Natalia Regina Souza Araujo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Natalia Regina Souza Araujo has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Rita de Cássia de Oliveira Sebastião
Universidade Federal de Minas Gerais
7 shared publications - Augusto César Teixeira Malaquias
Federal University of Minas Gerais - UFMG
3 shared publications - José Guilherme Coelho Baêta
Graduate Program in Mechanical Engineering, Universidade Federal de Minas Gerais (PPGMEC-UFMG), Belo Horizonte CEP 31270-901, MG, Brazil
3 shared publications - Oscar Ricardo Sandoval
Graduate Program in Mechanical Engineering, Universidade Federal de Minas Gerais, Belo Horizonte 31270-901, Brazil
3 shared publications - Fabrício José Pacheco Pujatti
Department of Mechanical Engineering, Federal University of Minas Gerais, Minas Gerais, Brazil
2 shared publications - Lucimar Venâncio Amaral
Programa de Pós-Graduação em Engenharia Mecânica, Universidade Federal de Minas Gerais, Av. Antônio Carlos 6627, Belo Horizonte CEP 31270-901, MG, Brazil
2 shared publications
Latest Publications
- Open Access
- 29 September 2026
- Published by Universidade Federal do Parana in Extensão em Foco
- Vol. 1 (39) , 1-20
Impact Assessment of University Extension Projects
This study proposes and validates a multidimensional framework for evaluating the impact of university extension (third mission) projects, addressing limitations in existing models that prioritize operational outputs and institutional metrics. The objective is to develop a structured assessment tool capable of capturing pedagogical value, community relevance, and long-term educational outcomes in extension initiatives. The methodology combined a critical analysis of two widely adopted frameworks (IBEU and Carrión et al.) with a Delphi process involving extension specialists to define and refine 21 strategic indicators organized across seven analytical dimensions. The framework was empirically applied to the 1000 Future Scientists (1000FC) program in Brazil. The results indicate that the model enables the assessment of scope, efficiency, pedagogical impact, community engagement, and innovation dimensions through the integration of quantitative and qualitative measures. The study concludes that the proposed framework provides methodological support for planning and evaluating university extension projects, contributing to greater coherence between institutional missions and social demands.
- 29 September 2026
- Published by Universidade Federal do Parana in Extensão em Foco
- Vol. 1 (39) , 1-20
Open AccessImpact Assessment of University Extension Projects
This study proposes and validates a multidimensional framework for evaluating the impact of university extension (third mission) projects, addressing limitations in existing models that prioritize operational outputs and institutional metrics. The objective is to develop a structured assessment tool capable of capturing pedagogical value, community relevance, and long-term educational outcomes in extension initiatives. The methodology combined a critical analysis of two widely adopted frameworks (IBEU and Carrión et al.) with a Delphi process involving extension specialists to define and refine 21 strategic indicators organized across seven analytical dimensions. The framework was empirically applied to the 1000 Future Scientists (1000FC) program in Brazil. The results indicate that the model enables the assessment of scope, efficiency, pedagogical impact, community engagement, and innovation dimensions through the integration of quantitative and qualitative measures. The study concludes that the proposed framework provides methodological support for planning and evaluating university extension projects, contributing to greater coherence between institutional missions and social demands.
- 9 September 2026
- Published by IOP Publishing in Biomedical Physics & Engineering Express
- Vol. 12 (5) , 055010
Evaluation of the physical and radiological properties of different Fricke gel formulations as water-equivalent materials and biological tissues
Fricke gel dosimeters, also known as hydrogels, have emerged as attractive tools in clinical dosimetry due to the stringent requirements of quality assurance protocols and radiation protection programs. The performance of these materials strongly depends on the characteristics of the polymeric matrices that constitute them. Their ability to mimic reference materials, such as water and biological tissue, has been widely investigated for applications as simulators in radiotherapy procedures. This study aims to evaluate the physical and radiological properties of Fricke gel formulations prepared with different polymeric matrices: porcine skin gelatin (GEL), sodium carboxymethylcellulose (CMC), guar gum (GGU), and poly(vinyl alcohol) (PVA). Key-parameters, including the linear and mass attenuation coefficients ( μ and μ / ρ ), the mass energy-absorption ( μ en / ρ ), the electronic stopping power (Stp e ), and the effective atomic number ( Z eff ), were determined using both experimental and computational methods. The physical investigations included the evaluation of thermal decomposition, determined by thermogravimetric analysis and kinetic analysis, incorporating isoconversional models based on multilayer perceptron (MLP) neural networks. The results obtained experimentally and through computational tools such as Phy-X, XCOM, TOPAS, EGSnrc, and ESTAR, for the different Fricke gel formulations, revealed that all gel formulations exhibit radiological parameters comparable to those of reference materials (water and biological tissue), in accordance with ICRU Report 44. Thermal analysis revealed distinct stability characteristics among the Fricke gel compositions. The GGU and PVA formulations exhibited greater thermal resistance compared with the GEL and CMC compositions. Kinetic analysis indicated that the average activation energy values for all formulations were below 40 kJ mol − 1 , suggesting that the thermal degradation processes initiate at relatively low temperatures. The application of MLP mathematical method proved valuable for accurately modeling the observed thermal kinetics. This study demonstrates that Fricke gel formulations (GEL, CMC, GGU, and PVA) effectively mimic the radiological response of biological tissue and water. Their performance in the field of radiotherapy is relevant for applications in three-dimensional dosimetry and radiological simulations due to their tissue-equivalent properties.
