Magnetic and structural properties of Gd-implanted zinc-blende GaN

Abstract

Zinc-blende GaN layers grown by molecular beam epitaxy were uniformly focused-ion-beam implanted with 300keV Gd3+ ions for doses ranging from 1×1012to1×1015cm−2, and their structural and magnetic properties were studied. The implanted samples were not subjected to any annealing treatment. Only Gd incorporation into zinc-blende GaN was observed by x-ray diffraction. Magnetic investigations using superconducting quantum interference device magnetometry reveal a (super)paramagneticlike behavior with an ordering temperature around 60K for the sample with the highest implantation dose. Our experimental studies indicate that the spontaneous electric polarization in wurtzite GaN is the crucial mechanism for its ferromagneticlike behavior upon Gd doping.

Limitations of measuring small magnetic signals of samples deposited on a diamagnetic substrate

“…Nevertheless, the artifacts are visible in other data of low signal measurements recorded with various MPMS machines, e.g., on Cr-doped InN [10], GaAs single crystals [11] or Gd ion implanted cubic GaN [12]. Therefore, they can be taken as representative for the entire family of the MPMS instruments.…”

Tracking defect-induced ferromagnetism in GaN:Gd

“…[8][9][10][12][13][14] Only few exceptions show partly ferromagnetic samples or no ferromagnetism. 15,16 The origin of the ferromagnetic order in GaN:Gd is still unknown. It was shown that cation vacancies can provide local moments which exchange directly by their overlapping wave functions and provide a long-ranged coupling mechanism.…”

Sensitive SQUID magnetometry for studying nanomagnetism

“…Nevertheless, the artifacts are visible in other data of low signal measurements recorded with various MPMS machines, e. g., on Cr-doped InN [17], GaAs single crystals [18] or Gd ion implanted cubic GaN [19], GaN doped with Fe [15], GaN doped with Mn [20]. Therefore they…”

“…Secondly, there would be no magnetic signal from an infinitely (= sufficiently compared to the size of the gradiometer) long object, providing it is magnetically uniform, independently whether it moves or not and what kind of magnetic response dominates its properties. Practically, anything of a length of [15][16][17][18][19][20] cm what allows to fix the sample close to its center and has got an adaptor connecting it with the main (long) sample rod will work for sensitive magnetometry, however, the least 'magnetic', the better. The MPMS XL system manufacturer advices to use clear drinking straws as sample holders.…”