Skip to content

Spectroscopy_Regauge

chemicha edited this page Mar 22, 2016 · 3 revisions

Regauge dI/dV Spectroscopy Data

Regauging dI/dV data is one of the core features of SpectraFox, and one of the reasons for its initial development.

In scanning tunneling spectroscopy one is interested in the conductance of the tip-sample junction, so in the change of the current with respect to a change of the sample bias (dI/dV(Vbias)). A calculation via the numeric derivative of the current-versus-bias signal would be prone to noise. Therefore one uses a lockin amplifier to record this signal. This enhances the signal quality significantly.

The disadvantage of this technique is the lost physical dimension of the data. The output of the amplifier is just a simple voltage. If the parameters of the lockin amplifier are known, then one can recalculate the real physical units. If the parameters are unknown, SpectraFox is still able to reconstruct the real physical dimensions by fitting the lockin signal to the numeric derivative of the current.

Regauging by Lockin Parameters

If the parameters of the lockin are known, the physical dimensions of the lockin signal can be easily calculated from these parameters. Still, be carefull that the lockin signal has no offset included.

Please enter the lockin parameters on the left side of the tool. Be careful, that SpectraFox expects the root-mean-square value of the lockin modulation voltage.

Regauging to the Numeric Derivative of the Current

If the lockin parameters are not at hand, or otherwise not available, SpectraFox can recalculate the physical dimensions of the lockin signal by fitting the data to the numeric derivative of the current.

In the tool you have to select in (A) the current recorded together with the dI/dV-signal of the lockin amplifier. The current will get smoothed in (B), and then the numeric derivative is calculated. If the current signal is very noisy, the smoothing may need a larger averaging window. The dI/dV-signal of the lockin amplifier (C) is then fitted by a quick least square fit to the numerically derived current. It is assumed, that the lockin amplifier adds an offset to the data (y0 in D), and stretches the data by a linear factor (m in D). The output can and should be checked in (E). A quick check may be done in linear current-versus-voltage areas, where the order of magnitude on dividing the current by the voltage should match with the corresponding regauged value in (E).

Clone this wiki locally