The Synthesis And Investigation Of Fe-based Superconductor LiFeP And New Diluted Magnetic Semiconductors

The discovery of Fe-based superconductors haLs generated extensive reseatches. Over the past seven years,the list of FePh/Ch superconductors has expanded rapidly.Scientists have employed various experimental and theoretical meth. ods to investigate the superconduting mechanism.And they have realized that superconductivity has been closely related to spin fluctuatioas. In this thesis, we conducted nuclear magnetic resonance measurements on the"111"type Fe-ba.sed superconductor LiFeP.After the synthesis of the polycrystal sample though solid.state reaction method,we measured the temperature dependence of resis-tivity and magnetization.The sharp drop in resistivity at lower temperatures, together with the diamagnetic signals from the magnetization,indicate that the superconductivity takes place at～5 K in LiFeP.The T-dependent resistivity in the temperature range between 5 K and 40 K shows a relation of ρ∝T2.We investigated the 31P NMR lineshapes and spin-lattice relaxation time T1.The Knight shift,31K,hardly changes from 4.2 K to 280 K.The spin-lattice relax-ation rate 1/T1 divided by T,1/T1T,is～0.065s-1K-1,which is almost an order smaller than～0.45 s-1K-1 at75 As sites of LiFeAs,implying much weaker spin fluctuations in LiFeP. In addition,1/T1T display a slight enhancement toward low temperature,indicating that antiferromagnetic spin fluctuations do exist in LiFeP.The Korringa relation,T1TKs2=4πκB/Hγ2n/γe2β,has been used evaluate quan-titatively the strength of the electron correlations.The values of βin the whole measured temperature range are smaller than 1,but close to 1,again indicating the weaker electron correlations,which may be related to the much lower Te of LiFeP.If Fe atoms of the FePn/Ch superconductors are fully replaced by Zn,one will obtain LaZnAso,LiZnAs,BaZn2As2 etc.,which are all direct-gap semiconduc-tors.Ferromagnetic ordering develops when substituting Zn by Mn to introduce local moments and introducing carriers at a different site. Recently,a series of Zn-based diluted magnetic semiconductors that are derivatives of Fe-based su-perconductors have been synthesized. These systems are named as "1111" [C. Ding et al., Phys. Rev. B 88,041102（R） （2013）], "111" [Z. Deng et al., Nat. Commun.2,422 （2010）], "122" [K. Zhao et al., Nat. Commun.4,1442 （2013）], "32522" [H. Y. Man et al., Europhys. Lett.105,67004 （2014）] DMS families. Furthermore, if substituting all the Fe atoms of the Fe-based superconductors with Mn, antiferromagnets will be obtained, such as LaMnAsO [N. Emery et al., Phys. Rev. B 83,144429 （2011）], LiMnAs [W. Bronger et al., Z. Anorg. Allg. Chem.539,175 （1986）], BaMn2As2 [Y. Singh et al., Phys. Rev. B 80, 100403（R） （2009）]. We should note that the superconducting, ferromagnetic and antiferromagnetic systems have the same lattice structure. The excellent lattice matching opens the possibilities to make junctions between these systems through the As layer. Since the spins and carriers are introduced by different elements, these new DMS systems have the advantage of decoupled carrier and spin doping, which enables us to investigate the influence of carrier density or spin density on the ferromagnetism.In this thesis, I have synthesized several new bulk form DMSs and character-ized them-with various methods. The ZF and WTF μSR results for Li（Zn,Mn）P have confirmed the gradual development of ferromagnetic ordering below Tc with a nearly 100% magnetic ordered volume, implying that static magnetic order develops in almost the entire sample volume. The good agreement of the re-lation between the static internal field parameter as and Curie temperature Tc for Li（Zn,Mn）P, other bulk form DMSs and （Ga,Mn）As, implies that all these systems share a common mechanism for the ferromagnetic