- 9 September 2026
- Published by IOP Publishing in Biomedical Physics & Engineering Express
- Vol. 12 (5) , 055010
Evaluation of the physical and radiological properties of different Fricke gel formulations as water-equivalent materials and biological tissues
Fricke gel dosimeters, also known as hydrogels, have emerged as attractive tools in clinical dosimetry due to the stringent requirements of quality assurance protocols and radiation protection programs. The performance of these materials strongly depends on the characteristics of the polymeric matrices that constitute them. Their ability to mimic reference materials, such as water and biological tissue, has been widely investigated for applications as simulators in radiotherapy procedures. This study aims to evaluate the physical and radiological properties of Fricke gel formulations prepared with different polymeric matrices: porcine skin gelatin (GEL), sodium carboxymethylcellulose (CMC), guar gum (GGU), and poly(vinyl alcohol) (PVA). Key-parameters, including the linear and mass attenuation coefficients ( μ and μ / ρ ), the mass energy-absorption ( μ en / ρ ), the electronic stopping power (Stp e ), and the effective atomic number ( Z eff ), were determined using both experimental and computational methods. The physical investigations included the evaluation of thermal decomposition, determined by thermogravimetric analysis and kinetic analysis, incorporating isoconversional models based on multilayer perceptron (MLP) neural networks. The results obtained experimentally and through computational tools such as Phy-X, XCOM, TOPAS, EGSnrc, and ESTAR, for the different Fricke gel formulations, revealed that all gel formulations exhibit radiological parameters comparable to those of reference materials (water and biological tissue), in accordance with ICRU Report 44. Thermal analysis revealed distinct stability characteristics among the Fricke gel compositions. The GGU and PVA formulations exhibited greater thermal resistance compared with the GEL and CMC compositions. Kinetic analysis indicated that the average activation energy values for all formulations were below 40 kJ mol − 1 , suggesting that the thermal degradation processes initiate at relatively low temperatures. The application of MLP mathematical method proved valuable for accurately modeling the observed thermal kinetics. This study demonstrates that Fricke gel formulations (GEL, CMC, GGU, and PVA) effectively mimic the radiological response of biological tissue and water. Their performance in the field of radiotherapy is relevant for applications in three-dimensional dosimetry and radiological simulations due to their tissue-equivalent properties.
- Open Access
- 26 May 2026
- Published by Elsevier in SSRN Electronic Journal
A Composite Index for Innovation-Oriented Programs: A Governance and Decision-Support Tool for Impact Assessment
Assessing the impact of innovation-oriented programs remains a persistent challenge, particularly when activities are multidimensional, socially embedded, and developed over extended periods. This work develops a composite index designed as a governance and decision-support tool for evaluating innovation-oriented programs. The index was constructed through a multi-stage process combining literature review, expert-based indicator refinement, interval-based normalization, and equal-weight aggregation. It is structured around seven dimensions, each composed of two indicators. Its analytical logic is demonstrated through application to a university-based innovation program, showing how heterogeneous monitoring data can be translated into dimension-level diagnostics and an overall impact score. In this case, the index generated a composite score of 0.79, classified as high impact, while also revealing internal asymmetries across dimensions. The strongest results were observed in Resources and Budget and Community Engagement dimensions, both reaching the maximum average score, followed by Quality and Institutional Support and Innovation and Technology, both with scores of 0.88. By contrast, Social Inclusion and Academic Participation dimensions showed comparatively lower scores (0.50), indicating areas requiring greater strategic attention. A sensitivity analysis using alternative weighting, aggregation, and normalization scenarios produced only modest variation in the composite score and did not alter the overall interpretive pattern. The work contributes to innovation management by offering an interpretable and adaptable tool to support impact assessment, strategic prioritization, and organizational learning in complex innovation settings.
