Well I can only speculate about this now. Basically the reaction between graphene and NPD is a redox-reaction. Frankly speaking the diazonium salt is cationic while it is no longer cationic after grafting. This implies that there is an electron transfer process in place. Most ppl (me included) believe that the electron is transferred from the graphene to the diazonium cation which will subsequently cleave N2 and react. If this happens by a single electron transfer before grafting or in a concerted process has not been investigated to the best of my knowledge so far. Lets assume for the sake of simplicity there is an electron transfer before grafting, which will leave a hole or a positive charge on the graphene and form a phenyl-sigma-radical from the diazonium compound. Graphite is known to readily react with radicals which to me makes this mechanism plausible enough to talk about it in public here. Lets now further assume the phenyl radical is grafted onto the graphene - what happens to the hole? If the process continues there will be more and more positive charge being accumulated on the graphene which at some critical point will stop the initial electron transfer step because it will become energetically increasingly less favourable. Additionally there will be coulomb repulsion preventing fresh diazonium salt approaching the meanwhile also positively charged graphene surface and the reaction will eventually stop.

Now the speculation begins: In the presence of a nucleophilic solvent like water, the solvent may eventually approach the holes on the graphene and lead to hydroxide groups being formed and protons being released into the solution. Hence the solvent will acidify. In the absence of a nucleophile in the solvent (and usually the counter ions for the diazonium salts are not nucleophilic as this would precent their very synthesis from the first place) there is no mechanism in place to reduce the positive charge on the graphene that stopped the reaction. In your case ACN would have to do this job but is not very good in doing so. If there is PF6 anions present there might be a route for grafting fluoride as a secondary functional group - however I am not aware of any literature discussing this.

One suggestion from Sandmeyer chemistry to provide the neccessary reducing species for the formation of the phenyl radicals would be adding Cu(I) salts to the solution, which get oxidized to Cu(II) and provide electrons. Would be intersting to see their effect tbh. Another possibility would be to do the reaction on a surface capable to provide electrons or compensation for the charge by backside functionalization - surface-OH groups on most oxides should be able to do the same trick as water does.

Hope I could provide a clearer picture about what is going on - if you need more information please feel free to ask further questions - its always the questions asked - not those kept by yourself - that helped to advance science.

My very best Regards,

JME

Are the papers you looking at using thermodegradation or electrolysis of the diazonium salt?

No, just different methods of chemical modification of graphene. So in principal, just one has to immerse a graphene flake into acetonitrile+nitrobenzenediazonium + this tetrabutylamonium. So in principle it should work without the substance in question but most of them put it without explanation.

Well, thanks you very much!

But the point is that they do not modificate graphene electrochemically. So they just immerse the flake into this solution. :) Maybe this way was just inherited from electrochemical modification of HOPG and people just didn't care and put it.

Many thanks.

Thanks a lot. That's right. Most authors describe the washing procedures as you said - just with acetone. Some of them use consequently acetonitrile, acetone, propanol.

Somebody also did the same chemical functionalization in water (4-NBD+water+surfactant). I tried - it didn't work at all. :)

However, concerning the adsorbed substances, if you characterize it by Raman, you won't be able to see the increase of D peak if it is just physadsorbtion. It has to be chemical bonding. So, Raman is the best tool for characterization in that sense.

I also do not see any point in putting an inert electrolyte into the ACN solution of a diazonium-salt if no electrochemistry is conducted. I did a lot of covalent diazonium approaches all without support electrolyte and sometimes they worked better than targeted. Just skip the TBA-PF6, less junk and less trouble.

I suggest the substance migrated into non electrochemical approaches by just blindly copying procedures from e-chemistry papers - maybe as simple as that.

Jan, many thanks for your reply. I tried to do it in ACN with and without TBA-PF6, and I doesn't work very well as far as I see from Raman. And I totally don't understand why. However it does work with water+4-NBD+surfactant (SDBS). So far I achieved an increase in I(D) / I(G) ratio from 0.1 to 0.4, but still it is not the same as people write in articles where this increase is from 0.01 to 1. So, for me it is still the secret what do I do wrong. :)

Well I can only speculate about this now. Basically the reaction between graphene and NPD is a redox-reaction. Frankly speaking the diazonium salt is cationic while it is no longer cationic after grafting. This implies that there is an electron transfer process in place. Most ppl (me included) believe that the electron is transferred from the graphene to the diazonium cation which will subsequently cleave N2 and react. If this happens by a single electron transfer before grafting or in a concerted process has not been investigated to the best of my knowledge so far. Lets assume for the sake of simplicity there is an electron transfer before grafting, which will leave a hole or a positive charge on the graphene and form a phenyl-sigma-radical from the diazonium compound. Graphite is known to readily react with radicals which to me makes this mechanism plausible enough to talk about it in public here. Lets now further assume the phenyl radical is grafted onto the graphene - what happens to the hole? If the process continues there will be more and more positive charge being accumulated on the graphene which at some critical point will stop the initial electron transfer step because it will become energetically increasingly less favourable. Additionally there will be coulomb repulsion preventing fresh diazonium salt approaching the meanwhile also positively charged graphene surface and the reaction will eventually stop.

