Semiconductor Tech

International Journal of Semiconductor Science & Technology (IJSST) ISSN 2250-1576 Vol.2, Issue 2 June 2012 1-17 © TJPRC Pvt. Ltd.,

INVESTIGATION ON MILD CONDITION

PREPARATION AND STRUCTURAL, OPTICAL AND

THERMAL PROPERTIES OF PVP CAPPED CDS

NANOPARTICLES

N.S. NIRMALA JOTHI, G. RAMALINGAM, R. GUNASEELAN, A.R.

BABY SUGANTHI AND P. SAGAYARAJ

Department of Physics, Loyola College, Chennai- 600 034, India.

ABSTRACT

Polyvinlypyrrolidone (PVP) capped cadmium sulphide (CdS)

nanoparticles are synthesized using a simple hydrothermal method. The

powder X-ray diffraction (XRD) result indicates that the nanoparticles are

crystallized in hexagonal phase. The optical properties are characterized

by Ultraviolet-Visible (UV-Vis) absorption and Photoluminescene (PL)

spectra. The transmission electron microscope (TEM) reveals that the

nanoparticles of CdS posses well defined morphology and high

crystallinity. The d-spacing measured from well resolved lattice fringes of

HRTEM ascertains the structure of CdS nanocrystals. The morphology

and composition of CdS nanoparticles are investigated using Scanning

electron microscope (SEM) and Energy dispersive x-ray analysis (EDAX)

respectively. The thermal behavior of the as prepared nanopowder has

been studied by Thermo gravimetric analysis (TGA).

2 N.S. Nirmala Jothi, G. Ramalingam, R. Gunaseelan, A.R. Baby Suganthi and P. Sagayaraj

KEYWORDS: Nanostructure; Chemical synthesis; Electron

microscopy; Powder diffraction; Optical properties;

1. INTRODUCTION

Recently, the synthesis of inorganic nanocrystals has attracted

much interest due to their strong size dependent, special optical,

electronic properties and potential applications in solar cells, light

emitting diode, nonlinear optical materials, optoelectronic and electronic

devices, biological labelling, thermoelectric coolers, thermo-electronic

and optical recording materials, etc. [1, 2]. These properties and the

applications are largely dependent on the size, shape and the impurities of

the nanomaterials.

*CORRESPONDING AUTHOR

Dr. N.S.Nirmala Jothi, Assistant Professor of Physics, Loyola College, Chennai – 600 034, India

Email: [email protected]

As a typical semiconductor material of the II-VI group, cadmium

sulphide (CdS) nanocrystals have been widely investigated [3, 4]. CdS is

a direct band gap semiconductor with the band gap of 2.4 eV for bulk

hexagonal (Wurtzite) structure and 2.38 eV for bulk cubic (zinc blend)

structure [5, 6]. CdS nanopartilces are considered to be one of the model

systems for investigating the unique optical and electronic properties of

quantum confined semiconductors. However, the lack of adequate

synthetic methods for producing the desired high quality nanoparticles is

Investigation on mild condition preparation and structural, 3 optical and thermal properties of PVP capped CdS nanoparticles

currently a bottleneck in this field. Over the past two decades, numerous

colloidal chemistry (or solution chemistry) methods have been developed

for the preparation of cadmium sulfide nanocrystals [7]. The solution

chemistry synthesis of CdS nanocrystals utilizes the organic stabilizers to

cap surface atoms of nanoparticles in order to control the growth process.

The kind of stabilizers is of great importance, since, it affects the

chemical as well as the physical properties of the semiconductor

nanocrystals, from stability to solubility to light emission. Modifying the

surfaces of nanoparticles with various organic, inorganic, species is

projected to remove their surface defects and subsequently, influence their

property. Organic capping of nanoparticles with surfactants would give

rise to a barrier to aggregation and electronic passivation of the particles.

The ability to segregate between the zinc blend (metastable phase) and

wurtzite structures is considered as the main key.

In aqueous phase synthesis of CdS nanoparticles, the

conventionally homogeneous phase arrested precipitation with the use of

phosphates, various thiols or hydrophilic polymer as capping reagents are

usually adopted [8]. Among the various traditional synthesis approaches

of the nanomaterials, the solvothermal/hydrothermal methods have been

widely applied to improve the crystallinity of nanosized particles.

Solvothermal synthesis is one of the most efficient methods used to

synthesize CdS with different morphologies. It is simple, convenient and

inexpensive. Some polymers such as PEG, PAA, PAN, PVP and PVA

have been used to modify the surface chemistry of the crystals and the

concentration of soluble species for crystal growing [9-12]. The

complexing agent (PVP), can cap the particle surface to prevent the

colloidal particle from agglomeration and play an important role in the

formation of the nanoparticles. The adsorption of PVP onto the

nanoparticles can solubilise the formed nanostructure into water. This


may be attributed to the hydrophilicity of the PVP molecules and the

repellence between positively charged PVP molecules adsorbed onto

separated nanoparticles. Competitive reaction kinetics between the growth

of CdS and termination of growth by capping the surface is considered to

induce the growth of CdS nanoparticles [13].

Recently, a wide range of cadmium sulfide (CdS) 3D

polycrystalline walnut-like nanocrystals were prepared by solvothermal

method with polyvinylpyrrolidone (PVP) as stabilizer [14]. The large

surface area-to-volume ratio, along with the ability to tune the band gap

makes the semiconductor nanoparticles like CdS to be used as sensitizers

and catalysts in photochemical reaction being universally accepted.