- 26 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessA Composite Index for Innovation-Oriented Programs: A Governance and Decision-Support Tool for Impact Assessment
Assessing the impact of innovation-oriented programs remains a persistent challenge, particularly when activities are multidimensional, socially embedded, and developed over extended periods. This work develops a composite index designed as a governance and decision-support tool for evaluating innovation-oriented programs. The index was constructed through a multi-stage process combining literature review, expert-based indicator refinement, interval-based normalization, and equal-weight aggregation. It is structured around seven dimensions, each composed of two indicators. Its analytical logic is demonstrated through application to a university-based innovation program, showing how heterogeneous monitoring data can be translated into dimension-level diagnostics and an overall impact score. In this case, the index generated a composite score of 0.79, classified as high impact, while also revealing internal asymmetries across dimensions. The strongest results were observed in Resources and Budget and Community Engagement dimensions, both reaching the maximum average score, followed by Quality and Institutional Support and Innovation and Technology, both with scores of 0.88. By contrast, Social Inclusion and Academic Participation dimensions showed comparatively lower scores (0.50), indicating areas requiring greater strategic attention. A sensitivity analysis using alternative weighting, aggregation, and normalization scenarios produced only modest variation in the composite score and did not alter the overall interpretive pattern. The work contributes to innovation management by offering an interpretable and adaptable tool to support impact assessment, strategic prioritization, and organizational learning in complex innovation settings.
- Open Access
- 13 April 2026
- Published by Elsevier in SSRN Electronic Journal
Understanding Ethanol Combustion by a Macro Kinetic Modelling in Spark Ignition Engines using Neural Network and Mass Fraction Burned data
This study presents a kinetic description of hydrous ethanol combustion in a spark-ignition single-cylinder research engine using experimental mass fraction burned (MFB) data and artificial neural networks (ANN). In-cylinder pressure measurements were obtained at indicated mean effective pressure (IMEP = 7 bar) and at three engine speeds (1900, 2200, and 3000 rpm). The normalized MFB curves were treated as a global conversion and analyzed through a nonlinear isoconversional method combined with ANN to determine the kinetic triplet as the apparent activation energy, pre-exponential factor, and combined mechanism, without relying on detailed chemical kinetic reactions. The apparent activation energy increased along conversion degree, ranging from approximately [[EQUATION]]at the onset of combustion to 24 [[EQUATION]]at about 70% conversion, followed by a slight decrease in the final stages until 16.5 [[EQUATION]], revealing a multistage process divided into three kinetic events associated with fuel activation, intermediate oxidation, and radical-driven reactions. The ANN results indicated that combustion is governed by a combination of Avrami–Erofeev and first-order models, with diffusion effects in intermediate stages. The proposed methodology was compared with the Wiebe model and validated against experimental MFB and heat release rate data, including operating conditions at engine speeds not used during the ANN training process. This approach provides a complementary and physically interpretable framework to correlate ethanol combustion stages directly with engine experimental data.
- 13 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessUnderstanding Ethanol Combustion by a Macro Kinetic Modelling in Spark Ignition Engines using Neural Network and Mass Fraction Burned data
This study presents a kinetic description of hydrous ethanol combustion in a spark-ignition single-cylinder research engine using experimental mass fraction burned (MFB) data and artificial neural networks (ANN). In-cylinder pressure measurements were obtained at indicated mean effective pressure (IMEP = 7 bar) and at three engine speeds (1900, 2200, and 3000 rpm). The normalized MFB curves were treated as a global conversion and analyzed through a nonlinear isoconversional method combined with ANN to determine the kinetic triplet as the apparent activation energy, pre-exponential factor, and combined mechanism, without relying on detailed chemical kinetic reactions. The apparent activation energy increased along conversion degree, ranging from approximately [[EQUATION]]at the onset of combustion to 24 [[EQUATION]]at about 70% conversion, followed by a slight decrease in the final stages until 16.5 [[EQUATION]], revealing a multistage process divided into three kinetic events associated with fuel activation, intermediate oxidation, and radical-driven reactions. The ANN results indicated that combustion is governed by a combination of Avrami–Erofeev and first-order models, with diffusion effects in intermediate stages. The proposed methodology was compared with the Wiebe model and validated against experimental MFB and heat release rate data, including operating conditions at engine speeds not used during the ANN training process. This approach provides a complementary and physically interpretable framework to correlate ethanol combustion stages directly with engine experimental data.
