetration of the injection pipette into the egg cell. The Velocity of the manipulator during injection is 120µm/sec. Before performing the experiment the holding pipette is aligned with the injection pipette, which facilitates the injection pipette to penetrate into the egg. The injecting and holding pipette are aligned by controlling the 6DOF of the cell manipulation system in such a manner that the tip of the injecting pipette contacts the center of the egg cell.
 |
 |
| Figure 1: Integration of the PVDF film with the nanomanipulator. | Figure 2 : Connector integrated with the PVDF film. |
 |
| Figure 3. The experimental set up for injecting egg cells: A: MMJR-Manual manipulator; B:Holding pipette; C: Injecting pipette; D: PVDF film; E: MP-285 Nanomanipulator; F:Vibration isolation table; G: Magnetic base; H: Manipulator Stand; I: The PHANToM haptic interface. |
Experimental Calibration: PVDF (Polyvinylidene fluoride) piezoelectric polymer film is used to develop the force sensor for measuring the cell injection forces. The PVDF Film is calibrated with the load cell (Model: GSO-10 of Transducer technology Inc., Maximum measurement range: 98.1mN and accuracy of 50µN). A vibration isolation table is used to remove ground vibrations and an enclosure is used to cover the whole calibration setup to eliminate the interference of acoustic waves on the output voltage from the PVDF film (since it has been observed to be highly sensitive). The load cell is mounted on a stationary platform and the PVDF film is mounted on the MP-285 Nanomanipulator.
 |
 |
| Figure 4: Overview of calibration setup: A: PVDF film setup; B: Load cell setup; C: Vibration isolation table; D: Enclosure. | Figure 5: Experimental calibration of the PVDF film: A: Stationary platform; B: Movable platform; C: Load cell; D: PVDF film. |
 |
| Figure 6: Calibration curve showing a linear relationship between the force measured by the load cell and the integrated voltage from the PVDF film |
Results: While performing the experiment the injection pipette is moved towards the egg cell with the help of MP-285 nanomanipulator, where as the manual manipulator is held stationary. To track down the pipette tip it is filled with a dye as shown in figure 7. Visible light is used for illumination. As the egg cells are transparent in nature, the dark colored dye can be seen when the pipette penetrates the egg cell as shown as shown in figures 7 and 8. Force feedback during cell injection is achieved by in real time by using the PHANToM haptic interface device. The forces measured during cell injection are amplified and displayed to the user in real-time. As a result, the user can perceive the cell injection as an apparent drop in injection force after puncture. The experiments were performed on 10 samples of each egg cell. From figure 9, it can be observed that there is a gradual increase in the force prior to puncture of the membrane. The maximum force is recorded when the cell membrane is punctured. From Figure 9, the puncturing force for one of the flying fish egg cell was observed to be 1.69mN. Based on 10 different samples of the flying fish egg cell, the average puncturing force was found to be 1.6057mN with a standard deviation of 0.33mN. During each of these cell injections, force feedback was provided to the user through the PHANToM and the user was able to discern when the pipette penetrated the egg cell. Similarly the typical force profile obtained for puncturing the membrane of the salmon fish egg cell is shown in Figure 10. The puncturing force for one of the salmon fish egg cell was observed to be 2.38mN. In all the trials, we observed that the slope of the force vs. time curve was less steep for the salmon fish egg cell compared to the flying fish egg cell. This indicated that the salmon fish egg cell underwent more deformation than flying fish egg cell, before its membrane was punctured. The average force for puncturing the salmon fish egg cell for 10 different samples was 2.2694mN with a standard deviation.
