This article by D.W. Lachenmier et al. is an important addition to a previous research by Duarte et al. that describes beer characterization using NMR spectroscopy. The researchers of this article have successfully improved Duarte's research by adding enough extra samples to be able to analyze quality control of beer.

By thoroughly reading the Abstract, it is easy to comprehend the conditions in which this research is developed. Readers can find quick mentions about the methods and techniques that were used to study quality control in beer samples in this analysis. However, it doesn't seem quite clear where this research aims to, since the authors don't include one specific objective. It is expected to acknowledge that the authors improved Duarte et al.'s NMR spectroscopy technique by increasing the amount of samples and quality control study.

The introduction mentions several analytic methods that are commonly used to study the quality of beer, claiming these consume a lot of time and money to be performed. However, considering beer's composition includes several high molecular weight compounds and also that NMR spectroscopy isn't a quite reliable method because of its sensitivity, multivariate statistical analysis must be applied on the research. The introduction also includes examples about other articles that have also studied this concept, like NMR spectra applied for apple and grape juice. The authors mention the main study in which this article is based on. It is very useful that the statistical analysis methods that are used on this research are very well explained, considering the presentation of data must be well understood in isolation. A solution is not provided because this research doesn't expect to solve any matter, but to study and expand the knowledge on compounds that are identified and allow to describe the quality of a beer sample. Based on a previous similar research, its purpose and singular objectives are stated.

This research provides both operational and conceptual analysis on the study of the quality of beer from NMR spectroscopy, since it applies chemical methods and develops multivariate statistical analysis. NMR spectroscopy is a previously established technique in the chemical and physical field of study, so there aren't extra steps and it isn't combined with any other method, but supplemented with statistical analysis that allows to an alternative approach for interpretation of the spectra.

As it is previously mentioned, this research improves the Duarte et al. study so it analyzes eighty samples, compared to the seventeen in the previous research that weren't enough to make comparisons. The sample set has two main kinds of fermented beer styles that represent a large population.

The Beer Samples division in the methodology section is an accurate demonstration on how many samples were obtained, who or which organization provided them and their very well complemented set. Samples were initially analyzed by organoleptic and methodological examinations, determining some relevant physical properties like density and OG, since both top-fermented and bottom-fermented kinds of samples were provided. Nevertheless, there is no flow chart that allows the reader to understand and visually classify the samples into the several aspects and properties they have. It might be useful if the authors provided a table with information such as common compounds' content based on every sample, division by groups like samples based on bottom- and top-fermented, and division by aromatic, aliphatic and mid-field region shown in the spectra.

In regard to the process and NMR measurements, samples were correctly prepared and measured according to Table 1, which concept is supplemented by Figure 1 by showing how the flow injection technique works. As Table 2 shows, there were a few samples excluded from the research according to their ppm range.

The Chemometrics division, however, mentions two data tables that are commonly studied through the PLS statistical method. Even though both PCA and PLS methods are applied for this research's data analysis, these X and Y tables are nowhere shown in the article, but their concept is described in paragraph 2 of the Chemometrics division in the Materials and Methods section. This lack of graphical data distribution turns the interpretation of this analysis into confusing. This research provides enough information to perform a replicate study, from environmental conditions of the place where it is performed, to softwares and their respectives instruments that were used on the multivariate statistical analysis.

The research is expected to provide statistical analysis because there are eighty samples that are examined and their individual and collective NMR spectras show plenty of overlap regions and peaks, so statistical measures like Principal components analysis (PCA) and Partial least squares regression (PLS) are applied to collect the data and distribute it to be analyzed through scatter plots.

The Results and Discussion section provides an enhanced methodology to prepare the samples, in order to avoid some overlaps due to pH differences and to only show real differences from the compounds. It is helpful that aromatic, aliphatic and mid-field region ranges are described on the NMR spectra division. Since NMR spectra won't show qualifications on beer types, scatter plots on PCA analysis are correctly labeled and the samples are very well classified. However, Figure 3 doesn't quite provide a characterization on which beer type each sample corresponds to. Besides labeling samples that were excluded, it should indicate which type they belong to.

The discussion section, especifically, doesn't lead to alternative interpretations that might get the article a little confusing. This research applies an operational method to complete the analysis, so the authors provide a conclusion based on interpretation about the multivariate analysis developed in the previous division. It isn't quite clear if the conclusion suggests a future research based on its characteristics, but it does mention some advantages, like both quantitative and qualitative significant outcomes, of using high-resolution NMR spectroscopy combined with statistical analysis instead of just simple spectroscopy.

In summary, this article has a few minor but significant points that need to be improved for a reader's good interpretation of the results and expanding of knowledge about quality and characterization of beer.

References used in this review include the reviewed article and the article it is based on:

Iola Duarte, António Barros, Peter S. Belton, Renton Righelato, Manfred Spraul, Eberhard Humpfer, and Ana M. Gil. Journal of Agricultural and Food Chemistry 2002 50 (9), 2475-2481

Lachenmeier, D.W., Frank, W., Humpfer, E. et al. Quality control of beer using high-resolution nuclear magnetic resonance spectroscopy and multivariate analysis. Eur Food Res Technol 220, 215–221 (2005). https://doi.org/10.1007/s00217-004-1070-7