Study of Hafnium dioxide (HfO2) film

Document Sample
Study of Hafnium dioxide (HfO2) film Powered By Docstoc
					Study of Hafnium dioxide (HfO2)
   Atomic Layer Deposition

                   EE 5657
     Principles of Thin Film Technology

             Literature Review

               Prepared by
              Yen Chin Woo


Content Page


Fabrication Method---------------------------------------------------------------9

Experimental Parameters-------------------------------------------------------10

Results and Discussion----------------------------------------------------------11
      Film Growth--------------------------------------------------------------11
      Film Structure------------------------------------------------------------13
      Film Composition-------------------------------------------------------16
      Dielectric Behavior------------------------------------------------------17




Hafnium Iodide, Hafnium chloride and Hafnium Tetrakis(enthymethylamide)
with oxygen were used as precursors for atomic layer deposition of hafnium
dioxide (HfO2) thin film on Silicon substrate. The surface reactions were
found to be complementary and self-limiting, as a result providing highly
uniform thicknesses and conformal coating. Some amount of cubic,
tetragonal or orthorhombic phase was observed in thin films. The annealing
procedure did not influence the surface roughness but improve the film
crystallinity. The growth rate was fairly more stable in iodide precursor

Complementary-Metal-Oxide-Semiconductor (CMOS) has been the most
widely used technology for decades. This success is largely due to the
excellent properties of the Si-SiO2 system. With a greater demand of
higher performance and reduced manufacturing costs, a lot of research
has been put into the scaling of CMOS transistor. As CMOS technology
tends towards ultra large scale integration (ULSI) for high switching
speed and faster logic operation, the gate oxide thickness is continuously
scaled down as shown in Figure 1. However, the physical limits of SiO2
thickness, which is 2.3nm, will be reached in a few years as predicted by
Semiconductor Industry Association (SIA) International Technology
Roadmap for Semiconductors1. For such thin layers, ultra thin gate oxide
has problems such as current leakage, result of electron tunneling and
other reliability issues as shown in Table 12-3. One possible solution is to
use materials with a much higher permittivity called high-κ materials
such as Si3N4, Al2O3, Y2O3, La2O3, Ta2O5, TiO2, HfO2, and ZrO2 as shown
in Figure 2. This would allow the use of thicker insulating layers. Hafnia
(HfO2) is the most promising candidate for substitute of SiO2 because it
has high dielectric constant (κ ≈ 25), high melting point, wide band gap
(≈ 5.8eV) for low leakage, good thermal mechanical and chemical stability,
relatively high refractive index, and high damage threshold7, 11.

    Figure 1. A technology roadmap for the required thickness of SiO2 gate

                                          Table 1. Physical Limitation28
                                     SiO 2
           Band Gap (eV)

                            7                 CaO
                              ZrSiO4        ZrO2
                            6 HfSiO 4 Y2 O3    HfO 2
                                               La 2O 3
                            5           SrO
                                Si 3 N4
                            4                        Ta 2O 5
                            3                                               TiO 2

                                0      10        20            30     40   50       60
                                          Figure 2. High-κ dielectric29

HfO2 has been fabricated using various techniques. Traditionally,
physical vapor deposition has been used to make such films. However,
processing of such films is difficult and poses uniformity and growth
issues. To overcome these problems, researchers are looking into other
methods of fabrication such as sputtering4-6, ion-assisted e-beam
evaporation7-8, sol-gel process9, photo-induced chemical vapor deposition
(UV-photo CVD)10-11, pulsed laser deposition12, metal-organic chemical
vapor deposition (MOCVD) 13-16 and atomic layer deposition (ALD)17-27; 29-
30. Among the deposition methods, ALD is an attractive method for

depositing thin film due to precise thickness control, 100% step coverage,
conformality and uniformity (Table 2). ALD provides several advantages
over conventional CVD and PVD techniques as shown in Table 3. ALD is
self-limiting growth process where thickness of the oxide film is only

