[AoisPokemaster]

Hey guys, I am again the person who is researching about curcumin. I have different solvents to test the the absorption of curcumin in, and so far I have used the following solvents: Ether, Acetone, Ethanol, Ethyl Acetate, isopropanol, propanol. I have also used hexane, but due to the low solubility in hexane, I added 4% of ethyl acetate to the hexane in order to increase the solubility of curcumin in the solvent. I did the same for water and also acetone. I tried finding literature elsewhere, but I cannot find any explanation for why compounds like curcumin exhibit positive solvatochromism - is it a generalised mechanism, or specific to each compound? I need to explain in the perspective of molecular interactions, but I have absolutely no idea how to do so. please send help. Additionally, I also measured the fluorescence of the compounds, and realised that the curcumin dissolved in hexane with 4% ethyl acetate exhibits a lower than expected fluorescence frequency. If you take a look at the attachment, there does seem to be a relationship between the points, but I really just cannot explain why hexane is lower than the others.  Thank you guys very much in advance.

[Babcock_Hall]

My experience is limited to the absorption spectra of a carotenoid.  I cannot provide an explanation, but my observations were that the wavelength of maximum absorbance were similar in hexane and methanol but shifted red in chloroform and dichloromethane.  This makes me think that other properties, perhaps polarizability, were important.

[rolnor]

If you look at Curcumin, its partially in the enol-form. How much is in the enol-form can be depending on the solvents ability to form hydrogen-bonds or polarity. Its very complex, especially if you use solvent mixtures. Its also interesting to see the correletion you have. Maybe you can run NMR in different solvents/solvent-mixtures and see if there is more or less of the enol-form?

https://en.m.wikipedia.org/wiki/Curcumin

[wildfyr]

In addition to the considerations of keto-enol and what can happen in protic or polar solvents, sometimes solvents can promote dimerization that totally changes the photophysical properties of a compound.

You are probably on your own to learn about this interplay of solubility and photophysics in curcumin which is a tough place to be, but with some study you can provide usefully to human knowledge on a reasonably common compound.

[rolnor]

I think if you do carefull search on Scifinder/Reaxys you will find papers on the subject, its a well-known old compound, nothing exotic.

[Corribus]

First things first.

1. What are the units in your plots? This isn't trivial. Plotting Stokes shifts in the wrong units result in misleading conclusions. And what is "relative polarity"?
2. You are plotting Stokes shifts, not absorption or fluorescence wavelength data - yet your post title refers to absorption and your post text refers to fluorescence frequencies. You're mixing a lot of related but different things here, so it's no wonder you (and we) don't know what's going on. You're also not being clear when you describe your data: you say hexane is "lower" than "the others", but what does "lower" mean? And what others?

I suggest first you clarify to yourself and to us what exactly you're doing, what you're plotting, and what you need to explain. Learning how to report and describe scientific results clearly is as important as learning the chemical and physical principles.

[AoisPokemaster]

First things first.

1. What are the units in your plots? This isn't trivial. Plotting Stokes shifts in the wrong units result in misleading conclusions. And what is "relative polarity"?
2. You are plotting Stokes shifts, not absorption or fluorescence wavelength data - yet your post title refers to absorption and your post text refers to fluorescence frequencies. You're mixing a lot of related but different things here, so it's no wonder you (and we) don't know what's going on. You're also not being clear when you describe your data: you say hexane is "lower" than "the others", but what does "lower" mean? And what others?

I suggest first you clarify to yourself and to us what exactly you're doing, what you're plotting, and what you need to explain. Learning how to report and describe scientific results clearly is as important as learning the chemical and physical principles.

Sorry if I had presented information in the wrong way. When I refer to hexane as lower, I meant that it has the lowest polarity, or is the most non-polar solvent out of the ones that were tested. I used polarities that are calculate relative to water, that are taken from an online data base, its accuracy is not meant to be discussed. I am looking at the relationship between the relative polarity of the solvent used on the absorbance and also fluorescence. I have little to no knowledge about chemistry and also how to present data and methodology, as it is my first time writing a paper with over 2000 words. The datapoint on the leftmost of the diagram shows hexane with 4% ethyl acetate, which is used to increase the solubility of curcumin in solution, but in this experiment I am assuming that the solvent system uses entirely hexane. The stokes shift was calculated by taking the wavelength at which fluorimetry was done, which was 405nm, and calculating the energy, after which subtracting fluorescence wavelength energy obtained from fluorimetry. I am not sure if this is correct, but this formula was what I assumed from looking at the jablonski diagram which models the stoke's shift. I am 16 years old and I still have a lot to learn, so I appreciate any feedback that all of you guys give.

