We have made electrical measurements on a system using carbon nanotubes as the dopant material. A semiconjugated, organic polymer was mixed with carbon nanotubes to form a wholly organic composite. Composite formation from low to high nanotube concentration increases the conductivity dramatically by ten orders of magnitude, indicative of percolative behavior. Effective mobilities were calculated from the spacecharge regions of the current-voltage characteristics for the 0-8 % mass fractions. After an initial rise these were seen to fall from 1-8 % doping levels as predicted by theory. From the values for conductivity and mobility, an effective carrier density was calculated. This was seen to decrease between 0% and 1%, before rising steadily. ͓S0163-1829͑98͒51536-6͔

ring overnight at room temperature, monomer and initiator were readily absorbed in the vesicle bilayer. Photoinitiated polymerizations were performed in a thermostatted quartz reactor using either an UV-lamp (HPR 125 W, Philips) or a pulsed excimer laser (Lambda Physics XeF, 351 nm, 2 Hz pulse frequency, 30 mJ energy per pulse) as irradiation source. Conversions were determined by HPLC analysis of the residual monomers.Details concerning the use of cryo-electron microscopy have been described earlier [19].

We have studied the effect of erbium-impurity interactions on the 1.54 μm luminescence of Er3+ in crystalline Si. Float-zone and Czochralski-grown (100) oriented Si wafers were implanted with Er at a total dose of ∼1×1015/cm2. Some samples were also coimplanted with O, C, and F to realize uniform concentrations (up to 1020/cm3) of these impurities in the Er-doped region. Samples were analyzed by photoluminescence spectroscopy (PL) and electron paramagnetic resonance (EPR). Deep-level transient spectroscopy (DLTS) was also performed on p-n diodes implanted with Er at a dose of 6×1011/cm2 and codoped with impurities at a constant concentration of 1×1018/cm3. It was found that impurity codoping reduces the temperature quenching of the PL yield and that this reduction is more marked when the impurity concentration is increased. An EPR spectrum of sharp, anisotropic, lines is obtained for the sample codoped with 1020 O/cm3 but no clear EPR signal is observed without this codoping. The spectrum for the magnetic field B parallel to the [100] direction is similar to that expected for Er3+ in an approximately octahedral crystal field. DLTS analyses confirmed the formation of new Er3+ sites in the presence of the codoping impurities. In particular, a reduction in the density of the deepest levels has been observed and an impurity+Er-related level at ∼0.15 eV below the conduction band has been identified. This level is present in Er+O-, Er+F-, and Er+C-doped Si samples while it is not observed in samples solely doped with Er or with the codoping impurity only. We suggest that this new level causes efficient excitation of Er through the recombination of e-h pairs bound to this level. Temperature quenching is ascribed to the thermalization of bound electrons to the conduction band. We show that the attainment of well-defined impurity-related luminescent Er centers is responsible for both the luminescence enhancement at low temperatures and for the reduction of the temperature quenching of the luminescence. A quantitative model for the excitation and deexcitation processes of Er in Si is also proposed and shows good agreement with the experimental results.

Electron paramagnetic resonance ͑EPR͒ measurements have been made of amorphous hydrogenated carbon ͑a-C:H͒ films grown by plasma enhanced chemical vapor deposition ͑PECVD͒ with negative self-bias voltages V b in the approximate range 10-540 V. For V b Ͻ100 V, as the film changes from polymerlike to diamondlike, the changes in linewidth and shape are interpreted in terms of changes to two contributions-one due to dipolar interactions between the unpaired spins and one due to unresolved lines arising from hyperfine interactions with H 1 . The former yields a Lorentzian line, the latter a Gaussian, and the resultant spectrum has the Voigt shape. The empirical relation ⌬B pp G ͑in Gauss͒ϭ͑0.18Ϯ0.05͒ϫ͑at. % H) between the peak-to-peak Gaussian contribution ͑in Gauss͒ ⌬B pp G and the hydrogen content in atomic percentage is obtained. For V b Ͼ100 V the linewidth is shown to be dominated by the dipolar interactions and exchange and it decreases as V b increases; the change is shown to arise primarily from a change in the exchange interaction. Evidence for this comes from measurements which show that the spin-lattice relaxation time appreciably shortens and the spin-spin relaxation time lengthens as the bias voltage is increased. The magnitude and variation with bias of the linewidth are consistent with the EPR signal originating from the -type radicals.

We have described a one-step, high-yield, nondestructive purification and processing method for multiwalled carbon nanotube (MWNT) containing soot using a conjugated organic polymer host. This host selectively suspends nanotubes relative to impurities. The fraction of available MWNTs suspended can be measured using electron paramagnetic resonance (EPR) and rises with increasing polymer mass before saturating at approximately 50% by mass for a soot to polymer mass ratio of 1:5. Thermogravimetric analysis can then be used to calculate the mass of nanotubes suspended and hence the purity of the original soot. Furthermore, this allows the calculation of numerical constants relating the EPR signal intensity to the nanotube mass, allowing the routine calculation of nanotube content. Finally the host polymer was removed by filtration, giving 91% pure nanotube material. In this case a yield of 17% pristine nanotubes was reclaimed from the soot. Full optimization of this process could lead to yields of up to 40%.