dependent on the number of deposition cycles, so it has good thickness
control. There is no need for reactant flux homogeneity because it has a
large area capability, excellent conformality and good reproducibility. No
gas phase reactants take place in ALD chamber because there is
separate dosing of precursors. Advantages of ALD also include good thin
film properties such as high crystalline quality and very smooth surface.
With ALD, it is simple to change the deposition sequence and hence new
materials can be created, superlattice structures fabricated and it also
makes atomic substitution for desired layers possible. Furthermore,
atomic modification at an interface of heterostructure and atomic
displacement by strained superlattice is made possible as well. It is
simple to achieve conformal growth on good quality surface.
Disadvantages of ALD are the relatively low throughput and it is hard to
control precise deposition parameters.

ALD reactions typically carried out in range between 200°C to 400°C. If
the temperature is too high, chemical bonding cannot be maintained. If
the deposition temperature is too low, thermally activated chermisorption
and film forming reactions rates decrease reducing deposition rates.

          Method           ALD MBE CVD Sputter Evaporation PLD
                           good    fair   good    good        fair         fair
    Film Density           good good good         good        poor         good
    Step Coverage          good poor varies poor              poor         poor
    Interface Quality      good good varies poor             good      varies
    Number of
                           fair   good poor       good        fair         poor
    Low Temp.
                           good good varies good             good          good
    Deposition Rate        fair   poor good       good       good          good
                           good    fair   good    good       good          poor
          ALD = atomic layer deposition, MBE = molecular beam epitaxy.
          CVD = chemical vapor deposition, PLD = pulsed laser deposition
           Table 2. Comparison of Thin Film Deposition Methods29

  Characteristic     Inherent Implication for
                                                         Practical Advantages
  feature of ALD          film deposition
Self-limiting        Film thickness is                   Accurate and simple
growth process       dependent only on the                thickness control
                     number of deposition
                     No need for reactant                Large-area capability
                     flux homogeneity                    Large-batch capability
                                                         Excellent conformality
                                                         No problems with
                                                          inconstant vaporization
                                                          rates of solid precursors
                                                         Good reproducibility
                                                         Straightforward scale-
                     Atomic Level control of             Capability to produce
                     material composition                 sharp interfaces and
                                                         Possibility to interface
Separate dosing of   No gas phase reactions              Favors precursors
reactants                                                 highly reactive towards
                                                          each other, thus
                                                          enabling effective
                                                          material utilization
                     Sufficient time is                  High-quality materials
                     provided to complete                 are obtained at low
                     each reaction step                   processing
Processing           Processing conditions of            Capability to prepare
temperature          different materials are              multilayer structures in
windows are often    readily matched                      a continuous process
   Table 3. Characteristic features of ALD, implications on film growth and
                             practical advantages30

Precursor chemistry plays an important role in ALD. There are some
precursors’ requirements:
    Precursors must be volatile and thermally stable to ensure efficient
     transportation so that reactions will not be precursor flux
    The vapor pressure of the precursors must be high enough to
     completely fill the substrate area in the chamber so that monolayer

       deposition takes place in a reasonable length of time. The vapor
       pressure of precursors must be approximately 0.1 torr
      Precursors must chemisorb onto the substrate or react rapidly
       with surface groups and react with each other to keep deposition
       times short
      The Precursors should not experience self-decomposition, would
       lose self-limiting property and effect film contamination
      Precursors should not etch or dissolute into the film or substrate
      Precursors are preferably liquids and gases, but solids can also be

Most researchers use precursors like Hafnium (IV) Chloride, Hafnium (IV)
Iodide, Hafnium Tetrakis(enthylmethylamide), Hafnium (IV) Tert-butoxide,
Hafnium Tetrakis(diethylamido), and Hafnium Tetrakis(dimethylamido)
with water to produce HfO2 film.

This paper describes the fabrication of HfO2 by ALD and compares or
discusses the results of 3 different kind of metal precursors, which are
Hafnium Chloride (HfCl4)18-20,24,27, Hafnium Iodide (HfI4)19,21-22 and
Hafnium Tetrakis(enthylmethylamide) (Hf[N(CH3)(C2H5)]4)26. Water (H2O)
is in all cases as oxygen precursor.