 |
 |
| Figure 7: The figure on the top shows the injecting pipette approaching the flying fish egg cell and the figure on the bottom shows the injecting pipette penetrating into the egg cell. | Figure 8: The figure on the top shows the injecting pipette approaching the salmon fish egg cell and the figure on the bottom shows the injecting pipette penetrating into the salmon fish egg cell. |
 |
 |
| Figure 9: The top figure shows the variation of force with time during membrane puncture of a flying fish egg cell. The puncturing force is1.69 mN and the bottom figure shows the variation of the puncturing force for 10 samples of flying fish egg cells. The average puncturing force was 1.6057mN. | Figure 10: The top figure shows the variation of force with time during membrane puncture of a salmon fish egg cell. The puncturing force is 2.38 mN and the bottom figure shows the variation of the puncturing force for 10 samples of salmon fish egg cells. The average puncturing force was 2.2694mN. |
Conclusion: We have developed a force feedback interface for reflecting forces to the user during cell membrane puncturing tasks. The force sensing system is capable of measuring forces in the µN-mN range. The successful implementation and calibration of the force sensor has been presented . The average force values obtained for puncturing the outer membrane of flying fish egg cells and salmon fish egg cells was 1.6057mN and 2.2694mN respectively. During all membrane puncture tasks the user was clearly able to discern when the membrane was punctured through a rapid drop in the force felt through the PHANToM. The membrane puncture forces and the history of the force change prior to puncture can be used to develop cell models. However, the work presented in this paper cannot be directly applied to puncture smaller cells in the range of 50-100µm diameter. In our future work, we plan to use the PVDF based force sensing system to effectively calibrate a vision-based force sensing system to quantify and display cell injection forces. Also, human factors studies comparing automated (force and vision feedback) vs. manual (only vision feedback) cell injection to improve the cell viability after injection will be the natural future direction of this research.
B. Evaluating the Role of Force Feedback for Biomanipulation Tasks
Introduction: We have developed a force feedback interface, which has the capability of measuring forces in the range of µN and provide a haptic display of the cell injection forces in real time. Using this force feedback interface, we performed several human factors studies to evaluate the effect of force feedback on cell injection outcomes. We tested our system with 40 human subjects and our experimental results indicate that the subjects were able to feel the cell injection force and confirmed our research hypothesis that the use of combined vision and force feedback leads to a higher success rate in cell injection task compared to using vision feedback alone.
Materials and Methods: The biomanipulation system (see Fig. 1(a) consists of a nanomanipulator (Model: MP-285, manufactured by Sutter, Inc.) forming the cell injection unit and a micromanipulator (Model: TransferMan NK2, manufactured by Eppendorf, Inc.) forming the cell holding unit. The nanomanipulator as well as the micromanipulator has 3-DOF each and the inverted microscope (Model: IX81, manufactured by Olympus, Inc.) has a 2-DOF stage for positioning the sample. The travel range for MP-285 nanomanipulator (computer controlled) is 25mm along all three axes (X, Y & Z) with lowest resolution of 0.02 µm/step and highest resolution of 40nm/step. The travel range for TransferMan NK2 micromanipulator (joystick controlled) is 20mm in all three axes with a resolution of 40nm per step. The injecting pipette is connected to a pneumatic PicoPump (Model: PV830, manufactured by WPI, Inc.) for the purpose of injecting blue dye into the cell. A manual piston pump (Model: CellTram Air, manufactured by Eppendorf, Inc.) is used to apply suction for reliable holding of suspended zebrafish egg cells. The cell injection system is integrated with vision and haptic interface as shown in Fig. 2.
 |
 |
| Fig. 1. a) Experimental system for evaluating the role of force feedback in cell injection task. b) Magnified view of the cell injection system | Fig. 2. The Cell injection system with vision and force feedback interface. A: Injecting pipette, B: Holding pipette, C: Zebrafish egg cell, D: Visual Interface, E: PHANToM haptic interface. |
 |
|
| Fig. 3. Variation of force with time during membrane puncture of a zebrafish egg cell. The puncturing fore was 700 µN. |
Experimental Set Up and Research Protocol: We were interested in evaluating the role of force feedback in cell injection. The outcome of a transgenesis task depends not only on successful delivery of the desired gene into the cell but also on the successful integration of the genetic material into the genome within the nucleus as a stable transfection. Successful delivery of desired material into the cytoplasm is itself a challenging task. To judge the outcome of the injection process (success or failure), we chose to inject trepan blue dye in zebrafish egg cell. We chose to inject the dye and not a genetic material or fertilize the egg because genetic modification and change in the cell structure depends on various other biochemical factors once the genetic material is placed within the cytoplasm. Since in mechanical manipulation techniques, injection of the membrane and the consequent integrity of the membrane is crucial, we focused on human factors studies on evaluating the membrane integrity after injection using only vision feedback and vision + force feedback. We created two different scenarios in our experiment: (a) S1: the subject was prohibited from seeing the dye being injected and (b) S2: the subject was allowed to see the dye being injected. In a practical situation, the first scenario (S1) corresponds to injecting non-transparent cells whereby it is impossible to ascertain the presence of the injected material (colored or colorless) in the cell and the second scenario (S2) would correspond to injecting transparent cells (zebrafish eggs, for example) with a colored dye. Since we are working with only one type of egg, namely, zebrafish, which are transparent, we created experimental conditions for scenarios S1 and S2 by differentiating whether the subject cannot see (S1) or can see (S2) the injected colored dye within the cell. Since in conventional biomanipulation tasks, the cell may or may not be transparent, our experimental scenarios S1 and S2 described above cover the most general cases to study the effect of force feedback in biomanipulation tasks. We had 40 human subjects perform the experiments with vision (V) feedback alone and combined vision + force (V+F) feedback, with 20 subjects allocated for S1 and 20 subjects allocated for S2. All the subjects were non-surgeons having no previous experience in cell injection tasks. Each subject performed 5 trials with vision (V) feedback alone and 5 trials with vision + force (V+F) feedback. Details for both the (V) and (V+F) feedback protocol are presented below.
