CHEM340 Instrumental Analysis

Tutorial Prof. A. Kindness

The following questions are designed to make you go back into your notes and understand what
you have been taught so far. Some of these calculations will help you with practicals so it is
important that you attempt these questions. They are also examples of the type and complexity
of questions you may get in the exams.

1. If the solution used to calibrate an instrument was only 98% pure (in terms of the analyte)
what affect would this have on the accuracy and precision on the analyses of samples?
Simple little question that explores your understanding of precision (random error) and
accuracy (systematic error). In this case the answer would be consistently wrong (low
by 2%). It only affects the accuracy, does not influence the precision at all.
2. The following data was obtained for the analysis of arsenic in a liver sample. It does not
matter which technique, this is testing your ability to draw a calibration graph.
signal signal for blank
concentration
(absorbance @ 480nm) (absorbance
(μg/mL)
@ 480nm)
0.00 0.005  Blank signal 0.004
2.00 0.201 derived from 0.005
replicates 
4.00 0.399 0.003
6.00 0.601 0.006
8.00 0.799 0.008
10.00 0.999 0.004
What is the limit of detection for this technique?
Here I am looking to see if you know how to calculate the LoD. First of all work out the
standard deviation of the blank (= 0.00179). Plot graph to get the calibration line:
y = 0.0995x + 0.0031
LoD = 3 x sd, so blank signal would be 0.005 + (3 x 0.00179) = 0.0104, substitute this
into the equation of the line. LoD = 0.070 μg/mL
3. A mass of 2.473 g of an unknown sample was
Sample (μg K/mL) Relative emission
dissolved in 10 mL of conc. H2SO4 and the
mixture boiled for 5 minutes. The solution was Blank 0
cooled and the volume made up to 250 mL in a 5.00 124
volumetric flask. The emission of this solution
10.00 243
was measured at 404.3 nm. A series of
potassium standards gave the following emission 20.0 486
intensities at 404.3 nm. 30.0 712
Find the concentration of potassium in the Unknown 417
unknown.
Simple question asking you to draw a calibration graph. From graph y = 23.77x + 4.02
and ∴ 417 = 17.4 μg/mL.
We have 250 mL ∴ 17.4 μg/mL x 250 mL = 4344 μg. This was from 2.473g ∴
4344 μg /2.473g = 1757 μg /g or 1.757 mg/g 0.1757% (m/m)

Page 1 of 7
4. Li was determined by atomic emission using the method of standard additions. From the
data in the following table, find the concentration of Li in pure unknown.
The Li standard contained 1.62 μg Li/mL.

Unknown Standard Final Volume Emission intensity

(mL) (mL) (mL) (arbitrary units)
10.00 0.00 100.0 309
10.00 5.00 100.0 452
10.00 10.00 100.0 600
10.00 15.00 100.0 765
10.00 20.00 100.0 906
First of all identify that your are being asked to do a "method of standard additions".
First calculate the conc of the standards added. (5.00 mL x 1.62 μg/mL) /100 mL =
0.081 μg/mL and so on. The equation of the line = y = 1860.5x + 305
∴x = 0.164 μg/mL but this was derived from 10 mL of sample therefore the conc of
Li in sample 0.164 x (100/10) = 1.64 μg/mL.
5. Methanol was transported to South Africa from India by sea. The methanol was believed to
be contaminated by sea water during the voyage. The buyer of the methanol stipulated
that the chloride concentration must be less than 2 μg/mL. Several samples of methanol
was removed from various sections of the tanker and combined. Aliquots of this laboratory
sample were analysed in the following manner:
• 10 mL of methanol was added to a 25 mL volumetric flask.
• various amounts of a standard chloride solution was added
• 2 mL of a Fe (III) solution and 2 mL of a saturated Hg(SCN)2
• The solution was mixed and made up to the mark with pure methanol.
• The solutions were allowed to react for 15 minutes, an aliquot of this solution was then
transferred to a 1cm cell and the absorbance was measured at 580nm.
Assume that there was not a significant difference in the densities of the solutions.
The Hg(SCN)2 reacts with the chloride ion and releases the thiocyanate ion which
subsequently reacts with the Fe (III) solution.
The following results were obtained.

mL of standard
methanol
154 ppm Cl Absorbance
sample (mL)
solution
10 0 0.076
10 0.10 0.135
10 0.20 0.205
10 0.30 0.271

a) Why were the Fe (III) and saturated Hg(SCN)2 solutions added to the methanol.
To form a chromophore, i.e. to make it coloured.
b) What species was actually being measured by the spectrophotometer?
The actual species being measured is Fe(SCN)2+.
c) Is the methanol acceptable to the buyer? Calculate the concentration of chloride in the
methanol.
So first things first, identify that it is method of standard addition.