     Fabrication Method
     Atomic Layer Deposition (ALD), also known as Atomic Layer Epitaxy
     (ALE), was originated by T. Suntola in Finland. Using precursors with
     self-limiting mechanism22, films accurate to monolayer can be deposited.
     At any one time, only one type of precursor is in the chamber. Since the
     precursor has been designed to react with specific type of substrate, only
     a single layer is grown. When precursor gases are introduced onto the
     substrate surface, chemi-sorption or surface reactions take place at the
     surface. The ALD reactor is purged with an inert gas between the
     precursor pulses. Due to ALD has slow deposition rate, the thickness of
     the films are reducing to a few nanometers. A typical processes to grow
     substrate (S) – Hf(A)O2(B) has following steps. First, precursor A-(X) will
     be flowed into the chamber, causing a surface reaction and causing the
     formation of S-A-(X). Once the entire surface is coated, the chamber is
     purged of A-(X). After B-(Y) is flowed into chamber, this reacts with A-(X)
     interface to form S-A-B and (X)-(Y) is carried away. This process is
     repeated until the desired thickness is formed. Figure 3 illustrates the
     deposition of HfCl4. Since it is a layer-by-layer deposition, it advances the
     growth of uniform thickness and tremendous conformal films on large



          1. HfCl4 (gas)                2. HfCl4 (Surface Monolayer)                 Cl



3. HfCl4(surface layer) + H2O (gas)         4. HfO2 (Surface Film)
                       Figure 3. Atomic Layer Deposition of HfCl4

Experimental Parameters
To analysis the characteristic of growth films made by 3 different
precursors, the experiment preparation is will be described in the

HfCl4 and HfI4 precursors

Deposition was done on Single Crystal p-Si(100) substrate at 300°C in a
hot wall flow type ALD. Native silicon oxide was first removed by
Hydrofluoric acid bath for 30 seconds. HfI4 or HfCl4 was evaporated from
open boats held at 210°C and 160°C respectively inside the reactor. H2O
was generated using an external reservoir at room temperature. Nitrogen
was used both as the carrier gas and purge gas. Pulse lengths of 0.4s
and 0.5s were used for the metal and oxygen respectively. Purges times
of 0.5s was used19. The pressure of the reactor was approximately

Hafnium Tetrakis(enthymethylamide) Precursors

Deposition was done on borosilicate glass, indium-tin-oxide and Si(100)
substrate in the temperature range of 150°C to 325°C in a hot-wall
horizontal flow-type ALD reactor. Native silicon oxide of the silicon
substrate was first removed by a hydrofluoric acid bath for 30 seconds.
Hf[N(CH3)(C2H5)]4 was evaporated from an open boat at 60°C inside the
reactor. H2O vapor was generated in an external reservoir at room
temperature. Pulse lengths of 0.2 to 1.0s and 0.5s were used for the
metal and oxygen. Nitrogen was used as both the carrier gas and purge
gas. Purge times of 0.5s used26. The pressure of the reactor was
approximately 10mbar.

Results and Discussion

Films made for purpose of replacing SiO2 are usually characterized by
four parameters: films growth, film structure, film composition and
dielectric behavior.

Film Growth
(a) HfI4 + H2O                          (b) HfI4 + H2O

(c) HfI4 and HfCl4 + H2O                (d) Hf[N(CH3)(C2H5)]4 + H2O

Figure 4. (a) Growth rate of HfO2 as a function of growth temperature
determined for oxygen partial pressure at 13 and 40Pa21-22 by HfI4 (b) The
thickness of HfO2 films deposited by HfI4 at 620°C22 (c) HfO2 Film Thickness
versus growth cycles of HfI4 and HfCl419 (d) Growth rate versus Temperature for

From Figure 4(a), it can be seen that growth rate of 0.03nm/cycle, where
O2 partial pressure is 40Pa, on the Si(100) substrate at 400°C is very
slow. With a slight increase in temperature to 500°C, it was obvious that
the growth rate increased significantly. However, from 500°C to 800°C,
the growth rate is almost constant, which is 0.11-0.12nm/cycle.
Saturated thickness grew linearly with number of cycles. The result
shows that metal precursor undergo self-limiting reaction with metal
surface prepared by reaction with oxygen when it reaches the saturation.