1) Vision (V) Feedback Protocol: The first part of the experiment consisted of only vision feedback to perform the cell injection task. The subject was able to view injecting pipette through the eyepiece of the inverted microscope. An operator applied suction and fixed the zebrafish egg cell. At this moment, center of the egg was not aligned with the tip of the injecting pipette. With the help of the joystick (TransferMan NK2) the subject controlled the movement of the holding pipette and aligned the egg with the tip of injecting pipette. The egg and the tip of injecting pipette were maintained at a fixed distance apart by the subject. For the first test, a practice session was given to familiarize the subject with the alignment task. The subject was then able to view the alignment on a video screen and moved the injecting pipette forward with the help of computer controlled nanomanipulator. The subject observed the injecting pipette penetrating the egg membrane on a video screen and stopped the motion of injecting pipette when he/she was confident that pipette has penetrated the cell membrane. At this moment, trepan blue dye was injected by the operator by depressing the foot pedal switch. The injection was deemed successful, when the dye remained inside the cell and the cell did not collapse on removing the pipette. Since the dye is blue in color, it was straight forward to determine if the dye remained inside the cell after injection. The volume of dye injected was approximately 0.001 times the volume of the egg cell. Completion time for the injection task (including alignment task) was recorded. The process was repeated for five trials for each scenario S1 and S2.
2) Vision and Force (V + F) Feedback protocol: The second test was conducted by using both vision and force feedback to perform the cell injection task. This test was performed in the same way as the vision test, with the addition of force feedback. For this experiment the subject used the PHANToM® (haptic interface device manufactured by Sensable Technologies, Inc.) by holding its stylus. The forces were amplified by a factor of 800. The direction of the force feedback was horizontal and was acting towards the subject. The operator controlled the movement of the injecting pipette with the help of computer controlled nanomanipulator. If the subject contacted the cell membrane and pressed against it, he/she would perceive an apparent increase in force followed by a drop in force when the membrane was punctured. A typical force profile for cell membrane penetration is shown in Fig. 3. During cell injection, the PVDF film is subjected to a force which increases with time until the cell membrane is punctured. After feeling the drop in force, the subject communicated to the operator to stop the motion of the injecting pipette. The operator injected trepan blue dye after the subject confirmed that the cell membrane was punctured. The criteria mentioned in the vision test were used to judge the outcome of the injection process. Completion time for the injection task (including alignment task) was recorded. The process was repeated for five trials for each scenario S1 and S2.
Data collection and analysis: The experiment was performed by 40 subjects (20 subjects for each scenario S1 and S2) for a total of 400 trials (5 trials each for V and V+F; hence 10 trials by each subject). The data were collected in a qualitative fashion with cell injections characterized as either "success" or "failure" and denoted by a value 1 or 0 respectively since trepan blue dye is easily observed under the microscope. The data were then analyzed using a non-parametric equivalent of the paired t test i.e. Wilcoxon test. The test generates a p value (probability) for the null hypothesis (H0) and thus a probability for the research hypothesis (H1) to be tested. The lower the p value, the smaller the probability for the null hypothesis to be true and consequently higher is the probability that there is a significant statistical difference between the data sets (or the research hypothesis H1 is true). The level of significance (alpha) for our statistical analysis was chosen to be 0.05, meaning that our research hypothesis would be considered true if p < alpha.
