6 Random Errors in Chemical Analysis

6A The Nature of Random Errors

6A-1 Random Error Sources?
Fig. 6-1 Three-dimensional
plot showing absolute error in
Kjeldahl nitrogen
determination for four different
analysts.
Analyst Precise Accurate
1
2
3
4

Table 6-1 Possible combinations of four equal-sized uncertainties

Combinations of Magnitude of Number of Relative
Uncertainties Random Error Combinations Frequency
+ U1+ U2+ U3+ U4 +4U 1 1/16=0.0625
U1 + U2 + U3 + U4
+ U1 U2 + U3 + U4
+2U 4 4/16=0.250
+ U1 + U2 U3 + U4
+ U1 + U2 + U3 U4
U1 U2 + U3 + U4
+ U1 + U2 U3 U4
+ U1 U2 + U3 U4
0 6 6/16=0.375
U1 + U2 U3 + U4
U1 + U2 + U3 U4
+ U1 U2 U3 + U4
+ U1 U2 U3 U4
U1 + U2 U3 U4
2U 4 4/16=0.250
U1 U2 + U3 U4
U1 U2 U3 + U4
U1 U2 U3 U4 4U 1 1/16=0.0625

24
a b c
Fig. 6-2 Frequency distribution for measurements containing (a) 4 random
uncertainties; (b) 10 random uncertainties and (c) a very large number of random
uncertainties.
6A-2 Distribution of experimental data
Table 6-2 Replicate Data on the Calibration of a 10-mL Pipet
Trial mL Trial mL Trial mL Trial mL Trial mL Trial mL
1 9.988 10 9.990 19 9.980 28 9.991 37 9.988 46 9.978
2 9.973 11 9.980 20 9.994 29 9.981 38 9.971 47 9.983
3 9.986 12 9.989 21 9.992 30 9.969 39 9.986 48 9.980
4 9.980 13 9.978 22 9.984 31 9.985 40 9.978 49 9.983
5 9.975 14 9.971 23 9.981 32 9.977 41 9.986 50 9.979
6 9.982 15 9.982 24 9.987 33 9.976 42 9.982
7 9.986 16 9.983 25 9.978 34 9.983 43 9.977
8 9.982 17 9.988 26 9.983 35 9.976 44 9.977
9 9.981 18 9.975 27 9.982 36 9.990 45 9.986
Mean volume = 9.982 mL Median volume = 9.982 mL
Spread = 0.025 mL Standard deviation = 0.0056 mL

Table 6-3 Frequency Distribution of Data from Table 6-2

Volume Number % in
Range, mL in Range Range
9.969~9.971 3 6
9.972~9.974 1 2
9.975~9.977 7 14
9.978~9.980 9 18
9.981~9.983 13 26
9.984~9.986 7 14
9.987~9.989 5 10
9.990~9.992 4 8
9.993~9.995 1 2
Fig. 6-3

25
6B Statistical Treatment of Random Error
6B-1 Samples and Populations
6B-2 Properties of Gaussian Curves

a
b
Fig. 6-4 Normal error curves. The standard deviation for curve B is twice that for
curve A; that is, B = 2A. (a) The abscissa is the deviation from the mean in the units
of measurement. (b) The abscissa is the deviation from the mean in units of . Thus,
the two curves A and B are identical here.

Population Mean and sample Mean x

Population standard Deviation ()
N
( xi ) 2 (x )
z=
= i =1
N
Areas under a Gaussian Curve
1 68.3%
2 95.4%
3 99.7%

Sample standard deviation (s)

N

N N N
( xi ) 2
( xi x) 2 (d i ) 2 x 2
i i =1

N
s= i =1
= i =1
= i =1

N 1 N 1 N 1

26
 s
Standard Error of the Mean: sm =
N
N N
2
Variance (s )
( xi x) 2 (di ) 2
s 2 = i =1 = i =1
N 1 N 1

Coefficient of Variation (CV), s

CV = 100% (% RSD)
Relative Standard Deviation (RSD) x

Spread or Range (w)

Ex. 6-1 The following results were obtained in the replicate determination of the lead
content of a blood sample: 0.752, 0.756, 0.752, 0.751 and 0.760 ppm Pb.

