[shanu]

Hi All,

I am having difficulties in acquiring spectra of the proteins adsorbed on n-type silicon, I am using the standard procedure of adsorption; Immersion of the sample in the solution at 37C for an hour then washing with DI water and drying in nitrogen flow (concentration range from 1mg/ml to 10mg/ml, proteins Albumin and Fibrinogen). Unfortunately when i try to acquire the spectra i am unable to get anything (in terms of Amide I or II), i am using the ATR crystal accessory (Germanium crystal), If anyone can suggest anything that would be great  

[shanu]

Most of the time spectra look like this:

[marquis]

First, can you look at your graph?  It looks like the x axis is wavenumber (not the xlabel indicated).  Can you confirm this and give the y axis label (absorbance?)?

A quick glance at the spectrum doesn't show much.  All I see from the graph is some CO2 (at about 2200) that is probably from your background spectrum, a silicate or silicone (the broad peak at 1000) and some noise at 3500. The silicate or silicone (possibly silicon?) may be from your n-type silicon. 

First guess, the concentration of protein is not high enough to show by FTIR.

Good luck.

[shanu]

Thank you so much for the reply and yes my apologies axis are wrong, i guess i messed up with origin, however can you guide me about the concentration ? In the literature there is plenty of evidence that people use even less concentrations than that and after an hour in the protein solution they simply washed it with DI water and spectra clearly shows Amide I and Amide II peaks. I even tried with 30mg/mL but still unable to get anything.

Any suggestions

[marquis]

ATR covers a wide range of equipment.  My only suggestion would be to get a very detailed description of the instrument, method, and ATR apparatus used.  Then see how your equipment and methods compare.

[shanu]

ok Marquis,

Thank you so much for your valuable response, cheers !!!

[djt]

A thin film of sample should be more than plenty. However can you provide some additional details about your experimental set up.

In order to characterise films on silicon, be it in standard reflection mode or with ATR you need to be using a variable angle accessory either side of the grazing angle on silicon. 60-65º would be a good starting point. You will obviously need a germanium crystal as the refractive index of silicon is about 3.8 from memory.

If you are already trying variable angle work, all I can suggest is increasing the number of scans, resolution and possibly purging the accessory with N2, which will at least minimise the CO2 peak.

You could also try limiting the range (1500-4000 cm^-1) as your spectra might be being saturated by the silicon.

Are you working with the native oxide rather than Si/SiO2?

[shanu]

Thanks djit,

I am using Varian-640IR and the ATR accessory is combined with the microscope so this is the different setup compared to conventional ATR accessory, not only this one is laterally resolved but also provide you the information by inspecting the surface with the microscope. ATR accessory containing the crystal is inserted from the top and sample is placed on a sample holding plate/glass slide facing upward. A live spectrum runs and we gradually move the sample upward untill it came in contact with the crystal, at that moment we we stop the live spectrum and take the real scan.

I understand from the signals that they are very much saturated, i tried to reduce the range from 1000 to 2000 as i am only interested in Amide I (1500-1600) and amide II (1600-1700) but all it gave is a straight line with no peaks at all. Other techniques like xps and AFM confirms the presence of a mono layer and even AFM clearly shows the globular form of ALBUMIN but for some reason FTIR is not giving the result.

My usual procedure is background (100 to 200 spectra), then Silicon and then silicon with proteins . I subtract Silicon from Silicon + proteins to get the protein signal. I tried with other standard sample like polystyrene etc and they provide very sharp peaks.

Protein Concentration i am using for FTIR is 10mg/ml however in the literature people used very less concentrations and they still have strong signals, If you do have any experience in detecting adsorbed proteins on solid samples and can suggest anything further, it will be greatly appreciated.

Regards

[djt]

Shanu,

Firstly, I should point out that I am no expert in FT-IR analyse of films, or even proteins for that fact. I can just give you some advice based on my own experience of characterising films on silicon.

I am not familiar with your exact set up, but imagine it is similar to most microscope FT-IR experiments. Can you control the angle of incidence using said set-up? Is this how other groups generally record spectra of proteins on silicon? If you put the references up, I don't mind having a read through.

I would try recording a background with bare silicon, then running your sample, this is generally the accepted method. Also can you confirm if you are using bare silicon, or thermally grown silicon oxide on silicon? You've got next to no chance of recording a spectrum on a thermal oxide.

The other advantage of grazing angle work is the spot size of the IR source is increased.

You might find the following links useful.

http://www.harricksci.com/

Particularly their Seagull and VariGATR accessories.

I'm guessing by the time difference, you are based in the states, I contacted and sent samples to Harrick, who were more than happy to run spectra on their accessories (with a view to a sale).

Pike also make a range of accessories,

http://www.piketech.com/home.php

I'm also guessing that you are using a DTGS detector, do you have access to an MCT or other which is more sensitive than a DTGS?

Hope that helps a bit, if theres anything else, just post and I'll try to help.

