Magneto-Seebeck effect in $R\mathrm{FeAsO}$ ($R=\mathrm{rare}$ earth) compounds: Probing the magnon drag scenario

We investigate the Seebeck effect in $R\mathrm{FeAsO}$ ($R=\mathrm{rare}$ earth) compounds as a function of temperature and magnetic field up to 30 T. The Seebeck curves are characterized by a broad negative bump around 50 K, which is sample dependent and strongly enhanced by the application of a magnetic field. A model for the temperature and field dependence of the magnon drag contribution to the Seebeck effect by antiferromagnetic (AFM) spin fluctuation is developed. It accounts for the magnitude and scaling properties of such bump feature in our experimental data in LaFeAsO. This analysis accounts for the apparent inconsistency of literature Seebeck effect data on these compounds and has the potential to extract precious information on the coupling between electrons and AFM spin fluctuations in these parent compound systems, with implications on the pairing mechanism of the related superconducting compounds.

Figure 1

Seebeck coefficient curves of $R\mathrm{FeAsO}$ ($R=\mathrm{Sm}$, Pr, La, Ce) polycrystals.

Figure 2

Seebeck effect values of $R\mathrm{FeAsO}$ compounds at 300 K (upper panel) and at 50 K (lower panel) collected by Refs. [2, 12, 4, 13, 14].

Figure 3

Seebeck coefficient curves of LaFeAsO samples of our data in comparison with data taken from Refs. [2] and [12]. Inset: Resistivity curves of the same samples, normalized to the room temperature value.

Figure 4

Seebeck coefficient curves of the LaFeAsO sample measured at zero and 9 T.

Figure 5

$S$ curves versus the magnetic field of the LaFeAsO sample performed at $T=30$, 45, 60, and 77 K.

Figure 6

$MS\left(T,B\right)$ as a function of the magnetic field for the longitudinal field $B_{\parallel }$ (top panel) and transverse field $B_{\perp }$(bottom panel). Note the different vertical scales of the two panels. The different colors correspond to different temperatures according to the top legend. Other parameters are the magnon gap ${\Delta }_0=90$ K and magnon velocity $v=3.5\times {10}^4$ m/s. Inset: The same data plotted as a function of $B/T$. Note that the longitudinal contribution follows the scaling laws and instead the transverse one does not. See text for details.

Figure 7

(a) $MS\left(T,B_{\parallel }\right)$ as a function of $B_{\parallel }/T$ evaluated for $T=30$ K and $T=60$ K, magnon velocity values of $v=1.4\times {10}^4\mathrm{m}/\mathrm{s}$ (square) and $4.6\times {10}^4\mathrm{m}/\mathrm{s}$ (triangle), and magnon gap values of ${\Delta }_0=90$ K (filled markers) and 70 K (empty markers). (b) $MS\left(T,B_{\parallel }\right)$ as a function of $B_{\parallel }/T$ evaluated for $g=2$, 3, and 4 at $T=30$ and 60 K, $v=4.6\times {10}^4\mathrm{m}/\mathrm{s},$ and ${\Delta }_0=90$ K.

Figure 8

Experimental $S$ curves of our LaFeAsO sample and of a LaFeAsO sample taken from Ref. [12], plotted together with the calculated diffusive contribution $S_d$ (see text). Inset: Drag contribution evaluated by subtracting $S_d$ by the experimental curve as explained in the text.

Figure 9

$M{S}^{\exp }\left(T,B\right)=\left[S\left(T,B\right)-S\left(T,0\right)\right]/S_{\mathrm{DRAG}}\left(T\right)$ extracted from the experimental $S$ curves of Fig. 5 and plotted as a function of $B/T$. Inset: $\frac{\Delta S\left(T,B\right)}{S\left(T,B=0\right)}= \left[S\left(T,B\right)-S\left(T,0\right)\right]/S\left(T,0\right)$ vs $B/T$.

Figure 10

Specific heat curves of the LaFeAsO sample measured in zero and 14 T magnetic fields.