Synthesis of Novel Earth-Abundant Cu<SUB>2</SUB>BaSn(S,Se)<SUB>4</SUB> Chalcogenide and its application to Thin-Film Solar Cells

Donghyeop Shin(신동협) 한국신재생에너지학회 2017 한국신재생에너지학회 학술대회논문집 Vol.2017 No.9

영상 캡션과 장면 그래프 생성을 위한 맥락 정보 활용

신동협(Donghyeop Shin),김인철(Incheol Kim) 한국정보과학회 2018 한국정보과학회 학술발표논문집 Vol.2018 No.6

Solution-Processed Ag Nanowires + PEDOT:PSS Hybrid Electrode for Cu(In,Ga)Se₂ Thin-Film Solar Cells

Shin, Donghyeop,Kim, Taegeon,Ahn, Byung Tae,Han, Seung Min American Chemical Society 2015 ACS APPLIED MATERIALS & INTERFACES Vol.7 No.24

<P>To reduce the cost of the Cu(In,Ga)Se<SUB>2</SUB> (CIGS) solar cells while maximizing the efficiency, we report the use of an Ag nanowires (NWs) + poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) hybrid transparent electrode, which was deposited using all-solution-processed, low-cost, scalable methods. This is the first demonstration of an Ag NWs + PEDOT:PSS transparent electrode applied to CIGS solar cells. The spin-coated 10-nm-thick PEDOT:PSS conducting polymer layer in our hybrid electrode functioned as a filler of empty space of an electrostatically sprayed Ag NW network. Coating of PEDOT:PSS on the Ag NW network resulted in an increase in the short-circuit current from 15.4 to 26.5 mA/cm<SUP>2</SUP>, but the open-circuit voltage and shunt resistance still needed to be improved. The limited open-circuit voltage was found to be due to interfacial recombination that is due to the ineffective hole-blocking ability of the CdS film. To suppress the interfacial recombination between Ag NWs and the CdS film, a Zn(S,O,OH) film was introduced as a hole-blocking layer between the CdS film and Ag NW network. The open-circuit voltage of the cell sharply improved from 0.35 to 0.6 V, which resulted in the best cell efficiency of 11.6%.</P><P><B>Graphic Abstract</B> <IMG SRC='http://pubs.acs.org/appl/literatum/publisher/achs/journals/content/aamick/2015/aamick.2015.7.issue-24/acsami.5b02989/production/images/medium/am-2015-02989b_0008.gif'></P>

2016 경주지진 단층파열방향성 추정

설동협(Donghyeop Seol),강태섭(Taeseob Kang) 대한지질학회 2018 대한지질학회 학술대회 Vol.2018 No.10

Effects of the incorporation of alkali elements on Cu(In,Ga)Se₂ thin film solar cells

Shin, Donghyeop,Kim, Jekyung,Gershon, Talia,Mankad, Ravin,Hopstaken, Marinus,Guha, Supratik,Ahn, Byung Tae,Shin, Byungha Elsevier 2016 Solar energy materials and solar cells Vol.157 No.-

<P><B>Abstract</B></P> <P>This study describes in detail the effects of sodium and potassium on Cu(In,Ga)Se<SUB>2</SUB> (CIGS) absorbers and solar cells. We report on the influence of these species on the surface and bulk composition as well as bulk defect structure of CIGS films as revealed by X-ray photoelectron spectroscopy (XPS), secondary ion mass spectroscopy (SIMS), and photoluminescence (PL). From the XPS studies it is found that Na and K promote oxygen absorption onto the CIGS films. Furthermore, potassium accelerates the formation of indium and gallium oxides on the film surface, making the surface Cu-deficient. Low temperature PL studies suggest that (i) Na and K help passivate non-radiative recombination centers, presumably at the grain boundaries, and (ii) Na further impacts the bulk defect structure inside of CIGS grains, which is not observed with K. This change in bulk defect structure is attributed to the greater diffusivity of Na in CIGS relative to K due to the smaller atomic size. This in-depth study (integration of XPS, SIMS, PL, and device characteristics) reveals that the surface chemistry and the grain boundary passivation have stronger influences on the device performance than the bulk defect structure.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Effects of alkali elements-Na and K-onCIGS are studied using SIMS, XPS, and PL. </LI> <LI> Alkali elements passivate non-radiative recombination centers at grain boundaries. </LI> <LI> The device with Na and K shows the highest efficiencies due to <I>surface</I> passivation. </LI> <LI> Passivation of both <I>external interfaces</I> and <I>internal grain boundaries</I> is crucial. </LI> </UL> </P> <P><B>Graphical abstract</B></P> <P>[DISPLAY OMISSION]</P>

