We have succeeded in detecting nonradiative recombination (NRR) centers in InAlGaN multiple quantum wells (MQWs) for the sterilization wavelength at around 265 nm by our scheme of two-wavelength excited photoluminescence (PL). Samples studied are InAlGaN multiple quantum well structures with InAlGaN electron blocking layer grown on sapphire (0001) substrates by metal-organic chemical vapor deposition (MOCVD) technique at the growth temperature of 880 8C (sample A) and 920 8C (sample B). The MQW consists of three InAlGaN wells of 2 nm sandwiched by 7 nm InAlGaN barrier layers. With the addition of the below-gap excitation (BGE) light of 1.17 eV, the PL intensity decreased for the sample A but increased for the sample B. Both change in the PL intensity implies the existence of NRR centers, which were activated by the BGE. We attribute both intensity change to two-levels model and one level model, respectively. Based on rate equation analysis, a set of NRR parameters of sample B was determined by utilizing a saturating tendency of the PL intensity change. Spectroscopic and quantitative advantages of the method enable us to clarify energy distribution of NRR centers without providing electrode.
An investigation of two-wavelength excited photoluminescence on GaPN alloys containing 0.56% nitrogen was conducted to directly excite intermediate band (IB) states and monitor its impact on photoluminescence (PL) properties. The 22 K PL due to above-gap excitation (AGE) showed broad peak emission induced by the IB states. With the use of below-gap excitation (BGE) of 1.17 eV energy in addition to the AGE, the PL peak intensity was found to decrease linearly with increasing the BGE power, which suggests that the BGE perturbs the bound exciton recombination mechanism by exciting electrons from the IB states through the dissociation of excitons.
This paper addresses a novel unsupervised algorithm to rank numerical observations which is important in many applications in computer science, especially in information retrieval (IR). The proposed algorithm shows how correlation coefficients between attribute values and the concept of magnetic properties can be explored to rank multi-attribute numerical objects. One of the main reasons of using correlation coefficients between attribute values and the concept of magnetic properties is that they are easy to compute and interpret. Our proposed Unsupervised Ranking using Magnetic properties and Correlation coefficient (URMC) algorithm can use some or all the numerical attributes of objects and can also handle objects with missing attribute values. The proposed algorithm overcomes a major limitation of the state-of-the-art technique while achieving excellent results.
Metallic samples of Be and Si pair ␦-doped GaAs structures which undergo a metal-insulator transition with a decrease in the hole concentration are investigated by Hall resistance and magnetoresistance measurements. The anomalous Hall effect and negative magnetoresistance are observed from the samples in a temperature range above 70 K. Magnitudes of negative magnetoresistance and anomalous Hall resistance significantly vary among the samples, although their doping conditions are close to one another. Dependence of anomalous Hall resistance on the temperature and applied magnetic field is closely correlated to that of negative magnetoresistance for each sample. Their dependence is explained on the basis of a paramagnetic state of localized magnetic moments coexisting with itinerant holes in these samples. Both anomalous Hall effect and negative magnetoresistance decrease with lowering the temperature from 150 K and vanish at a temperature around 70 K, a possible origin of which is discussed.
Coronavirus disease 2019 (COVID-19) has made a huge pandemic situation in many countries of the world including Bangladesh. If the increase rate of this threat can be forecasted, immediate measures can be taken. This study is an effort to forecast the threat of present pandemic situation using machine learning (ML) forecasting models. Forecasting was done in three categories in the next 30 days range. In our study, multiple linear regression performed best among the other algorithms in all categories with R2 score of 99% for first two categories and 94% for the third category. Ridge regression performed great for the first two categories with R2 scores of 99% each but performed poorly for the third category with R2 score of 43%. Lasso regression performed reasonably well with R2 scores of 97%, 99% and 75% for the three categories. We also used Facebook Prophet to predict 30 days beyond our train data which gave us healthy R2 scores of 92% and 83% for the first two categories but performed poorly for the third category with R2 score of 34%. Also, all the models’ performances were evaluated with a 40-day prediction interval in which multiple linear regression outperformed other algorithms.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.