Chemical Forums
Chemistry Forums for Students => Organic Chemistry Forum => Topic started by: tkboarder63 on April 05, 2011, 05:25:54 PM
-
Hey guys, i'm developing an experimental procedure for the separation of 2 unknown solid compounds using column chromatography. The possible unknowns include biphenyl, p-dimethoxybenzene, p-dibromobenzene, acetanilide, benzanilide, 1-indanone, 2-acetonaphthone, benzil, and 9-fluorenone. Now, the initial step would be to dissolve the compound in something like dichloromethane and run TLCs of different solvents to find the ones that give the best separation. Here is where the problem is occurring for me: as I increase the polarity of the solvent from hexane to 10% ethyl acetate/hexane etc., the spot moves up a bit, but no separation occurs. In other words, for every TLC i've run, only one spot is visible. Now, my rationale for this is that they are both fairly polar compounds with relatively similar polarities. So, i tried increasing the solvent polarity all the way to 80% ethyl acetate/hexane and then 100%. However, this proved to be useless because once again, only one spot migrated with the solvent front, so there was still no separation. I'm really baffled by this and am certain that there is something I am missing. Please, if you have any answers, advice, explanations, or any comments, I would really appreciate the help, Thanks everybody
-
First off: are you sure there's two compounds in there? Even if its a standard for a lab course, sometimes people make mistakes. Are you using UV, I2, or another stain to visualize your spots? Everything on this list should be UV active. I2 is usually pretty good but sometimes it can be capricious. Always worth it to confirm your spots with at least two methods.
How big are your spots? Large splotchy spots can easily run together and hide the truth. I prefer to have the smallest TLC spots possible, about 2 - 3 mm in diameter. If they're the size of dimes, your TLC standard is too concentrated and you should dilute it significantly.
Finally, sometimes its also worth it to just completely change your solvent system. When Hexane/EtOAc doesn't cut it for me, I usually switch over to 2 - 5% MeOH in DCM.
-
In addition to Honclbrif's advice, I'd suggest acetone/toluene (or ethyl acetate/toluene) as an eluent system to try your TLCs with. I find this especially good at separating aromatics that otherwise co-spot in ethyl acetate/hexane. As a very rough guide, 20% acetone in toluene should give you an Rf similar to what you'd get in 50% ethyl acetate in hexane. Possibly an obvious comment, but remember to dry your TLC plate properly after running it in toluene otherwise you won't see anything under UV.
As Honclbrif pointed out, your visualisation method may be the issue so look into that first.
-
Hey thanks a lot guys. So, to address one issue, there are certainly two compounds present, as the goal in the lab is separate the starting mixture given. As for the size of my spots, they are not alarmingly big or undefined. It just seems quite dense and also there is no spot visible at the start, indicating that both compounds are migrating together. As for my visualization methods, I have been using mostly UV as the benzene rings in all of the compounds would make them all visible. Also, shouldn't the dissolved standard be concentrated in order to get the best separation? One last quick general question: say I begin to use exceedingly more polar solvents (i.e., 100% ethyl acetate, 80% ethyl acetate, or even 60%), would it be safe to say that the more polar compound would have the higher rf being that the majority of the solvent is polar rather than polar or would the less polar compound still migrate farther due to its lack of binding to the silica gel plate? Thanks so much for your help so far, any more comments on my situation would be greatly appreciated, thanks
-
I was just wondering how you are going to flash the unknown to separate the compounds ?
-
It just seems quite dense and also there is no spot visible at the start, indicating that both compounds are migrating together.
I don't understand the italicised bit, can you explain?
Also, shouldn't the dissolved standard be concentrated in order to get the best separation?
Not for TLC. More concentrated means bigger, streakier spots. If you're talking about the column, then yes you need to load the sample at high concentration (low volume) to achieve narrower bands (better separation) - but the two cases are different, the equivalent for TLC is making your spots small.
say I begin to use exceedingly more polar solvents (i.e., 100% ethyl acetate, 80% ethyl acetate, or even 60%), would it be safe to say that the more polar compound would have the higher rf being that the majority of the solvent is polar rather than polar or would the less polar compound still migrate farther due to its lack of binding to the silica gel plate?