Now the speculation begins: In the presence of a nucleophilic solvent like water, the solvent may eventually approach the holes on the graphene and lead to hydroxide groups being formed and protons being released into the solution. Hence the solvent will acidify. In the absence of a nucleophile in the solvent (and usually the counter ions for the diazonium salts are not nucleophilic as this would precent their very synthesis from the first place) there is no mechanism in place to reduce the positive charge on the graphene that stopped the reaction. In your case ACN would have to do this job but is not very good in doing so. If there is PF6 anions present there might be a route for grafting fluoride as a secondary functional group - however I am not aware of any literature discussing this.

One suggestion from Sandmeyer chemistry to provide the neccessary reducing species for the formation of the phenyl radicals would be adding Cu(I) salts to the solution, which get oxidized to Cu(II) and provide electrons. Would be intersting to see their effect tbh. Another possibility would be to do the reaction on a surface capable to provide electrons or compensation for the charge by backside functionalization - surface-OH groups on most oxides should be able to do the same trick as water does.

Hope I could provide a clearer picture about what is going on - if you need more information please feel free to ask further questions - its always the questions asked - not those kept by yourself - that helped to advance science.

My very best Regards,

JME

Jan, many thanks for such a great reply! It helped a lot to understand deeper what happens indeed in the process. It is very useful. Thank you very much.

I just realized that it was your article "Covalent Bulk Functionalization of Graphene" that I was very interested in. Very impressive work. I would like to try something similar because in principle in the end I want to get functionalized graphene paper ( also called as graphene laminates or membrane), so I have to functionalize graphene in suspension. The first step was just to try it with flake.

So, basically, what I do is I dissolve graphite powder in NMP, sonicate it, centrifuge the suspension, and filter it through the alumina membrane. But if I add some 4-NBD in suspension it will precipitate. So it has to be done, for instance in water with surfactant. Then it won't precipitate. So I am still thinking how to make it efficiently. Perhaps I will try you method.

P.S. In fact, I just need covalently functionalize graphene in suspension independently of the functional groups. The only requirement is that they have to be not just single atoms (preferably big groups) as in graphene oxide or fluorographene or hydrogeneted graphene.

Anyway, many thanks again for your reply!

Dear Dennis,

thank you very much for reading my paper - I really hoped that you would find the article. In this work I played several tricks to overcome the limitations of the diazonium functionalization outlined above and I encourage you try it even though I have to warn you as it requires some decent anaerobic working techniques and VERY pure and oxygen free solvents. I want to clarify this a little more. In order to overcome the limitations of running into saturation early I used to reduce the graphite beforehand by treatment with alkali metals like potassium or sodium-potassium alloy, which is highly air sensitive and will spontaneously ignite in contact with water/air/oxygen. The reduced graphitic specimen will a) harbour a significan amount of excessive negative charge which will grant higher functionalization ratios as there will be more electrons available for reducing the diazonium compounds and b) the reduced graphitic species will stabilize themselves in solution by coulomb repulsion and hence will be hindered from reaggregation. This however only works under the strict exclusion of air.

Another approach to obtain phenyl radicals is the thermal or photolytic cleavage of dibenzoylperoxides, that will cleave CO2 upon thermal treatment and produce the neccessary radicals. This route seems to me more compatible to what you are planning to do and might also be possible in NMP dispersions.

I would be very happy to recieve a citation from you anyways - please keep me posted about the outcome of your experiments if you find the time to do so.

Best Regards,

Jan

Dear Jan, thanks a lot for your comments. I'll take all of this into account. Of course, I'll let you know if I get smth interesting. Undoubtedly, you will receive a citation if I manage to perform that.

Many many thanks for your help. Keep in touch. ;)

Denis.

Dear M. Farooq Wahab, thanks a lot for you comment. I haven't tried the XPS before. Actually that is a good idea indeed to get rid of substances present after flakes preparation. Let's say to anneal the sample before functionalization. I'll do that.

Also thanks for the suggestion of increasing the concentration. You are right. I guess I have to try different doses of 4-NBD. Make sense.

About electrochemical approach - I though absolutely in the same way. I am goind to to use the graphite membrane as an electrode. It is also should work.

Many thanks for your help. I'll let you know about my progress.