You are running an ICP-MS (Inductively coupled plasma mass spectrometry), with bracketing standards of 10 ppm(mg/L) K+ and 100 ppm K+ and sample of 500 ppm, prepared 5 mL diluted to 100 mL. Shouldn’t you adjust your upper limit bracketing standard, instead of just diluting your sample? Your original sample concentration of K+ (500 ppm) should not change adjusted for the dilution factor. What am I not understanding mathematically?
As Ed said you are following the correct protocol by diluting the sample. The key reason is for doing so is to avoid a non-linear region in the calibration curve. This is why one would dilute the sample rather than make more concentrated standards. Several interesting effects happen with alkali metals at high concentrations in the flame or plasma. One of them is the concept of self-absorption.
Imagine you are monitoring the resonance line of K emission. Once you introduce lot of potassium atoms in the plasma (by using a concentrated standard) what may happen is that the ground state K atoms may absorb the light emitted by the excited potassium atoms? What is the result? You get less intensity than expected.
Try watching a sodium street lamp with a diffraction grating, or take a CD outside at night. You will see a prominent dark line in the emission spectrum of the lamp like this at 589 nm. This is the same self absorption phenomenon. There are so many Na atoms in the ground state that that they are self absorbing the light emitted by excited sodium atoms in the electrical discharge.