Lowering the partial pressure of O2 did not change the shape of the curve,
it is only effected the growth rate extensively, increasing the lowest
possible growth temperature from 400°C to 500°C and it is flattened out
the growth rate to 0.08-0.09nm/cycle at higher temperature of 620°C.

According to Figure 4(b), the experiment showed 1000 deposition cycles
for both cases. The growth rate was found to be proportional to the
number of deposition cycles for each film. In addition, the growth rate is
increased by approximately 30% for higher O2 partial pressure. Figure
4(c) shows the graph data for HfCl4 and HfI4 done by another group.
Compared the data from Figure 4(b) with Figure 4(c), HfI4 in Figure 4(c)
has lower growth rate. Although Kukli19 did not specify the O2 partial
pressure in his article, I believe the O2 partial pressure must be lower
than 13 Pa based on the analysis from Figure 4(b). From the thickness
versus growth cycles graph in Figure 4(c), the slopes of both precursors
are approximately the same. Since both HfCl4 and HfI4 are halides, it is
reasonable result and one would expect this convergence in data. From
the result of Figure 4(c), I would expect that the growth rate versus
substrate temperature would also have the same characteristics as
Figure 4(b). However, there is insufficient data to verify it. In addition,
there is different study of HfCl4 shows that growth rate at lower growth
temperature is higher than higher growth temperature due to the
significance and higher density of active adsorption sites (OH-groups)
regenerated in hydrolysis steps and participating in surface reactions27.

In Figure 4(d), the rate limiting mechanism is also seen in the graph.
Between 200°C to 250°C, the growth rate is almost self-regulating of the
growth temperature. Notice that from 150°C to 200°C and 250°C to
400°C, there is an increase in growth rate. Since ALD is self-limiting
process, I would expect that growth is constant throughout with varying
temperature. However, this increase would mean the growth rate cannot
be self-limiting. There are other factors that are changing this rate. Two
possible explanations for the increase in growth rate at lower
temperature: it can be either an increase in density of functional
adsorption sites or molecular adsorption of the precursors. Also at
elevated temperatures, thermal decomposition of the Hf precursors could
enhance the growth and account for the higher growth rate26.

Film Structure
(a) HfI4                                           (b) HfI4

(c) HfCl4 and HfI4                                 (d) HfCl4 and HfI4

                                                                        df = 9.5nm

                          df = 29nm

                                                                        df = 9.5nm

                          df = 34nm

(e) Hf[N(CH3)(C2H5)]4

                                      df = 170nm

                                      df = 150nm

                                      df = 132nm

                                      df = 132nm

Figure 5. The GI-XRD patterns for HfI4 at different temperatures using O2
partial pressure at (a) 13Pa and (b) 40Pa22. (c) XRD and (d) GI-XRD patterns for
HfCl4 and HfI4 in as-deposited and post-annealed state (it is under atmospheric
pressure in forming gas for 30mins at 400°C19. (e) XRD for Hf[N(CH3)(C2H5)]4 at
various temperature26.

The phase composition of films was determined in Figure 5(a), 5(b), and
5(d) by grazing incidence X-Ray Diffractometer (GI-XRD), whereas the
phase composition of films was determined in Figure 5(c) and Figure 5(e)
by X-Ray Diffractometer (XRD). Figure 5(c) and 5(d) also demonstrated

after the annealing state. As of the Figures, the phase composition was
slightly dependent on the precursor used.

