The system LiHo_<em>x</em>Y_1-<em>x</em>F₄ is a nearly perfect example of a dilute, dipolar-coupled Ising magnet and, as such, it is an ideal testing ground for many theories in statistical mechanics. At low holmium concentration (<em>x</em> = 0. 045) an unusual spin liquid or "anti-glass" state was discovered in previous work [1]. This state does not exhibit a spin glass freezing transition as is expected for a long-range interaction. Instead, it shows dynamics which are consistent with a collection of low-frequency oscillators [2]. It was also seen to have sharp features in its specific heat [3]. <br /><br /> We present heat capacity measurements on three samples at and around the concentration of the spin liquid state in zero magnetic field and in a temperature range from around 50 mK to 1 K. In contrast to previous measurements, we find no sharp features in the specific heat. The specific heat is a broad feature which is qualitatively consistent with that of a spin glass. The residual entropy as a function of <em>x</em>, obtained through a numerical integral of the data, however, is consistent with numerical simulations which predict a disappearance of spin glass ordering below a critical concentration of dipoles [4]. <br /><br /> Also presented here, is ac susceptibility data on an <em>x</em> = 0. 45 sample which exhibits a paramagnetic to ferromagnetic transition and is found to be consistent with previous work.

Nanocrystalline silicon (nc-Si:H) is an attractive material for fabrication of low temperature, large area electronic devices due to superior properties versus the traditional amorphous silicon (a-Si:H) and polycrystalline silicon (polySi). Nanocrystalline silicon possess higher carrier mobility and better stability than a-Si:H and better device uniformity and lower fabrication cost than polySi. This thesis looks at the application of nc-Si:H material in fabricating two different diodes used for rectification and light generation. Optimization of n-type nc-Si:H deposited via plasma enhanced vapor chemical deposition (PECVD) was achieved through adjusting the concentration ratio of phosphine (PH3) dopant source gas versus silane (SiH4). Optimizing for dark conductivity, n+ nc-Si:H material with dark conductivity of 25.3 S/cm was deposited using a [PH3]/[SiH4] ratio of 2%. Using the optimized n+ nc-Si:H film, a p-n junction diode utilizing an undoped and an n+ nc-Si:H layers was fabricated designed for rectification use. The diode achieved a current density of 1 A/cm2, an ON/OFF current ratio of 106 and a non-ideality factor of 1.9. When the 200*200µm2 nc-Si:H diodes were employed in a full-wave bridge rectifier, a 2.6 V direct current voltage could be generated from an input sine wave signal with amplitude 2 VRMS and frequency of 13.56 MHz, thus demonstrating the feasibility of using nc-Si:H to fabricate diodes for using on radio frequency identification (RFID) tags. Nanocrystalline silicon was also applied in fabrication of a light emitting diode (LED), by utilizing the nanocrystals embedded inside nc-Si:H, inside which recombination of carriers could result in radiative recombination. By limiting the deposition time of the nc-Si:H, 10 – 20 nm thick films of nc-Si:H were used to fabrication a p-i-n structure LED with average crystallite size between 7.5 nm to 13.7 nm corresponding to an theoretical emission wavelengths in the near infrared region of 875 nm to 963 nm. Unfortunately, light emission from the nc-Si:H LED were not detected using two different methods. Undetectable emission could have been due to a combination of low recombination efficiency due to carriers recombining in defects in the a-Si:H matrix and majority of current travelling completely through the nc-Si:H films without recombining. A study of the thin intrinsic nc-Si:H films used in the LED was carried out. The thin films were found to be highly defected, with large variation in current-voltage relationship measured and hysteresis observed in the IV characteristic. Annealing the nc-Si:H films were found to cause a drop in conductivity explained through hydrogen effusion from the nc-Si:H film during annealing. Passivation of defects was achieved through the use of hydrogen plasma which resulted in a lowering of activation energy measured in the film. Oxygen plasma was also trialed for passivating the nc-Si:H film but the effect was only a temporary increase in current conduction attributed to oxygen ions chemisorbing temporarily at the film surface.

