Abstract

The weight percent oxygen in commercial gasoline samples has been determined by using partial least-squares (PLS) regression analysis combined with either FT-Raman, FT-IR, or dispersive near-IR spectroscopy. Calibration models were constructed with the use of 33 MTBE oxygenated commercial gasolines. The minimum standard errors of validation with the use of leave-one-out validation are 0.156, 0.188, and 0.119 wt % oxygen for FT-Raman, FT-IR, and near-IR, respectively. An independent test set of 36 MTBE oxygenated commercial gasolines was used to further validate the PLS models. The minimum standard errors of prediction for the test set are 0.155, 0.143, and 0.131 wt % oxygen for FT-Raman, FT-IR, and near-IR, respectively. The wt % oxygen in all samples ranges from 0.2 to 3.262%.

Distinguishing octane grades in gasoline using terahertz metamaterials

J. Li, Z. Tian, Y. Chen, W. Cao, and Z. Zeng
Appl. Opt. 51(16) 3258-3262 (2012)

Recent applications of FT-IR spectroscopy to polymer systems

J. L. Koenig and M. K. Antoon
Appl. Opt. 17(9) 1374-1385 (1978)

Multicomponent blood analysis by near-infrared Raman spectroscopy

Andrew J. Berger, Tae-Woong Koo, Irving Itzkan, Gary Horowitz, and Michael S. Feld
Appl. Opt. 38(13) 2916-2926 (1999)

Terahertz quantitatively distinguishing gasoline mixtures using multiparameter-combined analysis

Yi-nan Li, Zhou-mo Zeng, Jian Li, Zhen Tian, Li-jun Sun, and Nan Zhou
Appl. Opt. 52(30) 7382-7388 (2013)

Two-color-absorption sensor for time-resolved measurements of gasoline concentration and temperature

Sung Hyun Pyun, Jason M. Porter, Jay B. Jeffries, Ronald K. Hanson, Juan C. Montoya, Mark G. Allen, and Kevin R. Sholes
Appl. Opt. 48(33) 6492-6500 (2009)