The text you've cited correctly states that luminescence intensities vary with lumophore concentration, but not always in a linear (or predictable) manner. From a practical perspective, you can certainly use emission intensities to determine analyte concentrations, but it's necessary to first construct a calibration curve (i.e. by measuring luminescence intensities from serial dilutions of an analyte solution whose concentration is known) using the same solvent, temperature, pH, etc. as the system you're hoping to study. You can then determine the lumophore concentration in your system of interest by comparing your emission measurement(s) to your calibration curve and extrapolating.

Beer-Lambert law allows the calculation of the concentration from absorbance data. The same is not valid for the fluorescence. Higher concentrations of fluorophore give dimers, excited dimers, excimer formation. Stability of the excited state can be influenced by polarity, pH. Other reasons: fluorescence quenching (static or dynamic),.

Therefore, if you are interested in determining concentration, fluorescence is not the way.

In my opinion, both the fluorescence and absorption are roughly dependent on the concentration, and the absoption is easy be quantized, such as 20%, 50%.....，you can easily to say which concentration corresponding which absortption, but you can not quantize the fluorescence, so it is difficult to answer how amout of fluorescence corresponding which absorption.

The text you've cited correctly states that luminescence intensities vary with lumophore concentration, but not always in a linear (or predictable) manner. From a practical perspective, you can certainly use emission intensities to determine analyte concentrations, but it's necessary to first construct a calibration curve (i.e. by measuring luminescence intensities from serial dilutions of an analyte solution whose concentration is known) using the same solvent, temperature, pH, etc. as the system you're hoping to study. You can then determine the lumophore concentration in your system of interest by comparing your emission measurement(s) to your calibration curve and extrapolating.

Yes, I agree with Zach Erno's comments. Fluorescence does varies flurophore concentration. there is a necessity to establish a calibration curve so that quantification of analyte may be possible precisely.

The point is that an absorption measurement is always a relative measurement where you compare the transmitted light intensities with and without solute. This relative measurement (ideally) eliminates all instrumental factors. The transmittance or absorbance then depends only on the molar extinction coefficient and the light path length through the sample. Knowing these two parameters you can determine the concentration with a single absorbance measurement, independently on the instrument you use. This is not the case in fluorescence, where you measure the absolute emitted light intensity, which depends linearly on the fluorophore concentration (at low concentrations) but also on instrumental factors. So you need to calibrate your individual instrument (as Zach and Suriya say) or use a reference compound. But of course it is perfectly possible to determine concentrations from fluorescence intensities.

Beer-Lambert law works only in diluted solvent (solvent is dilute if the Beer-Lambert law works). fluorescence is something similar but unlike in UV/visible spectroscopy, ‘there are lot of factors influence and distort the spectra.

Measured fluorescence depend on

- excitation light intensity (it changes during the lifetime of the source) and wavelength

- the environment of the dye (solvent effect, quenching, absorbtion, excited dyes reactivity could be higher and could react with the environment)

- quantum yield of the dye

- concentration of the dye (higher concentration could cause FRET, dimer, tetramer formation, self absorbtion, inner-filter effects, and of course if the concetration is high, Beer-Lambert law is not true so flourescence can't be depend on concentration.

Even so, there is a technique called Fluorometry use for Determining concentration from fluorescence. The simplest method is using working curve as Zach Erno described. It is important that you have to use the same solvent (same environment) when you take the calibration. there must be a range where the plot is linear. In this range you can determine an unkown concentration.

As others have pointed out, the measurement is dependent on many factors that are "in" the sample but also on the instrument stability

Some work on the subject was done in the last few years (see the NIST example http://nvlpubs.nist.gov/nistpubs/jres/106/2/j62gai.pdf)

More options are available from BAM (sold by Sigma Aldrich) and you can get some information right here https://www.researchgate.net/publication/230949311_Linking_fluorescence_measurements_to_radiometric_units

Article Linking fluorescence measurement to radiometric units

you can solve this in Matlab with matrix equations. It is not completely accurate as noise contribution lead to some regions of the image having seemingly negative concentration (which is meaningless) but is a good approximation. I will check my notes later and post the equations if you are interested (it is based on beers law).

There have been several good answers, but just to chime in as well, the fluorescence of a sample is proportional to fluorophore concentration at low enough concentrations, but also proportional to the fluorescence quantum yield (will not vary if we are comparing solutions of the same fluorophore), and instrumental factors, like what wavelength or band of wavelengths is being used to excite the sample, and how much of the fluorescence is captured by the detection optics and registered by the electronics. So, as has been pointed out, one can do a set of standards to calibrate the fluorescence of a particular chromophore in a particular instrument, being careful that each solution is identically placed relative to the excitation and detection optics, that the lamp intensities for all runs are the same, and that the photomultiplier voltage and excitation and detection slits (bandwidths) are the same also.

A calibration curve should be established for the analyte.