- 1 January 2026
- Published by Sociedade Brasileira de Quimica (SBQ) in Journal of the Brazilian Chemical Society
Combustion Kinetics of Gasoline in a Spark-Ignition Engine Using an ANN-Based Analysis of Mass Fraction Burned Data: Effect of a Commercial Nanoparticle Additive
Improving the efficiency of gasoline-fueled engines remains important for reducing fuel consumption during the energy transition. However, the kinetic effects of nanoparticle-based fuel additives under actual engine conditions remain insufficiently understood. This study investigated the influence of a commercial nanoparticle additive on the combustion kinetics and performance of Brazilian commercial E27 gasoline. Experiments were conducted in a single-cylinder research engine at a constant indicated mean effective pressure of 7 bar and engine speeds of 1900, 2200, and 3000 rpm. Experimental mass fraction burned profiles were analyzed using a previously validated artificial neural network (ANN) based isoconversional methodology to determine the complete apparent kinetic triplet, activation energy, frequency factor, and relative contributions of kinetic models, throughout combustion. Unlike conventional black-box ANN applications, the proposed framework provides a physically interpretable description of combustion kinetics directly from engine data. An objective kinetic-compensation and segmented-regression analysis identified two apparent kinetic events, with the transition shifting from a conversion degree of 0.40 for commercial gasoline to 0.65 for gasoline containing the additive. Nanoparticle addition increased the duration of the first event and reduced the second event. The additive also redistributed the contributions of the apparent kinetic models, demonstrating a modification of the combustion pathway. These kinetic changes were accompanied by a 4.1-5.3% increase in combustion efficiency and a 2.5-5.1% reduction in indicated specific fuel consumption. The proposed methodology therefore advances ANN-based combustion analysis from macroscopic prediction toward physically interpretable kinetic assessment.
- 1 January 2026
- Published by Sociedade Brasileira de Quimica (SBQ) in Journal of the Brazilian Chemical Society
Combustion Kinetics of Gasoline in a Spark-Ignition Engine Using an ANN-Based Analysis of Mass Fraction Burned Data: Effect of a Commercial Nanoparticle Additive
Improving the efficiency of gasoline-fueled engines remains important for reducing fuel consumption during the energy transition. However, the kinetic effects of nanoparticle-based fuel additives under actual engine conditions remain insufficiently understood. This study investigated the influence of a commercial nanoparticle additive on the combustion kinetics and performance of Brazilian commercial E27 gasoline. Experiments were conducted in a single-cylinder research engine at a constant indicated mean effective pressure of 7 bar and engine speeds of 1900, 2200, and 3000 rpm. Experimental mass fraction burned profiles were analyzed using a previously validated artificial neural network (ANN) based isoconversional methodology to determine the complete apparent kinetic triplet, activation energy, frequency factor, and relative contributions of kinetic models, throughout combustion. Unlike conventional black-box ANN applications, the proposed framework provides a physically interpretable description of combustion kinetics directly from engine data. An objective kinetic-compensation and segmented-regression analysis identified two apparent kinetic events, with the transition shifting from a conversion degree of 0.40 for commercial gasoline to 0.65 for gasoline containing the additive. Nanoparticle addition increased the duration of the first event and reduced the second event. The additive also redistributed the contributions of the apparent kinetic models, demonstrating a modification of the combustion pathway. These kinetic changes were accompanied by a 4.1-5.3% increase in combustion efficiency and a 2.5-5.1% reduction in indicated specific fuel consumption. The proposed methodology therefore advances ANN-based combustion analysis from macroscopic prediction toward physically interpretable kinetic assessment.
