Energetically it`s more efficient for NPs to stick to the surface instead of being in free solution.

According to my experience it`s very hard to make the NPs nicely cover the surface. Try with dilution, should be better (thou I don`t know what do you want to obtain...).

I was considering trying to deposit in a sonicator, or using the sonicator after deposition to see if these islands break apart. I also will try an extreme dilution of np solution. Any other ideas?javascript:

How about trying step-by-step deposition?

How about if u add some base (e.g, NaOH) to keep pH near 8 or 9, it will keep the AuNPs well dispersed. Did you observe the same phenomenon?

Energetically it`s more efficient for NPs to stick to the surface instead of being in free solution.

According to my experience it`s very hard to make the NPs nicely cover the surface. Try with dilution, should be better (thou I don`t know what do you want to obtain...).

Yes I agree with Robin. There are many factors that can affect this agglomeration. The 2 that spring to mind is self agglomeration of the stock solution (due to shelf-life) and DLS could help you determine this. Alternatively APTMS/APTES monolayer formation is very sensitive to conditions and you could be seeing some polymerization of the APTMS molecules, thus having multiple layers/ridges in random areas. The colloids would then attach accordingly and this is what you probably are seeing.

I recommend you check carefully the monolayer formation step.

Good luck.

Robin (and others) thank you for your thoughtful responses. I will attempt to answer the good questions you guys brought up.

We functionalize with 3-aminopropyltrimethoxysilanes (have also had success with a similar amine functionalized substance, and currently retrying old work with 3-mercaptapropyltrimethoxy silane).

I user pure ethanol without water. I know there are some instances where water was shown to improve formation, but have not tried it with our current protocol.

We actually dip an optical fiber vertically into a cuvette of stirred nanoparticles and I watch the formation of this layer in real time via the uvvis reflectance from the fiber. This process takes between 3-20 minutes, depending on np concentration and fiber-to-fiber/batch-to-batch differences.

We only rinse by submersion in h20 under stirring (eg not sonication, which I think could improve this if the np's are loosely bound).

I'm confident that the np's are not aggregated in this manner in free solution via the clean uvv-vis spectrum with no long-wavelength signature. Also, the aggregates generally seem to appear later in the formation of the fiber. If I dip a fiber into np's for a short period, it is very much monodisperse.

Also, I am only confident in my monolayer in as much as that it's working better now than it has in the past. I have no analytical tools to monitor it, so it may be the culprit here. One worry I have is that the annealing step, I'm forced to do in air in an oven instead of in vacuum. Oddly enough, I'm using an almost identical protocol to a paper that showed fairly monodisperse layers on glass, but who knows the black magic of it.

Current protocol:

1. Rinse fibers in pirhana.

2. Immersion in 1-1.5% APTMS in ethanol under sonication for one hour. We then rinse in ethanol under sonication for ten minutes and anneal at 120-150C for 30min.

3. Fibers are left submerged in ethanol until use.

This might involve with "the magic number", the smallest number the structure can exist. AuNP also follows the rule, leading to the pie up no matter how neat your process is.

I could make the following suggestions if all other options fail.

1. Scale back the APTMS percentage (0.01 to 0.1%). I even had to use a mixed monolayer to prevent agglomerations (see Anal Bioanal Chem (2008) 391:1813–1820 for exact conditions)

2. Try the formation of the monolayer in dry toluene. Most chemists I know invariably quote this way over the alcoholic route with heating.

Two new ideas:

1. try to play with temperature of your bath with NPs - decreasing should make your structures better

2. The annealing step you can do in an oven with Ar flow instead of air. I don`t know if vacuum would be better.

Thanks Marta. Unfortunately, I don't have access to an oven with this capability. I'll have to ask around campus though just to be sure.

Also, wouldn't lower temperature makes np's more likely to bind (eg less thermal energy to distrupt bonds)

Should be easy to make such device if you are using e.g. tube furnance - just make two "caps" for both sides of the tube, make holes on them and one connect with gas bottle. If I understood well you don`t need to have oxygen-free ambient, am I right?

About the temperature case... It is only my long shot. According to our theory guys, the lower the temperature, the nicer self-assembly patterns they obtain in their Monte Carlo simulations. I don`t know why is it so, but they claim those numbers do not lie. It is really not intiutive but worth to try IMO.

i have tried depositing 24nm citrate capped nanoparticle on APTES functionalized glass slide, i am not sure abt other solvents, but i dissolved 2% APTES in isopropanol and after rinsing with isopropanol i washed the slides with pure water to remove physiabsorbed APTES ( i have read in literature APTES form jelly like layer,of course u can control that with its concentration) but in my case this water rinsing really helped to get a homogeneous layer of gold on glass slide after rinsing , i cured the slide at 120C for 10min and then incubated with gold particle, if u r interested u can chk my paper

Hi guys. Thanks for sharing your observations and papers. I actually just tried a new linker that was suggested to me in another thread and it did seem to improve the quality at full coverage (at least for the first batch). See the attached slide (some of hte larger things on the image are debris from handling after np deposition and not np aggregates).

Oops (attachment here)

Coffee-ring effect ? NPs clust together when drying

I don't know. It's possible (I sure hope not though) but I have indirect evidence to believe this effect occurs during monolayer formation. Here's a recent fiber with a new crosslinker that seems improved. Ignore the major gunk that is due to handling after sacrificing the fiber.

I used Raman to detect the monolayer on Ag NPs in the wet state. If the Ag NPs dispersed well, there is SPR gap mode. So the raman scatter will be very strong. I do'nt like SEM result, because NPs will change it morphology after drying and electron charge in SEM

I guess glutaraldehyde should be a good way to link the gold to the substrate. Its a pretty popular process and quite simple too

Bean-jon, do you use Raman to inspect monolayer on slides/chips or on optical fibers? Unfortunately, my fibers can't hold the wavelength of light required for Raman, and I don't have a Raman setup in our lab, so I'd still have to dry out the fibers to attempt to examine them. In regard to the SEM, you're probably right that the drying and charging in the SEM do impact the fiber; however, as you can see from the second picture I posted, it can't explain entirely the aggregation.

Tejabhiram, to my knowledge, gluteraldehyde is involved in crosslinking biomolecules to gold. I've never seen it used to affix np's to a substrate. I'd be interested to learn more. Also, I think the most recent crosslinker I tried did do a good job (see most recent image)

Adam, Monolayer on Au can be detect by Raman. It need SPR enhancemnet. My experiment is done by Ag NPs on Al.

and CRYO sample holder in a FEG-TEM..could be an alternative way to study them.

Adam Hughes , Did you coat the amine monolayer following pre-established protocols or did you use commercially available amine-functionalized glass slides? Were you able to achieve monolayer of AuNP's repeatedly using the "new" linker? Could you please share the name of the linker, the reference and/or your paper that describes this protocol? Thanks in advance!!