First, you didn't say whether the spectra you showed are excitation or emission spectra. If you fixed the excitation wavelength and scanned the emission wavelength, they are emission spectra. If you scanned the excitation wavelength and fixed the emission wavelength, they are excitation spectra. (The excitation spectrum usually resembles the absorbance spectrum measured with a spectrophotometer.) You should characterize your compounds by collecting both excitation and emission scans.

Fluorescence intensity is measured in relative fluorescence units (RFU), which is just the amount of light collected by the instrument. Because the instrument is not equally sensitive to all wavelengths of light, the spectra should be corrected for instrument effects. The instrument software may include correction files for excitation and emission spectra that consists of multiplicative factors for each wavelength.

The textbook "Principles of fluorescence spectroscopy" by Lakowicz is an extremely useful guide.

You would have to check the absorption spectra on a UV-VIS spectrometer first, or look up in the literature where the is the maximum absorption of the two compounds. Then try the emission spectra of each compound using the wavelength of maximum absorption as excitation wavelength. This would be my recommendation, step by step. You can later also try to record Excitation Emission Matrices (EEMs) and use PARAFAC to extract the contribution of individual compounds to the total fluorescence spectrum.

First, you didn't say whether the spectra you showed are excitation or emission spectra. If you fixed the excitation wavelength and scanned the emission wavelength, they are emission spectra. If you scanned the excitation wavelength and fixed the emission wavelength, they are excitation spectra. (The excitation spectrum usually resembles the absorbance spectrum measured with a spectrophotometer.) You should characterize your compounds by collecting both excitation and emission scans.

Fluorescence intensity is measured in relative fluorescence units (RFU), which is just the amount of light collected by the instrument. Because the instrument is not equally sensitive to all wavelengths of light, the spectra should be corrected for instrument effects. The instrument software may include correction files for excitation and emission spectra that consists of multiplicative factors for each wavelength.

The textbook "Principles of fluorescence spectroscopy" by Lakowicz is an extremely useful guide.

Hi Naveen:

In order to determine the Stokes shift (SS) you've to look for the difference b/w the peak frequencies in abs. and em. spectra; say your abs. peak is at x nm and em. peak at y nm, the SS should be the difference  x - y converted to freq. units.

If you're having two em. peaks you've to first ensure which one you're actually referring to for calculating the SS; this would in turn depend on the origin of the two em. peaks and your requirement accordingly.

Usually it does not make a genuine sense to assign a specific unit to em. spectral intensity; it would vary a s a function of many factors, even the grating of your fluorimeter and the way the software programming is done. This is why we usually assign an arb. unit to em. spectral intensity or you may chose to represent the intensity of one compd. in relation to another, i.e., relative one another. You may chose to normalize the max. intensity of one compd to = 1 and represent the other w.r.t. this normalized value.

Hope this makes sense! As for otherwise please feel free to ask. I would be happy to elaborate on the issue.

Cheers!

Hello Naveen, please tell us the excitation wavelength you used for this measurements. I have a suspicion...

For learning about optical spectroscopy, you might start reading from the following links: http://www.olympusmicro.com/primer/techniques/fluorescence/fluorescenceintro.html

https://webfiles.uci.edu/mdigman/USIBR/Lecture_1_Fluorescence.pdf

https://en.wikipedia.org/wiki/Fluorescence_spectroscopy

Relative Fluorescence Units means that the measurement does not provide physical units, instead it is related to instrument response. They might be normalized, e.g. to a reference signal level, or not, e.g. digital counts generated by the electronics of the detector. In order to get a physical fluorescence spectrum you need to do a radiometric calibration, but this (unfortunately) is seldom done.

Your result is the emission spectrum, but it might be deformed with respect to the physical one if the instrument response is not constant throughout the spectral range (which is typically the case).

For the absorption spectrum, you should measure a clear solution first, then your dissolved compounds. Dividing the second by the first should give you the absorption spectrum. Make sure that you measure in a spectral range shorter than the emission. This way of measuring has a limitation in those cases where absorption and emission overlap.

If you can use a monochromator as light source you could measure the excitation-emission matrix (EEM).