ATB
djt

[shanu]

djt,

first of all thank you so much for such a detailed reply and few details which i would like to share with you are as follows:

1/ Sample is clean Si and not silicon oxide

2/ Yes most the groups around the world use same method of adsorption and detection is almost always done with the ATR (few references below), however i am not familiar with the other setups as you was mentioning that if i have the freedom to control the incident angle i can increase the sensitivity

http://www.sciencedirect.com/science/article/pii/S0927776505000986

http://pubs.acs.org/doi/abs/10.1021/ac061341j
 
3/ I cannot control the angle of incidence as this setup is fixed, kindly see the pictures below; ATR accessory comes from the top and sample is placed upside









4/ I am familiar with the PIKE and infact recently bought an accessory from them for doing proteins in liquid cell, however harrick is something new for me and they have good range of accessories.

5/ Yes the detector is a DTGS and i do not have an access to an MCT detector.

lastly i am based in UK, Northern Ireland to be specific

thank you so much for your time and valuable input.

kind regards

shanu

[djt]

I think this could be the main source of your problems.

http://www.sciencedirect.com/science/article/pii/S0927776505000986

http://pubs.acs.org/doi/abs/10.1021/ac061341j
 
5/ Yes the detector is a DTGS and i do not have an access to an MCT detector.

In both of these papers, they use an MCT detector, which is probably why they can get away with using a microscope FT-IR, rather than grazing angle accessory. Essentially as you adjust the angle of incidence, the number of reflections varies until you reach a maximum just before the grazing angle, as you go through the grazing angle, the signal inverts sharply and hence the most sensitive region is either side of the grazing angle for a given substrate.

The graph at the end of the first page in the following link shows how the sensitivity varies as a function of angle of incidence using their Veemax accessory.

http://www.piketech.com/files/pdfs/VeeMAXIIPDS611.pdf

If you have recently purchased an accessory, it is likely you have a good rapport with the sales person. It is worth speaking to them and trying to arrange a demo of the accessory.

As I mentioned in my previous post Harrick also manufacture accessories. However, they are based in the states and you will pay a premium for them, expect them to be 2-3 more expensive than the Pike, although I believe in this case you genuinely get what you pay for.

With reference to the papers, I can't quite work out what they have done in terms of collecting the data. Have the recorded the spectra with 100-200 scans or, have they recorded 100-200 samples with x number of scans per sample?

Looking at their spectra (I like the way they have deconvoluted the amide bands) and yours, is it possible you have a very weak signal in the 1500-1800 cm^-1 region?

It looks like you have a fairly nice IR there, I would try to record the following spectra with the highest practical resolution and a minimum of 1024 scans.

Background
Substrate
Substrate + Sample.

Baseline correct, smooth and normalise all the spectra in the IR software and export as CSV or similar.

Use origin to plot the individual spectra and then use origin to subtract you references. This is where 1024 scans will help. Origin is also quite good at doing this. I haven't for origin on this computer, but if you need help here, can check and let you know the steps.

If that doesn't work, I would really follow up with Pike/Harrick, especially if you don't have access to an MCT.

Good luck
djt

[shanu]

Thanks djt once again for your detailed reply, well i dont know how those guys collected the data but i assume that it is collected on 5 different spots at 100-200 scans and then average out. The subtraction procedure you mentioned i follow exactly the same protocol and i know how to subtract the reference even with the origin.

I believe that the best thing will be to speak with the supplier as you suggested and if i find anything useful which will solve the problem i will try to post a reply here.

Many thanks once again
cheers

[djt]

Shanu,

Did you get anywhere with this?

[shanu]

hi djt,

Well i am still very much there i guess, the things you had suggested are very much true as i get the same reply from Agilent Technologies which is as follows:

''The ATR is working as we can see the polystyrene, but this does not mean the crystal is fully centred.
But more importantly for a hard sample such as a Si-wafer the contact area between Ge-ATR and wafer will only be approximately 20µ. In my experience the coating on Silicon wafers are not particularly homogeneous, and it may be that contact is not made with a part of the wafer with sample on it.
I would say though that generally I would not use ATR for looking at what I believe are Self Assembled Monolayers (SAMs). I would not expect the data to be very good with ATR. If it was me with samples like this I would probably choose either a Grazing Angle (in bench) accessory or a Grazing Angle Objective for the microscope. Even then I would probably want to use the accessory/objective with variable angle polarisers so that I could look at the s and p polarized components of the sample.
Don’t know if this helps but please feel free to pass it on''

unfortunately we don't have this grazing angle accessory with us and most likely i have to go someplace else to use the system 

I also double check the detector and it is MCT not DTGS.
These days i am working much of AFM and hopefully will get the permission to use the FTIR at some other place

Thank you so much once again for your support, cheers 

[alimeeabey]

Most of the time this graph comes right when proteins are adsorbed on p-type silicon Wafers, Check with your supplier or what kind of material you are using. May be that can help.