Page 2 of 7
 Calculate the "amount" of Cl added and plot the graph. Get the equation of the line,
y = 0.1063x + 0.0735, therefore the conc in the sample = 0.691 μg/mL.
Take into account dilutions and sample volume etc and you should get 1.73 μg/mL.
Yes the methanol is acceptable to the buyer, ignoring any errors.

d) When the solutions were reacting a colloidal precipitate (due to the excess Hg(SCN)2)
formed. Why did the analyst have to remove the colloidal suspension?
This would have scattered the light and you would have an error in the analysis, you
would have measured absorbed light + scattered light.
e) What is the name of this type of calibration and when is it generally used.
Standard addition is appropriate when the sample matrix is unknown or complex
and affects the analytical signal.

6. Glucose in blood serum can be determined Amount

Abs
spectrophotometrically by formation of a coloured glucose
complex with o-tolidine. Six identical 50.0 μL serum added μg
samples were treated with various amounts of 0 0.230
glucose and taken through the procedure, 10 0.272
producing these results. Prepare a standard addition 25 0.340
calibration plot and calculate the glucose conc. in 40 0.416
the serum in units of mg/100 mL. 60 0.507
75 0.568
Plot the data and you should get y = 0.00458x + 0.229, so the amount glucose present
is 49.7 μg of glucose in the 50 μL sample. No dilution factors as the volume was given
and was constant, so simply convert the 49.7 μg/50 μL to mg/100 mL = 99.4 mg/100 mL

7. A municipal water company was having problems with its water analysis. The problem lay
in the AAS determination of Fe at 248.3 nm. The absorbance of the water, after 5 fold
dilution was 0.646 at 248.3 nm. A standard solution prepared by dissolving 0.1483g of iron
wire in acid, diluting to 250 mL. After a further x100 dilution the solution had an
absorbance of 0.813. Calculate the ppm in the water sample. What is wrong with this
analysis?
On a proper investigation of the above problem a series of iron standards were prepared by
taking various volumes of a 0.0593 mg Fe per mL and diluting up to 100 mL. The
absorbance of the solutions were as follows:
Volume of Absorbance
What is the true value of the Fe content in the municipals Fe taken (mL) @ 248.3 nm
water (use absorbance in part 1). 0.00 0.000
Calculate the percent error between the results. 1.00 0.113
What assumption was made in part one that was not 3.00 0.334
valid? (This assumption should never be made in 5.00 0.530
atomic spectroscopy and at the very least three standard 7.00 0.672
readings should be recorded. I prefer using 5!) 10.00 0.813
Work out concentration of standard 0.1483g Fe in 250 mL is equivalent to 0.5932g/L =
593 ppm. A x 100 dilution was carried out on the solution ∴ 5.93 ppm Fe.
Only one standard solution prepared (bad analysis) so direct comparison of sample with
5.93ppm χ ppm
standard. = χ = 4.71ppm Fe
0.813 0.646

Page 3 of 7
 take into account × 5 dilution ∴ water solution contains 23.55 ppm or 23.55 μg/mL
Convert the data on volumes of standard solution into ppm (or whatever units you want)
draw graph.

0.8
0.7
0.6
Absorbance

0.5
0.4
0.3
0.2 3.93ppm
0.1
0
0 1 2 3 4 5 6
ppm Fe
3.93 ppm is equivalent to (5 × 3.93) 19.65 ppm in the water sample.
Difference 23.55 - 19.65 = 3.9 as a % 3.9/19.65 x100 = 19.8%
The assumption was that the calibration was linear. In AAS/FES graphs normally curve
to some extent. I always use at least 5 standards, some people use 3!
8. Six samples of raw milk, each 5 mL, were treated with 0, 5.00, 10.00, 15.00, 20.00 and
25.00 μg of Zn2+ and diluted to 50 mL. The absorbance of these solutions were 0.235,
0.292, 0.346, 0.397, 0.458 and 0.510. How much zinc is in the milk?
Another method of standard addition. Draw the graph y = 0.0110x + 0.236, solve for
y = 0 and you should get 21.4 μg of Zn in the 5 mL of milk or you can express it as
4.3 μg/mL. You will note that I didn't convert to concentrations etc, if all the volumes
are constant then you don't need to. Try it; all you do is multiply and divide by the
same answer.
9. Explain why errors associated with spectrophotmetry are worse at high and low
absorbance measurements.
Simply go into your notes and you will see why. At High absorbances the detector is
struggling to measure the small amount of light reaching the detector, at low
absorbance virtually no light is absorbed and the detector finds it difficult to
differentiate between the light intensity. Remember Absorbance is a ratio
measurement. Also at high absorbance values any stray light entering the
spectrometer will give rise to problems. Also look at the graph of %T versus conc.
Also think about signal to noise.
10. Which will be the most sensitive method of spectrophotometric analysis for Fe: dithizone
(ε = 4.16×103 cm-1 mol-1 L), o-phenanthroline (ε = 1.06×104 cm-1 mol-1 L) or thiocyanate
(ε = 7.90×103 cm-1 mol-1 L)? Briefly explain why.
The most sensitive method will use o-phenanthroline, it has the highest ε, this will
result in the highest Absorbance value for any given concentration.

