[Mr_Man]

topic

what is the melting point for a metal? when something starts to melt or when it has fully melted?

[FeLiXe]

any pure substance has one distinct melting point. only with mixtures you'd have melting intervals

[Yggdrasil]

To determine the melting point of a substance you heat it very slowly and watch it closely as it melts.  You then record the temperature that it begins to melt and the temperature that it has been completely melted.  That range of temperatures that you record is the melting point.  For a pure substance the melting point can be as small as a few tenths of a degree (if you heat really slowly).  For a mixture/impure sample, the range can be fairly large (10 or more degrees).

[Mr_Man]

To determine the melting point of a substance you heat it very slowly and watch it closely as it melts.  You then record the temperature that it begins to melt and the temperature that it has been completely melted.  That range of temperatures that you record is the melting point.  For a pure substance the melting point can be as small as a few tenths of a degree (if you heat really slowly).  For a mixture/impure sample, the range can be fairly large (10 or more degrees).

i dont get the bit in bold. how can a 'range' be a 'point'

and would it be best to heat a metal up till it melts, record, then record to the temp that it freezes at to get the most accurate results?

[Mitch]

Most materials actually have a range of temperatures where it melts instead of a single point. This is usually due to impurities in the material.

[allanf]

and would it be best to heat a metal up till it melts, record, then record to the temp that it freezes at to get the most accurate results?

This has difficulties of its own, in that you can end up measuring a hysteresis.  The temperature at which the solid melts can be different from the temperature at which the liquid freezes due to super-cooling of the liquid or super-heating of the solid.