# Searching Optimum Self-Brazing Powder Mixtures Intended for Use in Powder Metallurgy Diamond Tools—A Statistical Approach

## TL;DR

This paper uses statistical methods to optimize self-brazing powder mixtures for diamond tools, focusing on their chemical composition and resulting hardness.

## Contribution

A statistical model is developed to predict matrix hardness based on powder composition, aiding in the design of diamond tools.

## Key findings

- Hardness is primarily influenced by phosphorous content, with nickel also having a significant effect.
- Tin bronze addition improves sinterability and reduces porosity, enhancing diamond retention.
- The statistical model achieves an 80% determination coefficient in predicting matrix hardness.

## Abstract

This paper presents a study on optimising self-brazing powder mixtures for powder metallurgy diamond tools, specifically focusing on wire saws used in cutting natural stone. The research aimed to understand the relationship between the chemical composition of powder mixtures and the hardness of the sintered matrix. The experimental process involved the use of various commercially available powders, including carbonyl iron, carbonyl nickel, atomised bronze, atomised copper, and ferrophosphorus. The samples made of different powder mixtures were compacted and sintered and then characterised by dimensional change, density, porosity, and hardness. The obtained results were statistically analysed using an analysis of variance (ANOVA) tool to create linear regression models that relate the material properties to their chemical composition. The investigated materials exhibited excellent sintering behaviour and very low porosity, which are beneficial for diamond retention. Very good sinterability of powder mixtures can be achieved by tin bronze addition, which provides a sufficient content of the liquid phase and promotes the shrinkage during sintering. Statistical analysis revealed that hardness was primarily affected by phosphorous content, with nickel having a lesser but still significant impact. The statistical model can predict the hardness of the matrix based on its chemical composition. This model, with a determination coefficient of approximately 80%, can be valuable for developing new metal matrices for diamond-impregnated tools, particularly for wire saw beads production.