Review Reports - Development and Testing of a Dual-Driven Piezoelectric Microgripper with High Amplification Ratio for Cell Micromanipulation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper, the authors have introduced a dual-driven piezoelectric microgripper with high displacement amplification ratio, large stroke and parallel gripping. The design incorporates three types of flexure-based mechanisms connected in series to achieve three-stage displacement amplification. Additionally, the use of a parallelogram mechanism isolates parasitic rotational movements and ensures parallel movement of the gripping jaws. The authors have established the kinematics, statics, and dynamics models of the microgripper using pseudo-rigid body and Lagrange methods, and have optimized key geometric parameters. Finite element simulations and experimental tests have been conducted to verify the performance of the microgripper. Overall, this study presents valuable findings and is well-suited for publication in Machines. I would, however, like to raise the following points for further clarification:

1. It appears that the maximum displacement value presented in Figure 12 (a) is different from that in Figure 12 (c). Moreover, the maximum displacement value presented in Figure 12 is also different from that in Figure 10 (b). Could the authors briefly explain the reasons for these variations?

2. In Figure 14 (a), the meaning of the red dashed lines is unclear. It is recommended that the authors provide clear definitions of these lines in the figure caption for better understanding.

3. The authors state that the nondestructive micromanipulation of zebrafish embryos was successfully achieved using the developed microgripper. To strengthen this claim, I recommend that the authors include clearer images of the cell both before and after gripping to facilitate a more direct comparison. From the current images in the manuscript, it is difficult to determine whether the cell remains unchanged after manipulation.

Comments on the Quality of English Language

Minor editing of English language required.

Author Response

Comment 1: It appears that the maximum displacement value presented in Figure 12 (a) is different from that in Figure 12 (c). Moreover, the maximum displacement value presented in Figure 12 is also different from that in Figure 10 (b). Could the authors briefly explain the reasons for these variations?

Response: In Fig. 12 (a) and (c), a step signal and a sinusoidal signal with different amplitudes are selected as examples to test the closed-loop tracking performance of the microgripper, which is just for the testing purpose. Their amplitudes can be specified randomly within the motion stroke of gripper. As for the Fig. 10 (b), we want to test the open-loop output performance of the gripper. Since the maximum input voltage of 150V is applied to the PEA, the corresponding maximum output displacement is generated, which is the reason why the maximum value is different from that in Fig. 12.

Comment 2: In Figure 14 (a), the meaning of the red dashed lines is unclear. It is recommended that the authors provide clear definitions of these lines in the figure caption for better understanding.

Response: In Fig. 14 (a), the red dashed lines represent that the closed-loop control bandwidth of the gripper is 23.1 Hz under the magnitude of -3dB. According to the Reviewer’s suggestion, we have provided explanation in the figure caption and highlighted them in red.

Comment 3: The authors state that the nondestructive micromanipulation of zebrafish embryos was successfully achieved using the developed microgripper. To strengthen this claim, I recommend that the authors include clearer images of the cell both before and after gripping to facilitate a more direct comparison. From the current images in the manuscript, it is difficult to determine whether the cell remains unchanged after manipulation.

Response: Considering the Reviewer’s suggestion, we have updated the Fig. 13 to make it more clear. All these changes are highlighted in red. The following figures are the original microscopic images （Please see the attachment）.

Fig. 13 (a)

Fig. 13 (b)

Fig. 13 (c)

Special thanks to you for your good comments.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript is generally interesting. However, it needs major revisions before it can be accepted for publication.

The compliant mechanisms with piezoelectric stack actuators are a good way to obtain a precise motion for positioning.

Pay attention to the representation of the equivalent pseudo-rigid model in the G point from Figure 2 and Figure 3. There is one kinematic element HI and G doesn’t split the HI; in this case, the representation is similar to D.

Please increase the font size in Equation 37.

Figure 9. (a) Experimental setup for cell microgripping, is hard to see because the words cover the image.

Author Response

The manuscript is generally interesting. However, it needs major revisions before it can be accepted for publication.

Comment 1: The compliant mechanisms with piezoelectric stack actuators are a good way to obtain a precise motion for positioning.

Response: Thanks for the Reviewer’s positive evaluation.

Comment 2: Pay attention to the representation of the equivalent pseudo-rigid model in the G point from Figure 2 and Figure 3. There is one kinematic element HI and G doesn’t split the HI; in this case, the representation is similar to D.

Response: Thanks for the Reviewer’s kind suggestion, and we have checked Figs. 2 and 3 carefully to make the representation of points G and D correct. All modifications have been highlighted in red.