exchange interaction. In addition, Li1+y/（Zn1-xMnx）P has the advantage of decoupled carrier and spin doping, which enables us to investigate the influence of overdoped Li on the fer-romagnetic ordered state. For Li1+y（Zn0.93Mn0.07）P and Li1+y（Zn0.9Mn0.1）P, the Curie temperature （Tc）. Weiss temperature （θ）, the effective moment （Meff）, and the saturation moment （Msat） initially increase with y, reach a maximum, and then start to decrease, indicating more carriers are detrimental to the ferro-magnetic exchange interaction for a certain amount of spins, and that a delicate balance between charge and spin densities is required to achieve highest Tc-To try a different route of synthesizing bulk form DMS, where Mn and Cu or Co are codoped into the Zn sites of BaZn2As2, a p-type DMS Ba（Zn,MnCu）2As2 and a n-type DMS Ba（ZnMnCo）2As2 have been successfully obtained. The codop-ing Mn and Cu or Co haven’t changed the layered tetragonal lattice structure, but have changed the electronic and magnetic properties. In Ba（Zn1-2xMnxCux）2As2, heavy carriers suppress the resistivity, but the spin fluctuations induced by Mn increase the possibility of the scattering of the carriers, which enhances the resistivity. No ferromagnetic order occurs with （Zn,Mn） or （Zn,Cu） substitu-tion. Only when Zn are substituted by both Mn and Cu simultaneously, can the system undergo a ferromagnetic transition below Tc～70 K with the coer-civity Hc～1600 Oe. As large as-53% negative magnetoresistance has been observed in Ba（Zn0.75Mn0.125Cu0.125）2As2. AC susceptibility measurements for Ba（Zn0.75Mn0.125Cu0.125）2As2 show that the spin freezing temperature Tf strongly depends on the applied frequency and DC magnetic field with v0/v= （Tf/19— 1）-8.3 and Tf（H） ∝1-bH0.55, indicating the spin glass transition at Tf.Furthermore, a new DMS system, （Sr3La205）（Zn1-xMnx）2As2, has been syn-thesized and characterized, which represents the fifth DMS family that Has a direct counterpart with identical/similar structure in the Fe-based superconduc-tors. Mn substitution for Zn in （Sr3La2O5）Zn2As2 results in ferromagnetic or-dering with Curie temperature reaching-40 K and saturation moment reaching 0.5μB/Mn. The magnetization measurements for different doping levels indicate that appropriate amount of Mn is necessary for developing the ferromagnetic or-dering, but overmuch Mn will lead to antiferromagnetic exchange interaction be-tween spins from nearest-neighbor Mn sites, which will destroy the ferromagnetic ordering. This explains why ferromagnetic ordering disappears for the doping level of x= 0.30.Ba（Zni₂xMnxCox）2As2 is the first n-type bulk form DMS with Curie tem-perature Tc～40 K and the competition of ferromagnetic and antiferromagnetic coupling interactions. Large negative magnetoresistance has been observed. The Hall effect measurements of Ba（Zn,MnCo）2As2 show n-type carriers, which has confirmed that the n-type carriers in Ba（Zn,MnCo）2As2 system come from the （Zn2+,Co3+） substitution. The difference of the ionic radius of Zn2+, Mn2+ and Co3+ makes it difficult to substitute Zn2+ with both Mn2+ and Co3+. To further understand the effect of Mn and Co, we need to improve the quality of the sam-ples. Additional measurement approaches are on the way to investigate whether the magnetic transition in Ba（Zn,MnCo）aAs2 is intrinsic.The only way to explore new bulk form DMSs is to keep tirelessly try-ing. The materials I have tried include "111" type Li（Zn,Fe）P and Li（Zn,Fe）As, "1111" type （La,Ca/Sr）（Zn,Co/Cr）AsO, （Ba,K）F（Zn,Mn）As, （Ba,K）F（Cu,Mn）S and La（Zn,Mn）As（O,F）, "122" type （Ba,La）（Zn,Mn）2As2, Ba（Zn,MnNi）2As2 and Ba（Zn,Mn）2（As,P）2. Theses materials have failed to be good bulk form DMS, but only through endless trial and error, can one successfully synthesize the required materials. So all the efforts and experiences have a certain reference value for this field.