From the XRD patterns in Figure 5(c) and 5(e), the film growth is in
monoclinic HfO2 phase, which is stable polymorph of HfO2 under normal
conditions. In the case of the HfCl4 and Hf[N(CH3)(C2H5)]4, the film growth
shows evidence of mixed phase since there is an addition 111T peak that
occurs at 30.4°. These are reflections from tetragonal or orthorhombic
phases, in addition to monoclinic peaks. This indicates a more
heterogeneous film growth in terms of phase composition19,22,26.

Figure 5(c) also has the XRD patterns for annealed HfI4 and HfCl4. It
seems that annealing both samples did not change the crystallinity
except for introduction of a peak at 33°. This reflection is mostly a 200
reflection of the Si which is normally not present. This peak does not
correspond with any well known d-value of HfO2. However, it matches to
a certain degree with tetragonal hafnium. The origin of this peak after
annealing is still unclear. I believe that layer relaxation during annealing
could have introduced some sort of lattice strain, cause grain growth or
migrated some dislocations.

Analyzing GI-XRD pattern in Figure 5(d), there is no diffraction peaks
due to the films crystallized or recrystallized upon annealing. It is clear
that film growth from HfCl4 was crystallized at as-deposited state.
Compared between HfCl4+H2O process with HfI4+H2O process, HfI4
nucleation density is higher because it has greater disorder and more
amorphous structure due to the decomposition and fairly high surface
mobility of HfIx species adsorbed19. The dominating phase was
monoclinic HfI4 as shown in Figure 5(a) and 5(b). Furthermore, there is a
peak did not belong to the monoclinic phase at 2θ = 30.4°, where
tetragonal and cubic modifications give the first strongest 111 reflection,
whereas the cubic phase has the second strongest reflection 220 at 2θ =
50.4° as circled in red.

In addition, the annealed films show there is no substantial change in
the film structure. Both monoclinic and tetragonal or orthorhombic
peaks are noticeable in the films growth from HfCl4; however, only
monoclinic peaks existed in the film growth from HfI4.

(a) HfCl4 + H2O   300 °C                 (b) HfI4 + H2O   300 °C

(c) Hf[N(CH3)(C2H5)]4 + H2O   200 °C     (d) Hf[N(CH3)(C2H5)]4 + H2O   300 °C

Figure 6. AFM images taken with a sample size of 200 χ 200nm for various
samples (a) HfCl4 + H2O at 300 °C19 (b) HfI4 + H2O at 300 °C19 (c)
Hf[N(CH3)(C2H5)]4 + H2O at 200 °C26 (d) Hf[N(CH3)(C2H5)]4 + H2O at 300 °C26

Films deposited with HfCl4 with H2O at 300°C, HfI4 with H2O at 300 °C,
Hf[N(CH3)(C2H5)]4 (200°C) with H2O at 200°C and Hf[N(CH3)(C2H5)]4 (300 °C)
with H2O at 300°C were evaluated by Atomic Force Microscopy (AFM)19,26.
This was done so that the surface topology can be studied. The
measurements revealed a root-mean-square (rms) of 2.3nm, 1.8nm,
7.9nm and 3.7nm, respectively. Also interestingly for deposition using
Hf[N(CH3)(C2H5)]4, surface roughness decreases with an increased
deposition temperature. Also from XRD image, we can see that
crystallinity also increases with temperature. Lastly from Figure 6, it can
be seen that grain sizes are smaller with increase temperatures. This
results in a decrease in surface roughness. A possible reason for this is
the increase of nucleation density at 300°C. Comparing the three
precursors, HfI4 precursor has the best surface smoothness based on the
images obtained.