Page 4 of 7
11.
0.8

0.7

0.6

0.5
absorbance

0.4

0.3

0.2

0.1

0
300 350 400 450 500 550 600 650
wavelength (nm)

Above is the UV-Vis spectrum for a solution containing 7.50 ppm permanganate ion.
a) What type of material was the cell constructed from for recording the spectrum?
Since it is recorded in both the UV then it must have been a quartz cell, this can
also be used in the visible range.
b) What is λmax of the strongest absorption peak?
λmax = 525 nm
c) What is εmax (cm-1 mol-1 L) of the strongest absorption peak, given that cell length
was 5 mm?
A
Simple manipulation of Beer's law. From A = εcl ε=
cl
7.5 ppm ≡ 7.50 mg/L, c = 7.5 × 10-3/118.9 = 6.31 × 10-5 mol/L
ε525 = 0.68/(6.31 × 10-5 × 0.5)
ε525 = 2.15 × 104 mol-1 L cm-1
d) What type of transition is involved?
The high molar absorptivity suggests a charge transfer complex
12. The molar absorptivity for a complex formed between Bi(III) and thiourea is 9.32 × 103
L cm-1 mol-1 at 470 nm. Calculate the range of permissible concentrations for the complex if
the absorbance is to be no less than 0.10 nor greater than 0.90 when the measurements are
made in 1.00 cm cells.
A
Again simple manipulation of Beer's law. A = εcl c =
εl
So if A = 0.1, ε = 9.32 × 103 L cm-1 mol-1 and l = 1.0 cm minimum c= 1.01 x 10-5 M and
maximum c = 9.66 x 10-5 M
13. The ligand 4,4',4"-triphenyl-2,2',2"-terpyridine forms a 1:1 complex with Fe2+ whose molar
absorptivity is 3.02 × 103 mol-1 L cm-1 at 583 nm. If the lowest detectable absorbance is
0.005, what is the lowest concentration of Fe2+ that can be detected with this complex in a
10.0 cm cell?
Again simple manipulation of Beer's law =1.66x 10-7 M

Page 5 of 7
14. In the following diagram, a graphical representation of an AAS signal is given, in both cases
the concentration of analyte being measured was the same.
S = signal and N = noise, both are in arbitrary units.
Briefly discuss the two diagrams below in terms of sensitivity, signal:noise ratio and
which diagram represents the scenario that will offer the lowest (i.e. best) detection
limit?

The sensitivity is clearly better for C twice the signal of D for the same
concentration. However, the S:N ratio is 8 for both signals therefore the
detection limit will be the same in both cases.
15. The following equations for calibration plots were obtained for the calibration of Cd by ICP-
OES, Line A = y = 6.02x and Line B = y = 18.06x.
What line is the most sensitive? Justify your answer?
Line B is the most sensitive
If the mean noise for the blank signal for A = 0.05 counts per second and B = 0.18, which
line will have the lowest detection limit? Briefly explain.
You need to look at the 3 x s so for A LoD = 0.025 and for B = 0.030
16. The following chromatograms have been generated by gas chromatography (GC) and are
for a mixture of chloroform, m-xylene, pentanal and nonane.

compound mass peak area Sample

chloroform 0.251 3617 5421
pentanal 0.205 7452 1264
nonane 0.185 13687 1939
m-xylene 0.175 12044 4007
What is the composition of the sample?

compound peak area % component

chloroform 5421 75.9%
pentanal 1264 7.0%
nonane 1939 5.3%
m-xylene 4007 11.8%

17. When 1.06 mmol of propanol and 1.53 mmol of hexanol were separated by gas
chromatography, they gave peak areas of 992 and 1570 units respectively. When 0.57 mmol
of propanol was added to a solution containing an unknown amount of hexanol the peak
areas from the chromatogram were propanol = 843 and hexanol = 816.
How much hexanol did the solution contain?
Using relative response factors as in the practical manual, you get a factor of 0.848
then the concentration of hexanol = 0.47 mmol.

Page 6 of 7
18. The following data were obtained for analysis by ICP- Emission
[Cu] Sc 200 ppb
OES. Intensity
(μg/L) Emission
@324.8 nm
What is the Cu concentration in sample #1? 0 21 3500
What is the Cu concentration in sample #2 if the 25 599 2857
amount of Sc added was 400 ppb? 50 998 2476
100 2800 3457
Scandium is being used as an internal standard so we 250 7032 3365
plot the ratio of the signals. 500 17036 4056
Sample
4.5
#1 5850 3458
4 y = 0.0084x - 0.0099
#2 4263 6356
3.5
3
ratio Cu:Sc

2.5
2
1.5
1
0.5
0
0 100 200 300 400 500 600
Cu ppm

So for sample #1, 203 ppm

For sample #2 the internal standard is twice as much so we need to divide by 2; so the
internal signal is 6356/2, conc = 161 ppm

Page 7 of 7