Doping means adding a foreign element to alter a host material's composition and, with it, the material's properties. The doping concentration always measures some content against a chosen base — and the choice of base is the first thing worth pinning down.
In a formula such as A1−xMxB, x counts sites on the A/M sublattice rather than atoms overall: x = 0.05 means M sits on 5% of those sites, not on 5% of every atom in the material. Mass fraction, atomic percentage taken over all atoms, and site fraction each divide by something different, so one and the same number can stand for quite different compositions.
Nor should the nominal figure in a formulation be read as what actually ends up inside the host. Elements are lost during processing, or they precipitate as a second phase. In one series with nominal composition Te0.98M0.02 (M = As, Sb, Bi), atom probe tomography found average solute contents in the Te matrix of just 0.06, 0.03 and 0.09 at.% — most of the added element was sitting in the corresponding telluride precipitates.
Figure 1. Phases, elemental distribution and matrix atom probe analysis in samples of nominal composition Te0.98M0.02 (M = As, Sb, Bi).
Loading more into the charge in such a system may mostly grow the second phase rather than raise the solute in the matrix in proportion. What that second phase then does depends on its own properties and on how it meets the matrix: in the Te system above, the regions carrying telluride precipitates were the more conductive ones, so the gain cannot be credited entirely to substitutional doping in the Te lattice. Limited solid solubility and better composite performance can be true at the same time.
And even when the element does enter the lattice, chemical content and electrically active content stay separate quantities. Donors hand electrons to the conduction band; acceptors take electrons from the valence band and leave holes behind. Whether they ionize fully, and whether they throw up compensating defects, decides how many free carriers actually exist. An atomic-percent doping level and a carrier concentration quoted in cm−3 are therefore in different units — and can even trend in opposite directions as the formulation is changed.
Count only electrons, holes and singly charged donors and acceptors, and charge neutrality demands n + NA− = p + ND+, where n and p are the free electron and hole concentrations and NA− and ND+ the ionized acceptor and donor concentrations. A dopant atom turns into an extra carrier only after the ionization and compensation bookkeeping has been settled.
Compensation cancels what defects of the opposite sign contribute. An acceptor, say, favours holes, but a donor arriving alongside it cancels part of that gain, and some dopant atoms also distort the local structure and create defects of the other electrical character. The more dopant goes in, the smaller the return in free carriers can become.
In one study of As-doped CdTe single crystals, both crystal types showed very low dopant activation — around 1–3% in one of them. Pairing defect spectroscopy with first-principles calculations, the authors laid the main compensation on a donor-like defect called the AX centre: the lattice relaxes substantially around the original acceptor and forms a configuration that can cancel it. That is an explanation for those crystals and those growth conditions, not a fixed activation rate for As-doped CdTe as a class.
Once carriers are plentiful, mobility enters the conductivity as well. For an electron-dominated material σ ≈ neμ, with e the elementary charge and μ the mobility. Doping brings charged centres and local structural disorder, both of which strengthen scattering and pull μ down; if the rise in n does not cover the fall in μ, σ stops rising altogether. Compensation sets how many carriers are available to conduct; scattering sets how well they drift.
Ionic conductors face the same contest between how many defects there are and how easily they move. Where Gd3+ replaces Ce4+ in CeO2 and oxygen vacancies are the only compensation, every two Gd ions bring one oxygen vacancy. A vacancy gives neighbouring oxygen ions somewhere to jump to, so more vacancies mean more chances to migrate — but every jump must still clear a local barrier.
At higher concentrations the association between dopant ions and oxygen vacancies can no longer be ignored, and vacancies feel one another as well. Some sites fill up with vacancies more readily, while leaving those sites costs more energy. A larger total number of vacancies therefore does not guarantee more long-range migration per unit time: where the defects sit shapes diffusion too.
A simulation of Gd-doped CeO2 containing dislocations separates these spatial effects. At an average Gd fraction of 8%, equilibration at 1300 °C left enriched and depleted zones around the dislocations. In the molecular dynamics that followed at 900 °C, the enriched zones held more vacancies and yet still showed slower local oxygen diffusion than dislocation-free bulk, while the depleted zones were held back by, among other things, a shortage of vacancies. The comparison is between local diffusion coefficients, not the measured conductivity of a bulk ceramic.
In that simulation oxygen vacancies seldom left the enriched zones and seldom crossed the depleted ones, while regions well away from the dislocations came closer to isotropic diffusion. A high local vacancy count and fast long-range diffusion are two different questions: the first counts defects inside a region, the second also asks whether oxygen ions can keep moving on into other regions.
Thermoelectrics turn a temperature difference into electrical energy, and electrical conductivity on its own says little about how well they do it. The dimensionless figure of merit is ZT = S2σT/κ, where S is the Seebeck coefficient — the open-circuit voltage generated per unit temperature difference — T the absolute temperature and κ the total thermal conductivity. A large S2σ helps the electrical output, a small κ limits the heat that simply conducts through; both move ZT.
Doping shifts the carrier concentration and the band structure, and it changes phonon scattering too (a phonon being one quantum of lattice vibration). Carriers and band-structure effects act on S and σ, while phonon scattering acts on the heat carried through the lattice. None of these has to move the same way, so lower conductivity at one composition does not by itself settle whether ZT also falls.
In a concentration series of Al- and Sb-co-doped SnTe, raising the co-doping level lowered the hole concentration and the mobility, and with them the conductivity, while the Seebeck coefficient climbed. Part of that gain in S comes with the loss of holes, part from Al-driven changes in the electronic states; alloying with AgBiTe2 then moved the bands further and touched S again. The performance change across this series cannot be put down to carrier numbers alone.
These compositional changes also bring in point defects and dislocations that scatter phonons and lower the lattice thermal conductivity. Less heat transport, working together with the altered electrical transport, lets some doped samples reach a higher ZT in spite of lower conductivity. What competes here is not only the flow of electrons but the flow of heat.
With the Al:Sb ratio held fixed, taking the AgBiTe2 alloying fraction from 4% to 6% in the same paper went on lowering the room-temperature Hall hole concentration, yet in the mid-temperature range the 6% sample had a lower ZT than the 4% sample. XRD across this concentration series found no impurity phase, so the difference in performance had already appeared in a range where no new phase gave a clear diffraction signal.
The ZT gap between the two also varies with temperature, which means the better ratio comes with a temperature window. In such narrow-gap materials high temperature can also excite minority carriers, strengthening bipolar transport and altering both the Seebeck coefficient and the heat flow; the balance among band structure, carrier number and their scattering shifts with it. A carrier concentration that pays off at one temperature need not give the highest ZT at another.
SAT NANO is a best supplier of nano material in China, we can supply the nitride doped material like nitride doped tio2, nitride doped carbon nanotube, silver doped tio2 powder. If you have any enquiry, please contact us at sales03@satnano.com .