Second-order and higher-order multivariate calibration methods applied to non-multilinear data using different algorithms

We discuss and evaluate the current state of second-order and higher-order multivariate calibration methods devoted to the determination of compounds in non-multilinear data systems. We examine possible causes of multilinearity deviations:

(1)

a non-linear relationship between signal and analyte concentration;

(2)

a signal for a given sample that is non-multilinear; and,

(3)

component profiles that are not constant across the different samples.

We discuss the advantages and the limitations of the algorithms available to cope with these different situations.The review covers relevant analytical problems found in samples of environmental and biological interest, highlighting some significant examples, and evaluating the advantages and the limitations of the different algorithms available.     

Challenging applications offered by direct analysis in real time (DART) in food-quality and safety analysis

Direct analysis in real time (DART) is an ambient ionization technique undergoing rapid development. With minimal sample pre-treatment, ionization of analyte molecules outside the mass spectrometry (MS) instrument in the ordinary atmosphere is feasible. This ionization approach relies upon the fundamental principles of atmospheric pressure chemical ionization.The current review highlights and critically assesses application of DART (and some related desorption/ionization techniques) coupled to various types of MS analyzers for both target and non-target analysis of complex food matrices. Based on existing studies, DART-MS is presented as a simple, high-throughput tool for:

(i)

qualitative confirmation of chemical identity;

(ii)

metabolomic fingerprinting/profiling; and,

(iii)

quantification of low-molecular-weight food components, including some trace organic contaminants.

With regard to regulatory requirements, we mention practical aspects of DART-MS use, as well as performance characteristics that can be attained.     

Imaging in the modern age

This report of the 2011 James L. Waters Symposium at Pittcon 2011 highlights the powerful imaging technologies of electron microscopy (EM) and ion microscopy (IM). The four speakers each provided a window into a specific subset of the field:

•

David Bell described the history, development, application, and commercialization of transmission EM (TEM) and scanning TEM (STEM);

•

David Martin presented the challenges and methodologies of imaging ordered polymers and biomaterials with TEM;

•

Joseph Michael explained the history of the commercialization of scanning EM (SEM) and its modern applications; and,

•

David Joy, who submitted his talk in absentia, provided a history of EM and a summary of the advantages of IM versus EM.

     

Detecting “bad” regression models: multicriteria fitness functions in regression analysis

Regression models with good fitting but no predictive ability are sometimes chance correlations and often show some pathological features such as multicollinearity, overfitting, and inclusion of noisy/spurious variables. This problem is well known and of the utmost importance. The present paper proposes some criteria that are to be fulfilled as conditions for model acceptability, the aim being to recognize linear regression models with pathology. These criteria have been thought of in order to face the following problems:

•

model instability due to outliers and influential objects;

•

predictor multicollinearity;

•

redundancy in explanatory variables;

•

overfitting due to chance factors.

A multicriteria fitness function based on the maximization of the Q² statistics under a set of tests is proposed here. This new fitness function can also be used in model searching by variable selection approaches in order to obtain a final optimal population of models. Computations on the Selwood data set are reported to illustrate the use of this multicriteria fitness function in model searching.     

The caesium oxygen interactions in the crystalline solids

Through the study of the Cs-O bonds registered in the literature one observes that:

(i)

the bond lengths range from 2.46 to 3.60 Å;

(ii)

the preferential coordination numbers adopted by the caesium ions are 8, 9 and 10 but values from 1 to 12 also exist;

(iii)

the average bond lengths increase with the coordination (CN) of the caesium ions with the following values: 2.714 Å (CN=1), 2.98 Å (CN=2), 3.057 Å (CN=3), 3.104 Å (CN=4), 3.149 Å (CN=5), 3.188 Å (CN=6), 3.224 Å (CN=7), 3.245 Å (CN=8), 3.261 Å (CN=9), 3.269 Å (CN=10), 3.293 Å (CN=11) and 3.323 Å (CN=12).

A new R_ij=2.469 constant is determined with all the caesium coordination polyhedra to compute electrostatic bond valence sums.The U_eq values of caesium in crystal structure are about 0.03 and generally less than 0.06 similar to those of the oxygen atoms and often bigger.     

Synthesis of 5,6-dihydrobenz[c]acridines: a comparative study

5,6-Dihydrobenz[c]acridines were synthesized by the reaction of 1-chloro-3,4-dihydro-2-naphthaldehyde with aromatic amines under three different conditions:

a.

Thermolysis of 1-chlorovinyl-(N-aryl)imines prepared from 1-chloro-3,4-dihydro-2-naphthaldehyde.

b.

Acid catalyzed cyclization of 1-(N-aryl)amino-3,4-dihydro-2-naphthaldehydes.

c.

Thermolysis of N-arylenaminoimine hydrochlorides derived from 1-chloro-3,4-dihydro-2-naphthaldehyde in DMF medium.

All the three approaches exclusively yielded only 5,6-dihydrobenz[c]acridines and not the isomeric 7,8-dihydrobenzo[k]phenanthridines.The structures of these products have been unambiguously established by detailed NMR spectral study and by independent synthesis as well as by single crystal XRD study.     

NdHO, a novel oxyhydride

A new metal oxyhydride; neodymium oxyhydride, NdHO, has been synthesized from a reactant mixture of metal hydride (CaH₂ or NdH₃) and neodymium oxide (Nd₂O₃). The unit cell dimensions decrease smoothly in the series from LaHO, CeHO, PrHO to NdHO, in line with the lanthanide contraction. The crystal structure of NdHO is described on the basis of Rietveld refinement on neutron powder diffraction data:

Space group: P4/nmm (no. 129, D_4h⁷).

Axis lengths: a=7.8480(5) Å, c=5.5601(8) Å.

Volume: V=342.46(6) Å³.

The tetragonal structure is derived from the fluorite structure, showing complete ordering of hydride and oxide ions over the anion sublatttice. The formation of NdHO was further substantiated by Raman spectroscopy.     

Molecular weight influence on viscosimetric parameters of poly(4-vinylpyridine) polymers

This work describes the effect of the molecular weight on the viscosimetric parameters of poly(4-vinylpyridine) (P4VP) polymers in ethanolic solution. Numerous studies concerning this question have been reported in very separate intervals of molecular weight. We have observed a discordance (discontinuity) in the variation of the intrinsic viscosity as a function of the molecular weight of these polymers ([η]=f(M_w)). In order to establish a general relationship between viscosimetric parameters and M_w, we have considered 10 P4VP samples in a wide interval of molecular weights: 0.75×10⁴ to 153×10⁴. These results have been compared and completed with that of the literature. We have observed that:

(i)

All viscosimetric parameters (intrinsic viscosity [η], Huggins constant k_H, second virial coefficient, viscosimetric expansion coefficient α_η, and critical concentration) change according to a continuous function without a break.

(ii)

The lower is the molecular weight of P4VP; the higher are the variations of the expansion coefficient and the interaction effects.

(iii)

The variation of the intrinsic viscosity versus the molecular weight follows a unique relation in the whole M_w range. In fact, the Berkowitz equation (1), described for a limited range of relatively high M_w (10⁵ to 18.5×10⁵) is extended for all M_w interval values.

(iv)

Empiric laws for [η], k_H, A₂ and C^* and variations as a function of molecular weight were proposed for the P4VP in ethanol.