Comparison of PDMS and NOA Microfluidic Chips: Deformation, Roughness, Hydrophilicity and Flow Performance

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript investigates a promising material, Norland Optical Adhesive (NOA), for microfluidic fabrication. This study compares the compliance and deformation properties of three different characteristic sized microfluidic devices made of PDMS and NOA on the basis of the Young's modulus, roughness, contact angle, channel width deformation, flow resistance and compliance. This work could potentially encourage NOA devices to become the more prevalent option in microfluidic research and in high pressure microfluidic flow systems. Therefore, I’d like to recommend a “minor revision” based on a few comments to consider for improvement of the manuscript.

1. Specific issues addressed with the NOA microfluidic devices should be illustrated in the abstract and introduction.

2. Will bubbles be generated in NOA 63 during processing?

3. Description of Figure 6 should be adjusted to the position near the Figure 6.

4. Symbols in formulas should be unified and standardized, for instance, symbols in line 275, 284, 288 are not in italic format; the difference of pressure is defined repeated in line 236 and 251.

5. Scale bar should be added in Figure 5.

Comments on the Quality of English Language

Minor editing of English language required

Author Response

October 17, 2023

 

Dear Editor, 

 

We would like to thank the reviewers for taking the time to read this paper and providing valuable comments. We have revised the manuscript and have made modifications in relation to the comments provided. Specific modifications are highlighted in the text through the “Track Changes” function. In addition, English has been reviewed, with a particular focus on favoring precision of language and prioritizing clarity (highlighted in blue).

 

Please see below for our detailed answers to each comment. 

 

REVIEWER 1: 

1. Specific issues addressed with the NOA microfluidic devices should be illustrated in the abstract and introduction.

 

The introduction has been enhanced by incorporating additional context and providing examples of situations where precise and consistent dimensions in microfluidic channels are crucial.

 

“Constant and precise dimensions of microfluidic channels are crucial for various applications that require precise control of fluid flow and interactions. The dimensions of microfluidic channels directly influence fluid behavior, including flow rate, pressure drop, mixing, and diffusion [1]. Any variation in channel dimensions can lead to inconsistent results and unreliable experimental outcomes. The flow rate of a fluid through a microfluidic channel is directly proportional to the channel dimensions, such as width, height, and length [2]. By precisely controlling these dimensions, researchers can manipulate the flow rate and achieve desired fluid velocities [3]. This is crucial for applications such as drug delivery, where precise control of flow rate is necessary to ensure accurate dosing. In microreactors, where chemical reactions occur within microfluidic channels, precise dimensions ensure efficient mixing and reaction rates [4]. Similarly, in biological applications, such as cell culture and analysis, constant channel dimensions allow for consistent fluid flow and controlled interactions between cells and reagents [5]. Compliance is the change in volume for any given applied pressure. When using a syringe pump as the system’s input source, the time it takes for the system to reach steady flow conditions can vary from seconds to hours depending on the fluidic resistance and compliance. This time, between the initial state and when steady flow conditions are reached, is called the response time or the characteristic time.”

 

2. Will bubbles be generated in NOA 63 during processing?

The formation of bubbles in NOA 63 during processing can occur if it is not handled properly. Bubbles can be introduced during mixing, dispensing, or curing. When handling samples we carefully poured NOA on the glass slide and moved the glass slide to spread evenly the NOA.  No apparent bubbles were present. During the measurement of the characteristic time and pressure-flow, chips were primed with water prior measurement and a microscope examination confirmed the absence of any bubbles.

 

 

3. Description of Figure 6 should be adjusted to the position near the Figure 6.

The modification was made in reviewed manuscript (line 361)

4. Symbols in formulas should be unified and standardized, for instance, symbols in line 275, 284, 288 are not in italic format; the difference of pressure is defined repeated in line 236 and 251.

The modifications were made in reviewed manuscript, lines 256, 263, 264, 266, 267, 271, 278-280, 283, 285, 304, 305.

5. Scale bar should be added in Figure 5.

The modifications were made in reviewed manuscript, line 354.

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Recommendation: Publish after major revision noted.

Comments:

The manuscript presents an overview of the study comparing microfluidic devices made from polydimethylsiloxane (PDMS) and Norland Optical Adhesive (NOA) with a focus on compliance, deformation properties, and various other characteristics. However, considering the below suggestions and opinions, the reviewer suggests the paper should have a careful revision before being accepted by Micromachines.

1.     Why did the author choose to use NOA63 as the experimental material instead of other types of NOA adhesives? Are there any differences of deformation, roughness, hydrophilicity and flow performance between different types of NOA adhesives?

2.     In Figure4, why PDMS has a higher channel increase as the flow rate increase for chip A and B than NOA, while the NOA has a higher increase in the channel width than PDMS for Chip C?

 

3.     The authors mentioned NOA devices has potential applications in high pressure microfluidic flow systems. The study on the maximum pressure that microfluidic chips made with NOA adhesive can withstand is highly meaningful.

Comments on the Quality of English Language

The sentences are generally well-structured, making the content easy to follow. However, there are some lengthy sentences that could be broken down for improved readability.