- 1 January 2026
- Published by Sociedade Brasileira de Quimica (SBQ) in Journal of the Brazilian Chemical Society
Stability Assessment of HKUST-1 Metal-Organic Framework with Gold Nanoparticles
HKUST-1 is a metal-organic framework (MOF) with the formula [Cu3(BTC)2(H2O)3]n (BTC: benzene-1,3,5-tricarboxylate), which exhibits relevant properties due to its high porosity. However, HKUST-1 is highly sensitive to water, collapsing in aqueous media or under highhumidity conditions. The modification of this MOF with metal nanostructures has seen significant growth in recent years, with gold nanoparticles (AuNPs) among the most widely used. This article reports the association between gold nanoparticles (AuNPs) and MOF HKUST-1 and investigates the influence of water from AuNPs on the degradation of HKUST-1, in addition to proposing a modification that maintains the stable structure, based on the transfer of gold nanoparticles to a non-aqueous medium (N,N-dimethylformamide) before the modification of HKUST-1. An analysis of the influence of the ethanol washing process on gold-modified HKUST-1 samples was also performed.
- 1 January 2026
- Published by Sociedade Brasileira de Quimica (SBQ) in Journal of the Brazilian Chemical Society
Stability Assessment of HKUST-1 Metal-Organic Framework with Gold Nanoparticles
HKUST-1 is a metal-organic framework (MOF) with the formula [Cu3(BTC)2(H2O)3]n (BTC: benzene-1,3,5-tricarboxylate), which exhibits relevant properties due to its high porosity. However, HKUST-1 is highly sensitive to water, collapsing in aqueous media or under highhumidity conditions. The modification of this MOF with metal nanostructures has seen significant growth in recent years, with gold nanoparticles (AuNPs) among the most widely used. This article reports the association between gold nanoparticles (AuNPs) and MOF HKUST-1 and investigates the influence of water from AuNPs on the degradation of HKUST-1, in addition to proposing a modification that maintains the stable structure, based on the transfer of gold nanoparticles to a non-aqueous medium (N,N-dimethylformamide) before the modification of HKUST-1. An analysis of the influence of the ethanol washing process on gold-modified HKUST-1 samples was also performed.
- 1 January 2025
- Published by Sociedade Brasileira de Quimica (SBQ) in Journal of the Brazilian Chemical Society
Determination of Thermal Process Kinetics by Artificial Neural Networks: A Review Considering Isothermal and Non-Isothermal TG and DSC Data
The integration of artificial neural networks (ANNs) with thermal analysis techniques, such as thermogravimetry (TG) and differential scanning calorimetry (DSC), presents great potential to accurately determine kinetics of thermal processes and thermal behavior. This review examines the application of ANNs, including Multilayer Perceptron (MLP) and Hopfield Neural networks (HNN), to analyze both isothermal and non-isothermal TG and DSC data. These networks provide robust tools for predicting thermal behavior, determining kinetic parameters, and modeling complex reaction mechanisms, significantly improving accuracy when compared to traditional methods. The review highlights advances in ANN-based methodologies, including the ability to integrate multiple kinetic models and manage noisy experimental data, thereby offering new insights into thermal decomposition and reaction kinetics. Moreover, the use of HNN in distributed activation energy models (DAEM) is explored, indicating their potential in understanding heterogeneous systems. This study emphasizes the transformative potential of ANNs for improving kinetic studies, suggesting future directions to explore machine learning in material characterization and thermal process optimization.
- 1 January 2025
- Published by Sociedade Brasileira de Quimica (SBQ) in Journal of the Brazilian Chemical Society
Determination of Thermal Process Kinetics by Artificial Neural Networks: A Review Considering Isothermal and Non-Isothermal TG and DSC Data
The integration of artificial neural networks (ANNs) with thermal analysis techniques, such as thermogravimetry (TG) and differential scanning calorimetry (DSC), presents great potential to accurately determine kinetics of thermal processes and thermal behavior. This review examines the application of ANNs, including Multilayer Perceptron (MLP) and Hopfield Neural networks (HNN), to analyze both isothermal and non-isothermal TG and DSC data. These networks provide robust tools for predicting thermal behavior, determining kinetic parameters, and modeling complex reaction mechanisms, significantly improving accuracy when compared to traditional methods. The review highlights advances in ANN-based methodologies, including the ability to integrate multiple kinetic models and manage noisy experimental data, thereby offering new insights into thermal decomposition and reaction kinetics. Moreover, the use of HNN in distributed activation energy models (DAEM) is explored, indicating their potential in understanding heterogeneous systems. This study emphasizes the transformative potential of ANNs for improving kinetic studies, suggesting future directions to explore machine learning in material characterization and thermal process optimization.