Accurate determination of extra-column band broadening using peak summation

Computational Fluid Dynamics Study of the Dispersion Caused by Capillary Misconnection in Nano-Flow Liquid Chromatography

It is well known that high-speed/high-efficiency separations in nano-flow liquid chromatography (LC) are very sensitive to the quality of the connections between the column and the rest of the instrument. In the present study, two types of connection errors (capillary misalignment and the occurrence of an inter-capillary gap) have been investigated using computational fluid dynamics. Interestingly, it has been found that large degrees of capillary misalignment (assuming an otherwise perfect contact between the capillary end-faces) can be afforded without introducing any significant dispersion over the entire range of investigated relative misalignment errors (0 ≤ ε/dcap ≤ 75%), even at the largest flow rates considered in nano-LC. On the other hand, when an inter-capillary gap is present, the dispersion very rapidly increases with the radial width Dc of this gap (extra variance ∼Dcn with n even reaching values above 4). The dependency on the gap length Lc is however much smaller. Results show that, when Dc ≤ 30 μm and Lc ≤ 200 μm, dispersion losses can be limited to the order of 1 nL2 at a flow of 1.5 μL/min, which is generally very small compared to the dispersion in the capillaries (20 μm i.d.) themselves. This result also reconfirms that zero-dead volume connectors with a sufficiently narrow bore can in theory be used without compromising peak dispersion in nano-LC, at least when the capillaries can be matched perfectly to the connector in- and outlet faces. The results are also indicative of the extra dispersion occurring inside microfluidic chips or in the connections between a microfluidic chip and the outer world.

Fundamental investigation of the dispersion caused by a change in diameter in nano liquid chromatography capillary tubing

We report on a Computational Fluid Dynamics (CFD) study of the extra dispersion caused by the change in diameter when coupling two pieces of capillary tubing with different diameter. In this first investigation into the problem, the focus is on the typical flow rates (0.25≤F≤2μL/min) and diameters (d≤40μm) used in nano-LC, considering both the case of either a doubling or halving of the diameter. The CFD simulations allow to study the problem from a fundamental point of view, i.e., under otherwise perfect conditions (perfect alignment, zero dead-volume). Flow rates, capillary diameters, diffusion coefficients and liquid viscosities have been varied over a range relevant for nano-LC (Reynolds-numbers Re ≤ 1), with also an excursion made towards high-temperature nano-LC conditions (Re ≥ 10 and more). The extra dispersion caused by the change in diameter has been quantified via a volumetric variance σ²_conn, defined in such a way that the overall dispersion across the entire capillary system can be easily reconstructed from the known analytical solutions in the individual segments. When the two capillaries are longer than their diffusion entry length, covering most of the practical cases, σ²_conn converges to a limiting value σ²_conn,∞ which varies to a close approximation with the square of flow rate. Under the investigated nano-LC conditions, the σ²_conn,∞-values are surprisingly small (e.g., on the order of 0.01 to 0.15 nL² in a 20 to 40μm connection) compared to the dispersion occurring in the remainder of the capillaries.

Deep convolutional autoencoder for the simultaneous removal of baseline noise and baseline drift in chromatograms

Enhancement of chromatograms, such as the reduction of baseline noise and baseline drift, is often essential to accurately detect and quantify analytes in a mixture. Current methods have been well studied and adopted for decades and have assisted researchers in obtaining reliable results. However, these methods rely on relatively simple statistics of the data (chromatograms) which in some cases result in significant information loss and inaccuracies. In this study, a deep one-dimensional convolutional autoencoder was developed that simultaneously removes baseline noise and baseline drift with minimal information loss, for a large number and great variety of chromatograms. To enable the autoencoder to denoise a chromatogram to be almost, or completely, noise-free, it was trained on data obtained from an implemented chromatogram simulator that generated 190.000 representative simulated chromatograms. The trained autoencoder was then tested and compared to some of the most widely used and well-established denoising methods on testing datasets of tens of thousands of simulated chromatograms; and then further tested and verified on real chromatograms. The results show that the developed autoencoder can successfully remove baseline noise and baseline drift simultaneously with minimal information loss; outperforming methods like Savitzky-Golay smoothing, Gaussian smoothing and wavelet smoothing for baseline noise reduction (root mean squared error of 1.094 mAU compared to 2.074 mAU, 2.394 mAU and 2.199 mAU) and Savitkzy-Golay smoothing combined with asymmetric least-squares or polynomial fitting for baseline noise and baseline drift reduction (root mean absolute error of 1.171 mAU compared to 3.397 mAU and 4.923 mAU). Evidence is presented that autoencoders can be utilized to enhance and correct chromatograms and consequently improve and alleviate downstream data analysis, with the drawback of needing a carefully implemented simulator, that generates realistic chromatograms, to train the autoencoder.

RpeakChrom: Novel R package for the automated characterization and optimization of column efficiency in high-performance liquid chromatography analysis

Characterization of chromatographic columns using the traditional van Deemter method is limited by the necessity of calculating extra-column variance, issue particularly relevant when modeling asymmetrical peaks eluted from monolithic columns. A novel R package that implements Parabolic Variance Modified Gaussian approach for accurate peak modeling, van Deemter equation and two alternatives approaches, based on van Deemter, has been developed to calculate the height equivalent to a theoretical plate (HETP). To assess package capabilities conventional packed reverse-phase and monolithic HPLC columns were characterized. Peaks eluted from the monolithic column showed a high value of factor asymmetry due, in part, to the contribution of extra-column factors. Such deviation can be circumvented by the two alternatives approaches implemented in the R-package. Furthermore, increased values of eddy diffusion and mass transfer kinetics terms in HETP were observed for the packed column, while accuracy was below 9% in all cases. These results showed the usefulness of the R-package for both modeling chromatographic peaks and assessing column efficiency. The RpeakChrom package could become a helpful tool for testing new stationary phases during column development and to evaluate column during its lifetime. This R tool is freely available from CRAN (https://CRAN.R-project.org/package=RpeakChrom).