No, silica gel will pretty much always elute less polar before more polar. The stationary phase is generally more important - I have run TLCs in 5:3:2 BuOH/EtOH/H2O, which is a very polar system, and the polarity order is still the same. If the Rfs (and therefore polarities) are very close (as yours seem to be), changing the solvent system can cause a switch in elution order. I've only seen it a couple of times, but I've had cases where, say, compounds A and B have Rfs 0.50 and 0.48 (respectively) in ethyl acetate/hexane but switch to 0.50 and 0.55 (respectively) in acetone/toluene.
-
It just seems quite dense and also there is no spot visible at the start, indicating that both compounds are migrating together.
I don't understand the italicised bit, can you explain?
By this I just meant that no spot appeared where I had originally "spotted" the mixture, which would indicate that the entire mixture had moved away from the starting position.
Also, shouldn't the dissolved standard be concentrated in order to get the best separation?
Not for TLC. More concentrated means bigger, streakier spots. If you're talking about the column, then yes you need to load the sample at high concentration (low volume) to achieve narrower bands (better separation) - but the two cases are different, the equivalent for TLC is making your spots small.
say I begin to use exceedingly more polar solvents (i.e., 100% ethyl acetate, 80% ethyl acetate, or even 60%), would it be safe to say that the more polar compound would have the higher rf being that the majority of the solvent is polar rather than polar or would the less polar compound still migrate farther due to its lack of binding to the silica gel plate?
No, silica gel will pretty much always elute less polar before more polar. The stationary phase is generally more important - I have run TLCs in 5:3:2 BuOH/EtOH/H2O, which is a very polar system, and the polarity order is still the same. If the Rfs (and therefore polarities) are very close (as yours seem to be), changing the solvent system can cause a switch in elution order. I've only seen it a couple of times, but I've had cases where, say, compounds A and B have Rfs 0.50 and 0.48 (respectively) in ethyl acetate/hexane but switch to 0.50 and 0.55 (respectively) in acetone/toluene.
-
So, I think that my main problem has been a combination of a too concentrated solvent standard and too thick spotting technique. So I will certainly decrease the concentration and the size of my spotting. I am also wondering if my previous sleuth of TLCs reveal any information as to what effective solvents may be, despite the concentrated standard. For example, pure hexanes yielded no migration and as I increased the polarity, the single spot's rf increased. So, could that spot's rf be indicative of the rf of one of the compounds(less or more polar), while the other compound is, for lack of better words, "stuck" with it. In other words, can I learn anything from my previous TLCs based on the migration of this "mutual spot." This has really been a huge help so I really appreciate it, can't thank everyone enough. Please comment if you have any more advice. Thanks
-
I was just wondering how you are going to flash the unknown to separate the compounds ?
Based on the results I get from my TLCs, I will choose my two solvents. One aimed at migrating the less polar compound through the column and into a set of fractions, then another solvent driven to eluent the more polar compound into a different set of fractions. I would then do a TLC for the fractions to visualize which compound is in which set of fractions. Then I would boil off the solvent after pooling the fractions together(wind up with one pool of less polar compound and another with more polar), leaving the solid.
-
It is possible that you have two ketones or an acetanilide and benzanilide which would produce similar polarities and be more difficult to separate as a result.
-
If pure hexanes didn't move anything off the baseline (yielded no migration), I think you can safely eliminate the non-polar aromatics from your list of unknowns.
-
Hey, I've just been told about the possibility of Beer's law taking effect where I would have to increase the concentration of the solvent standard in order to get an effective absorption visibility under UV. This kinda of threw me for a loop because now i'm dealing with two extreme possible answers. To dilute my standard in order to get a good separation, which may be a possibility. Or i'm dealing with a case where my previous TLCs with one spot visible, in reality has two spots, but only one can be seen because the concentration is not high enough. I thought I had everything worked out, but now i'm just confused. Any advice, please comment
-
There is a happy medium.