In spite of the concentrated efforts to develop high quality CdS

nanoparticles, there are still challenges to further simplify the synthesis

procedures so as to encourage the mass production and the use of

hydrothermal synthesis route is one of the best options to move towards

this goal. Further, the use of water as a solvent offers several advantages,

compared with non-aqueous synthesis, aqueous is more reproducible, low

cost, environment friendly and the “as-prepared” samples are more water

soluble and bio-compactable [8]. In-situ surface modification under

hydrothermal conditions helps significantly in preparing nanocrystals

with a highly controlled size, shape and dispersibility. This article deals

with the preparation of CdS nanoparticles by a simple hydrothermal

method and the influence of water soluble polymeric capping agent like

polyvinylpyrrolidone (PVP) on the crystalline quality, morphology, size

and crystalline phase is investigated. The sample was systematically

characterized by powder XRD, UV-Vis absorption and

photoluminescence spectroscopy, SEM and TEM, and thermal analysis.


2. EXPERIMENTAL

2.1. Materials

Cadmium nitrate (Cd (NO3)2.4H2O) Merck and thiourea

(H2NCSNH2) are used as the starting materials. All chemical are of high

purity and no further purification is done. Polyvinylpyrrolidone (PVP) is

the capping ligand.

2.2. Synthesis of CdS nanoparticles modified with

polyvinylpyrollidone

The stoichiometric ratio of the starting materials (Cd (NO3)2.4H2O

and H2NCSNH2) was kept as 1:3. Polyvinylpyrollidone (PVP) is a water

soluble polymer used here as the capping ligand. Initially, 2 g of PVP was

taken and dissolved in 20 ml of water and then cadmium nitrate solution

was added and mixed thoroughly in 75 ml of water. In a similar way

thiourea was dissolved in 75 ml of water and stirred magnetically. These

two solutions were mixed together for about 1 h. Ammonia was directly

added in the solution until the pH of the solution reaches 12. The final

solution was transferred into a 200 ml Teflon coated autoclave. The

autoclave was placed in an electrical oven and maintained at 180 °C for 6

hours. After that the autoclave is removed from the oven and naturally

allowed to cool down to room temperature. The yellow precipitate was

washed continuously with water and ethanol several times so as to

remove the excessive thiourea and other by products. The as prepared

powder is grounded in a mortar and dried with a temperature of 80 °C

nearly for 4 hours.

2.3. Characterization

The room temperature powder XRD pattern for the as prepared

sample was done using RICH SEIFER with monochromatic nickel


filtered CuKα (λ=1.5461 Å) radiation. The UV-Vis absorption spectral

studies were carried out using VARIAN CARY 5E UV-Vis-NIR

SPECTROPHOTOMETER in the spectral region of 200 and 800 nm. The

photoluminescence spectra of the samples were recorded with a VARIAN

CARY 5E UV-Vis-NIR SPECTROPHOTOMETER. Particles size and

distribution analyses were carried out with TEM model JEOL JEM 3010

at an accelerating voltage of 200 kV. For the TEM observations, the

sample was dispersed in ethanol and ultrsonicated for 30 minutes and then

it was kept on a carbon coated grids. Scanning electron microscope

(SEM) was employed for morphological study using JEOL JSM 6310

operated at 10 kV with Energy Dispersive X-ray analyzer (EDX).

Thermogravimetric (TG) and Differential thermo gravimetric analysis

(DTG) for the air dried sample was performed on a SDT Q600 with a

heating rate of 20 °C min-1.

3. RESULTS AND DISCUSSION

3.1 Powder XRD analysis

Fig. 1 shows typical XRD pattern of obtained CdS nanocrystals

with Millipore water as a solvent and polyvinylpyrrolidone (PVP) as

capping reagent. All the diffraction peaks are conveniently indexed to

hexagonal structure. Compared with the standard card (JCPDS card (PDF

No.80-0006), the (0 0 2) diffraction peak, the second strongest peak in

bulk hexagonal CdS, were unusually strong and narrow, which may be

ascribed to the preferential growth along [0 0 1] hexagonal CdS

crystallites. The relatively broad peaks probably resulted from the smaller

dimensions of the other surfaces [15]. The corresponding lattice constants

are a = b = 4.121 Å and c = 6.682 Å. There are reports on the role of


different dosages of PVP on the resulting structure and orientation growth

[15]. In the present study, the preparation carried out with 2 g of PVP/(20

ml) resulted in good quality nanoparticles with best orientation growth.

Fig1:XRD pattern of PVP capped CdS nanocrystals.

3.2 Energy dispersive X-ray analysis (EDAX)

EDAX is an important technique to analyze the composition of

elements quantitatively and solve the chemical identity of any

nanomaterial. It is inferred from the result of the EDAX spectrum (Fig. 2)

obtained for nanoparticles prepared using PVP, that the sample is

composed of only Cd and S which are exactly CdS nanoparticles and no

trace of other elements is observed. From the Fig. 2 it is clear that the

sample is generally cadmium rich even though a relatively higher

concentration of sulphur was used than the cadmium to synthesize CdS

nanoparticles. The fact that the nanoparticles are richer in cadmium than


in sulphur suggests that their surface should be mainly composed of

cadmium atoms. From the EDAX and XRD analyses, it is clear that the

obtained product is pure cadmium sulphide in wurtzite phase.

Fig 2.EDAX pattern of CdS nanoparticles modified with

PVP capping agent.

3.3 UV-Vis absorption spectroscopy

The controlling and tuning of band edge emission and surface traps

state emission of CdS nanocrystals are obviously very important to realize

the tunable optical properties and laser emission [16]. The UV-Vis

spectral analysis was carried out between 200 nm and 800 nm. Fig. 3

shows the absorption spectrum of CdS nanoparticles prepared with

surfactant PVP. The absorption band edge was shifted to 490 nm and the


corresponding band gap is 2.53 eV which is higher compared to bulk CdS

band gap (2.4 eV). Thus it is clear from the optical absorption study that

the capping of CdS with PVP modifies the band gap of the CdS

nanoparticles and the sample is blue shifted when compared with the bulk

CdS (512 nm).