Results: We performed the experiments to test the validity of our research hypothesis, namely, providing vision + force feedback simultaneously leads to higher success rate in cell injection task than only vision feedback ((V + F) > V). The ">" sign denotes "is better than" in the hypothesis. Vision feedback alone was tested before combined vision + force feedback because of the presence of more than one sensory cue. If the above approach is not followed, the subjects may link the drop in force to the visual cue of cell membrane puncture and that may contribute to a learning effect. The trials were, thus, presented in the following order: vision feedback alone followed by vision + force feedback. The individual results for each subject performing the cell injection task in each of the 2 methods (V and V + F) for two different scenarios, S1 and S2, are shown in Fig. 4 based on the research protocol outlined in section II.D. As seen from Fig. 4, the outcome of cell injection with combined vision + force feedback is superior to vision feedback alone for each subject. The effect of learning is shown in Fig. 5. As observed from the figure, there is a significant learning effect using vision feedback in S2 compared to using combined vision + force feedback. There was no learning effect observed in S1 for both V and V + F feedback. Overall the average success for S1 and S2 is shown in Fig. 6. Paired t test was performed to evaluate whether there was significant difference in each trial between 2 data sets (V and V + F) for different scenarios S1 and S2. Table 1 shows the p value generated when comparing the data sets for each trial and average success, for scenarios S1 and S2. The p value was less than alpha value (0.05) for all trials in S1 and for trials 1 & 2 in S2. Thus comparing the 2 data sets (V & V+F) in S1 for each trial and in S2 for trials 1 & 2, there exists a probability of greater than 95%, that there was a significant difference between the two data sets. There was no significant difference when comparing the data sets for trials 3, 4 & 5 in S2, which shows that the subjects had a learning effect. The p value obtained when comparing the average success for the 2 data sets for S1 and S2 was less than 0.0001 and equal to 0.005 respectively, leading to a probability of greater than 99.99 % that there was a significant difference between the data sets. Hence, there was a significant improvement in success rate using combined vision + force feedback compared to using vision feedback alone for S1 as compared to S2. Also there is a large standard deviation in vision feedback alone for S1 as compared to S2. Thus force feedback plays a major role in improving the success rate while injecting non transparent egg cells with a color/colorless dye or injecting transparent egg cells with colored material (trepan blue dye). Fig. 7 shows one of the injection tests which was unsuccessful and Fig. 8 shows one of the injection tests which was successful based on the feedback protocol outlined in section II.D. In most of the unsuccessful cell injection tasks the subjects perceived that they had penetrated the cell membrane while in reality they had not. Obviously the range of pipette movement by the subject cannot be very large since that could lead to cell rupture or the pipette reaching the other end of the cell membrane. The obvious advantage of force feedback is the perceived drop in injection force after the membrane is punctured, which resulted in a successful cell injection task. The average completion time taken by each subject to perform experiments with V and V+F feedback for S1 and S2 scenarios is shown in Fig. 9. As seen from the figure there was no significant difference in the completion time for each subject in V and V+F feedback. However there was a significant variation in completion time across subjects.
 |
(a) (b)  |
| Fig. 4. The percentage of successful cell injection for each individual using (V) and (V+F) feedback. a) In S1, the subject was not allowed to see the dye being injected. b) In S2, the subject was allowed to see the dye being injected. | Fig. 5. Percentage of successful injections for all five trials using only vision feedback and using combined vision + force feedback : (a) for S1 (b) for S2 . Learning effect is clearly observed for vision feedback in S2, as expected. |
(a) (b)  |
(a) S1
(b) S2
|
||||||||||||||||||||||||||||
| Fig. 6. In two experimental scenarios, S1 and S2, the percentage of successful cell injection for all 20 subjects. (a) For S1: Average success using only vision feedback (37%) and vision + force (V + F) feedback (81%). (b) For S2: Average success using only vision feedback (75%) and vision + force (V + F) feedback (89%). | Table 1: The p value generated when comparing the 2 data sets (V & V+F) for each trial and average success for: a) S1 and b) S2 ( * indicates p < alpha, where alpha = 0.05 ) |
 |
 |
| Fig. 7. Example of an unsuccessful injection of zebrafish egg cell. (A) The injecting pipette approaching the cell. (B) The subject stopped the forward motion of the pipette being certain that the cell membrane is punctured, at this moment trepan blue dye was injected into the cell. (C) The pipette being withdrawn. (D) The trepan blue dye remained outside the cell. | Fig. 8. Example of a successful injection of zebrafish egg cell. (A) The injecting pipette approaching the cell. (B) The injecting pipette in contact with the cell membrane. (C) The subject stopped the motion of the pipette when perceiving an apparent drop in force; at this moment trepan blue dye was injected into the cell. (D) The trepan blue dye remained inside the cell. |
(a)
|
(b)
|
| Fig. 9. The average completion time taken by each subject to perform cell injection with vision (V) feedback and vision + force (V+F) feedback for: a) S1 and b) S2. | |
Conclusion: We have developed a cell injection system with force feedback capability along with visual display. The force sensing system is capable of measuring forces in the µN range. As we performed experiments only on zebrafish egg cells (transparent), we created two different scenarios: S1 and S2, which covered the most general cases in biomanipulation tasks. Our results confirm that subjects had a higher degree of success in injecting the desired material (trepan blue dye) into the cell with simultaneous vision + force feedback compared with vision feedback alone. Overall, considering all 40 subjects, the research hypothesis was validated through our experimental results. Statistical analysis proved that there is a significant difference between the 2 data sets (V and V+F). Our findings confirm that a system with force feedback capability when combined with vision feedback can lead to potentially higher success rates in transgenesis, specifically where mechanical manipulation techniques are involved. In our future work, we plan to develop microelectromechanical systems (MEMS) based cell injection system to hold the cell and fixate the nucleus so that the genetic material can be injected directly into the nucleus for efficient transgenesis.