Author Response

October 17, 2023

Dear Editor, 

We would like to thank the reviewers for taking the time to read this paper and providing valuable comments. We have revised the manuscript and have made modifications in relation to the comments provided. Specific modifications are highlighted in the text through the “Track Changes” function. In addition, English has been reviewed, with a particular focus on favoring precision of language and prioritizing clarity (highlighted in blue).

 

Please see below for our detailed answers to each comment. 

 

REVIEWER 2:

1. Why did the author choose to use NOA63 as the experimental material instead of other types of NOA adhesives?Are there any differences of deformation, roughness, hydrophilicity and flow performance between different types of NOA adhesives?

 

“NOA63 and NOA81 were both considered, as they were both used in previous studies as microfluidic device materials.  NOA63 was chosen due to its increased viscosity 2,500 cps at 25C and low shrinkage of approximately 1.5%.”

 

This information has been added in the text line 118.

 

 

2. In Figure 4, why PDMS has a higher channel increase as the flow rate increase for chip A and B than NOA, while the NOA has a higher increase in the channel width than PDMS for Chip C?

 

NOA chip C has in fact a higher increase in the channel width percentage for the flow rate of 100 µl/hr with a percentage of 8.6% for NOA compared to 7.8% for PDMS (p=0.000001). This difference is probably due to the measurement of the channels. For the flow rates of 25 and 50 µl/hr, PDMS has a higher increase in the channel width percentage than the NOA (at 25 µl/hr: 3.6% for NOA and 4.6% for PDMS and at 50 µl/hr: 4.2% for NOA and 6.0% for PDMS).

 

3. The authors mentioned NOA devices has potential applications in high pressure microfluidic flow systems. The study on the maximum pressure that microfluidic chips made with NOA adhesive can withstand is highly meaningful.

 

This information was added in the introduction.

 

“This allows NOA to be used in high pressure flow systems with minimal compliance as shown by Elodie Sollier et al. when comparing PDMS to other forms of polymer-based materials, including NOA, in high pressure flow systems [6]. Their findings revealed that the maximum pressure (Pmax) at which delamination occurs in NOA is approximately 74-79 PSI, while PDMS bonded to glass experienced delamination at pressures as low as 36 PSI [6].”

 

Please see the attachment. 

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript “Comparison of PDMS and NOA Microfluidic Chips: Deformation, Roughness, Hydrophilicity and Flow Performance” by Turcitu, et al. describes the comparative properties of PDMS and NOA microfluidic chips in terms of their compliance, roughness, hydrophilicity and flow performance. This is achieved using by designing three different microfluidic chips differing in their geometries. The manuscript is well written in an easy and clear language to make it easy to understand by the non-experts of the field. The authors have explored the topic thoroughly and made an honest attempt to present the results obtained in a fair manner.

 

Because all four parameters has not been studied together with respect to PDMS and NOA microfluidic devices before, there is novelty in the manuscript that makes it potentially interesting to both the microfluidics and polymer communities. As difficult it is to work with PDMS due to its sticky and viscous nature, this work provides an alternate solution with reasonable arguments for its use and thus is appropriate for publication. However, a few concerns should be addressed prior to full acceptance. First, a few more examples should be mentioned where device compliance and flow performance are important criteria for experimental validation. Second, the importance of surface roughness is not explained at any point in the manuscript. Why is it important to measure surface roughness? Third, there should be some discussion about how the presented comparison would or would not be challenged by the characteristics of a stiffer PDMS for example, base:curing agent ratio of 5:1.

Other minor comments:

·      Labelling of ports E and G is not clear in Figure 1.

·      Why samples of different dimensions for PDMS and NOA63 are tested for mechanical strength in section 2.4?

·      In figure 4, PDMS should be replaced with NOA for initial channel widths of 101.7μm, 38.2 μm and 25.6 μm.

·      In figure 7, caption for (d) should be corrected as 100 μl/hr instead of 25 μl/hr.

·      Figure 8 caption should be corrected for “three” different flow rates instead of “four”.

Author Response

October 17, 2023

 

Dear Editor, 

 

We would like to thank the reviewers for taking the time to read this paper and providing valuable comments. We have revised the manuscript and have made modifications in relation to the comments provided. Specific modifications are highlighted in the text through the “Track Changes” function. In addition, English has been reviewed, with a particular focus on favoring precision of language and prioritizing clarity (highlighted in blue).

 

Please see below for our detailed answers to each comment. 

 

REVIEWER 3:

1. First, a few more examples should be mentioned where device compliance and flow performance are important criteria for experimental validation.

As mentioned to reviewer 1, the introduction has been enhanced by incorporating additional context and providing examples of situations where precise and consistent dimensions in microfluidic channels are crucial.