200 300 400 500 600 700 800

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

Abs

orba

nce

W ave length (nm )

Fig 3.Uv-vis absorption spectrum of PVP capped CdS nanoparticles.

3.4 Photoluminescence (PL) study

One of the interesting aspects of the photo-physical properties is

the photoluminescence (PL) of CdS nanocrystallites. The PL studies on

CdS nanocrystallites have been investigated by various research groups

[17]. In general, the reported emission spectrum consists of two broad

bands in the range 400-520 nm and then 520-800 nm. Usually, the

peaking is observed at around 480 and 650 nm respectively [18].

Recently, Li et al investigated the room temperature luminescence of CdS


nanostructured materials prepared with various sulphur sources and

studied the influence of S sources on the green and red emission spectra

of the nanopowder [19]. The PL property is influenced by the structure,

composition, particle size and morphology of the CdS nanoparticles. In

addition, the method of preparation has marked influence. Fig. 4 shows

the PL spectrum of CdS nanoparticles prepared with PVP. It is evident

from the PL spectra that the emission peak at 505 nm for the sample can

be assigned to the surface trap induced fluorescence which involved the

recombination of electrons trapped inside a sulphur vacancy with the hole

in the valence band of CdS nanoparticle. The emission peak observed at

505 nm for CdS synthesized by PVP assisted solvothermal method goes

well with the earlier report [15].

Fig4. PL spectrum of PVP capped CdS nanoparticles.


3.5 SEM analysis

The SEM image of CdS nanocrystals with water as a solvent and

polyvinylpyrrolidone as capping reagent is shown in Fig. 5. The

solvothermal temperature along with the capping agent can influence the

nanoparticle size. At the lower temperature around 80oC only irregular

nanospheres of large diameter are usually formed. A close observation of

the SEM image of the present case suggests that the surfaces of the

nanospheres are relatively smooth and there are few prolated spheres as

well and this could be attributed to the relatively high reaction

temperature employed. It reveals that PVP played an important role in

controlling the size and mono-dispersion of the CdS nanocrystals in this

process. The absence of agglomerates is attributed to the role played by

PVP. In the formation process of the shape evolution of CdS architectural

structure, the capping agent PVP is adsorbed onto the different planes of

the incipient CdS nuclei and it not only prevents the particles from

agglomeration, but also influences the growth of these planes [14].

Fig5. SEM image of PVP capped CdS nanoparticles.


3.6 Transmission electron microscopy (TEM)

The transmission electron microscopic analysis allows one to

visualize particles at nanosize regime with high degree of accuracy, it

offers better understanding about growth aspects and helps to analyze the

actual size of the particles, shape and growth pattern.

The TEM micrograph of the PVP capped CdS nanopowder is

shown in Fig. 6, which suggests the formation of spherical as well as

short rods of nanoparticles. Further, the side faces of the product are not

smooth and the particles and the level of agglomeration are slightly on the

higher side and the short rods are not clearly visible. In the present case,

the aspect ratio of the nanorods formed is only 3-4 and this is possibly due

to the short duration of the reaction time employed. In the HRTEM

image, a representative nanorod of width 6 nm and 25 nm length is clearly

visible. There are reports on the capping mechanism of PVP in growth

process of CdS nanoparticles and also the influence of PVP dosage in

tailoring the shape and size of the CdS nanostructures. The best dosage of

PVP for the orientation growth of CdS nanocrystal was optimized as 0.8 g

for 50 ml by Qingqing et al [15]. The HRTEM image shows the presence

of few short nanorods which are seen crosslinked due to agglomeration, in

spite of this, we notice in Fig. 6 (iv), the formation of well arranged lattice

fringes in a single nanorods. Thus the problem of agglomeration and

improving the aspect of the nanorods are the issues yet to be addressed.

The SAED pattern of the CdS nanoparticles is presented which confirms

the hexagonal (Wurtzite) phase of CdS. The pattern consists of diffraction

rings corresponding to particle size and morphology.


Fig.6. TEM and HRTEM images of CdS nanoparticles capped with

polyvinylpyrrodoline (PVP). (i) Nanoclusters with small CdS

nanoparticles (Inset SAED pattern), (ii, iii and iv) HRTEM images of one

dimensional CdS with lattices fringe


3.7 Thermal analysis

The thermal behavior of the semiconductor nanoparticles of CdS

was studied by employing TG/DTA analysis. Since the temperature plays

an important in the formation of nanostructured materials, temperature

induced phase changes are important for the utility of these nanoparticles

for various applications. The TG analysis of the sample CdS obtained

with PVP was done at the heating rate of 20°C/min. The TGA graph (Fig.

7) of the sample synthesized using PVP as capping agent shows the sharp

weight loss of about 3.277 % at 525 °C.A gradual weight loss of about

1.246 % was obtained between 525 °C and 840 °C.

Fig.7. TG/DTA traces of PVP capped CdS nanoparticles

4. CONCLUSION

Semiconductor nanoparticles of CdS are successfully prepared

under solvothermal/hydrothermal conditions in surfactant assisted


synthesis. The powder XRD result show that particles are purely

crystallized in hexagonal phase with the broadening of diffraction peaks is

attributed to nanoscale size of the particles. From the optical absorption

spectra, the blue shift of 490 nm as compared to bulk counterpart is due to

the quantum confinement effect. The broad emission band observed in the

PL spectrum for PVP capped CdS nanoparticles is assigned to the surface

trap induced fluorescence which involved the recombination of electrons

trapped inside a sulphur vacancy with the hole in the valence band of CdS

nanoparticle. From TEM, it is evident that the nanoparticles of CdS

exhibit well defined morphology and high crystallinity. The HRTEM

result reveals the well resolved lattice fringes of CdS nanoparticles. Thus,

the present study demonstrates that the hydrothermal synthesis method

with capping ligand is one of the successful routes for obtaining good

quality CdS nanoparticles.