C. Micro Robotic Gripper for Cell Manipulation
Motivation: Micromanipulation of individual biological cells using pipette technology is prevalent in the field of molecular biology. The technology requires the operator to undergo long training and the success rates are low. Moreover the poor control of pipette suction force to hold an individual cell may rupture the cell membrane. Therefore, there exists a need to have a micro tool for handling biological cells.
Technical Approach: Polypyrrole actuators are based on the reversible volume change of the conducting polymers upon oxidation and reduction. The volume change of the polymers is controlled by electrochemical processes, which cause ion insertion and expulsion. For a polymer (P) doped with a large immobile anion A- in contact with an electrolytic solution containing a small mobile cation C+, the electrochemical reaction is given by (Smela 1999, Jager et al 2000):
P+ (A-) + C+ (aq) + e- = P0 (A- C+)
The microactuators presented in our work consists of PPy(DBS) (polypyrrole doped with dodecylbenzene sulfonate anions) in a bi-layer configuration with gold (Au) acting as both a structural layer and an electrode. When a negative potential (greater than -2 V but less than 0 V) is applied on Au, PPy is reduced to its neutral state [PPyo (Na+DBS-)] due to the insertion of cations (Na+) from an ion source (NaDBS electrolytic solution) into PPy, resulting in expansion. When a positive potential is applied (>= 0 V), the PPy is oxidized [PPy+ (DBS-)] because the cations (Na+) now move out of PPy back into the electrolyte solution and the material contracts. Since Au does not change volume on applying potentials, the expansion and contraction of PPy results in an out of plane bending motion. In a microfabrication process, PPy-Au mircoactuators can be released using two methods, namely, sacrificial layer or differential adhesion (Smela 1999, Jager et al 2000) technique. Sacrificial layer may result in thin gold layer at the edge of the cantilever, resulting in maximum bending at the edge. A second drawback is that for large areas, underetching takes a significantly long time. This process increases the chance of damaging PPy. We therefore chose the differential adhesion method, which is based on the property of poor adhesion between gold (Au) and silicon (Si). A pattern of adhesion promoting layer (Chromium/Titanium) results in adhesive and non-adhesive areas for depositing Au layer. Polypyrrole is electrochemically deposited and patterned over the Au layer. A minor part of the PPy/Au bi-layer is in contact with adhesive Cr layer and functions as an anchor holding the actuator to the surface. Major part of the bi-layer is in contact with the Si, to which Au does not adhere. As a result, activating the bi-layer causes it to pull itself free from the surface. The fabrication process is outlined below:
 |
 |
| Fig 1. Fabrication sequence of the cantilever bi-layers (PPy-Au) by differential adhesion method: (a) Deposition and patterning of Cr (50 Å) followed by Au (250 Å) (b) Deposition of second layer of Au (2000 Å) (c) Electrodeposition and patterning of polypyrrole by RIE etching (d) Etching Au electrode followed by etching Cr. | |
Experimental Setup: The LOTUS microgripper was electrochemically controlled by using a potentiostat, EA161 (e-Daq, Inc.). The gold electrode of the LOTUS microgripper acts as a working electrode. We used a gold coated Si wafer as a counter electrode and an Ag/Agcl electrode as a reference electrode. The LOTUS microgripper was actuated in a solution containing 0.1M aqueous sodium dodecylbenzenesulfonate (NaDBS). As the DBS anion is immobile, it does not move during redox reaction and the volume change is determined by cation (Na+) transport. The cantilever bi-layers were actuated within the limits of -1.8V to 0V against Ag/Agcl reference electrode. It has been observed that during the first reduction scan, there is no movement of the bi-layers even though one would expect the PPy to expand and bend backwards due to the insertion of cations. During the first oxidation, the PPy(DBS) contracts and the bi-layers bend away from the wafer. After the initial reduction and oxidation, the bi-layers bend upon oxidation (Na+ ions are expelled from PPy) and straighten upon reduction (Na+ ions are inserted into PPy). The potential was varied smoothly by cyclic voltammetry and the bi-layer motion was observed to be smooth and unidirectional.
