Research Article

A Statistical Estimation Approach for Quantitative Concentrations of Compounds Lacking Authentic Standards/Surrogates Based on Linear Correlations between Directly Measured Detector Responses and Carbon Number of Different Functional Groups

Table 7

Results of the best projected RF for individual compound when matching the best fit equation with each of the six original VOC functional groups representing all 18 reference compounds.

OrderGrouping codeaCompoundCarbon numberActual RFProjected RFbPDc

1Aldehyde (I)dPA311,98414,217 18.6
2BA443,52039,053 10.3
3IA565,98163,888 3.17
4VA559,56363,888 7.26

5Aromatic (II)B6131,520133,598 1.58
6T7167,211163,054 2.49
7S8189,954192,510 1.35
8p-X8186,274192,510 3.35
9m-X8195,478192,510 1.52
10o-X8196,258192,510 1.91

11Carboxylic
(II + III + VI)
PPA326,26927,588 5.02
12BTA469,54660,582 12.9
13IVA597,01593,576 3.54
14VLA579,28293,576 18.0

15Ketone (I + II + III + IV)MEK448,77349,955 2.42
16MIBK6117,515120,904 2.88

17Alcohol (II + V)i-BuAl493,22389,250 4.26

18Ester (III + VI)BuAc6119,453119,773 0.27

Mean5.60
 SD    5.63

Predictive equations ( slopes and intercepts) developed for 29 arbitrary groups (codes) in Table 6 are used.
Slope: Eqn (I) = 24,836, Eqn (II) = 29,456, Eqn (II + III + VI) = 32,994, Eqn (I + II + III + IV) = 35,474, Eqn (II + V) = 25,484, and Eqn (III +VI) = 29,615.
Intercept: Eqn (I) = −60,290, Eqn (II) = −43,139, Eqn (II + III + VI) = −71,393, Eqn (I + II + III + IV)) = −91,942, Eqn (II + V) = −12,686, and Eqn (III+VI)=−57,914.
bThe best projected RFs are derived by taking the minimum PD value for each compound (out of 18) after testing against 29 linear regression equations (between the number of carbon (x-axis) and actual RF values (y-axis)).
cPercent difference (PD) = × 100/RF (Actual): here, the AA data are excluded due to the eccentricity.
dBest fit equation (Roman letter) for a given chemical group is shown in the parenthesis.