ACKNOWLEDGMENTS

The authors are grateful to UGC for the instrumentation facility

provided at Loyola College through a project (F38-119/2009(SR)). The

authors thank Prof. B.S.Murty, Department of Metallurgy and Materials

Science, Indian Institute of Technology, Madras for TEM facility and for

useful suggestions.

REFERENCES

[1] L.E. Brus, J. Chem. Phys. 80 (1984) 4401.

[2] T.S. Ahmadi, Z.L. Wang, T.C. Green, A. Henglein, M.A. El-

Sayed, Science, 272 (1996) 1924.


[3] Z.A. Peng, X. Peng, J. Am. Chem. Soc.123 (2001) 183.

[4] B.A. Korgel, H.G. Monbouquette, J. Phys. Chem. 100 (1996)

346.

[5] L. Zeiri, I. Patla, S. Acharya, Y. Golan, S. Efrima, J. Phys.

Chem. 111 (2007) 11843.

[6] H. Cao, G. Wang, S. Zhang, X. Zhang, D. Rabinovich, Inorg.

Chem. 45 (2006) 5103.

[7] L. Zou, Z. Fang, Z. Gu, X. Zhong, J. Lumin. 129 (2009) 536.

[8] J. Yang, C. Xue, Y. Qian, Angew. Chem. Int.Ed. 41 (2002) 4697.

[9] Qito Zhao, Lisong Hou, Ruin Huan, Inorganic Chem. Commun.,

6 (2003) 971.

[10] W. Qingqing, X. Gang, H. Gaorong, J. Solid state Chem., 178

(2005) 2680.

[11] Titipum Thongtem, Anukorn Phuruangrat, Somachi Thongtem,

Ceramics International, 35 (2009) 2817.

[12] XU Guo-yue, Wang Han, Cheng Chuan-wei, Zhang Hai-qian,

CAO Jie-ming, JI Guang-bin, Trans. Nonferrous Met. Society of

China, 16 (2006) 105.

[13] Y. D. Wu, L. S. Wang, M. W. Xiao, X. J. Huang, J. Non-cryst.

Solids, 354 (2008) 2993.

[14] Q. Xia, X. Chena, K. Zhao, J. Liu, Mater. Chem. Phys. 111

(2008) 98.

[15] W. Qingqing, Z Gaoling, H Gaorong, Mater. Lett. 59 (2005)

2625.


[16] Z. Jun, D. X. Wei, L. Z. Liang, X. Gang, X. S. Ming, Z. X. Ping,

Trans.Nonferrous Met.Soc.China, 17 (2007) 1367.

[17] P.Zhao, K Huang, Cryst. Growth Des. 8 (2008) 717.

[18] S. Karan, B. Mallik, J. Phys. Chem. C. 111 (2007) 16734.

[19] F. Li, W. Bi, T. Kong, C. Wang, Z. Li, X. Huang, J. Alloys

compd. 479 (2009) 707.

FIGURE CAPTION

Fig.1. Powder XRD pattern of PVP capped CdS nanoparticles

Fig.2. EDAX pattern of CdS nanoparticles modified with PVP capping

ligand

Fig.3. UV-Vis absorption spectrum of PVP capped CdS nanoparticles

Fig.4. PL spectrum of PVP capped CdS nanoparticles

Fig.5. SEM image of PVP capped CdS nanoparticles

Fig.6. TEM and HRTEM images of CdS nanoparticles capped with

polyvinylpyrrodoline (PVP). (i) Nanoclusters with small CdS

nanoparticles (Inset SAED pattern), (ii, iii and iv) HRTEM

images of one dimensional CdS with lattices fringes.

Fig.7. TG/DTA traces of PVP capped CdS nanoparticles

International Journal of Semiconductor Science & Technology (IJSST) ISSN 2250-1576 Vol.2, Issue 2 June 2012 19-41 © TJPRC Pvt. Ltd.,

STUDY AND MODELING OF THE TRANSPORT MECHANISM IN A SCHOTTKY DIODE ON THE BASIS OF GaAs SEMI INSULATING.

MMrr..AA..RREESSFFAA ** .. BOURZIG Y.SMAHI. BRAHIMI.R.MENEZLA.

Corresponding Author Tel:213 6 67 60 17 80 Email address: [email protected]

Laboratory of modeling and conception of the circuits electronic,

department of electronics. University Djillali Liabès. BP89, Sidi Bel Abbes 22000.ALGERIA.

ABSTRACT:

The current through a metal–semiconductor junction is mainly due

to the majority carriers. Three distinctly different mechanisms exist in a

schottky diode: diffusion of of the semiconductor carriers in metal,

thermionic emission-diffusion (TED) of carriers through a Schottky gate,

and a mechanical quantum that pierces a tunnel through the gate. The

system was solved by using a coupled Poisson-Boltzmann algorithm.

Schottky BH is defined as the difference in energy between the Fermi

level and metal band carrier majority of the metal - semiconductor

junction to the semiconductor contacts. The insulating layer converts the

MS device in an MIS device and has a strong influence on its current–

voltage (IV) and the parameters of a Schottky barrier from 3.7 to 15 eV.

There are several possible reasons for the error that causes a deviation of

the ideal behaviour of Schottky diodes with and without an interfacial

insulator layer. These include the particular distribution of interface

states, the series resistance, bias voltage and temperature. The GaAs and

20 MMrr..AA..RREESSFFAA,, BOURZIG Y.SMAHI AND BRAHIMI.R.MENEZLA

its large concentration values of trap centers will participate in an increase

of the process of thermionic electrons and holes, which will in turn the

IV characteristic of the diode, and an overflow maximum value

[NT=3×1020] is obtained. The I–V characteristics of Schottky diodes are

in the hypothesis of a parabolic summit.