Small but easily visible spots are better than huge blotchy spots or nearly invisible spots. Once you've diluted your standard, apply it to the plate. and check it with the spot lamp before you even run it. If you can see it, you're good. If you can't see it, spot it again until you can.
-
So after considering all of the evidence, I think that my problem has been my spotting technique. I have to ensure that the initial spots be small but concentrated enough with the standard unknown to be seen before developing. I looked over all of my TLCs again and I am beginning to think that what was happening was that my non-polar(or less polar) compound has not been visible due to a low concentration, hence leaving only one spot visible(the polar compound). This separation is not meant to be of high difficulty so I have since ruled out the idea that the two compounds have very similar polarities, resulting in two nearly identical rfs, which would give the illusion of one spot. Feel free to comment on my reasoning. Also, I am planning on dissolving my initial compound in acetone for the TLC and then possibly dichloromethane for the column chromatography, does this sound right? Or should I be dissolving my compound in an ethyl acetate/hexane mixture for the column chromatography? (Note: all eluent solvents will be varying proportions of ethyl actetate/hexane) Please comment, thanks
-
I don't think you want to dissolve the compound for column chromatography. If you did the sample would end up all over the place and you wouldn't get any separation. I usually put the pure/crude sample at the top of the column, drain the column down to the level of the silica, then fill it up with the eluent and go.
-
For TLC standards, I would dissolve them in whatever gives the best solubility. I find that it does not matter what your TLC standard is dissolved in, so long as it is suitably volatile the solvent will evaporate before you ever run the plate.
For columns, to avoid potential funny business, I prefer to dissolve my sample in the mobile phase with which I will be running the column. Obviously to get the tightest bands you want to keep the volume of solution you are loading to a minimum (i.e: high concentration). The number I've been quoted and which has worked well for me is "less than 2% of the bed volume".
To load: pack your column as usual (slurry pack please, dry packing is for delinquents), then drain the solvent down until it is level with the top of the bed (or sand if you are using a sand layer on top), then use a pipet to gently dribble your solution down the sides of the column. Start as close to the bed as you can get and be as gentle as possible, the last thing you want to do is to disturb the top of the bed (another reason why a sand layer is usually a good idea). Drain until the solvent is again even with the top of the bed, then dribble a few mL of mobile phase in and drain as before. Repeat with solvent 3 or 4 times to get it good and loaded, then gently pipet solvent in until you have about 5 cm of solvent above the bed, then pour more solvent in until the column is full. Now elute as you normally would.
-
I don't think you want to dissolve the compound for column chromatography. If you did the sample would end up all over the place and you wouldn't get any separation. I usually put the pure/crude sample at the top of the column, drain the column down to the level of the silica, then fill it up with the eluent and go.
Good advice if your material is an oil, but if it is a solid you've got to dissolve it in something first. Just throwing the solid in there and waiting for it to dissolve as the mobile phase runs would probably be a disaster if the material does not dissolve rapidly.
The only exception to this I know of is if the material is not particularly soluble in the mobile phase. In this case dissolve it in something which it is very soluble in, add a few grams of silica gel, and vac off the solvent. Pack the column, then add the pre-loaded gel to the top and elute as normal.
-
Oh yes I forgot about that lol. I'm in undergrad and have only been working with oils lately.
-
Re: Columns
I think it really depends on what you're working with and what you're trying to separate really. For me I've found the most important factor in getting a good separation is to run your column really, really, really slow. Most of my reactions end up giving me a spot that runs partially into my product spot (Rf difference less than 0.05). I've found though that if you run the column in a solvent system where your stuff runs at an Rf of 0.05 (basically barely nudging off the baseline), you can still get fairly good separation. Before you ask yes I tried different solvent systems, the resolution is just as bad.