맥락 정보를 이용한 시각 장면 이해

신동협(Donghyeop Shin),김인철(Incheol Kim) 한국정보과학회 2018 정보과학회논문지 Vol.45 No.12

본 논문에서는, 시각 장면 이해의 한 문제로서, 입력 영상들로부터 장면 그래프와 영상 캡션을 동시에 생성하는 문제를 다룬다. 장면 그래프는 영상 내 물체들과 그들 간의 관계들을 나타내는 정형 지식 표현이며, 영상 캡션은 주어진 영상에 담긴 장면을 서술하는 자연어 문장이다. 본 논문에서는 이러한 문제를 효과적으로 해결하기 위해, 맥락 정보를 서로 교환함으로써 서로 다른 두 가지 표현을 상호보완적으로 생성하는 새로운 심층 신경망 모델을 제안한다. 제안 모델은 물체 탐지, 관계 탐지, 캡션 생성 등 각기 다른 세 가지 계층들로 구성되며, 각 계층은 그 계층에 부여된 작업을 성공적으로 수행하기 위해 맥락정보를 적절히 활용한다. 제안 모델의 성능을 평가하기 위해, 대규모 벤치마크 데이터 집합인 Visual Genome을 이용한 다양한 실험들을 수행하였다. 이러한 실험들을 통해, 맥락 정보를 활용하는 제안 모델이 기존의 경쟁 모델들에 비해 높은 성능 향상이 있었음을 확인할 수 있었다. In this paper, as a visual scene understanding problem, we address the problem of generating corresponding scene graphs and image captions from input images. While a scene graph is a formal knowledge representation expressing in-image objects and their relationships, an image caption is a natural language sentence describing the scene captured in the given image. To address the problem effectively, we propose a novel deep neural network model, CSUN(Context-based Scene Understanding Network), to generate two different representations in a complementary way, by exchanging useful contexts with each other. The proposed model consists of three different layers, such as object detection, relationship detection, and caption generation, each of which makes use of proper context to accomplish its own task. To evaluate performance of the proposed model, we conduct various experiments on a large-scale benchmark dataset, Visual Genome. Through these experiments, we demonstrate that our model using useful contexts, achieves significant improvements in accuracy over state-of-the-art models.

Deep Image Understanding Using Multilayered Contexts

Shin, Donghyeop,Kim, Incheol Hindawi Limited 2018 Mathematical problems in engineering Vol.2018 No.-

<P>Generation of scene graphs and natural language captions from images for deep image understanding is an ongoing research problem. Scene graphs and natural language captions have a common characteristic in that they are generated by considering the objects in the images and the relationships between the objects. This study proposes a deep neural network model named the Context-based Captioning and Scene Graph Generation Network (C2SGNet), which simultaneously generates scene graphs and natural language captions from images. The proposed model generates results through communication of context information between these two tasks. For effective communication of context information, the two tasks are structured into three layers: the object detection, relationship detection, and caption generation layers. Each layer receives related context information from the lower layer. In this study, the proposed model was experimentally assessed using the Visual Genome benchmark data set. The performance improvement effect of the context information was verified through various experiments. Further, the high performance of the proposed model was confirmed through performance comparison with existing models.</P>

참조 표현 이해를 위한 물체간의 관계 모델링

신동협 ( Donghyeop Shin ),김인철 ( Inched Kim ) 한국정보처리학회 2017 한국정보처리학회 학술대회논문집 Vol.24 No.2