Film Composition

    Figure 7. Concentrations of H, C and N Residues versus Growth Rate 26

The composition of films grown from HfCl4, HfI4, Hf[N(CH3)(C2H5)]4 (200°C)
and Hf[N(CH3)(C2H5)]4 (300 °C) was analyzed using time-of-flight elastic
recoil detection analysis (TOF-ERDA). With films grown from HfI4, the
concentration of O, Hf, H and I were 66 atomic percent, 32 atomic
percent, 1.5 atomic percent and 0.4 atomic percent, respectively. In films
growth from HfCl4, the concentration of O, Hf, H and Cl were 66 atomic
percent, 32 atomic percent, 1 atomic percent and 0.4 atomic percent,
respectively. Lastly the films grown from Hf[N(CH3)(C2H5)]4 (175-300°C) , the
concentration of Hf and O were 33 atomic percent and 63 atomic percent,
respectively. Within the accuracy limits of TOF-ERDA, this results shows
that all the samples are stoichiometric dioxides26. Figure 7 shows the
concentration of H, C and N for Hf[N(CH3)(C2H5)]4 growth. Carbon is
important here as carbon sites in the oxide can lead to trap assisted
tunneling. If carbon causes trap assisted tunneling, the original problem
of leakage current is not solved and may worsen.

Dielectric Behavior

(a) HfCl4 + H2O                          (b) HfI4 + H2O

(c) Hf[N(CH3)(C2H5)]4 + H2O

Figure 8. C-V curves (a) HfCl4 + H2O19 (b) HfI4 + H2O19 (c) Hf[N(CH3)(C2H5)]426

The dielectric behavior of the samples was studied using the C-V curves.
From Figure 8, we can see that all the C-V curves demonstrate a
counterclockwise hysteresis when electrode voltage first goes from
positive to negative values and then backwards. The curve shows clear
charge accumulation at sufficiently high negative bias and negative
charge trapping in the oxide layer. The effective permittivities calculated
were 13.1, 14.1, 11.3 and 14.5 for film growth from HfCl4, HfI4,
Hf[N(CH3)(C2H5)]4 (200°C) and Hf[N(CH3)(C2H5)]4 (300 °C), respectively.

Oxide charge density can be calculated from Figure 8 using

             N ot  C acc 

where Cacc is the accumulation oxide capacitance, ΔVfb is the hysteresis
width, q is the electron charge and A is the effective electrode area.
Calculating the densities, the values are 1.7 χ 1012cm-2, 5 χ 1011cm-2, 4 χ
1012cm-2 and 1.1 χ 1012cm-2 for the films growth using HfCl4, HfI4,

Hf[N(CH3)(C2H5)]4 (200°C) and Hf[N(CH3)(C2H5)]4 (300 °C), respectively. Possible
reasons for this charge could include the degree of crystallization and
crystallography phase content. It could also be that the different
chemical composition of defects may have certain effects. Kukli19
suggests that the meta stable tetragon crystallites in the HfCl4, stabilize
but increase trap density. I believe this to be unlikely as the work done
with Hf[N(CH3)(C2H5)]4 does not seem to follow this trend.


It has been demonstrated that HfO2 films can be grown with HfCl4, HfI4
and Hf[N(CH3)(C2H5)]4 as precursors. While growth rate of the films for
HfCl4 and HfI4 were somewhat similar, whereas Hf[N(CH3)(C2H5)]4 had a
much higher growth rate. All of the growth rate curves (Figure 4)
exhibited a self-limiting mechanism; this is where the curves are
saturated or constant. Since Hf[N(CH3)(C2H5)]4 yields a higher growth rate
at a lower temperature, this would be a useful feature where underlying
circuitry and metal interconnects degradation is a critical issue.

From the XRD, HfCl4 and Hf[N(CH3)(C2H5)]4 (300 °C) resulted in monoclinic
and tetragonal or orthorhombic HfO2, while HfI4 and Hf[N(CH3)(C2H5)]4
(200°C) resulted in monoclinic polymorph HfO2. Moreover, it would seem
that Hf[N(CH3)(C2H5)]4 has larger grain size than halides even when
deposited at a lower temperature, possibly due to higher energy or higher
impingement of Hf[N(CH3)(C2H5)]4 on the substrate. The annealed films
show there is no significant change in the film structure, but improved
fairly the film crystallinity.

The results of the C-V analysis do not match each other closely. Factors
that seem to affect the halides do not seem to affect Hf[N(CH3)(C2H5)]4
much and vice versa. There is no one parameter that seem to affect the
C-V characteristics throughout the samples. More data and comparison
should be made before a more stable conclusion can be made about the
effects the precursor has on the C-V curves.

