Controlled doping by self-assembled dendrimer-like macromolecules

Doping via self-assembled macromolecules might offer a solution for developing single atom electronics by precisely placing individual dopants at arbitrary location to meet the requirement for circuit design. Here we synthesize dendrimer-like polyglycerol macromolecules with each carrying one phosphorus atom in the core. The macromolecules are immobilized by the coupling reagent onto silicon surfaces that are pre-modified with a monolayer of undecylenic acid. Nuclear magnetic resonance (NMR) and X-ray photoelectron spectroscopy (XPS) are employed to characterize the synthesized macromolecules and the modified silicon surfaces, respectively. After rapid thermal annealing, the phosphorus atoms carried by the macromolecules diffuse into the silicon substrate, forming dopants at a concentration of 1017 cm−3. Low-temperature Hall effect measurements reveal that the ionization process is rather complicated. Unlike the widely reported simple ionization of phosphorus dopants, nitrogen and carbon are also involved in the electronic activities in the monolayer doped silicon.

. GPC results of polyglycerol (X-axis is time and Y-axis is signal strength voltage). Water is the mobile phase and the peaks at around 39.00 min are related to some impurity substances from water. Figure S2. ESI-MS spectra of Polyglycerol at the molecular weight of 1000 to 5000. Owing to the limitation of ESI equipment, the polyglycerol molecules with large molecular weight may not be ionized.    where n si,surf is the surface density of Si atoms (7.8×10 14 cm -2 for Si(111) and 6.9×10 14 cm -2 for Si(100)), n Si,bulk is the atomic number density of Si atoms (5.0×10 22 cm -3 ), and l Si is the escape depth equal to λ Si ×sin(90º ) in which λ Si is 1.6 nm and 90º is the angle between the detector and sample surface for our instrument.     During the coupling process, only two coupling reagents are used, i.e. DCC (Dicyclohexylcarbodiimide) and DMAP (4-dimethylaminopyridine). The binding energy of N peak in the XPS spectra is 400.3 eV, which is different from N in DCC (399.00 eV) and N in DMAP (399.50 eV of dimethylamino and 399.00 eV of pyridine) 1 . As a result, the nitrogen doped in silicon must come from the intermediate of DCC and carboxyl group.
[1] NIST X-ray Photoelectron Spectroscopy Database, Version 4.1 (National Institute of Standards and Technology, Gaithersburg, 2012); http://srdata.nist.gov/xps/.   . Figure S15. IR transmission spectra of PG macromolecules (red) and PG-modified silicon (black).  To do TEM, a thin slab of the sample is cut by focused ion beam (FIB), during which a layer of Pt is deposited on the sample surface. However, P element mapping is problematic because the kα1 signal for P and Pt are located at the same energy position, resulting in the same purple dots in Panel c. Figure S22. SIMS N distribution from surface to bulk for samples annealed in N2 and Ar enviroment. Surprisingly, N elements are shown up in the blank sample which is undoped silicon as purchased. This is due to the physical absorption of N2 on the sample surface. Figure S23. Hall effect of samples pre-treated with oxygen ambient (550 ℃ for 5h). In order to remove the carbon contaminant, the PG modified silicon sample was pre-treated at 550 ℃ in oxygen ambient for 5h. After the pre-treatment, the silicon sample was annealed at 1050 ℃ in vacuum for 30s. This sample was measured by the PPMS to find the concentration of electrically active dopants. The results indicate that oxygen pretreatment significantly improves the phosphorus activation rate. The parameter f chooses a value from the following curve.

Equation II
Ration of R x to R y Reference: L. J. van der Pauw, A method of measuring specific resistivity and Hall effect of discs of arbitrary shape, Philips Res. Repts. 13, 1-9, 1958 (No.1).

The equation about dopants energy level in nitrogen doped silicon
For the N-doped silicon bulk, according to the charge neutrality, we have: Where n o is the electron concentration, N + d is the average concentration of ionized dopants and p 0 is the hole concentration. And in the n-type semiconductor, it follows: Thus, we have Also, it is known that the electron concentration is dependent upon the temperature, according with the equation Where N c is the effective density of states function, N c ≈ w*(kT) 3/2 in which W is constant for the silicon semiconductor, E F is the Fermi level, E c is the conductance band edge. And the average concentration of ionized dopants N + d follows the equation Where E d is the donor energy level. Combined eq.(1)~eq.(5), and by solving the 2-order equation (eq.7) related to n 0 , we have the equation (8) Where E a is the activation energy, with E a = E c -E d .
Note that phosphorus atoms remain complete ionization and the average concentration of completely ionized P donors, N p , remains constant in the temperature range of the measurement. Hence, the electron concentration of the the N and P doped sample is shown as follows.