참조 표현이란 영상 내의 특정 물체를 가리키는 자연어 문장을 의미한다. 그리고 이러한 자연어 참조표현을 기초로, 한 영상에서 실제로 대상 물체의 영역을 찾아내는 일을 참조 표현 이해라고 한다. 본 논문은 참조 표현 이해를 위한 새로운 심층 신경망 모델과 학습 방법을 제안한다. 본 논문에서 제안하는 모델은 효과적인 참조 표현 이해를 위해, 참조 표현에서 언급하는 대상 물체와 보조 물체를 모두 고려할 뿐만 아니라, 두 물체간의 관계정보도 활용한다. 또한, 본 논문에서 제안하는 모델은 이러한 다양한 맥락 정보들을 참조 표현 의존적인 방식으로 가중 결합함으로써, 참조 표현에 부합하는 대상 물체 영역을 보다 정확히 탐지해낼 수 있도록 설계하였다. 본 논문에서는 대규모 참조 표현 데이터 집합인 Google Reffixp를 이용한 성능 비교 실험들을 통해, 제안하는 모델의 우수성을 확인하였다.

Study of ZnS/CIGS Hetero-interface for Cd-free CIGS Solar Cells

신동협(Shin, Donghyeop),김지혜(Kim, Jihye),고영민(Go, Youngmin),윤재호(Yun, Jaeho),안병태(Ahn, Byungtae) 한국신재생에너지학회 2011 한국신재생에너지학회 학술대회논문집 Vol.2011 No.05

The Cu(In,Ga)Se2 (CIGS) thin film solar cells have been achieved until almost 20% efficiency by NREL. These solar cells include chemically deposited CdS as buffer layer between CIGS absorber layer and ZnO window layer. Although CIGS solar cells with CdS buffer layer show excellent performance, the short wavelength response of CIGS solar cell is limited by narrow CdS band gap of about 2.42 eV. Taking into consideration the environmental aspect, the toxic Cd element should be replaced by a different material. Among Cd-free candidate materials, the CIGS thin film solar cells with ZnS buffer layer seem to be promising with 17.2%(module by showa shell K.K.), 18.6%(small area by NREL). However, ZnS/CIGS solar cells still show lower performance than CdS/CIGS solar cells. There are several reported reasons to reduce the efficiency of ZnS/CIGS solar cells. Nakada reported ZnS thin film had many defects such as stacking faults, pin-holes, so that crytallinity of ZnS thin film is poor, compared to CdS thin film. Additionally, it was known that the hetero-interface between ZnS and CIGS layer made unfavorable band alignment. The unfavorable band alignment hinders electron transport at the heteo-interface. In this study, we focused on growing defect-free ZnS thin film and for favorable band alignment of ZnS/CIGS, bandgap of ZnS and CIGS, valece band structure of ZnS/CIGS were modified. Finally, we verified the photovoltaic properties of ZnS/CIGS solar cells.

A Study on the Determination of Export Activities of Korean Manufacturing Companies : Focused on R&D Activities

Taeyang Kim,Donghyeop Shin ASCONS 2019 INTERNATIONAL JOURNAL OF EMERGING MULTIDISCIPLINAR Vol.3 No.4

For this study, the 2018 Korean Innovation Survey (KIS) by Science and Technology Policy Institute (STEPI) was used to analyze the impact on the exports of Korean manufacturing companies. The analysis consists of two main themes. One is to look at the factors influencing the decision on the export by using the logit model, and the other is to look at the factors influencing the decision on the export by region in the major overseas markets and its marginal effects by using the multinomial logit model. According to our result, the factors influencing the exports of the companies were classified largely into factors of R&D activities, corporate and industrial characteristics, and government support benefits. As a result of the analysis, for the R&D activity factor, internal R&D activities, R&D investment per capita, and proportion of personnel dedicated to R&D activities were found out to have a statistically significant positive (+) effect on the export activity.