[Dolphinsiu]

Here is my experimental procedure:

1. Measure the %T of the 0.001 M Cr(III) solution using cuvette at the range of 400 nm to 750 nm

2. Select six wavelength to study the absorbance as function of concentration
(1)(2) Wavelength at two maxia (490,750)
(3) Wavelength %T at one minimum (400)
(4) wavelength at steeply rising portion (440)
(5) wavelength at steeply descending portion (520)
(6) 625 nm
Measure all transmittance %T for 0.01 M, 0.02 M, 0.03 M, 0.04 M, 0.05 M chromium (III) solution at these six wavelengths.

3. Determine %T and absorbance of the unknown Cr(III) solution.

I have measured all transmittance %T for 0.01 M, 0.02 M, 0.03 M, 0.04 M, 0.05 M chromium (III) solution at six wavelengths.

?                                                                  %T at wavelength of
Concentration/mol dm-3   490 nm   750 nm   400 nm   440 nm   520 nm   625 nm
              0.01                     79.0         85.5        63.7       69.0       75.0      73.6
              0.02                    72.9          85.0        46.1       56.2      65.1       62.8
              0.03                     67.9         84.8        33.8       46.5      56.9      53.9
              0.04                     63.1         84.8        24.8       38.3      49.7      46.4
              0.05                     59.2        84.8        18.6        32.3      44.0       40.5
Range of %T                     19.8         0.8        45.1        36.7       31.0        33.1

Why 440 nm is best wavelength for analysis? (I have explained why we use 20%T - 80%T, but why we use the highest range of %T? I have found many book but still fail)

Also, why there is 2 maximum peak appear in beer's curve (%T vs wavelength for 0.01 M Cr(III) solution in 400 nm to 750 nm)? (2 oxidation state Cr2+, Cr3+ can explain 2 peaks, why?)

Thank you!

[Borek]

Have you prepared plots?

[Dolphinsiu]

Yes. If not, I may not have questions.

Too many chemistry problems get stuck on me although lecturer has taught me a lot!

I find many problems when doing lab. So I would like you to help me answer!

[Dolphinsiu]

Does anyone do similiar experiment that can help me?

[ARGOS++]

Dear Dolphinsiu,

In case I have caught correctly what you did, you have only done the half of
the exercise you got, and that’s why you have problems.
First you did all your experiments in %T, as you had to do it, but then you
lost the red line.
Under 2. you write something about ABSORBANCE, but I’m not able
to see any of your result in this particular “UNITS”. Do you have an idea, why
you should do it in absorbance anyway?
If you would have done it, and even drawn a plot of absorbance vs. wavelength,
you should result in a similar plot as this one below, only with single wavelengths,
but with more dilution (conc.) steps:
   
With other words, if you translate ALL into absorbance, you result for each
wavelength in a plot as on the 2^nd Linearity Report Page (“Measurement”) on:
    "Linearity:  --  maybe a little more ?"
     http://www.p-forster.com/english/themes/Spectroscopy/Linearity.htm,
but each graph will show you a different slope (Why?), AND in case of
“full spectra” you get a similar/”quasi identical” graph as on the 3^rd Linearity
Report Page (“Raw Results”) of:
    "Linearity:  --  maybe a little more ?"
    http://www.p-forster.com/english/themes/Spectroscopy/Linearity.htm,
Now it’s easy to answer most of your questions, especially if you keep in
mind, that still, nobody knows why (I too!), WHY Everybody thinks in
OLD, NOT VERY STABLE AND NOT VERY REPRODUCIBLE spectrometers:
a.) Your 440 nm seams to be best, at most, because two facts:
     1. Everybody defines sensitivity as ?Signal divided by ?Concentration, and that
         means, THAT YOU HAVE (But it's NOT true!) to measure
         it “very near” the Lambda_max. 
     2. If you ADD the tangent at the spectrum at Lambda_max, and at a strong
         increasing/decreasing region of the spectrum, then you see it very easy, that at
         Lambda_max the tangent is a horizontal line, what is indicating, that
         near this region, you are nearly independent of the wavelength stability/reproducibility
         of your "very bad" spectrometer.     
         But what’s in the other situation?
     As you may have realised, THAT IS by far NOT LONGER TRUE, you have
     only to inspect the 4^th Linearity Report Page (“Validation”, see above),
     where all dilution spectra are conc. normalised, whoops!
     Yea, - Beer-Lambert tells you, that “HE” IS true for EVERY wavelength!     
     So it’s even not longer true, that you ONLY have to use the range 20%-80%T, or?

b.)  The two peaks will never be caused by the two oxidation state, because
      Complexation/Coordination will mostly go prior, as you can easily extract from
      the following table:
       
         A very big bouquet of thank you for the table, to:
             
His complete report you may find under:
       http://wwwchem.uwimona.edu.jm:1104/courses/Tanabe-Sugano/TSintro.html

An other source of information may be:
       http://digitalcommons.fau.edu/dissertations/AAI1380720/

I hope, it is of any help to you, too!
Good Luck:
                      ARGOS⁺⁺

[Dolphinsiu]

Thank you very much. ARGO. You do open my learning scope in my knowledge!
as I have never heard the terms 'wavelength stability' and 'reproducibility'

[ARGOS++]

Dear Dolphinsiu,

Thank you, that it seems I was of help to you.

The terms 'wavelength stability' and 'reproducibility' are easer to memories,
if you draw an imaginary picture:
First Picture:
   Think about you would have to do all the Day long, during 365 Days a year,
   only one single spectroscopic method (e.g. Cu²⁺ at 800 nm). How many times
   would you change the wavelength of your spectrometer? Yes, never!
   Now, what would your wavelength be at begin of the year and what at the end?
   (Most instruments are semi automated driven for wavelength adjusting, even!!!)
   If it pass, it would be called “Long Time Stability”, but most of all instruments will
   even fail what is called “Short Time Stability” (FDA: < ± 1.0nm UV/Vis, all times!).
   Quintessence: Think about the influence to your published Results in case of a
   horizontal tangent, and in case of the tangent in a strong slope of your spectrum!!!
Second Picture:
   If you publish your results taken/measured at a strong slope of the spectrum, and
   1’000 laboratories will reproduce your scientific work, and all send you their results.
   In case you try to calculate the standard deviation of all their “cooking’s”, - How BIG
   will the “outcome-ing” number be?
   And how big/small will it be in case of a horizontal tangent?

That’s why in earlier times a “Golden Rule” told you to measure and estimate ? at a
place  of your spectrum, where you have at the same time a horizontal tangent!!
But that’s NOT longer true for PDA’s, Photo Diode Array Spectrometers,
if you reference to the Page: "Linearity:  --  maybe a little more ?" from above.
Also the second part of the “Rule”: "to measure at Lambda_max to be most
sensitive" is NOT longer valid, toooo, because it’s mostly that Lambda_max can
have other, important disadvantages, too!!!

I hope, the pictures we drew, are not only pied to you, and others.

Good Luck!
                   ARGOS⁺⁺