The system LiHo_<em>x</em>Y_1-<em>x</em>F₄ is a nearly perfect example of a dilute, dipolar-coupled Ising magnet and, as such, it is an ideal testing ground for many theories in statistical mechanics. At low holmium concentration (<em>x</em> = 0. 045) an unusual spin liquid or "anti-glass" state was discovered in previous work [1]. This state does not exhibit a spin glass freezing transition as is expected for a long-range interaction. Instead, it shows dynamics which are consistent with a collection of low-frequency oscillators [2]. It was also seen to have sharp features in its specific heat [3]. <br /><br /> We present heat capacity measurements on three samples at and around the concentration of the spin liquid state in zero magnetic field and in a temperature range from around 50 mK to 1 K. In contrast to previous measurements, we find no sharp features in the specific heat. The specific heat is a broad feature which is qualitatively consistent with that of a spin glass. The residual entropy as a function of <em>x</em>, obtained through a numerical integral of the data, however, is consistent with numerical simulations which predict a disappearance of spin glass ordering below a critical concentration of dipoles [4]. <br /><br /> Also presented here, is ac susceptibility data on an <em>x</em> = 0. 45 sample which exhibits a paramagnetic to ferromagnetic transition and is found to be consistent with previous work.

Nanocrystalline silicon (nc-Si:H) is an attractive material for fabrication of low temperature, large area electronic devices due to superior properties versus the traditional amorphous silicon (a-Si:H) and polycrystalline silicon (polySi). Nanocrystalline silicon possess higher carrier mobility and better stability than a-Si:H and better device uniformity and lower fabrication cost than polySi. This thesis looks at the application of nc-Si:H material in fabricating two different diodes used for rectification and light generation. Optimization of n-type nc-Si:H deposited via plasma enhanced vapor chemical deposition (PECVD) was achieved through adjusting the concentration ratio of phosphine (PH3) dopant source gas versus silane (SiH4). Optimizing for dark conductivity, n+ nc-Si:H material with dark conductivity of 25.3 S/cm was deposited using a [PH3]/[SiH4] ratio of 2%. Using the optimized n+ nc-Si:H film, a p-n junction diode utilizing an undoped and an n+ nc-Si:H layers was fabricated designed for rectification use. The diode achieved a current density of 1 A/cm2, an ON/OFF current ratio of 106 and a non-ideality factor of 1.9. When the 200*200µm2 nc-Si:H diodes were employed in a full-wave bridge rectifier, a 2.6 V direct current voltage could be generated from an input sine wave signal with amplitude 2 VRMS and frequency of 13.56 MHz, thus demonstrating the feasibility of using nc-Si:H to fabricate diodes for using on radio frequency identification (RFID) tags. Nanocrystalline silicon was also applied in fabrication of a light emitting diode (LED), by utilizing the nanocrystals embedded inside nc-Si:H, inside which recombination of carriers could result in radiative recombination. By limiting the deposition time of the nc-Si:H, 10 – 20 nm thick films of nc-Si:H were used to fabrication a p-i-n structure LED with average crystallite size between 7.5 nm to 13.7 nm corresponding to an theoretical emission wavelengths in the near infrared region of 875 nm to 963 nm. Unfortunately, light emission from the nc-Si:H LED were not detected using two different methods. Undetectable emission could have been due to a combination of low recombination efficiency due to carriers recombining in defects in the a-Si:H matrix and majority of current travelling completely through the nc-Si:H films without recombining. A study of the thin intrinsic nc-Si:H films used in the LED was carried out. The thin films were found to be highly defected, with large variation in current-voltage relationship measured and hysteresis observed in the IV characteristic. Annealing the nc-Si:H films were found to cause a drop in conductivity explained through hydrogen effusion from the nc-Si:H film during annealing. Passivation of defects was achieved through the use of hydrogen plasma which resulted in a lowering of activation energy measured in the film. Oxygen plasma was also trialed for passivating the nc-Si:H film but the effect was only a temporary increase in current conduction attributed to oxygen ions chemisorbing temporarily at the film surface.