Based on the limited data that I have right now, it would suggest that
films grown from iodide precursors has better electrical stability than the
other films and therefore for the purpose of gate insulation. It would be a
strong argument why the recent researchers use iodide precursors over
the other precursors.

 The 2001Edition of the International Technology Roadmap for Semiconductors, San
Jose, California 10 October 2005. <>

2G. D. Wilk, R. M. Wallace, J.M. Anthony, “High-κ gate dielectrics: Current status and
materials properties considerations,” Journal of Appl. Phy. 89 (2001) 5243-5275

3J Sune, M Nafria, X Oriols, R Rodriguez and X Aymerich, “Failure physics of ultra-thin
SiO2 gate oxides near their scaling limit,” Semicond. Sci. Technol. 15 (2000) 445-454

4C. T. Kuo and R. KworK. M. Jones, “Study of sputtered HfO2 thin films on silicon,”
Thin Solid Films 213 (1992) 257-264

5 H. Grüger , Ch. Kunath , E. Kurth , S. Sorge , W. Pufe and T. Pechstein, “High quality
r.f. sputtered metal oxides (Ta2O5, HfO2) and their properties after annealing,”Thin Solid
Films 447-448 (2004) 509-515

6 S. Edlou, A. Smajkiewicz, and G. Al-Jumaily, “Optical properties and environmental
stability of oxide coatings deposited by reactive sputtering,” Appl. Opt. 32 (1993) 5601-

7M. Alvisi, F. De Tomasi, M. R. Perrone, M. L. Protopapa, A. Rizzo, F. Sarto and S.
Scaglione, “Laser damage dependence on structural and optical properties of ion-
assisted HfO2 thin films,” Thin Solid Films 396 (2001) 44-52

8M. Gilo and N. Croitoru, “Study of HfO2 films prepared by ion-assisted deposition
using a gridless end-hall ion source,” Thin Solid Films 350 (1999) 203-208

9 Toshikazu Nishide, Shinji Honda, Masaharu Matsuura and Mieko Ide, “Surface,
structural and optical properties of sol-gel derived HfO2 films,” Thin Solid Films
371(2000) 61-65

10Q. Fang, J. -Y. Zhang, Z. M. Wang, J. X. Wu, B. J. O'Sullivan, P. K. Hurley, T. L.
Leedham, H. Davies, M. A. Audier, C. Jimenez et al. “Characterisation of HfO2 deposited
by photo-induced chemical vapour deposition,” Thin Solid Films 427(2003) 391-396

11Q. Fang , J. -Y. Zhang , Z. Wang , M. Modreanu , B. J. O'Sullivan , P. K. Hurley , T. L.
Leedham , D. Hywel , M. A. Audier , C. Jimenez et al. “Interface of ultrathin HfO2 films
deposited by UV-photo-CVD,” Thin Solid Films 453-454(2004) 203-207

12C. Essary , J. M. Howard , V. Craciun , D. Craciun and R. K. Singh, “Kinetics of
interfacial layer formation during deposition of HfO2 on silicon,” Thin Solid Films
450(2004) 111-113

13Yoshio Ohshita, Atsushi Ogura, Asako Hoshino, Shigeki Hiiro, Toshie Suzuki and
Hideaki Machida, “Using tetrakis-diethylamido-hafnium for HfO2 thin-film growth in
low-pressure chemical vapor deposition,” Thin Solid Films 406(2002) 215-218

14Moon-Kyun Song , Sang-Woo Kang and Shi-Woo Rhee, “Direct liquid injection metal-
organic chemical vapor deposition of HfO2 thin films using Hf(dimethylaminoethoxide) 4,”
Thin Solid Films 450(2004) 272-275

15Yoshio Ohshita, Atsushi Ogura, Masato Ishikawa, Asako Hoshino, Shigeki Hiiro,
Toshie Suzuki and Hideaki Machida, “Hf1−xSixO2 deposition by metal organic chemical
vapor deposition using the Hf(NEt2)4/SiH(NEt2)3/O2 gas system,” Thin Solid Films
416(2002) 208-211