Hope this helps,

Luis

"Relative" means that the value is not absolute and it is not really necessary to know the absolute. If you use this to measure the amounts of DNA in a sample, for instance, you will need to make a calibration. From your graph, you can see that the red curve is about twice as high as the blue one, meaning that there is twice as much fluorescence from the sample and probably twice as much fluorophore (which is usually true for dilute solutions). What you see are emission spectra and you can't make any real conclusions from this about the absorption spectra - other than it being shifted to shorter wavelengths. Also, there is a certain probability that the absorption spectrum will be a more or less mirror image of the emission spectrum due to the nature of fluorescence itself - see Jablonski diagram - however, it is not always a mirror! Jablonski diagram also explains why the spectra are broad and not simple lines - all molecules emmit in a broad speactrum due to the existence of various energy levels in the molecule. Best read on fluorescence in my experience is this book: http://www.springer.com/la/book/9780387312781 - it's huge, but you'll find everything you need in there. Also, use wikipedia - the info might not be complete and not always totally correct, but it will help you get a general idea.

Adam has a pertinent, clear and succinct answer, and others have added good explanations and information too.

The "relative" part of the RFU indicates that the signal intensity measured by the instrument is dependent on the instrument itself. As Adam said, the instrument will have correction/calibration files so the the "relative" intensity measured today is consistent with the intensity measured tomorrow on "that" instrument. However, it is still relative and cannot be compared with the fluorescence intensity measured for the same sample by a different instrument.

So, instrument A might give you an RFU of 200 (at wavelength X) for a sample, and instrument B might give you an RFU of 5000 for the exact same sample (at the same wavelength). This is unlike an absorbance measurement which is (ideally) absolute, and you can expect a sample with absorbance of 0.5 (at X wavelength) to have the same absorbance of 0.5 (at that wavelength) on any other spectrophotometer. 

Firstly I am sincerely thankful to all senior scientists (Dr. and prof.) those kindly provided useful suggestions. Specifically I am very grateful to Dr. Adam B Shapiro. 

This was actually a absorption spectra but later I performed emission scan separately (since before I didn't know whether I have to perform the emission scan separately. Now I get the emission peak at  328 nm), that is the stokes shift of 36 nm. I have the question that relative fluorescence unit (RFU) is very high (800000 for compound one "C9") in absorption spectrum but in emission spectrum the value of RFU is low (nearly 23792)  (FILE ATTACHED). Now how can I normalize the relative fluorescence units of absorption spectra, so that I can forward with the future experiment. Do I need to normalize RFU in to fluorescence units (o to 1). How can I normalize these large RFU (of absorption spectra). I would be thankful for providing me solution. 

Since this is first time that I scanned or performed fluorescence spectra. During my course work at IIT, we have the course and I got many princpiples about fluorescence spectroscopy (Jablonski principle, fluorescence lifetime, correlation spectroscopy,  Kd determination etc), with some experiments as well. I have Lakowicz book as well, nevertheless I find difficult in transforming the RFU in to fluorescence units or normalization.

Sincerely thankful to all again.

One way of normalzing the emission is dividing by the integral of the absorpted energy.

This gives the fluorescence yield. Then you can compare with other compunds in fair terms.

Naveen,

The spectra you attached with your original post would in fact be "excitation spectra" NOT absorption spectra. While they are often similar (like Adam pointed out), they are not always the same, and you might want to make sure you understand the difference between the two. Remember: RFU (relative "fluorescence" units) deals with fluorescence (excitation and emission); Absorbance (A) is unit-less, and typical modern spectrophotometers will measure absorbance in a range of 0 to 2 A (with extended extrapolated ranges for shorter path lengths).

You have concluded that you see a Stokes shift of 36 nm from the emission spectra which you attached later. However the peaks in these spectra appear to be artifacts of using an emission filter. Such filers only allow light of specific wavelengths to pass through, and it seems that part of your emission peak is "cut off" by the filter. You need to either not use this filter, or use a different, more appropriate filter, before you can make a conclusion about the emission spectra.

How can you normalize these excitation and emission spectra? One way, like Luis has replied above, is to determine the fluorescence yield. However this might be unnecessary, depending on your needs and goals. Alternatively, for simple illustration purposes, you can normalize the spectra to a range of 0 to 1 like you mentioned (or 0 to 100 or any other arbitrary scale, as long as you clearly note that the spectra are normalized).

To expand on the process of normalization as explained by Nitesh, you simply identify the highest RFU value in the spectrum, and divide each measurement by that value. The shape of the spectrum is unchanged, but the vertical scale goes up to 1. This makes it easier to compare the shapes of spectra, but obscures differences in intensity.

The quantum yield of a fluorescent substance (the ratio of photons emitted to photons absorbed) can be measured by comparison to a standard. This is an absolute, unitless measurement, under a given set of environmental conditions.