Directly Printed Hollow Connectors for Microfluidic Interconnection with UV-Assisted Coaxial 3D Printing

Round 1

Reviewer 1 Report

The paper is very well organized with precise details and illustrations along with a good flow in the overall description of the project. Very well-written except for a few grammatical errors which are indicated below.

Line 23: Check grammar – “can be achieved”.

Line 25: Check grammar – “as size decreases” or “with decreasing size”.

Line 27: Avoid using apostrophes regularly. Sentence can be rewritten to make more sense.

Line 28: Start of the sentence can be rewritten.

Line 31: last word “those” can be replaced with “the following”

Line 36: Avoid using Microfluidics as a noun. Please use terms like “Microfluidic devices”

Line 46: “Advantages of being scalable”

Line 69: Please replace the word “applied” with another suitable term.

Line 89: Unwanted use of apostrophe

Line 224: I suggest replacing the word “Zoom-in” with “magnified”.

Line 226: Remove the term “was” at the end of the line.

Author Response

Please see the attachment. Thanks!

Author Response File: Author Response.docx

Reviewer 2 Report

This paper presents a way to improve microfluidic world-to-chip interfaces by a 3D direct printing method. By improving a commercially available stereolithographic system with a homemade coaxial nozzle, the authors printed hollow capillaries. They used water as sacrificial material. The experimental 3D setup, liquids, and manufactured materials are very well described and analyzed.

What is unclear in this paper is where the connector is supposed to be integrated in the manufacture process of a microfluidic chip. Figure 3 show a graphical suite of the microfluidic chip manufacture, but steps are missing in the end. Where is the PDMS bonding? This Figure should be exploited to explain where the 3D printed connector fit in the manufacturing process.

Presented as it is, it is unclear how the printed connectors are integrated in the fabrication process of microfluidic devices and therefore what is the interest of such connectors? You should explain more clearly, where in the process of microfluidic manufacturing such connectors are integrated?; what type of connectors they intend to replace?; how such connectors will facilitate chip to world interface?

The fact that you chose to integrate your connectors to a PDMS/glass microchannel is awkward. As a matter of fact, the main advantage of PDMS microchannels is that they don’t need any connectors...

I am convinced that the work you are presenting is very interesting to the field, and should be published, but you may not have choosen a good example. Connecting devices manufactured by hot embossing or injection molding, or laminated devices (microchannel defined by thickness of double sided tape) would make more sense? You could also insist on the fact that 3D printed connectors allow more versatility in the design of the microfluidic chip, as already for PDMS.

Also, could you think of a table comparing the advantages of your connector solution with the other connector available? (including the commercially available ones)?

Author Response

Please see the attachment. Thanks!

Author Response File: Author Response.docx

Reviewer 3 Report

The authors Xu, Lo, and Lee present in their manuscript “Directly Printed Hollow Connectors for Microfluidic

Interconnection with UV-assisted Coaxial 3D Printing” a technique to fabricate hollow microfluidic connectors that can be attached to PDMS-glass microchips. The manuscript demonstrates the 3D printing of straight microchannels made of a UV-curable adhesive using 3D printing. The optimization of curing conditions is presented.

The results are interesting and helpful for designing microfluidic connectors for glass chips, as an alternative to the time-consuming fabrication of chip holders. There are several unclear points in the manuscript. Therefore, I recommend that authors should clarify the following issues.