16Atsushi Ogura, Kimihiko Ito, Yoshio Ohshita, Masato Ishikawa and Hideaki Machida,
“Evaluation of HfO2 film structures deposited by metal-organic chemical vapor
deposition using Hf(N(C2H5)2)4/O2 gas system,” Thin Solid Films 441(2003) 161-164

17O. Renault, D. Samour, D. Rouchon, Ph. Holliger, A. -M. Papon, D. Blin and S.
Marthon, “Interface properties of ultra-thin HfO2 films grown by atomic layer deposition
on SiO2/Si,” Thin Solid Films 428 (2003) 190-194

18Jaan Aarik, Hugo Mändar, Marco Kirm and Lembit Pung, “Optical characterization of
HfO2 thin films grown by atomic layer deposition,” Thin Solid Films 466 (2004) 41-47

19Kaupo Kukli, Mikko Ritala, Timo Sajavaara, Juhani Keinonen and Markku Leskelä
“Comparison of hafnium oxide films grown by atomic layer deposition from iodide and
chloride precursors,”
Thin Solid Films 416(2002) 72-79

20J. Aarik, A. Aidla, A. -A. Kiisler, T. Uustare and V. Sammelselg, “Influence of
substrate temperature on atomic layer growth and properties of HfO 2 thin films,” Thin
Solid Films 340(1999) 110-116

21Jaan Aarik, Jonas Sundqvist, Aleks Aidla, Jun Lu, Timo Sajavaara, Kaupo Kukli and
Anders Hårsta, “Hafnium tetraiodide and oxygen as precursors for atomic layer
deposition of hafnium oxide thin films,” Thin Solid Films 418(2002) 69-72

22J. Sundqvist, A. Hårsta, J. Aarik, K. Kukli and A. Aidla, “Atomic layer deposition of
polycrystalline HfO2 films by the HfI4–O2 precursor combination,” Thin Solid Films
427(2003) 147-151

23H. Y. Yu, M. F. Li and D. L. Kwong, “ALD (HfO2)x(Al2O3)1−x high-k gate dielectrics for
advanced MOS devices application,” Thin Solid Films 462-463(2004) 110-113

24C. Wiemer , S. Ferrari , M. Fanciulli , G. Pavia and L. Lutterotti, “Combining grazing
incidence X-ray diffraction and X-ray reflectivity for the evaluation of the structural
evolution of HfO2 thin films with annealing,” Thin Solid Films 450(2004) 134-137

25H. Wong, N. Zhan, K. L. Ng, M. C. Poon and C. W. Kok, “Interface and oxide traps in
high-κ hafnium oxide films,” Thin Solid Films 462-463 (2004) 96-100

26Kaupo Kukli, Mikko Ritala, Timo Sajavaara, Juhani Keinonen and Markku Leskela,
“Atomic layer Deposition of Hafnium Dioxide Films from Hafnium
Tetrakis(enthymethylamide) and Water,” Chem. Vap. Deposition 8 (2002) 199-204

27Kaupo Kukli, Jaan Aarik, teet Uustare, Jun Lu, mikko Ritala, Aleks Aidla, Lembit
Pung, Anders Harsta, Markku Leskela, Arvo Kikas, and Vaino Sammelselg,
“Engineering structure and properties of hafnium oxide films by atomic layer deposition
temperature,” Thin Solid Films 479 (2005) 1-11

28Scott Thompson, Portland Technology Development, Intel Corp. Paul Packan,
Technology Computer Aided Design, Intel Corp. Mark Bohr, Portland Technology

Development, Intel Corp. Intel Technology Journal. MOS Scaling: Transistor Challenges
for the 21st Century. 15 October 2005.

29   Cambridge NanoTech Inc. 3 December 2005. <>

30Mikko Ritala and Markku Leskelä, “Atomic layer epitaxy - a valuable tool for
nanotechnology?” Nanotechnology 10 (1999) 19-24


Shared By: