Influence of oxygen doping on the sign of the slope of Tc vs pressure curves in Tl2Ba2CuO6+x
Introduction
Commonly it has been accepted that Tl2Ba2CuO6+x (Tl:2201) belongs to the overdoped branch of the copper oxide high-temperature superconductor family because increasing oxygen content, x, leads only to diminishing Tc [1], [2]. Resistivity and Hall coefficient measurements [3] have given grounds to associate oxygen doping with carrier (hole) doping. The excess in carrier content resulting in diminishing the electron–electron interactions due to Coulomb repulsion has been proposed to explain the decrease in Tc with doping.
Some experimental facts, however, make doubtful the point that the reduction of Tc with oxygen doping is solely related to increasing carrier content. Specific heat measurements [4] revealed no difference in the density of states in Tl:2201 samples with different Tc's. Similarly, Bazhenov and Rezchikov [5] have found that the plasma frequency, of the order of 1 eV, remains unchanged in going from the optimally doped Tl2Ba2CuO6+x single crystal with Tc≈90 K to the overdoped one with Tc≈30 K. They have inferred that, provided the mass of carriers is not specifically adjusted to the change in their concentration, the hole number is constant within some 5–7%, while according to the Hall coefficient data [3] it should change by more than 50%.
On the other hand, heat capacity [4], magnetic susceptibility [3], [6], and NMR [6] reveal the presence of magnetic impurities in the oxygen-doped Tl:2201, the higher it is doped, the more is the number of the impurities. The suggestion has been put forward [6], [7] that the drastic decrease in Tc upon oxygen doping (Tc goes from 90 to 10 K or even zero when x goes from 0 to 0.1–0.2 [3], [8]) is mainly caused by the pair-breaking effect of the impurities. The influence of magnetic impurities on superconductivity is a well known phenomena [9] resulting in reduction of the superconducting gap and lowering Tc.
Kresin et al. [7] have proposed the idea of the “intrinsic” Tc of the cuprates. They have postulated that this intrinsic Tc is higher than that of the optimally doped1 (oxygen-undoped, as is believed for the case of Tl:2201) compound. Once the system is oxygen-overdoped, the carrier content grows, but the magnetic impurities show up as well. This leads to downturn in Tc and its intrinsic value is never reached.
It is suggested that if the concentration of carriers in the oxygenwise optimally doped, magnetic impurity-free sample could be increased without adding new impurities, i.e. in some way other than oxygen doping, the Tc would go up and ideally reach its intrinsic value. The possible way to do this is apply external pressure which increases charge transfer by decreasing c-lattice parameter, hence increasing the hole content. No additional magnetic impurities are added, and Tc should grow.
In this paper we report on resistive measurements of Tc vs quasihydrostatic pressure, P, dependences for high quality single crystals of Tl:2201 with Tc≈40 K (oxygen-overdoped) and 89 K (optimally doped) at ambient pressure. The net result is that dTc/dP is negative for the overdoped sample and positive for the other one.
Section snippets
Experimental
The crystal growth technique has been described previously [10]. We emphasize that the crystals used in the measurements were as-grown ones. The oxygen content was controlled during the synthesis by adjusting oxygen pressure, not by the subsequent heat treatment. From the growth conditions we infer that the crystal with Tc≈40 K (we call it sample B) is more oxygen-rich than the one with Tc≈89 K (sample A). Since Tl:2201 is believed to hit only the right-hand (overdoped) side of the well known
Results and discussion
Fig. 1 shows normalized resistivity of samples A and B vs temperature for different values of the applied pressure. One immediately finds that in sample A, the pressure shifts the superconducting transition upwards while in sample B it shifts Tc downwards. Using the 50% criterion (middle-point of the transition), we determine Tc for each curve and replot Tc against pressure, P, in Fig. 2.
The positive sign of dTc/dP for sample A is exactly what Kresin’s “intrinsic Tc” approach [7] predicts, as
Summary
We have studied the effect of pressure on Tc in the optimum-doped, oxygen contentwise, and overdoped single crystals of Tl2Ba2CuO6+x. This effect is found positive for the optimum-doped sample, in agreement with the prediction of Kresin et al. about the “intrinsic” Tc of the cuprates [7]. This result implies that the oxygen contentwise system might in fact be an underdoped carrier contentwise.
For the overdoped sample, the dTc/dP is negative. We propose that this effect, like in the case of
Acknowledgements
The authors would like to thank Prof. M.A. Kulakov for his contribution to development of the sample preparation technique and fruitful discussions of the results.
References (17)
- et al.
Physica C
(1990) - et al.
Physica C
(1995) - et al.
Phys. Rev. B
(1991) - et al.
J. Supercond.
(1994) - A.V. Bazhenov, K. Rezchikov, 1995,...
- et al.
J. Phys. I France
(1996) - et al.
Phys. Rev. B
(1996) - et al.
Phys. Rev. Lett.
(1993)
Cited by (1)
The dependence of the critical temperature on pressure
2020, International Journal of Modern Physics B