Loading...
Publication
Exploratory analysis of pH reference materials for quality control acceptance by chemometrics algorithms
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 5.25 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
O National Institute for Standards and Technology (NIST) tem procedimentos definidos para a
aceitação dos seus Materiais de Referência Certificados (MRC) para medição de pH. Estes
procedimentos são morosos e destrutivos para a amostra. A sua substituição por métodos
quimiométricos recorrendo a técnicas de análise não destrutivas traria grandes benefícios, por um
lado em termos da rapidez e simplicidade do procedimento de aceitação de MRC.
Neste trabalho recolheram-se espectros no infravermelho próximo (NIR) e espectros Raman e
trataram-se os dados através de análise de componentes principais (PCA), a fim de realizar uma
análise qualitativa de três diferentes MRC de pH: carbonato de cálcio (CaCO3), tetraborato de
sódio decahidratado ( bórax) e hidrogeno ftalato de potássio (KHP).
O método foi testado para distinguir entre MRCs produzidos pelo NIST e outros materiais
candidatos a MRC, assim como amostras comerciais de menor pureza que serviram como
controlo negativo.
Uma coleção de 87 amostras de CaCO3, 52 amostras de bórax e 63 amostras de KHP foi
analisada por espectroscopia NIR e Raman. No caso do CaCO3, o PCA alcançou uma boa
discriminação, coerente com o método de referência NIST para aceitação de MRC de pH. No
caso do bórax, obteve-se boa discriminação entre as amostras, mas que se revelou inadequada
para controle de qualidade dos materiais candidatos. No caso do KHP, a discriminação entre
amostras foi insuficiente, mas não exclui a possibilidade de aplicar o método proposto ao
controle de qualidade de materiais candidatos visto o método de referência NIST também foi
incapaz de distinguir os controlos negativos.
A diferenciação obtida por PCA para o caso do CaCO3 foi explicada por difração de raios-X,
tendo-se verificado que os diferentes grupos observados por PCA correspondem a diferentes
polimorfos de calcite, aragonite e vaterite.
No caso do bórax, a análise termogravimétrica revelou que o bórax tende a perder parte de sua
água de cristalização ao longo do tempo de armazenamento e se transforma lentamente na forma
pentahidratada. A análise por PCA diferenciou grupos de acordo com o grau de hidratação das
amostras. As amostras de KHP não foram suficientemente separadas na análise por PCA, com exceção de
amostras com maior grau de cristalinidade, o que se explicou devido à presença de água oclusa
nos cristais de KHP.
Este estudo prova que o uso de quimiometria e sua capacidade de discriminar entre amostras
quimicamente diferentes é potencialmente uma ferramenta poderosa para garantir a identidade e
a qualidade de MRC para pH, e para simplificar o procedimento de aceitação de novos MRC.
Spectra collected with near infrared (NIR) and Raman spectroscopic methods were used along with unsupervised pattern recognition, namely, principle component analysis (PCA), to conduct a qualitative analysis of three different pH materials: calcium carbonate (CaCO3), sodium tetraborate decahydrate (borax) and potassium hydrogen phthalate (KHP). The NIR and Raman based fingerprinting with PCA were tested to distinguish between standard reference materials (SRM‘s) produced by the National Institute of Standards and Technology (NIST), and other potential candidate materials, and commercial samples of lower purity. A collection of 87 samples of CaCO3, 52 samples of borax and 63 samples of KHP were measured by NIR and Raman spectroscopy. In the case of CaCO3, PCA achieved a good discrimination coherent with the NIST reference method for acceptance of pH SRM‘s. In the case of borax, good discrimination was obtained between the samples but inadequate for quality control of candidate materials. In the case of KHP, insufficient discrimination was obtained between samples. This does not exclude the possibility of applying the proposed method to the quality control of candidate materials, because the NIST reference method was also unable to distinguish negative controls from SRMs. The PCA clusters were further explained by X-ray diffraction (XRD) in case of calcium carbonate, which revealed that the PCA discrimination is based on the phase transformation of polymorphs of calcite, aragonite and vaterite. In the case of borax, thermogravimetric analysis (TGA) revealed that borax is not a stable substance but it tends to lose some of its crystallization water and transforms slowly into the pentahydrate. This is in good agreement with the different groups distinguished on the PCA. KHP was characterized by the impurities within the samples. The dominant impurity was proven to be the occluded water. This study proves that the use of chemometrics and its ability to discriminate between chemically different samples could be a powerful tool to assure the identity and quality of the pH buffer materials and to streamline the acceptance procedure of new issues of SRM‘s.
Spectra collected with near infrared (NIR) and Raman spectroscopic methods were used along with unsupervised pattern recognition, namely, principle component analysis (PCA), to conduct a qualitative analysis of three different pH materials: calcium carbonate (CaCO3), sodium tetraborate decahydrate (borax) and potassium hydrogen phthalate (KHP). The NIR and Raman based fingerprinting with PCA were tested to distinguish between standard reference materials (SRM‘s) produced by the National Institute of Standards and Technology (NIST), and other potential candidate materials, and commercial samples of lower purity. A collection of 87 samples of CaCO3, 52 samples of borax and 63 samples of KHP were measured by NIR and Raman spectroscopy. In the case of CaCO3, PCA achieved a good discrimination coherent with the NIST reference method for acceptance of pH SRM‘s. In the case of borax, good discrimination was obtained between the samples but inadequate for quality control of candidate materials. In the case of KHP, insufficient discrimination was obtained between samples. This does not exclude the possibility of applying the proposed method to the quality control of candidate materials, because the NIST reference method was also unable to distinguish negative controls from SRMs. The PCA clusters were further explained by X-ray diffraction (XRD) in case of calcium carbonate, which revealed that the PCA discrimination is based on the phase transformation of polymorphs of calcite, aragonite and vaterite. In the case of borax, thermogravimetric analysis (TGA) revealed that borax is not a stable substance but it tends to lose some of its crystallization water and transforms slowly into the pentahydrate. This is in good agreement with the different groups distinguished on the PCA. KHP was characterized by the impurities within the samples. The dominant impurity was proven to be the occluded water. This study proves that the use of chemometrics and its ability to discriminate between chemically different samples could be a powerful tool to assure the identity and quality of the pH buffer materials and to streamline the acceptance procedure of new issues of SRM‘s.
Description
Keywords
Amostras de CaCO3 Amostras de KHP Infravermelho Termogravimétrica