Banca de DEFESA: FELIPE TEIXEIRA DE OLIVEIRA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : FELIPE TEIXEIRA DE OLIVEIRA
DATE: 25/08/2023
TIME: 16:00
LOCAL: Campus Araxá
TITLE:

GEOMETALLURGICAL MODEL FOR PREDICTING APATITE FROM CHEMICAL ANALYSIS

KEY WORDS:

Geometallurgical model; Linear regression; Phosphatic ore flotation and Factorial design.

 

PAGES: 49
BIG AREA: Engenharias
AREA: Engenharia de Minas
SUBÁREA: Tratamento de Minérios
SPECIALTY: Métodos de Concentração e Enriquecimento de Minérios
SUMMARY:

Geometallurgy is a combination of geological and metallurgical information from the mine, including
chemical analysis, capable of predicting the behavior of an ore, aiming at maximizing
the Net Present Value (NPV), which is an important financial indicator. This optimization can
be achieved via protection of metallurgical recovery and/or via reduction of operating costs,
mainly related to the consumption of reagents. Currently, geometallurgy at the company has
been carried out by the geology team, which classifies the ore into 5 (five) typologies monitored
in hole monitoring and chemical analysis. Each typology is indexed with metallurgical
information (recovery, reagent consumption, natural mud and contents of each oxide). However,
within the same typology it is possible to observe metallurgical differences. For this reason,
a newapproach is proposed in thiswork. With a database, obtained from drill holes, which contains
information from chemical and mineralogical analyses, using the Minitab software it was
possible to obtain a linear regression model with an R2(pred) of 92.76%, capable of predicting the
percentage of phosphorus (P) from apatites (P-apatite) in each sample. This approach, unlike
the previous one which classified the ore into classes, numerically determines the P-apatite
of each ore block through chemical analysis. Next, it is checked whether the P-apatite obtained
by the model has a significant effect on the geometallurgical model via laboratory tests.
The laboratory flotation tests were carried out, in bench columns, using P-apatite as a factor,
along with other important factors for flotation tests, which are reagent dosage and pH. As a
result of these tests, P-apatite showed a significant effect both for metallurgical recovery, with
a contribution of 33.65%, and for the non-concentrated P2O5 content, with a contribution of
66.84%. The geometallurgical model idealized with information from the P-apatite presented
an R2(pred) of 94.03% for the metallurgical recovery and 62.83% for the grade. These R2 values
show the statistical similarity of this proposed new approach, since there is concrete evidence
that p-apatite is a statistically relevant factor for a future geometallurgical model.

COMMITTEE MEMBERS:
Presidente - NATAL JUNIO PIRES
Interna - MICHELLY DOS SANTOS OLIVEIRA
Interno - MARIO GUIMARAES JUNIOR
Externo ao Programa - LEANDRO HENRIQUE SANTOS