• The manuscript contains several language errors in almost all sections. For example, 4th line in the abstract “microfluidic connectors with advantages of simple, rapid and scalable.” – the sentence makes no sense. How a connector could be “rapid”? Also, the adjective usage is not correct. Also, line 6 – In the English language “and” is not a word to start the language. Line 7 – “adhesives curing” should be corrected as “curing of adhesives”, I suppose. “neighborhood of the coaxial nozzle” sounds not correct. 
• Introduction 2nd paragraph, line 7 – remove “in summary”. 
• Introduction 2nd paragraph, line 8 – As a common practice, reversible fixtures can easily be connected to the PDMS chips by punching holes one size smaller than the tubing size. Therefore, the sentence does not reflect a correct description of the state of the current methodology. 
• Figure 1 – I suppose the system Is a custom-tailored one, so the authors should be able to change the parts easily. I wonder why the authors performed an entire series of experiments with only one type of nozzle with a fixed OD and ID. They tried different flow speeds, but for a more complete story, more dimensions would have been explored since the flow profile in jetting mode changes a lot by geometry. 
• Figure 2 –the panels c and d show Tracepo simulation results. The authors should add a section to report the input parameters, as at this moment this part is not reproducible by a reader who wants to do the same type of simulation. 
• Figure 4 – I am not sure what this image adds to the story. It should be moved to supplementary data. Actually, I did not quite understand the motivation of the authors to perform FTIR of Loctite 3491? It feels like “filler data” in the manuscript. If they would have studied “the effect of unpolymerized Loctite on cells” kind of work, I would understand the investigation of appearance or disappearance of C-O bonds – but now it is not a part of their story.
• “gel point” should be corrected as the “gelation point” in the entire manuscript, including the figure tags and legends.
• In Figure 5, intensities seem not to follow a logical order, assuming that higher UV intensities would result in quicker polymerization. For example, the storage and loss modulus of G’ at 1.5 W/cm^2 evolves faster than 3.0 W/cm^2. Why is that?
• Figure 5 f – where are the error bars? What is the n number, and standard deviation? The data looks like not trustable.
• In section 3.1.2 the authors mention that “But higher UV intensity could accelerate the UV curing reaction, as confirmed with more precipitous rheological curves and smaller gel times.” So, higher UV intensity leads to shorter polymer chains. The consequence of this would be that the material is more vulnerable to stress. Such a parameter would be crucial since the fabricated part would be continuously exposed to stress from inside (flow pressure) and outside (mounting/dismounting of tubing). Why the authors did not investigate this parameter? 
• Table 2- what determines the success rate? What does 50% success rate mean? The table legend should include the n number (how many times these tests repeated for each data point)?
• Page 8, second paragraph. The authors performed experiments under certain flow rates (4 mL/min and 8 mL/min). Why did they select to work with these values? What is the relation between the “success rate” and the flow rate? The manuscript should report such a relation. 
• If the authors aim at the inlets and outlets on the already bonded (if I understood correctly) PDMS-glass chips, do the water or leaking UV-curable adhesive block the channels?   
• Did the authors investigate the roughness of inner channels? The distorted channel walls can affect the flow profile in microchannels.

Author Response

Please see the attachment. Thanks!

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The idea and technology for direct printing of connectors is excellent. This paper is of interest to the field. 

The authors answered to the reviewer's comments fully in the answer to the reviewers letter. But not all suggestions were followed in the manuscript. For example, it would be a plus if a table comparing the proposed technology for connection with the other commonly used ones (lab solutions and commercial ones) was proposed and discussed.

Again, I understand that the connections were printed on top of holes on a glass substrate, and I understand that bonding glass to PDMS is easy and almost impossible to fail. But: using connectors on glass to connect channels manufactured in PDMS is not a clear way to demonstrate your excellent technology in my opinion.

Author Response

Please find the attachment, thanks!

Author Response File: Author Response.docx

Reviewer 3 Report

In the revised manuscript, the authors improved the overall quality, especially the introduction reads much better now. However, they did not answer a large number of previously asked questions, some of which need to be addressed in this work and not in future work. A parametric study and explanation of UV curing times vs polymer structure would be interesting aspects to add. The authors indicated that they would consider future work, I would be interested in finding more information on these parts later if not know. 

As a crucial step: All graphs in the manuscript need to be revised and error bars should be added. For example, there are still no error bars in figure 4f. Please note that interpreting results from only one measurement is scientifically not correct. 

Author Response

Please find the attachment, thanks!

Author Response File: Author Response.docx