 

“Constant and precise dimensions of microfluidic channels are crucial for various applications that require precise control of fluid flow and interactions. The dimensions of microfluidic channels directly influence fluid behavior, including flow rate, pressure drop, mixing, and diffusion [1]. Any variation in channel dimensions can lead to inconsistent results and unreliable experimental outcomes. The flow rate of a fluid through a microfluidic channel is directly proportional to the channel dimensions, such as width, height, and length [2]. By precisely controlling these dimensions, researchers can manipulate the flow rate and achieve desired fluid velocities [3]. This is crucial for applications such as drug delivery, where precise control of flow rate is necessary to ensure accurate dosing. In microreactors, where chemical reactions occur within microfluidic channels, precise dimensions ensure efficient mixing and reaction rates [4]. Similarly, in biological applications, such as cell culture and analysis, constant channel dimensions allow for consistent fluid flow and controlled interactions between cells and reagents [5]. Compliance is the change in volume for any given applied pressure. When using a syringe pump as the system’s input source, the time it takes for the system to reach steady flow conditions can vary from seconds to hours depending on the fluidic resistance and compliance. This time, between the initial state and when steady flow conditions are reached, is called the response time or the characteristic time.”

 

2. Second, the importance of surface roughness is not explained at any point in the manuscript. Why is it important to measure surface roughness?

 

This point has been incorporated into the introductory.

“It is known that more surface roughness increases the surface area which lead to absorption effect and possibility of trapped air bubbles during flow [20]. So, it is important to study the surface roughness of both materials due to the challenge of the micro-scale control, the interfacial properties, the complex boundary effects and the lack of theoretical characterization [20].

A precision was added in the introduction line 73.

 

“Surface roughness increases the surface area which leads to absorption effect and possibility of trapped air bubbles during flow [20]. So, it is important to study the surface roughness of both materials due to the challenge of the micro-scale control, the interfacial properties, the complex boundary effects and the lack of theoretical characterization [20].”

 

3. Third, there should be some discussion about how the presented comparison would or would not be challenged by the characteristics of a stiffer PDMS for example, base:curing agent ratio of 5:1.

 

PDMS ratio of 5:1 has a Young Modulus of 2.75MPa which is higher than the Young Modulus of PDMS ratio 10:1 [33]. By having a higher, Young Modulus, less compliance is expected due to the stiffness of the material. But, we can assume the benefit will not be significant compared to NOA that has a Young Modulus 800 times higher.

 

A precision was added in the discussion line 445.

 

“While is a well-established fact that preparing PPDMS with a 1:5 ratio increases its stiffness, resulting in a Young's modulus of approximately 2.7 MPa [33], this value remains significantly inferior to that of NOA, which is 800 times higher.”

 

 

4. Labelling of ports E and G is not clear in Figure 1.

The modifications were made in reviewed manuscript line 93.

5. Why samples of different dimensions for PDMS and NOA63 are tested for mechanical strength in section 2.4?

 

There were complications for making the NOA samples. We poured NOA on top of the PDMS sample and then cured it. Unfortunately, we had to avoid the margins of the PDMS sample so NOA does not drip. For the height, if too much NOA is poured, it also drips. The slight variation in dimension falls within the realm of uncertainty. Nevertheless, as our Young modulus measurements for both PDMS and NOA align with the values reported in the literature, we can confidently assert that the dimensions did not have much impact on our results. (see Table 3).

 

6. In figure 4, PDMS should be replaced with NOA for initial channel widths of 101.7μm, 38.2 μm and 25.6 μm.

 

The modifications were made in reviewed manuscript, line 337.

 

7. In figure 7, caption for (d) should be corrected as 100 μl/hr instead of 25 μl/hr.

The modifications were made in reviewed manuscript, line 386.

8. Figure 8 caption should be corrected for “three” different flow rates instead of “four”.

The modifications were made in reviewed manuscript, line 394.

 

 

Please see the attachment. 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I think the authors didn't response the question about "In Figure 4, why PDMS has a higher channel increase as the flow rate increase for chip A and B than NOA, while the NOA has a higher increase in the channel width than PDMS for Chip C?". In the response letter, they only revisited the results, and I believe it is necessary for them to add data explaining the variations in the results among chips A, B, and C.

Author Response

After a thorough review of your comments, we have identified an error in our data file. We conducted a comprehensive data verification process and found that there was a single numerical transcription error, as evidenced by the Excel file provided. The average value on the page labeled '100 ul,' highlighted in red, was inaccurately reflecting only the last cell's value instead of the average of all the measurements. This error had a cascading effect on our calculations, and the affected cells in the sheet named 'Results' have been highlighted in red. It's important to note that the p-values, indicating statistical significance, were not affected by this error because they were derived directly from the raw data using the GraphPad software.

We have rectified this mistake and updated the new value in Figure 4. To support this correction, we have attached the raw data files to this response. We believe this clarification addresses the situation and enhances the validity of our results. We want to thank you immensely because your vigilance helped uncover this unfortunate error.

The descriptive text was adjusted accordingly (line 347-349):

"For chip C, at 100 µl/hr, there is an increase of 7.1% for the PDMS device and of 6.8% for the NOA device. PDMS has a higher channel increase as the flow rate increase for chip A, B and C than NOA."

Author Response File: Author Response.pdf