Banca de DEFESA: RAIZA SILVA BEZERRA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : RAIZA SILVA BEZERRA
DATE: 31/08/2022
TIME: 09:00
LOCAL: Programa de Pós-Graduação em Engenharia Civil
TITLE:

Expansive Soils in the Brazilian Semi-Arid: a panoramic view

KEY WORDS:

Semiarid; Expansive Soil; BANDASE, Occurrence; Identification; Methods; Tests; Hydrogeomechanical.

PAGES: 120
BIG AREA: Engenharias
AREA: Engenharia Civil
SUMMARY:

The Brazilian Semiarid region covers 1262 municipalities, with an area of1,182,697 km², corresponding to more than 60% of the Northeast and more to the northern region of Minas Gerais. With a population of almost 23 million inhabitants, it is a region that, despite presenting some economic and social advances in recent decades, requires greater attention due to low economic dynamism, with social indicators below national and regional averages. It is a region that presents great susceptibility to the occurrence of expansive soils, that is, soils that present volume variation and that in periods of drought are easily recognized due to many cracks. These soils cause major problems for civil construction, and can even collapse buildings. From the consultation of the Special Soils Database - BANDASE, initiated by Ferreira (1990), geotechnical information of 74 occurrences of expansive soils in the Brazilian semiarid, found in the states of Pernambuco (26 points), Bahia (36 points) was identified, Piauí (2 points), Ceará (6 points), Paraíba (1 point) and Rio Grande do Norte (3 points) in small to large projects. Characterization data were collected and analyzed: physical, chemical, mineralogical and microstructural; test data: single and double edometrics, direct shear, SPT. The samples were classified by direct and indirect methods. Interpretive maps of susceptibility to the occurrence of expandable soils; Artificial neural networks (ANN) capable of simulating the behavior of expandable soils and soil improvement techniques were also analyzed. Samples were classified according to expansibility by the criteria of: Skempton (1953) as inactive to normal; Van Der Merve (1964) from low to high; Chen (1965), Daksanamurthy and Raman (1973), Seed, Woodword and Lundgren (1962) from low to very high; Holtz and Gibbs (1956) from inactive to very active; Priklonskij as not expandable to only 1 samples, the others were all expandable. The percentage of clay in the analyzed samples varies between 2.1% and 70%. The mineral clays identified were kaolinite, mica, illite, montmorillonite, vermiculite and chlorite. For the cation exchange capacity, all samples analyzed in this criterion were classified from high to very high. The samples showed hydrogenion potential (pH) from moderately acidic to alkaline. Only 2 samples from Pernambuco are dystrophic, the others are eutrophic. The analyzed samples are sodic and sodic according to the saturation of (100 Na+/T). The most abundant oxides are silicon (SiO2), aluminum (AL2O3) and iron (Fe2O3). According to the criterion of Jimenez Salas (1980), pathological damage to buildings varies from small cracks to demolition of the building. Soil improvement techniques were proven by additions to the expansive soil of materials: lime, rice husk ash and construction waste. Another improvement technique used was to avoid contact between the building and the ground to prevent cracks, as long as the foundation is laid on a non-expansive layer. The Artificial Neural Networks showed good performance in the identification of expansive soils through the parameters of sand and clay percentage, plasticity indices and activity. In this way, with the correct identification of the parameters of this soil, it is possible to improve and adapt it for the purposes of civil construction, which, without a doubt, will provide a better living condition for the population of the Brazilian semiarid region.

BANKING MEMBERS:
Presidente - 1066974 - ANALICE FRANCA LIMA AMORIM
Externo à Instituição - JOAQUIM TEODORO ROMAO DE OLIVEIRA - UNICAP
Externo à Instituição - SÉRGIO CARVALHO DE PAIVA - UNICAP