KEYWORDS: The electrocstatic potentiel and density of carriers, The

current thermionic emission-diffusion (TED) and The current tunnel

through the gate.The current-voltage(IV) characteristics of Schottky

diodes,and the temperature.

1.1. INTRODUCTION

The operation of semiconductor components at high temperature

and high frequency such as Schottky diodes and PN is usually described

by a set of features for implementation of voltage (IV) whose analysis

provides some information on electricity from the transport mechanism.

The determination of the model parameters that are fundamental to

the Schottky diode enable us ; to know the height of the gate, the factor of

ideality, and the resistance set, whish plays an important role in the

conception and the manufacture of the semiconductor devices such as

photopiles.

The goal of this thesis is to contribute to the mathematical

modeling and the simulation of greatly

inhomogeneous semiconductor devices.

Study of the physical origin of currents in the I–V to low

temperature of Schottky diode on the basis of a GaAs. One finds that

these excess currents are due to the generation of network shortcomings

STUDY AND MODELING OF THE TRANSPORT MECHANISM 21 IN A SCHOTTKY DIODE ON THE BASIS OF GaAs SEMI INSULATING

close to the metal–semiconductor interface at the time of irradiation, One

warming heat or of an external distortion. These shortcomings produce

levels of trap in some parts of the space load region.

This paper investigate the diffusive limit of the Boltzmann

equations, to get a second order approximation of the concentration of the

carriers. The drift-diffusion equations are unchanged, but a correction of

the boundaries of layers, which is proportional to current flow, appears in

the boundary conditions for concentration. The proportionality coefficient

is calculated by solving the spectral method.

These limiting and classical conditions are compared numerically

on a physical problem. Thi paper is also dedicated to the extention of the

particle simulation programs for the treatment of thes inhomogeneous

structures. In these devices, the dynamics are governed by the limiting

conditions that need to be taken precisely into account. Geometry is one-

dimensional in space and three-dimensional with axisymmetry in a wave

vector. It is therefore about solving the system coupled of Boltzmann–

Poisson for the modeling of a Schottky diode, in the cases united and

multi-dimensional respectively. What will permit to explain the numeric

results obtained are explained later.

The system was solved by using a coupled Poisson–Boltzmann

algorithm. Schottky BH is defined as the difference in energy between the

Fermi level and the metal band majority carrier of a metal semiconductor

junction to the semiconductor contacts[1-4]. These include the particular

distribution of interface states[2,5], the series resistance[6-8], bias

voltage[6-10] and temperature[1, 2, 6,9,11].


1.1.1. SCHOTTKY CURRENT DIODE

The theory of diffusion supposes that the driving force is

distributed olong the length of the layer of the depletion. The theory of the

thermionic emission-diffusion (TED), only applies to energetic carriers,

which have energy equal to or bigger than the energy of the conduction

strip to the interface of the metal–semiconductor. Quantum mechanics

that pierces a tunnel through the gate, into account the wave nature of the

electrons. In a given junction, a combination of all three mechanisms

could exist[12-14]. However, typically there is only one dominant current

mechanism. The analysis reveals that the diffusion and thermionic

emission-diffusion (TED) can be written in the following form[5,7,15]:

( )( )1expexp... −⎟⎠⎞

⎜⎝⎛−= Vt

VaVtBNcqJ n

φν . (1)

This expression affirms that the current is the product of the

electronic load, q, a speed, v, and the available carrier density in the

semiconductor located next to the interface. Speed equal to the mobility

multiplied by the electric field at the interface, the diffusion current and

the speed of Richardson and the current of the thermionic emission-

diffusion (TED) [16-18]. It ensures that the current is zero so no voltage is

applied in thermal equilibrium.

The current tunnel is of a similar shape, to know:

Θ= nqJn Rν , (2)

where the vR is the Richardson velocity, q the electronic charge, and n is

the density of carriers in the semiconductor. The current tunnel has the

term of the probability Θ, is added since the total current depends on the


carrier' flux arriving at the tunnel gate multiplied with the probability, Θ,

that it pierces a tunnel through the gate.

1.1.2. Diffusion current

This analysis supposes that the depletion layer is big compared to

the middle free trajectory, so that the concepts of movement and diffusion

are valid. The density currents are thus obtained as.

( ) ( )[ ]1expexp22

−⎟⎠⎞

⎜⎝⎛−−= Vt

VaVt

NdVaqVtDnNc

Jn B

s

iq φε

φ . (3)

The current depends on the applied voltage, Va, and exponentiality

on the height of the gate, φB, therefore the prefactor can be consisted more

easily if one rewrites it as function of the electric field on the interface of

the metal–semiconductor max:

( )

s

NdVaiqε

φε −= 2max , (4)

( )[ ]1expexp. max −⎟⎟

⎠

⎞

⎜⎜

⎝

⎛−= Vt

VaVtNcnqJn Bφεμ , (5)

so that the prefactor equals the current of the movement to the interface of

the metal–semiconductor.


1.1.3. Current thermionic emission-diffusion

The theory of thermionic emission-diffusion (TED) supposes that

electrons, with energy bigger than the summit of the gate, will cross the

well stocked gate that is displaced toward the gate. The real shape of the

gate is ignored by this. The current can be expressed as:

⎟⎠⎞⎜

⎝⎛ −= − 12* eeTAJ Vt

VaVt

MS

Bφ, (6)

where h

KmAq

3

2** ...4π= is the effective Richardson constant, q the

electronic charge, k Boltzmann constant,

T the absolute temperature, and φB is the height of the Schottky gate.