[AoisPokemaster]

First things first.

1. What are the units in your plots? This isn't trivial. Plotting Stokes shifts in the wrong units result in misleading conclusions. And what is "relative polarity"?
2. You are plotting Stokes shifts, not absorption or fluorescence wavelength data - yet your post title refers to absorption and your post text refers to fluorescence frequencies. You're mixing a lot of related but different things here, so it's no wonder you (and we) don't know what's going on. You're also not being clear when you describe your data: you say hexane is "lower" than "the others", but what does "lower" mean? And what others?

I suggest first you clarify to yourself and to us what exactly you're doing, what you're plotting, and what you need to explain. Learning how to report and describe scientific results clearly is as important as learning the chemical and physical principles.

additionally, I want to point out that the units are not really defined yet, as my calculation are in a mess. I can guarantee that my calculations are all correct and the datapoints are in line with the experimental results, just that I cannot use kJ as I am only considering a single photon, which would make it so small that excel or google sheets would consider it to be 0.

[Corribus]

You shouldn't be trying to interpret data until you are sure the data are presented correctly. Excel can easily handle energies of molecular scale processes in kJ. But if you don't want to use kJ, use eV. Or use kJ/mol. As far as the "degree of polarity" goes, I'm still not sure what unit you are using. Is it the dipole moment?

Beyond that -

Let's define some terms. You may already know, but it's good to be sure.

Solvatochromism is the shift of absorption or emission color of a chromophore when the solvent is changed. It's more of a qualitative idea than something you can quantitatively measure, because a solvent change can result in a change in the absorption or emission peak maximum, the bandwidth, and the intensity, all of which influence the perceived color.

Stokes shift is the difference in energy between the lowest energy absorption peak maximum and the highest energy emission peak maximum. Although it is frequently expressed in wavelength, it is better to express it in energy units. The reason is that wavelength is not linearly proportional to energy - a 10 nm Stokes shift in the 400 nm wavelength region is different from a 10 nm Stokes shift in the 600 nm wavelength region. The best way to calculate your Stokes shift is to measured the peak wavelength of your absorption band, measure the peak wavelength of your emission band, convert both from nm to eV (or whatever energy unit you like - inverse cm are also common), then take the difference.

Strictly speaking, you can't really say much about solvatochromism from the Stokes shift data you have presented. You might infer that some solvatochromic behavior is occurring but hypso and bathochromatic shifts can both result in the positive (or negative) change in the Stokes shift.

As to why solvents influence absorption and fluorescence colors, and the Stokes shift: electrons in molecules exist in orbitals, and in colored compounds these molecular orbitals tend to be broadly delocalized over large portions of the molecular framework. Nearby solvent molecules interact with these electrons, particularly when the solvent molecules and chromophore are both polar. This interaction changes the energy of the chromophore's electronic states. When certain electrons in chromophores are excited, they are promoted from a lower lying orbital to a higher lying orbital that has some antibonding character. The energy difference between these two orbitals* determines the absorption spectral properties, which leads us to see these compounds as colored if the absorption wavelengths are in the visible region. The excited states of chromophores tend to be more highly polarized than their ground state analogs, and therefore they tend to interact more strongly with nearby solvent molecules. Actually what happens is this: absorption of light by electrons is more or less instantaneous, so that solvent molecules immediately before and immediately after absorption are basically in the same spatial position. But suddenly the solvent molecules find themselves in a very different electric field environment, and so they shift their positions gradually (actually on the subpicosecond to picosecond timescale, but that might as well be eons) around the polarized chromophore. The polarization and reorientation of solvent molecules around the excited chromophore stabilize its excited state and can lead to electronic and nuclear structural changes (lowering the excited state energy), which can cause even more solvent reorientation.** Some nanoseconds later, when the excited state spontaneously relaxes back to the ground state (via fluorescence), the emitted photon has a lower energy than the photon originally absorbed. This is the usually the cause of the Stokes shift. Because "more polar" molecules usually interact more strongly with polarized excited (fluorescing) states, they usually lead to greater Stokes shifts as well as more red-shifted fluorescence peaks (and often weaker intensity). BUT as noted above, solvent polarity is a pretty vague concept. Polarity isn't the key here, it's a complex kinetic process of solvent reorganization around polarized chromophore excited states that primarily causes the solvatochromism and Stokes shift. The solvent molecular size, electron distribution/polarizability, local viscosity, the radiative lifetime of the chromophore, etc., all play roles what the observed emission and absorption energies will be. Since you're using all kinds of different solvents here, this is probably why your trends are pretty complicated.
 