One more issue, many fluorophores have photostability issues. If you increase the excitation power, view for longer times, or rescan in an effort to produce an increased signal, you may discover that the dye bleaches. One way to notice this is to move the sample about 1/2 frame sideways after viewing, and take one more image. If the old half of the image is noticeably dimmer than the fresh half, you may want to decrease the illumination power or time in order to reduce photobleaching. The photobleaching rate can change by orders of magnitude depending on the kind of fluorophore, so you may be able to choose a more stable dye.

Though Henry Benner's answer is correct scientifically but in real life situations it's almost impossible to use fluorescence as a routine reliable tool for quantification of target molecules using fluorescence . This is especially true in complex biochemical heterogenous mixtures like biological tissues used in immunofluorescence histochemistry. It is because of following basic reasons ( I a m re-emphasizing the point already mentioned in earlier replies)...

1.Quantum yield of fluorophore varies with pH, net charges surrounding fluorophore.

2. Complex biochemical substrates like biological tissues contain several known and unknown chemical that may act as quenchers or

3.quantum yeild may be enhanced by nonradiative energy tranfer mechanisms (eg. FRET) to the fluorophore giving more than predicted fluorescence.example Tryptophan fluorescence enhanced hugely in pressence of negatively complex amino sugars like heparan sulfate.

4.Stokes difference between excitaion and emission wavelengths i.e peak excitation and peak emssion wavelength vary considerably high enough to make huge difference between real fluorphore concentration to the measured fluorecence yield.

In solid state, the determination of luminescent ions beyond certain concentration leads to error. Since the concentration quenching plays significant role in reducing the fluorescence it is not desirable to rely on concentration beyond this point.

As suggested by Gary Holtom photostability is an issue that one should not ignore during the analysis of fluorophone concentration. Repeated measurements reduces emission intensity of fluorophore and hence does not produce reliable results. In view of this we have been studying the fluorescent nano diamonds (FND) for bilogical applications after appropriate surface functionalisation. FND has exceptional photostability even under laser excitation.

Thanks for all the answers. But I will ask one more thing anyway.

The excitation coefficient in Lambert-Beer law is also dependent on experimental conditions like solvent, maybe pH, for sure wavelength etc. This should suggest, that equivalent of excitation coefficient could be calculated in fluorescence for set experimental conditions. The only problem being non-linearity?

The extinction coefficient is an "intensive" property, i.e. it does not depend explicitly on the concentration of the chromophore, but is instead calculated from the ratio of transmitted intensity to incident intensity. On the other hand, fluorescence intensity depends on chromophore concentration. What you seem to be looking for is a parameter which you could multiply by the chromophore concentration to give you the observed fluorescence in your experimental apparatus. In principle one could calculate such a parameter, but one would have to know the light intensity hitting the sample, how big the sample was, the fluorescence quantum yield for the sample, how much of the light is captured by the detection optics and how this is converted into an electrical signal. Even then, the measured fluorescence intensity in your sample may depend on the presence of fluorescence quenching groups and local environment, and the effective "concentration" may be quite different from that found in a fluid solution of the same chromophore. In your original quoted text, the relative changes in fluorescence intensities in an environment such as a protein or membrane may not reflect changes in concentration, but instead could be due to a change in polarity of the environment, or a change in probability of Forster electronic energy transfer, or the approach of quenchers.

You are getting good advice here from people who provided very different answers, which only shows the number of effects that contribute to observed fluorescence intensity. When you consider absorbance spectra, what matters is the form of the ground state and excited states molecular orbitals that define the transition. The strength of the transition can be calculated, and a corresponding 'Einstein Coefficient' describes both the absorbance and stimulated emission probabilities. Changes in the extinction coefficient correspond to changes in the structure and energy of the electronic states of the chromophore. For example, the familiar red shift in polar solvents arises from the lowering of the energy of excited state orbitals.

Any effect on absorbance also affects fluorescence, but in addition you have to consider the effect of competing pathways on fluorescence yield. Once light energy is absorbed, a fluorophore can decay to ground by many different pathways. For each fluorophore there is a characteristic radiative lifetime that specifies the rate of emission, but there are many other paths to ground so that the lifetime of the excited state is very short. The ratio of these these lifetimes is the quantum yield, the fraction of molecules that decay via radiation. For example, riboflavin at moderate concentration has a quantum yield of around 30%, depending on solution conditions. Quenchers, which provide additional non-radiative paths to ground, decrease the lifetime and hence the quantum yield. It is difficult to correct for this.

In addition, inner filter effects and the screening of the fluorophore absorbance bands by other chromophores decrease the observed fluorescence intensity. In absorbance this can be corrected for using an appropriate reference sample. The only possible correction in fluorescence would be a calibration standard in identical conditions, difficult for biological material that may not even be homogeneous; tryptophan in different environments in proteins can have very different fluorescence yields.