Impurity effects in niobium triselenide: A transport and atomic force microscope study

AFM and transport measurements have been employed to study the effects of elemental impurity doping in Niobium Triselenide (NbSe3). These studies confirm that a variety of NbSe3's properties is sensitive to the type and concentration of impurities. The impurities used to dope NbSe 3 were Cr, Ir, Mn, Pd, Re, Ru, Si, V, and W.;AFM scans well above the CDW transitions have revealed modulations in the electronic structure with wavelengths that are not integral multiples of unit cell spacing. This and a definite phase relationship with neighboring chains may indicate that interchain interactions are essential for the stability of the modulations. A variety of modulation wavelengths was found in all samples studied and may indicate local variation in the concentration of the impurities or of their placement within a unit cell. Two different modulations can also appear simultaneously and even lock, which enhances their amplitude and makes them look like a CDW.;Two different crystal structures resulting from impurity doping have been partially characterized. Samples from Pd0.03- and Pd 0.10NbSe3 growths both reflect a stoichiometry of Pd:Nb:Se = 1:10:30, carry no CDWs, and remain metallic down to 4.2 K temperatures. The new structure found in growths of Si0.03- and Si 0.10NbSe3 has a structure identical to one found in a growth of Cr0.05NbSe3 with a 0.66 x 0.43 nm2 unit cell. Its stoichiometry is Nb:Se = 1:5 with only traces of Si. It undergoes a metal-insulator transition above 660 K and has a bandgap of approximately 0.46 eV.;Transport measurements monitored residual resistance ratios (RRRs), the existence and onset temperatures TC' s of the charge density waves (CDWs), and the frequencies of magneto-quantum-oscillations. Very low RRRs appear to completely suppress quantum oscillations, but this is most likely due to highly effective scattering and does not signal a change in the Fermi surface (FS). CDW TC' s behaved somewhat erratically, depending on the dopant. Low RRRs tend to cause the TC' s to shift down. No correlation exists between the TC shifts and frequency or crystal structure changes. Frequency reduction and the onset of different crystal structures are strongly correlated.