*It's easy to confuse orbitals and states. I'm using them pretty loosely here.
**Using time-resolved fluorescence experiments, you can actually observe these processes happening and measure how long they take.

[AoisPokemaster]

You shouldn't be trying to interpret data until you are sure the data are presented correctly. Excel can easily handle energies of molecular scale processes in kJ. But if you don't want to use kJ, use eV. Or use kJ/mol. As far as the "degree of polarity" goes, I'm still not sure what unit you are using. Is it the dipole moment?

Beyond that -

Let's define some terms. You may already know, but it's good to be sure.

Solvatochromism is the shift of absorption or emission color of a chromophore when the solvent is changed. It's more of a qualitative idea than something you can quantitatively measure, because a solvent change can result in a change in the absorption or emission peak maximum, the bandwidth, and the intensity, all of which influence the perceived color.

Stokes shift is the difference in energy between the lowest energy absorption peak maximum and the highest energy emission peak maximum. Although it is frequently expressed in wavelength, it is better to express it in energy units. The reason is that wavelength is not linearly proportional to energy - a 10 nm Stokes shift in the 400 nm wavelength region is different from a 10 nm Stokes shift in the 600 nm wavelength region. The best way to calculate your Stokes shift is to measured the peak wavelength of your absorption band, measure the peak wavelength of your emission band, convert both from nm to eV (or whatever energy unit you like - inverse cm are also common), then take the difference.

Strictly speaking, you can't really say much about solvatochromism from the Stokes shift data you have presented. You might infer that some solvatochromic behavior is occurring but hypso and bathochromatic shifts can both result in the positive (or negative) change in the Stokes shift.

As to why solvents influence absorption and fluorescence colors, and the Stokes shift: electrons in molecules exist in orbitals, and in colored compounds these molecular orbitals tend to be broadly delocalized over large portions of the molecular framework. Nearby solvent molecules interact with these electrons, particularly when the solvent molecules and chromophore are both polar. This interaction changes the energy of the chromophore's electronic states. When certain electrons in chromophores are excited, they are promoted from a lower lying orbital to a higher lying orbital that has some antibonding character. The energy difference between these two orbitals* determines the absorption spectral properties, which leads us to see these compounds as colored if the absorption wavelengths are in the visible region. The excited states of chromophores tend to be more highly polarized than their ground state analogs, and therefore they tend to interact more strongly with nearby solvent molecules. Actually what happens is this: absorption of light by electrons is more or less instantaneous, so that solvent molecules immediately before and immediately after absorption are basically in the same spatial position. But suddenly the solvent molecules find themselves in a very different electric field environment, and so they shift their positions gradually (actually on the subpicosecond to picosecond timescale, but that might as well be eons) around the polarized chromophore. The polarization and reorientation of solvent molecules around the excited chromophore stabilize its excited state and can lead to electronic and nuclear structural changes (lowering the excited state energy), which can cause even more solvent reorientation.** Some nanoseconds later, when the excited state spontaneously relaxes back to the ground state (via fluorescence), the emitted photon has a lower energy than the photon originally absorbed. This is the usually the cause of the Stokes shift. Because "more polar" molecules usually interact more strongly with polarized excited (fluorescing) states, they usually lead to greater Stokes shifts as well as more red-shifted fluorescence peaks (and often weaker intensity). BUT as noted above, solvent polarity is a pretty vague concept. Polarity isn't the key here, it's a complex kinetic process of solvent reorganization around polarized chromophore excited states that primarily causes the solvatochromism and Stokes shift. The solvent molecular size, electron distribution/polarizability, local viscosity, the radiative lifetime of the chromophore, etc., all play roles what the observed emission and absorption energies will be. Since you're using all kinds of different solvents here, this is probably why your trends are pretty complicated.
 
*It's easy to confuse orbitals and states. I'm using them pretty loosely here.
**Using time-resolved fluorescence experiments, you can actually observe these processes happening and measure how long they take.

Thank you for the information. the y axis is simply the energy the light emitted by the compound during fluorescence, calculated using the equation E=hv, which was calculated using the maximum fluorescence wavelength. the x axis is the polarity. The problem here is that I am considering the energy of only a single photon, therefore the energy that is absorbed and also emitted is really small. There are a lot of limitations to my setup, which makes it really hard  for me to consider a lot of variables.

[Corribus]

You're overcomplicating your Stokes shift calculations. Let's say your absorption maximum is at 500 nm and your emission maximum is at 510 nm. You could just report this as 10 nm - probably Ok for what you're doing. If you want to be good about it and plot it on an energy axis, just find a unit conversion website, like here or here. Put in your wavelength values, pull out the energy values. Take the difference. Takes about 15 s.