The expression for the current due to TED can also be written as function

of the middle speed with which the electrons approach the gate interface.

This speed is known as the Richardson speed and is given by:

mKT

R πν 2= . (7)

So that the current density becomes:

( )[ ]1expexp... −⎟⎟

⎠

⎞

⎜⎜

⎝

⎛−= Vt

VaVtNcqJn B

R

φν . (8)

1.1.4. Current tunnel

The current tunnel is obtained with the product of the velocity and

density. The velocity is the Richardson velocity. The carrier's density


equals the available electron density, n, multiplied with the probability of

piercing a tunnel, Θ, to give: ... Θ= nqRnJ ν (9)

Here the probability tunnel is obtained by:

( )⎟⎟⎟

⎠

⎞

⎜⎜⎜

⎝

⎛−=Θ ε

φ 2/3*..234exp Bmq

h, (10)

and the electric field equals ε = φB/L. Therefore the current tunnel

exponentially depends on the height of gate, φB, to the 3/2.

1.2. The ohmic contact

A metal–semiconductor junction results in an Ohmic contact (a

contact with voltage independent resistance) if the Schottky gate height,

φB, is zero. In such case, the carriers are free to flow in or out of the

semiconductor so that there is minimal resistance through the contact. For

an n-type semiconductor, it means that the workfunction of metal must be

close to or smaller than the electron affinity of the semiconductor. A p-

type semiconductor, requires that the workfunction of the metal must be

close to or bigger than the sum of the electron affinity and the bandgap

energy. It can be problematic to find a metal that provides a p type Ohmic

contact with a semiconductor with a large bandgap such as GaN or SiC.

1.2. The ohmic contact

A metal–semiconductor junction results in an Ohmic contact (a

contact with voltage independent resistance) if the Schottky gate height,

φB, is zero. In such case, the carriers are free to flow in or out of the

semiconductor so that there is minimal resistance through the contact. For

an n-type semiconductor, it means that the workfunction of metal must be

Moses

Rectangle

repeating (refer previous paragraph)


close to or smaller than the electron affinity of the semiconductor. A p-

type semiconductor, requires that the workfunction of the metal must be

close to or bigger than the sum of the electron affinity and the bandgap

energy. It can be problematic to find a metal that provides a p type an

Ohmic contact with a semiconductor with a large bandgap such as GaN

or SiC.

1.2.1. The tunnel contact

A more convenient contact is a tunnel contact. Such contacts have

a positive gate to the metal semiconductor interface. If the width of the

region of the depletion to the metal–semiconductor interface is very thin,

of the order of 3 nm or less, the carriers can pierce a tunnel comfortably

through such a gate. The required doping density for such contact is a 1019

cm–3 or higher.

1.2.2. Resistance of contact

All sample or semiconductor structures are inevitably joined to

metallic lines of current transportation. It is indispensable that contacts

between the lines of transportation and the semiconductor allow the

current in the two direction to pass and present the weakest resistances

possible.

The resistance of a contact is defined by:

R=RC/S Ω.

Rc = specific resistor of contact (Ω·cm2 ) (resistor of contact) ; S = surface

of contact.

One can decrease this resistance while increasing the contact surface.

Moses

Rectangle


Fig. 1. Ohmic contact on semiconductor N.

Fig. 2. Ohmic contact on semiconductor P.


An ohmic contact on a SC'N' and SC'P' (Figs. 1, 2) is theoretically

possible with a metal working less than the output of a semiconductor.

Unfortunately this ideal situation is rarely achieved. In practice, one

decreases the resistance of a contact superficially by overdoping the

region where one wants to achieve contact: we obtain a degenerate buffer

layer (of 1019 to 1020 cm–3). An ohmic contact on a SC'N' and SC'P' (Figs.

1, 2) is theoretically possible with a metal working less than the output of

a semiconductor. Unfortunately this ideal situation is rarely achieved. In

practice, one decreases the resistance of a contact superficially by

overdoping the region where one wants to achieve contact: we obtain a

degenerate buffer layer (of 1019 to 1020 cm–3).

22.. RESULTS AND DISCUSSION

SSttrruuccttuurree ddiiooddee SScchhoottttkkyy ooff ttyyppee nn

We have used an SIM 3D simulator in our study,which is designed

the study of devices with small geometry. The I–V characteristics of

Schottky diodes in the hypothesis of a parabolic summit of the potential

and of a distribution of Boltzmann of the electrons.

2.1.1. Potential electrostatic and electric field

The electric field generated by a voltage of polarization presents an

intensity and a specific direction in every point in the ZCE. However it is

important to first determine all points of the electrostatic potential,

potential interest is the fact that the value of the electric field strength at a

point in drifts directly affects the variation of the potential (Fig. 3).


Fig. 3. Distribution of the potential to the out thermodynamic balance of Schottky diode of N type in GaAs (Plan XOY, Z=0.65μm))..

TThheerrmmooddyynnaammiicc bbaallaannccee rreeggiimmee

Fig. 4. Distribution of the potential to the thermodynamic balance of a

Schottky diode of N type in GaAs (Plan XOY, Z=0.65μm).


2.1.2. Distribution of the potential lines

Fig. 5. Distribution of the potential lines of the thermodynamic balance of

an N type Schottky diode in GaAs (Plan XOY, Z=0.65μm).

The electric field is characterized in every point of the domain by a

vector E(x, y, z) with a direction and an intensity (Fig. 4). In a three-

dimensional reference mark, it is marked by its three components scalar

Ex(x, y, z), Ey(x, y, z), Ez (x, y, z). The potential lines are generally given

by closed lines (Fig. 5). They include the loads and are perpendicular to

the lines of field.