This thesis will present research that studies the role of disorder, geometric frustration and the long range dipolar interaction on the collective behaviour of several insulating, rare earth magnets. Experiments were performed at low temperatures to measure the specific heat and magnetic susceptibility of several materials. Susceptibility was measured with a SQUID magnetometer that has been designed and constructed primarily for the study of slow dynamics in glassy systems. Specifically, this thesis will discuss three distinct topics. The first is the series of materials LiHo(x)Y(1-x)F(4), which are manifestations of the dilute, dipolar coupled Ising model. The low-x portion of the phase diagram has become a rather contentious issue in recent years with both theoretical and experimental groups disagreeing on the existence of a spin glass freezing transition and one experimental group arguing for the existence of an exotic "antiglass'' or spin liquid state resulting from quantum entanglement at x=0.045. We present specific heat and dynamical susceptibility measurements on four stoichiometries in this series: x = 0.018, 0.045, 0.080 and 0.012. No evidence of an unusual antiglass state is observed. Instead, our results show evidence, at all dilution levels studied, of a spin glass freezing transition. Interpretation of experimental data is found to be complicated by the anomalously slow dynamics in these materials. The relaxation time scales are found to increase as the concentration of Ho(3+) ions is reduced, an effect which can be attributed to single-ion physics and the importance of the nuclear hyperfine coupling in this system. A second set of materials studied here is a series of several Gd garnet materials, the most famous of which is Gd(3)Ga(5)O(12) (GGG), a material previously argued to be a disorder-free spin glass. Our specific heat experiments reproduce previous experiments on GGG and show that the homologous Gd garnets Gd(3)Te(2)Li(3)O(12) and Ga(3)Al(5)O(12) do not share the same glassy physics but exhibit sharp ordering features. By experimenting with the introduction of random site dilution, it is concluded that a 1-2% off-stoichiometry inherent in GGG is likely a special kind of disorder that is particularly effective in inducing random frustration and the formation of a spin glass. Finally, specific heat measurements on the pyrochlore antiferromagnet Gd(2)Sn(2)O(7) (GSO) are presented. While GSO has generally been found to be a well behaved and well understood model magnet, with long range order developing at around 1 K, like many other geometrically frustrated magnets, it has been discovered to possess persistent spin dynamics down to very low temperatures as measured by μSR and Mössbauer spectroscopy. Measurement of the low temperature limit of the specific heat when compared with linear spin-wave theory, however, presents a consistent picture of gapped magnon excitations that freeze out at low temperatures and make the existence of the proposed dynamic ground state unlikely.

This thesis presents research of two novel magnetic materials, LiHoF4 and Tb2Ti2O7. Experiments were performed at low temperatures and in an applied magnetic field to study thermal expansion and magnetostriction using a capacitive dilatometer designed during this project. This thesis presents 3 distinct topics. This manuscript begins with a thermodynamic description of thermal expansion and magnetostriction. The design of a capacitive dilatometer suitable for use at ultra-low temperatures and in high magnetic fields is presented. The thermal expansion of oxygen free high conductivity copper is used as a test of the absolute accuracy of the dilatometer. The first material studied using this dilatometer was LiHoF4. Pure LiHoF4 is a dipolar coupled Ising ferromagnet and in an applied transverse magnetic field is a good representation of the transverse field Ising model. An ongoing discrepancy between theoretical and experimental work motivates further study of this textbook material. Presented here are thermal expansion and magnetostriction measurements of LiHoF4 in an applied transverse field. We find good agreement with existing experimental work. This suggests that there is some aspect of LiHoF4 or the effect of quantum mechanical fluctuations at finite temperatures which is not well understood. The second material studied is the spin liquid Tb2Ti2O7. Despite theoretical predictions that Tb2Ti2O7 will order at finite temperature, a large body of experimental evidence demonstrates that spins within Tb2Ti2O7 remain dynamic to the lowest temperatures studied. In addition Tb2Ti2O7 also exhibits anomalous thermal expansion below 20K, giant magnetostriction, and orders in an applied magnetic field. Thermal expansion and magnetostriction measurements of Tb2Ti2O7 are presented in applied longitudinal and transverse fields. Zero-field thermal expansion measurements do not repeat the previously observed anomalous thermal expansion. A large feature is observed in thermal expansion at 100mK, in rough agreement with existing experimental work. Longitudinal and transverse magnetic fields were applied to Tb2Ti2O7. Longitudinal magnetostriction measurements show qualitatively di erent behavior than previous observations. These measurements were taken along di erent crystal axes so direct comparison cannot be made. Thermal expansion measurements in an applied transverse field show evolution with the strength of the applied field. This evolution may relate to an ordering transition, however difficulties in repeatability in a transverse field require that these results be repeated in an improved setup.

