No abstract

Increasingly stringent requirements for aerospace propulsion system performance, reliability, and operability motivate quantitative connections between fuel composition, physical characteristics, and system performance. Chemically accurate assessment of aviation turbine fuels (Jet-A, JP-8, etc.) and kerosene-based rocket propellants (RP-1 and RP-2) is requisite to mature these models. Comprehensive two-dimensional gas chromatography with time-of-flight mass spectrometry (GC × GC–TOFMS) is an excellent analytical tool for measuring detailed chemical information contained in complex fuels. Additionally, multivariate data analysis methods, referred to as chemometrics, are ideally suited to relate detailed chemical information contained within the GC × GC–TOFMS data to fuel properties and performance in a predictive manner. Herein, we apply these techniques to a chemically diverse set of 74 distillate and multicomponent aerospace fuels, resulting in an improved understanding of the chemical compositional basis for physical and thermochemical behavior. Informed by GC × GC–TOFMS data, highly reliable partial least squares (PLS) models are developed and employed in the prediction of physical properties (measured separately using conventional test methods). Root-mean-square errors of cross-validation (RMSECV) were relatively low: values of 0.0450 cSt, 41.3 Btu/lbm, 0.130 mass %, and 0.0064 g/mL were obtained for viscosity, heat of combustion, hydrogen content, and density, respectively. The corresponding normalized root-mean-square errors of cross-validation (NRMSECV) were 6.01, 10.3, 8.71, and 7.12%, respectively. Investigation of the linear regression vectors (LRVs) provides valuable insight into the relationship between the chemical composition and physical properties, enabling, in principle, the model-informed selection of fuel chemical composition to achieve desired performance criteria.

An innovative form of Fisher ratio (F-ratio) analysis (FRA) is developed for use with comprehensive two-dimensional gas chromatography/time-of-flight mass spectrometry (GC × GC−TOFMS) data and applied to the investigation of the changes in the metabolome in human plasma for patients with injury to their anterior cruciate ligament (ACL). Specifically, FRA provides a supervised discovery of metabolites that express a statistically significant variance in a two-sample class comparison: patients and healthy controls. The standard F-ratio utilizes the between-class variance relative to the pooled within-class variance. Because standard FRA is adversely impacted by metabolites expressed with a large within-class variance in the patient class, "control-normalized FRA" has been developed to provide complementary information, by normalizing the between-class variance to the variance of the control class only. Thirty plasma samples from patients who recently suffered from an ACL injury, along with matched controls, were subjected to GC × GC−TOFMS analysis. Following both standard and control-normalized FRA, the concentration ratio for the top 30 "hits" in each comparison was obtained and then t-tested for statistical significance. Twenty four out of 30 metabolites plus the therapeutic agent, naproxen (24/30), passed the t-test for the control-normalized FRA, which included 8/24 unique to control-normalized FRA and 16/24 in common with the standard FRA. Likewise, standard FRA provided 21/30 metabolites passing the t-test, with 5/21 undiscovered by control-normalized FRA. The complementary information obtained by both F-ratio analyses demonstrates the general utility of the new approach for a variety of applications.

No abstract

No abstract

A new approach is presented to determine the probability of achieving a successful quantitative analysis for gas chromatography coupled with mass spectrometry (GC-MS). The proposed theory is based upon a probabilistic description of peak overlap in GC-MS separations to determine the probability of obtaining a successful quantitative analysis, which has its lower limit of chromatographic resolution R at some minimum chemometric resolution, R*; that is to say, successful quantitative analysis can be achieved when R ≥ R*. The value of R* must be experimentally determined and is dependent on the chemometric method to be applied. The approach presented makes use of the assumption that analyte peaks are independent and randomly distributed across the separation space or are at least locally random, namely, that each analyte represents an independent Bernoulli random variable, which is then used to predict the binomial probability of successful quantitative analysis. The theoretical framework is based on the chromatographic-saturation factor and chemometric-enhanced peak capacity. For a given separation, the probability of quantitative success can be improved via two pathways, a chromatographic-efficiency pathway that reduces the saturation of the sample and a chemometric pathway that reduces R* and improves the chemometric-enhanced peak capacity. This theory is demonstrated through a simulation-based study to approximate the resolution limit, R*, of multivariate curve resolution-alternating least-squares (MCR-ALS). For this study, R* was determined to be ∼0.3, and depending on the analytical expectations for the quantitative bias and the obtained mass-spectral match value, a lower value of R* ∼ 0.2 may be achievable.

No abstract