The ZCE due to the metallurgic contact presents a width that

varies from inversely proportional to the concentration the doping of the

integrated layer. This zone is therefore important since it also presents a

intensity considerable electric field intensity due to the density of state of

the center traps condensed at the surface of the metallurgic diode.


2.1.3. The density of state of the center traps on I–V characteristics

Fig. 6. Variation of the NTD according to the voltage of polarization.




The simulation of the equilibrium state, shows the influence of

deep centers on I–V characteristics.

The density of electrons evolving between [of 6×1016 to 3×1020] is

inversely proportional JTED that evolves between [2.1×10–8 and 8.1×10–8

A] (Fig. 6/Fig. 7/Fig. 8).

The deep centers involved in a recombination mechanism such as

Shockley-Read is characterized by four new parameters that can vary

independently from each other: n1t (and p1t) (cm–3), which are functions of

the energy level in the forbidden gap; τnt and τpt (s–1) are related to the

capture efficient sections for electrons and holes, and also to the density

Nt (cm–3) of the centers.

The application of a forward biais voltage shows a reduction of the

transistion region ZCE, its variation depends to the deep trapping level


Δnt. Because N type semiconductors are always fully ionised, the second

trap center has a negligible influence on the capture of carriers due to its

low density. It can interfere in the recombination process, because of its

large capture coefficients due to its short lifetime (τnt and τpt: 10–8 to 10–10

s). An increase of the density of electrons through the N+ contact leads to

an increase of Δnt and consequently to an increase of the density of free

holes through the P+ contact leads to an increase of Δpt.

To achieve contact: one achieves a degenerated layer buffer (of

8×1016 to 3×1020 cm–3). The ZCE of the gate formed between the layer

buffer and the contact metal is so fine that the carriers can cross it by a

tunnel effect. The contact is no longer a rectifier and the characteristic

I(V) is symmetrical.

2.2. The current thermionic emission-diffusion (TED)

Fig. 9. Variation of the JTED current according to

the voltage of polarization.


Fig. 10. Variation of the JTED current according to

the voltage of polarization.

Fig. 11. Variation of the JTED current according to the

voltage of polarization.


The current through a metal–semiconductor junction is mainly due to majority carriers. Three distinctly different mechanisms exist in a Schottky diode: diffusion of the semiconductor carriers in metal, thermionic emission-diffusion (TED equation (6)) and a mechanical quantum that pierces a tunnel through the gate.

We conclude that the I–V characteristics of Schottky diodes in the hypothesis of a parabolic summit of potential.

Two mechanisms can cause breakdown, namely avalanche multiplication or impact ionization of carriers in the high electric field. Neither of the two breakdown mechanisms is destructive. However heating is caused by the breakdown voltage and diode may be destroyed unless sufficient heat sinking is provided. The breakdown in silicon can be predicted.

The introduction of deep centers in a semiconductor causes a disturbance of the characteristic I (V)

2.2.1. The current tunnel

Fig. 12. Variation of the JTunnel current according to the

voltage of polarization.


The GaAs and its large concentration values of traps centers, will participate in an increase of thermionic electrons and holes, which will in turn act on the I (V) characteristic of the diode, and it is the overflow

maximum value [NT=3×1020]. The characteristic I (V) is shown in Figs.

9–11. One notices that the Jtunnel current varies between 1×10–13 and

5×10–13 .This range of variation is very small in relation to the one of

JTED.

The current of thermionic emission-diffusion JTED is between

[6×10–8 A and 5×10–13 A]. The JTunnel current figure 12 (Eq. (9)) is

inversely proportional to the density of traps NT, whose values are

located between [8×1016 and 3×1020 cm-3 ] and is going to influence the

region of zone of desertion of carriers, act very quickly under the electric field effect the effect of the mechanism of transport of the Schottky diode. The electrons pulled out of the crystalline structure will be filled by other pairs of electrons holes created by the same phenomenon, this process repeats itself several times, until thermal saturation occurs, which leads to the straining of the diode.

Fig. 13. Variation of the JTED current according to the temperature.


3. THE TEMPERATURE

The ln(I)– V plots are generated at various temperatures and

ideality factors, and are estimated by a (of 1 to 2) fitting of simulated at

all temperatures (for 50 to 250 K) in Figure13, this is the range of

temperature that the thermionic emission-diffusion JTED current [of

6×10–8A and 9.5×10–8 A] remained steady at for small values. Beyond

this temperature, the current progresses in a brutal way until the Schottky

diode are destroyed. Electron transport may occur via shallow traps,

which are fewer in number, thus leading to relatively low ideality factor.

The interesting observation that the ideality factor increases above

unity depending on decreasing temperature is almost identical to the

reported variation of ideality factor obtained from the experimental data

on MIS contacts[19, 20]. Thus, an increase in ideality factor with decreasing

temperature is only possible if the interface state density is assumed to

have inverse temperature dependence. Inverse temperature dependence

implies that the interface states are effective at low temperature.

It can be further related to the available energy levels of interface

states. At higher energies there is a lower density of states, while at low

energies more states are available.

Thus, more of the energy levels of the interface states will be at

low energies. At low temperature, electron transport may occur through

deep level, traps states, whereas at high temperature due to the high

energies of electrons, shallow traps may participate in the conduction

process at the interface.

Therefore, at low temperatures electron transport occurs via deep traps states, of which there are more, so the effective density of states is higher and hence the resulting ideality factor arising due to potential drop


across the layer increases. On the other hand, at high temperature electron transport may occur via shallow traps, which are fewer in number, thus leading to relatively low ideality factor. The similar inverse temperature dependence of interface state density derived from the experimental work on MIS Schottky diodes is also reported in Refs[21-24]. The actual temperature dependence of interface state density at the MIS junction governs the rising trend of ideality factor with decreasing temperature. The exact distribution of interface state density will shed more light on under standing the behaviour of MIS Schottky diodes, which requires more investigation and is open for future discussion.