1. Calculate the mean and the standard deviation of this set of data.

Sample xi xi2
x=
xi = 3.771 = 0.7542
1 0.752 0.565504 N 5
2 0.756 0.571536 0.754 ppb Pb
3 0.752 0.565504
4 0.751 0.564001 ( xi ) 2 (3.771) 2 14.220441
= =
5 0.760 0.577600 N 5 5
xi = 3.771 xi = 2.844145
2
= 2.8440882

2.844145 2.8440882 0.0000568

s= = = 0.00377 0.004 ppm
5 1 4

2. Calculate the (a) variance, (b) relative standard deviation in parts per thousand,
(c) coefficient of variation and (d) spread.
(a) s2 = (0.0038)2 = 1.4 10-5
(b) RSD = 0.0038/0.754 1000 ppt = 5.0 ppt
(c) CV = 0.0038/0.754 100 % = 0.50 %
(d) w = 0.760 0.751 = 0.009 ppm Pb

27
Ex. 6-2 Glucose levels are routinely monitored in patients suffering from diabetes.
The glucose concentrations in a patient with mildly elevated glucose levels were
determined in different months by a spectrophotometric analytical method. The
patient was placed on a low-sugar diet to reduce the glucose levels. The following
results were obtained during a study to determine the effectiveness of the diet.
Calculate a pooled estimate of the standard deviation for the method.

Mean Sum of Squares

Time Glucose concentration, mg/L Glucose of Deviation S. D.
mg/L from Mean
Month 1 1108, 1122, 1075, 1099, 1115, 1083, 1100.3 1687.43 16.8
1100
Month 2 992, 975, 1022, 1001, 991 996.2 1182.80 17.2
Month 3 788, 805, 779, 822, 800 798.8 1086.80 16.5
Month 4 799, 745, 750, 774, 777, 800, 758 771.9 2950.86 22.2
Total No. = 24, Total sum of squares = 6907.89

Sum of squares of Month 1 = (1108-1100.3)2 + (1122-1100.3)2 + (1075-1100.3)2 +

(1099-1100.3)2 + (1115-1100.3)2 + (1083-1100.3)2 +(1108-1100.3)2 =1687.43

6907.89
pooled standard deviation: s pooled = = 18.58 19 mg/L
24 4

6C Standard Deviation of Calculated Results

Table 6-4 Error Propagation in Arithmetic Calculation
Type of Calculation Example Standard Deviation of y
+ or y=a+bc s y = sa2 + sb2 + sc2
sy s s s
or y = a b/c = ( a )2 + ( b )2 + ( c )2
y a b c
sy sa
Exponentiation y = ax = x( )
y a
s
Logarithm y = log10 a s y = 0.434 a
a
sy
Antilogarithm y = antilog10 a = 2.303 s a
y

28
6C-1 Standard Deviation of a Sum or Differences
*y = a ( Sa) + b ( Sb) - c ( Sc) s y = sa2 + sb2 + sc2
+ 0.02 + 0.03 + 0.05 = + 0.10
+ 0.50 ( 0.02)
+ 0.02 + 0.03 - 0.05 = - 0.02 - 0.03 + 0.05 = 0
+ 4.10 ( 0.03)
- 1.97 ( 0.05) - 0.02 - 0.03 - 0.05 = - 0.10
2.63 s y = (0.02) 2 + (0.03) 2 + (0.05) 2 = 0.06
the sum should be reported as 2.63 ( 0.06)

6C-2 Standard Deviation of a Product or quotient

4.10( 0.02) 0.0050( 0.0001)
*y = (a b)/c ; = 0.010406 ( ?)
1.97( 0.04)

sy s s s 0.02 2 0.0001 2 0.04 2

= ( a )2 + ( b )2 + ( c )2 = ( ) +( ) +( )
y a b c 4.10 0.005 1.97

= (0.0049) 2 + (0.0200) 2 + (0.0203) 2 = 0.0289

sy = y ( 0.0289) = 0.0104 ( 0.0289) = 0.000301

The answer is 0.0104 ( 0.0003)

Ex 6-4. Calculate the standard deviation of the result of

[14.3(0.2) 11.6(0.2)] 0.050(0.001)
= 1.725( ?) 10 6
[820(10) + 1030(5)] 42.3(0.4)

14.3 - 11.6 = a , sa = (0.2) 2 + (0.2) 2 = 0.283

820 + 1030 = b, sb = (10) 2 + (5) 2 = 11.2

2.7(0.283) 0.050(0.001)
= 1.725 10 6
1850(11.2)] 42.3(0.4)
sy 0.283 2 0.001 2 11.2 2 0.4 2
= ( ) +( ) +( ) +( ) = 0.107
y 2.7 0.05 1850 42.3
Sy = y 0.107 = 1.725 10-6 ( 0.107) = 0.185 10-6
round the answer to 1.7( 0.2) 10-6

