Chemical Forums
Chemistry Forums for Students => Analytical Chemistry Forum => Topic started by: orgo814 on August 02, 2014, 07:46:47 PM
-
I'm trying to troubleshoot here.. What are some of the reasons my standard curve could have come out bad? My r^2 was terrible at 0.3. My cuvettes may have been contaminated since they've been used before although I cleaned them. Could that be cause?
-
That's certainly a possible reason. However, that is really a horrible r2. I suspect many things -- could you have calculated it wrong, and of course there be another serious fault. Can you give a a
tale table of values, and possibly a chart, because really, how do you call this a line. Maybe you need a higher order polynomial for your data, if that's permissible.
-
Telling us what technique/experiment you're doing could also be helpful.
-
I redid it and got a 0.85 by changing the cuvettes I was using but that still is not good enough
-
The concentrations are: 1650, 825, 412.5, 206.25 (ug/mL)
The absorbances I got through UV-Vis were: 0.090, 0.078, 0.08, 0.073
I redid it using 20 mL standards instead of 10 mL. My r^2 ended up worse..
Absorbances: 0.194, 0.172, 0.012, 0.005
I always struggle once I put the 412.5 solution in. I am using 15 uL of each of the solutions (concentrations above) with 1 mL of a coumassie blue reagent I made up.
-
The R2 isn't your primary problem. I'm not sure exactly what you're doing, but if you double the concentration, you should double your absorbance - as long as you're in the linear range. You're not even close. In fact, between 412.5 and 825 ug/mL, your absorbance is going down. This is basic Beer's Law.
Are you doing a single read measurement, or are you acquiring the entire spectrum? If the former, are you sure you're reading the right wavelength (i.e., where there's a peak)? I would suggest you acquire an entire spectrum to be sure.
-
Here's a quick and dirty chart. Sorry I didn't take the time to label axes, or the series. You should be able to tell by the numbers, but pink is your "20 ml" sample, and blue is your "10 ml" sample. I'm sorry you don't like the "10 ml" sample, it gives a pretty straight line -- although not a useful one, as absorbance doesn't change much as a function of concentration.
I'm wondering if you can come up with some conclusions, looking at the data thi way.
-
I've done beers law before and know how to make standards well... The only thing I can think of is my pipettes weren't calibrated correctly or my coumassie blue went bad. This is so frustrating! :/
-
The latter can be ruled out (mostly) by taking an entire spectrum of a dilute solution of your coumassie blue and comparing it to literature. Are the peaks in the right place? Are there any extra ones that shouldn't be there? Do your extinction coefficients roughly match. UV-Vis is fairly tolerant of impurities.
I think your pipettes would have to be pretty terrible in order to be off by so much.
What wavelength are you measuring at?
-
It seems the extinction coefficient at 595 nm at pH 7 is around 43,000. This should help you estimate what your absorbance should be at the concentrations you are using.
-
I've been working at 595
-
I'm just completely confused at what could cause this
-
Based on an extinction coefficient of 43,000 M-1 cm-1, which seems about right, what do you expect your absorbance to be for a 206 ug/mL solution?
-
I know beers law is A = EBC, b being path length (1 cm). How would i convert the concentration to fit this formula
-
I may have solved my problem. Did research and the coumassie I had was the reddish kind with max absorbance at 470 and I was doing mine at 595.
-
I'm just completely confused at what could cause this
Really? Your "10 ml" samples look give essentially the same number in each case. Those look to me like you are below the ability of the system to detect an absorbance. How do the "10 mL" samples compare to a blank, just coomassie blue and water? If there's no difference, then you can't use your procedure at that level. Your "25 mL" seem to take a sudden drop as you go to lower levels, even those samples may not be concentrated enough for your system you quantitate them.
Any of a number of sample handling problems could also have affected that sudden drop. Are these serial dilutions? Or did you prepare each one individually?
-
These are serial dilutions. I picked the wrong max absorbance wavelength to blank and work at. Could this have caused this?
-
I may have solved my problem. Did research and the coumassie I had was the reddish kind with max absorbance at 470 and I was doing mine at 595.
This would certainly be a problem, which is why I suggested taking an entire spectrum.
-
Well I will test it out when back in lab but I have to think this is it. Thanks for your help I appreciate it.
-
I diluted the coumassie before using it but I still think the max wavelength is still lower than what I set it as. There are multiple coumassies.
-
Some comments
-please dont only use r^2 as the only factor to judge your curve. there are bad curves with good r^2
-ALWAYS look at the absorption spectrum of your compound if you do a new experiment. it takes no time.
-look at your blanks
-
I keep repeating it and getting same r^2 (around 0.85). Question: Are some standard curves going to have a bad r^2 and still be what would be acceptable for them? I was looking at some standard curves for BSA and not all of them are linear. I guess it depends on my concentrations?
-
I keep repeating it and getting same r^2 (around 0.85). Question: Are some standard curves going to have a bad r^2 and still be what would be acceptable for them? I was looking at some standard curves for BSA and not all of them are linear. I guess it depends on my concentrations?
You haven't answered some of the questions I asked, and my concerns were repeated by Irlanur:
Are your results significantly different from a blank? If not, you don't have enough sample to run your assay. The "10 ml" samples look all the same, they have a quite good r2 value, for all the wrong reasons. unless your sample isn't even reaching the beam of the instrument -- unless this "10 ml" sample isn't "10 mls of sample diluted to the same volume as the 25 mL sample" We're not there, so its up to you to give us all the information on your assay: what is the sample, what is its initial concentration, how are you diluting it, and with what. What is your coomassie blue solution, how much of it is in each sample, what are you diluting your final samples with, and what wavelengths are you using.
-
The sample is BSA. I make up solutions with the concentrations I listed earlier (through serial dilutions). I took 15uL of each solution made up and added it to 1 mL of my coumassie reagent. My coumassie reagent was made from 1 mL of the coumassie and 4 mL of nanopure water. The coumassie was actually red but when diluted had a greenish tint. My blank was greenish while the others were a brilliant blue (after addition of the coumassie reagent to 15uL... I think it turns that blue color when it binds to the protein).
-
Also, I worked at 595 nm.
-
So, given the shift in wavelength given below, blanking isn't as useful with coomassie blue as in other situations. This is problematical, but not a disaster. You mention working with a 1 ml sample size, that should put the fluid level in the beam path, but do be sure. For example, a serial dilution of the Coomassie Blue, without any sample, should show a linear absorbance, even if the wavelength with maximum absorbance isn't the same with protein bound. You could try evaluating that, to conform the instrument is working correctly and you're preparing things correctly.
-
I mean, I did it a second time and had my absorbances 0.778, 0.698, 0.533, and 0.328, respectively. Which is improved. There's a downward trend which is what I am looking for but my linearity is still not great.