Results: The fabricated LOTUS microgripper was tested on zebrafish egg cells (diameter ~ 700 µm). The egg cells were obtained from adult female zebrafish. Initially, an egg cell was placed on the anchoring site of the gripper (Fig. 2b). A cyclic voltage (trapezoidal wave) was then applied to the LOTUS microgripper with a peak voltage of negative 1.5V. As expected the cantilever bi-layers did not actuate and laid flat during the first reduction scan. When the voltage varied from negative 1.5V to 0V, the bi-layers curled up. After the initial reduction and oxidation, the bi-layers remained in the curled position at 0 V. We ran a couple of cycles from 0 to -1.5V at a rate of 0.9V/sec to activate the PPy/Au bi-layers to their maximum range. Fig. 3c shows the maximum bending of the bi-layers holding the zebrafish egg cell at 0V. The bi-layers lay flat when the voltage varies from 0 to -1.5V (Fig. 3d). The bi-layers curl in again and grab the egg as voltage is varied from -1.5V to 0 V (Fig. 2e & 2f). The experiment demonstrated successful actuation of the gripper grabbing the zebrafish egg cell without rupturing the cell membrane. Fig. 3 shows the actuation of the LOTUS microgripper without zebrafish egg cell. In this case a trapezoidal wave potential was applied to activate the LOTUS microgripper. The voltage was varied from negative 1.8V to 0V at a rate of 0.9V/sec for couple of cycles to achieve the maximum bending (Fig. 3a).
 |
| Fig. 2. (a) LOTUS microgripper with six cantilever bi-layers, (b) Zebrafish egg cell placed on the anchoring site of the gripper. (c) LOTUS microgripper holding the egg cell (d) The bi-layers curl out at -1.5V. (e) & (f) The bi-layers curl in again and grab the egg as voltage is varied from -1.5V to 0V. |
 |
| Fig. 3. Actuation of the LOTUS microgripper after initial reduction and oxidation. (a) The bi-layers are curled up at 0V i.e. the oxidized state, (b) The bi-layers uncurl upon applying - 1.8V at a rate of 0.9V/sec (c) The bi-layers are in reduced state at -1.8V. The bi-layers curl up when the voltage varies from -1.8V to 0V as shown in (d) to (f). |
Conclusion: We have developed a LOTUS microgripper comprising of six PPy/Au cantilever bi-layers capable of holding an individual biological cell. Polypyrrole was chosen as an actuation material because it actuates in aqueous media at low voltages (less than 2V) and at room temperature making it ideal for biomanipulation. In this article, polypyrrole was doped with dodecylbenzenesulfonate anions (PPy(DBS)) and fabricated with gold (Au) to form a bi-layer configuration. Differential adhesion technique was used to fabricate the PPy/Au bi-layers. The volume change of polypyrrole was determined by cation (Na+) transport. Experiments on zebrafish egg cell demonstrated the successful actuation of the LOTUS microgripper. The egg cell was grabbed by the LOTUS microgripper without rupturing the cell membrane. In our future work, we will calibrate the actuation voltage with the force generated from the LOTUS microgripper. Calibration is a major step in quantifying the force required to hold an individual biological cell.
For further information, please contact:
Prof. Jaydev P. Desai
Director, RAMS Laboratory
Department of Mechanical Engineering
Room 0160, Building 088
Glenn L. Martin Hall
University of Maryland
College Park, MD, 20742
Email: jaydev (at) umd.edu
Phone: 301-405-4427
Fax: 301-314-9477