E.g. 500 nm ~ 2.48 eV (20,000 cm-1). 510 nm ~ 2.43 eV (19607 cm^-1). Therefore your Stokes shift is about 0.05 eV, or just under 400 cm^-1.

(By way of comparison, suppose your absorption and emission max were 700 and 710 nm respectively. Your SS is still 10 nm, but now on the energy axis its, 14285-14084 cm^-1 = ~200 cm^-1. This is why converting your spectral measurements to energy units is important especially if you are comparing transitions that occur at very different wavelength values.)

You still haven't said what your polarity units are or how these numbers were measured, so this axis is meaningless to me.

[wildfyr]

Corribus, that is an absolutely brilliant explanation of solvatochromatism. I've never had it spelled out so clearly. I took this simple seeming phenomenon that I have encountered a hundred times for granted. Thank you! That explanation will probably stick in my head for the rest of my career.

[Babcock_Hall]

There is a good introductory discussion of several solvent polarity scales on pp. 318-320 in Advanced Organic Chemistry, 3rd ed. by Jerry March.
EDT The Grunwald-Weinstein Y-scale can be used for mixtures of two solvents.

[Corribus]

Thanks, glad you found it helpful. I wrote it while kind of glazy-eyed past my bedtime. Reviewing, I should not have said that polarity isn't the key. All things being equal, chromophores do usually have larger abs and PL redshifts and PL Stokes shifts in more polar solvents because they offer greater ability to stabilize (lower the energy) the polarized fluorescent excited state. But things are rarely always equal. You can imagine situations where solvent molecules are prevented from reorienting around the excited state (really low temperature, say), in which case you may see every little effect of polarity on the Stokes shift. You can take 5 solvents and rank them by some metric of polarity, say dipole moment. But solvent molecules aren't generic blobs with a polarity number attached to them. They have different shapes, functional groups, etc. which can influence the magnitude of their interactions with nearby electric fields generated by highly polarized chromophores - which themselves exhibit a lot of variety. The fact that professional molecular physicists still study these effects in "simple" chromophore and solvent systems using fast spectroscopic techniques tells you that there's a lot of complexity that still isn't fully understood.

I only bring it up so the OP doesn't think their data should have an easy explanation. We can confidently explain broad trends in general terms but more precise explanations may be difficult without more sophisticated equipment.

By the way, when spectroscopists want to study solvatochromism in particular chromophores, they will often use more controlled solvent systems. Instead of using a wide array of solvents that can have many confounding factors, the OP may consider using a system of only two miscible solvents and study the chromophore's properties in mixtures of the two solvents. E.g., pick a low polarity solvent A and a higher polarity solvent B and study absorption and emission properties as the ratio A:B changes from 100% A to 100% B.  And of course, you don't need sophisticated equipment to include units on your plots and clearly describe what you're doing (what instrument you're using, what the temperature is, what controls you have, what your chromophore concentration is, etc.). These are hallmarks of good experimental science and 16 isn't too young to learn them.

[AoisPokemaster]

You're overcomplicating your Stokes shift calculations. Let's say your absorption maximum is at 500 nm and your emission maximum is at 510 nm. You could just report this as 10 nm - probably Ok for what you're doing. If you want to be good about it and plot it on an energy axis, just find a unit conversion website, like here or here. Put in your wavelength values, pull out the energy values. Take the difference. Takes about 15 s.

E.g. 500 nm ~ 2.48 eV (20,000 cm-1). 510 nm ~ 2.43 eV (19607 cm^-1). Therefore your Stokes shift is about 0.05 eV, or just under 400 cm^-1.

(By way of comparison, suppose your absorption and emission max were 700 and 710 nm respectively. Your SS is still 10 nm, but now on the energy axis its, 14285-14084 cm^-1 = ~200 cm^-1. This is why converting your spectral measurements to energy units is important especially if you are comparing transitions that occur at very different wavelength values.)

You still haven't said what your polarity units are or how these numbers were measured, so this axis is meaningless to me.

Noted. My x axis are relative polarities that are calculated using the dielectric constant of the different solvents. I have no idea how it is calculated, as I was unable find the exact formula from which the numbers are derived. I am only able to tell you that these numbers are calculated using the dielectric constant. Here is where I retrieved the values. If there are any literature that you can find regarding the mechanism of solvatochromism, I would highly appreciate it, thank you so much for all your help so far.
https://sites.google.com/site/miller00828/in/solvent-polarity-table