Comment 3: Please increase the font size in Equation 37.

Response: According to the Reviewer’s suggestion, we have increased the font size of Eq. (37). All modifications have been highlighted in red.

Comment 4: Figure 9. (a) Experimental setup for cell microgripping, is hard to see because the words cover the image.

Response: To clearly demonstrate the Fig. 9 (a), the original image without words is given as follows (Please see the attachment).

Fig. 9 (a)

Special thanks to you for your good comments.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Please find my review attached.

Comments for author File: Comments.pdf

Author Response

Comment 1: Page 4 Figure 2 and Line 132: Duplicate use of capital letter K

In the figure A, B, … , K represent the center of rotation of the flexure hinges.

Page 6 Figure 5 and Line 163: Duplicate use of capital letter K

… corresponding rotation angles αA, αC, …, αK of the flexure hinges A, C, … , K can be derived …

Page 7 Line 173 and equations 20 and 22: Duplicate use of capital letter K

… (e.g. H, I, J, K)

Page 8 above Line 212 equation 37: Duplicate use of capital letter K

.. i = H, I, J, K

Page 9 Line 251/252 Duplicate use of capital letter K

(i.e. H,I,J,K)

Compare usage of capital letter K above to usage below:

Page 9 Line 228 and equations 43 and 47: Duplicate use of capital letter K

Where M, K, F represent equivalent mass, stiffness and generalized force of the microgripper.

Response: Thanks for the Reviewer’s kind suggestion. In this paper, the capital letter K has two formats. When it represents the flexure hinge, it is denoted by upright type; but when it is for the stiffness, it is italic type. So, it is clear to identify the meaning of capital letter K.

Comment 2: Page 7 Line 191 Replace statement: During the process of operating the targets, the movements of the microgripper usually includes two stages, i.e. approaching stage and grasping stage.

with statement: During micromanipulation of biological cells, the movements of the microgripper consists of two stages, i.e. approaching stage and grasping stage.

Response: According to the Reviewer’s suggestion, we have replaced this statement. All modifications have been highlighted in red.

Comment 3: Page 7 Line 192 to Line 201

Provide more detailed description of the approaching stage and grasping stage Are during the approaching stage gripping jaws open, are they closing? When do gripping jaws close, is it during approaching or during grasping stage? What is the position of the biological cell during the approaching and grasping stage?

Biological cells could be considered as elastic objects ( see: Kuznetsova et all., Atomic force microscopy probing of cell elasticity, Micron 38, 2007, pp. 824-833. Kodera et all. Microgripper Using Soft Microactuators for Manipulation of Living Cells, Micromachines 2022, 13(5), 794 open access)

Please, explain possible effects on grasping stage.

Response: According to the Reviewer’s suggestion, we have provided more detailed description of the approaching and grasping stages. Besides, during the approaching stage, the gripping jaws are open, and when the gripping jaws close, it is during the grasping stage. During the approaching and grasping stage, the cell lies at the middle of the gripping jaws.

It is really true as Editor pointed out that, biological cells could be considered as elastic objects. So, during the grasping stage, the cell will gradually deform under the gripping force. Once the cell deformation exceeds the maximum value it can withstand, the cell will break down.

Comment 4: Page 9 Line 233 and Line 253: Explain why is the microgripper natural frequency improved if it is increased and why your optimization goal is to maximize the first order natural frequency.

Response: In the application of cell micromanipulation, we hope the developed microgripper can operate within as high bandwidth as possible. So, if the natural frequency can be increased, we think the performance of the microgripper is improved. The reason why the optimization goal is to maximize the first order natural frequency, lies in that the 1-st natural frequency can reflect the operation bandwidth of the microgripper, and in lots of researches on the microgrippers, the natural frequency is often chosen as one of the performance optimization indices. Generally, the higher the natural frequency, the wider the operation bandwidth. But due to the physical constraints, such as the size of flexure hinges, the stiffness of the gripper, and the output displacement, the natural frequency can not be selected very high as we want. Thus, the optimization goal is set to maximize the first order natural frequency subjected to specified physical constraints.

Comment 5: Page 9 Line 234 You state: …, natural frequency of the microgripper can be improved by increasing the equivalent stiffness or decreasing the equivalent moment of inertia.

You are referring to equation 47, where M is equivalent mass, therefore change your statement to:

…, natural frequency of the microgripper can be improved by increasing the equivalent stiffness or decreasing the equivalent mass.

Response: According to the Reviewer’s suggestion, we have changed this statement. All modifications have been highlighted in red.

Comment 6: Page 9 Lines 248 to 259 optimization constraints

Explain: Is it possible to put constraint on the microgripper weight?