This research describes an investigation on the fabrication of CuInSe2-based thin-film solar cells with the device structure of Al/ZnO:Al/ZnO/CdS/CIS/Mo/SLG at the substrate temperature of 450oC, which is at least 100oC below the temperature currently used for depositing CIS thin films. A great advantage for the low temperature process is that the polymer material can be used as substrate and it is feasible to make lightweight and flexible thin-film solar cells. In this work, we used a co-evaporation technique with an introduction of Sb during the film deposition process to modify the film growth mechanisms and produce the CIS film with compact grain structure and smooth surface morphology. In most cases, there was only tiny amount of Sb existed in the film as a p-type dopant. In some cases, second phases of Sb compounds could be detected in the film as the Sb flux was kept too high during the film deposition stage. The I-V characteristics measured under the AM1.5 condition for the solar cell using a CIS:Sb film as the absorber showed that the open circuit voltage (Voc) was 0.364 V, short circuit current (Jsc) was 48.16 mA/cm2, fill factor (FF) was 44.5%, and energy conversion efficiency (£b) was 8%. The device with the same layer structure except the use of CIS film prepared without the addition of Sb and at a higher substrate temperature of 550oC had a comparable device performance but a slightly lower efficiency, i.e. Voc=0.325 V, Jsc=48.54 mA/cm2, FF=45.1%, £b=7.4%. It is clear that a lower temperature process using Sb to modify the growth process can be successful to obtain a device quality CIS layer. In addition, a CIGS thin-film solar cell was also fabricated and its device properties were Voc=0.392 V, Jsc=37.28 mA/cm2, FF=46.2%, and £b=7.0%. We see that the addition of Ga to increase the bandgap do increase the Voc and decrease the Jsc. However, a low efficiency of this cell indicates that further improvement in fill factor of the cell is a necessary.

Using low-temperature (650¢J) Ce3+:YAG doped glass (LTCeYDG) phosphor layer instead of conventional Ce:YAG doped silicone phosphor layer applied to high-power phosphor-converted white-light-emitting diodes (PC-WLEDs) is demonstrated.The glass transition temperature (Tg) of silicone is 150¢J but glass is 750¢J,it shows the glass were employed in high power LED than silicon. The uniformity of phosphor powder doped glass is an important item to discriminates between good and bad. Quantize the uniformity of glass phosphor by image processing software and Distribution Uniformity (Du). Calculate the uniformity of phosphor powder mix with glass powder which has different particle size and measurement optical properties of glass phosphor which has different uniformity. The Du of glass phosphor are 64.46%, 84.65%, 85.24% , 91.85% and the quantum efficiency are 18.49%, 28.31%, 29.73%, 28.56% ,respectively. By using Ceramic tube and low temperature glass powder sintering glass phosphor is a new fabrication. Compare with last fabrication, new fabrication reduce 100¢Jfabrication temperature from 750¢J to 650¢J, 70% material savings and high luminous efficiency. The quantum efficiency and lumen per watt were improved about 7 percentage point from 22.3% to 29.1% and 4.2 lm/W from 36.4 lm/W to 40.68 lm/W. We used the XRD to analyze the glass phosphor of last fabrication and new fabrication and the results show that the higher thermal stress destroys the structure of YAG, lower fabrication temperature used to get higher luminous efficiency.