4. CONCLUSIONS

The I–V characteristics of Schottky diodes in the hypothesis of a parabolic summit of the potential.The I–V characteristics of Schottky diodes with an interfacial insulator layer are studied by numerical simulation. The I–V data of the MIS Schottky diode are generated using the TED equation considering an interfacial layer parameter. The calculated I–V data are fitted into an ideal TED equation (6) to see the apparent effect of an interfacial layer on barrier parameters. It is shown that mere presence of an interfacial layer at the MS interface makes the Schottky diode behave as an ideal diode of high apparent BH. The apparent BH is shown to decrease linearly with decreasing the temperature. However, the ideality factor and series resistance remains the same as considered for a pure Schottky contact without an interfacial layer. It is shown that the bias coefficient of the tunneling barrier, however, increases the ideality factor, but makes the ideality factor decrease with decreasing temperature. It is considering that the potential drop across an interfacial layer also gives rise to high ideality factor, which remains constant at all temperatures. The inverse temperature dependence of interface states is suggested to be a possible reason for causing an increase in ideality factor with decreasing device temperature.


REFERENCES

[1] Sze S M. Physics of semiconductor devices. 2nd ed. New York:

Wiley, 1981

[2] Rhoderick E H, Williams R H. Metal–semiconductor contacts. 2nd

ed. Clarendon Press, Oxford University Press in Oxford

[England], New York, 1988

[3] Card H C, Rhoderick E H. Studies of tunnel MOS diodes I.

Interface effects in silicon Schottky diodes. J Phys D, 1971, 4:

1589

[4] Turut A, Yalcin N, Saglam M. Metallic polythiophene/inorganic

semiconductor Schottky diodes. Solid State Electron, 1992, 35:

835

[5] Cowley A M, Sze S M. Surface state and interface effects in

Schottky barriers at n type silicon surfaces. J Appl Phys, 1965, 36:

3212

[6] Tseng H H, Wu C Y. Interpreting the nonideal reverse bias C–V

characteristics and importance of the dependence of Schottky

barrier height on applied voltage. Solid State Electron, 1987, 30:

383

[7] Hudait M K, Kruppanidhi S B. The analysis of lateral distribution

of barrier height in identically prepared Co/n- Si Schottky diodes.

Solid State Electron, 2000, 44: 1089

[8] Borrego, Guttmann R J. High barrier GaAs/metal Schottky

junctions produced by electrochemical metal deposition . JM Solid

State Electron, 1979, 22: 621


[9] Altindal S, Karadeniz S, Tugluoglu N, et al. The role of the

interface insulator layer and interface states on the current-

transport mechanism of Schottky diodes in wide temperature

range. Solid State Electron, 2003, 47: 1847

[10] Singh A, Reinhardt K C, Anderson W A. InP tunnel metal–

insulator–semiconductor devices irradiated with 1 MeV electrons.

J Appl Phys, 1990, 68: 3475

[11] Chattopadhyay P, Daw A N. Energy distribution of interface state

charge density in Cu-nSi Schottky diode with thin interfacial oxide

layer. Solid State Electron, 1986, 29: 555

[12] Hatori K, Torn Y. A new method to fabricate Au/n-type InP

Schottky contacts with an interfacial layer. Solid-State Electron,

1991, 34(5): 527

[13] Dokme I, Altindal S. Temperature-dependent current–voltage

characteristics of the Au/n-InP diodes with inhomogeneous

Schottky barrier height. Semicond Sci Technol, 2006, 21: 1053

[14] Akkal B, Benamara Z, Bideux L, et al. Characterization of

interface states at Au/InSb/InP(100) Schottky barrier diodes as a

function of frequency. Microelectron J, 1999, 30: 673

[15] Cova P, Singh A. Metallic polythiophene/inorganic semiconductor

Schottky diodes. Solid State Electron, 1990,33:11

[16] Saxena A N, Surf. Temperature dependence of the Schottky barrier

height in gallium arsenide.Solid State Communiction, 1972, 11(5):

669

[17] S Chand, J Kumar. Current transport in Pd2Si/n-Si(100) Schottky

barrier diodes at low temperatures. Appl Phys A, 1996, 63: 171


[18] Werner J H. Pressure dependences of silicide/silicon Schottky

barrier heights. Appl Phys A, 1988, 47: 291

[19] Karatas S, Altindal S. Temperature dependence of barrier heights

of Au/n-type GaAs Schottky diodes. Solid-State Electron, 2005,

49(6): 1052

[20] Karatas S, Altindal S. Temperature dependence of characteristic

parameters of the Au/SnO2/n-Si (MIS) Schottky diodes. Mater Sci

Eng B, 2005, 122: 133

[21] Hudait M K, Kruppanidhi S B. The analysis of lateral distribution

of barrier height in identically prepared Co/n- Si Schottky diodes.

Solid State Electron, 2000, 44: 1089

[22] Altindal S, Karadeniz S, Tugluoglu N, et al. Analysis of interface

states and series resistance of MIS Schottky diodes using the

current–voltage (I–V) characteristics. Microelectron Eng, 2008,

85(1): 233

[23] Dokme I, Altindal S. Simulation studies of current transport in

metal–insulator–semiconductor Schottky barrier diodes. Physica

B, 2007, 390: 179

[24] Altindal S, Dokme I, Bulbul M M, et al. The distribution of barrier

heights in MIS type Schottky diodes from current–voltage–

temperature (I–V–T) measurements. Microelectron Eng, 2006, 83:

499

Semiconductor Tech

Documents

resolved lattice

scanning electron

mild condition

gaas semi

degenerate

inset saed

electrons

si schottky