Explain: Why is stiffness kIN constrained to 0.2kPEA? Don’t you want to maximize it?

Response: It is possible to consider the constraint on the microgripper weight. But in the paper, it is not considered, and only the flexure hinges’ size, stiffness, and strength constraints are considered.

As for the second question, the input stiffness of the microgripper can not be too large because it will conversely suppress the output displacement of piezoelectric actuator. In fact, the output displacement of piezoelectric actuator is affected by the external load. The bigger the external load, the smaller the output displacement. Thus, to guarantee a big output of piezoelectric actuator (or big motion range of the microgripper), the input stiffness can not be too big, and is finally determined to be constrained to 0.2kPEA by repeatedly tests.

Comment 7: Page 12 Figure 8 If your microgripper’s construction is symmetrical, Explain: Why is there a slight difference in natural frequencies of 1st , 3rd , 5th mode versus 2nd, 4th, 6th mode?

Response: The structure of the microgripper is designed symmetrically, but due to the unavoidable computation errors in finite-element simulation software, like the mesh generation and calculation accuracy, it is difficult to guarantee the natural frequencies of 1st , 3rd , 5th mode all the same as the 2nd, 4th, 6th mode. In fact, this slight difference is acceptable.

Comment 8: Page 14 Figure 12: Write frequency of sine signal on the Figure 12 c and d, or below in Caption.

Response: According to the Reviewer’s suggestion, we have added the frequency of the sine signal in the figure caption. All modifications have been highlighted in red.

Comment 9: Page 15 Figure 13: Explain what is seen in Figure, by putting description in Figure. Is there on the left side of Figure 13 (a,b) is glass from glass slide, and is there on the right side of Figure 13(a,b) is water?

Response: According to the Reviewer’s suggestion, we have put description in the figure caption.

Besides, the left side of Fig. 13 is glass and the right side is water.

Comment 10: Page 15 Line 389 to 394 and Figure 14: Explain: What means grasping at bandwidth of 20Hz or at 40 Hz? Sine with frequency 20Hz has a period of 50 ms. Is it possible for a human to move in such a short time the microgripper forward to grasp a cell? At 40 Hz the period is even shorter.

Response: The bandwidth of 20Hz represents that the microgripper can track the reference signals well within the frequencies of 20Hz, and simultaneously guarantee the firm and successful grasping of cells. When the frequencies of reference signals exceeds 20Hz, the tracking accuracy of the microgripper gets worse, and the amplitude of output gradually reduces, leading to the cell can not be grasped firmly and the success rate also becomes low. Once the frequencies exceeds 40Hz, the amplitude of output is suppressed severely, and the microgripper is totally unable to grasp the cell.

In our experiments, the microgripper does not need the operator to move it forward to grasp a cell, and the cells are pre-placed at the middle of the gripping jaws. Although the human can not achieve so fast movement, it is possible for automated robotics micromanipulation in the future. Thus, the developed microgripper has the potential for future applications of robotic cell micromanipulation.

Comment 11: Page 15 Figure 13: Zebrafish embryo is quite squeezed after the cell is gripped. Explain why.

Response: Since the cell is an elastic object, it is squeezed after gripped by the microgripper. In Fig. 13, we want to demonstrate the clamping and releasing performance of the microgripper without mechanical damage, so a relatively large step signal is applied to make the cell deformation more obvious.

Comment 12: Page 15 Provide images of successful grasping results using the microgripper for smaller biological cells, for instance cells with radius 50 μm (10 times smaller than zebrafish embryo).

Response: In this paper, zebrafish embryos are selected as samples to validate the performance of developed gripper, because they are a class of widely used cell models in biomedical fields. It is a very kind suggestion for us to use the microgripper to grasp smaller cells. But, under the current experimental conditions we have, it is unavailable for us to obtain smaller biological cells with radius 50 um. In fact, from the experimental results, it is obviously found that the microgripper is effective for cell micromanipulation, and we believe that it is also possible to successfully grasp smaller cells thanks to the sufficient clamping stroke of 2180 um.

Comment 13: Page 16 Table 2: Add microgripper weight into the Table 2 where it is available. When a microgripper is attached to the robot the weight of the gripper represents load of a robot and therefore it is important that microgripper is as light as possible. I assume cell (zebrafish embryo) weight is negligible in comparison to microgripper weight.

Response: Thanks for the Reviewer’s kind suggestion. The total weight of our developed microgripper is about 123.8 g. According to the Guest Editor’s suggestion, we deleted the Table 2. To make it more clear, we have added the weight information of the gripper in the experimental setup.