The study focus on low temperature process with doping antimony to refine the quality of the CI(G)S thin film, and doping gallium to increase energy band gap in two-stage co-evaporation process. Furthermore, we discuss about the variety of crystal structure, and recognize the value of energy band gap in transmission spectra. It has been achieved to increase the energy band gap of material with doping gallium. Recognizing the shift of XRD pattern and research result from papers, I estimate the content ratio of gallium in ¢»A atoms is 0.28~0.29, near my establishment ratio 0.3. By tuning the molecular beam flux of antimony effusion cell from 1.1¡Ñ1013 atoms/cm2second to 2.2¡Ñ1014 atoms/cm2second , to find out the property content of antimony involving of co-evaporation to optimize the quality of the CI(G)S polycrystalline thin film. We just observed that the thin film with antimony involving make effect of smoother and denser surface morphology. In our study, we also try discontinue supplying the antimony vapor to reduce the amount of antimony which involves the reaction process, and make low content of antimony leaved in the CI(G)S thin film. Here, We found out a special effect of the grain- growth of the CI(G)S thin film supplying antimony continually or not in the process. It should be strong (112) prefer orientation when we deposit the thin film using SLG substrate. However, we found out that antimony enhance the (220/204) .

Microfluidics emerged during the early 1990s with channel networks in silicon or glass. Microprocessing of these materials is labor-intensive and time-consuming, it requires sophisticated equipment in a clean room, and often involves hazardous chemicals. The subsequent use of polymer greatly simplified the fabrication of microchips and led to the rapid development of the field. Polymer such as poly(dimethylsiloxane) (PDMS), has other attractive properties, such as being elastic (easy to make efficient microvalves), permeable to gases, and compatible with culturing biological cells. Despite these advantages, applications of PDMS chips are severely limited by a few drawbacks that are inherent to this material: (i) strong adsorption of molecules, particularly large biomolecules, onto its surface; (ii) absorption of nonpolar and weakly polar molecules into PDMS bulk; (iii) leaching of small molecules from PDMS bulk into solutions; and (iv) incompatibility with organic solvents. To overcome all these problems, Teflon plastics seem to be the perfect solution. They are well-known for their superior inertness to almost all chemicals and all solvents; they also show excellent resistance to molecular adsorption and molecule leaching from the polymer bulk to solutions. However, Teflon has a high chemical inertness of the surface, which is restricted the bonding temperature (>260¢XC).It is not conducive to the low-temperature packaging process. This study presents a simple and rapid process for sealing Teflon-based microfluidic chip at a temperature of 140oC which is lower than typical bonding temperature of 260oC. A simple ammonium plasma treatment is used to enhance the surface energy of Teflon substrates such that the bonding temperature can be greatly reduced. Results indicate that the ammonium plasma treated Teflon substrates can be sealed using hot press bonding at a temperature of 140oC for 20 min. The measured iv bonding strength for the Teflon-based microfluidic devices is higher than those bonded at a reported temperature of 260oC for 60 min. It shows the measured contact angle for the Teflon substrates treated with different plasmas. Results indicated that the ammonium hydroxide plasma exhibited the best wettability property and the contact angle reached the minimum value of 45o after 5 min of treatment. The ESCA analysis showed the best Defluorination by ammonium plasma. The fluorine/carbon atomic ratio degraded from 1.96 to 1.10 by 5 minutes. The measured bonding strength for the Teflon substrates bonded with different surface activation protocols. Results showed that the bonding strength was enhanced upto 93% after the plasma treatment. The plasma treatment not only enhanced the bonding strength but also reduced the bonding temperature and time. The measured surface roughness only increased 15¡Ó5 nm (Ra) after the plasma treatment, which is acceptable for most applications in microfluidic systems. Finally, the fluorescence optical architecture and cross-chip successfully detected and isolated £XX-174 fragment of DNA samples confirmed the Teflon substrate for the emerging microfluidic plastic chip. The developed method provides a simple and rapid way to fabricate Teflon-based microfluidic devices.