29
6C-3 Standard Deviations in Exponential Calculations
Sy S
y = ax = x a
y a
Ex. 6-5 The standard deviation in measuring the diameter d of a sphere is 0.02 cm.
What is the standard deviation in the calculated volume V of the sphere if d
= 2.15 cm?
3 3
4 4 d 4 2.15
V = r 3 = = = 5.20 cm
3

3 3 2 3 2
SV S 0.02
= 3 d = 3 = 0.0279 , SV = 5.20 0.0279 = 0.145, V= 5.2 ( 0.1) cm3
V d 2.15

Ex. 6-6 The solubility product Ksp for the silver salt AgX is 4.0 ( 0.4) 10-8. The
molar solubility of AgX in water is
Solubility = (Ksp)1/2 =(4.0 10-8) 1/2 = 2.0 10-4 M
What is the uncertainty in the calculated solubility of AgX in water?
y = solubility, a = Ksp, x =
S a 0.4 10 -8 Sy 1 0.4
= , = = 0.05, Sy = 2.0 10-4 0.05 = 0.1 10-4
a 4.0 10 -8 y 2 4.0
Solubility = 2.0 ( 0.1) 10-4 M

6C-4 Standard Deviations of Logarithms and Antilogarithms

s
Logarithm y = log10 a s y = 0.434 a
a
sy
Antilogarithm y = antilog10 a = 2.303 s a
y
Ex. 6-7 Calculate the absolute standard deviations of the results of the following
calculations.
(a) y = log [2.00( 0.02) 10-4] = -3.6990 ? = -3.699 (0.004)
4
0.02 10
s y = 0.434 = 0.004
2.00 10 4
(b) y = antilog [1.200( 0.003)] = 15.849 ? = 15.8 0.1
sy
= 2.303 0.003 = 0.0069, Sy = 0.0069 15.849 = 0.11
y

(c) y = antilog [45.4( 0.3)] = 2.5119 1045 ? = 3 ( 2) 1045

sy
= 2.303 0.3 = 0.69, Sy = 0.69 2.5119 1045 = 1.7 1045
y
30
6D Reporting Computed Data
6D-1 Significant Figures
all of the certain digitals + the first uncertain digit
*Rules for significant figures

1. Disregard all initial zeros.

2. Disregard all final zeros unless they follow a decimal point.
3. All remaining digits, including zeros between nonzero digits, are significant.

*Sums and Differences

3.4 + 0.020 + 7.31 = 10.73 = 10.7

*Products and quotients

24 4.52 24 4.02
= 1.08 = 1.1 = 0.965 = 0.96
100.0 100.0

*Logarithms and Antilogarithms

1. In a log. of a no., keep as many digits to the right of the decimal point as there
are significant figures in the original no.
2. In an antilog. of a no., keep as many digits as there are digits to the right of the
decimal point in the original no.

Ex. 6-8 (1) log 4.000 10-5 = - 4.3979, (2) antilog 12.5 = 3 1012
(3) log 9.57 104 = 4.981

Ex. 6-9 A 3.4842-g sample of a solid mixture containing benzoic acid,

C6H5COOH (122.123 g/mol), was dissolved and titrated with base to a
phenolphthalein end point. The acid consumed 41.36 mL of 0.2328 M
NaOH. Calculate the percent benzoic acid (HBz) in the sample.
41.36 mL 0.2328 mmol/mL 1 122.123 mg/mmol
% HBr = 100%
3.4842 g 1000 mg/g
= 33.749 %

(1) buret : 0.02 mL the S.D. of the volume will be

[( 0.02)2 + ( 0.02)2] = 0.028 mL

the relative uncertainty = ( 0.028)/41.36 1000 ppt = 0.68 ppt

31
 (2) analytical balance : uncertainty = 0.0001 g
0.0001/3.4842 1000 ppt = 0.029 ppt
(3) the A.U. in the molarity of the reagent soln = 0.0001
0.0001/0.2328 1000 ppt = 0.43 ppt
(4) the R.U. in the molar mass of HBz is several orders of magnitude smaller
than that of the three experimental data and is of no consequence.
(5) 100 % and 1000 mmol are exact numbers

The answer is rounded so that its relative uncertainty lies between 0.2 and 2
times the largest relative uncertainty of the input data.

2(0.68) ~ 0.2(0.68) = 1.36 ~ 0.136 ppt

33.7 relative uncertainty = (0.1/33.7) 1000 ppt = 3 ppt
33.75 relative uncertainty = (0.01/33.75) 1000 ppt = 0.3 ppt

32